diff options
652 files changed, 11764 insertions, 5883 deletions
diff --git a/.github/workflows/bazel-checks.yml b/.github/workflows/bazel-checks.yml index 65d5164..7c3db4e 100644 --- a/.github/workflows/bazel-checks.yml +++ b/.github/workflows/bazel-checks.yml @@ -30,3 +30,26 @@ jobs: - name: Run Buildifier run: | buildifier --mode=check $(find ./utils/bazel -name *BUILD*) + + bazel-build: + name: "Bazel Build/Test" + runs-on: llvm-premerge-linux-runners + if: github.repository == 'llvm/llvm-project' + steps: + - name: Fetch LLVM sources + uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8 # v5.0.0 + # TODO(boomanaiden154): We should use a purpose built container for this. Move + # over when we have fixed the issues with using custom containers with Github + # ARC in GKE. + - name: Setup System Dependencies + run: | + sudo apt-get update + sudo apt-get install -y libmpfr-dev libpfm4-dev + sudo curl -L https://github.com/bazelbuild/bazelisk/releases/download/v1.27.0/bazelisk-amd64.deb > /tmp/bazelisk.deb + sudo apt-get install -y /tmp/bazelisk.deb + rm /tmp/bazelisk.deb + - name: Build/Test + working-directory: utils/bazel + run: | + bazelisk test --config=ci --sandbox_base="" \ + @llvm-project//llvm/unittests:adt_tests diff --git a/bolt/lib/Core/BinaryBasicBlock.cpp b/bolt/lib/Core/BinaryBasicBlock.cpp index d680850..a6d0ca9 100644 --- a/bolt/lib/Core/BinaryBasicBlock.cpp +++ b/bolt/lib/Core/BinaryBasicBlock.cpp @@ -22,8 +22,6 @@ namespace llvm { namespace bolt { -constexpr uint32_t BinaryBasicBlock::INVALID_OFFSET; - bool operator<(const BinaryBasicBlock &LHS, const BinaryBasicBlock &RHS) { return LHS.Index < RHS.Index; } diff --git a/bolt/lib/Core/DynoStats.cpp b/bolt/lib/Core/DynoStats.cpp index 1d98187..64a6d12 100644 --- a/bolt/lib/Core/DynoStats.cpp +++ b/bolt/lib/Core/DynoStats.cpp @@ -51,8 +51,6 @@ PrintDynoOpcodeStat("print-dyno-opcode-stats", namespace llvm { namespace bolt { -constexpr const char *DynoStats::Desc[]; - bool DynoStats::operator<(const DynoStats &Other) const { return std::lexicographical_compare( &Stats[FIRST_DYNO_STAT], &Stats[LAST_DYNO_STAT], diff --git a/bolt/lib/Profile/DataAggregator.cpp b/bolt/lib/Profile/DataAggregator.cpp index 8554683..cafe4bf 100644 --- a/bolt/lib/Profile/DataAggregator.cpp +++ b/bolt/lib/Profile/DataAggregator.cpp @@ -159,8 +159,6 @@ std::vector<SectionNameAndRange> getTextSections(const BinaryContext *BC) { } } -constexpr uint64_t DataAggregator::KernelBaseAddr; - DataAggregator::~DataAggregator() { deleteTempFiles(); } namespace { diff --git a/bolt/lib/Rewrite/RewriteInstance.cpp b/bolt/lib/Rewrite/RewriteInstance.cpp index 77e5688..ab3431e 100644 --- a/bolt/lib/Rewrite/RewriteInstance.cpp +++ b/bolt/lib/Rewrite/RewriteInstance.cpp @@ -295,7 +295,6 @@ cl::bits<GadgetScannerKind> GadgetScannersToRun( } // namespace opts // FIXME: implement a better way to mark sections for replacement. -constexpr const char *RewriteInstance::SectionsToOverwrite[]; std::vector<std::string> RewriteInstance::DebugSectionsToOverwrite = { ".debug_abbrev", ".debug_aranges", ".debug_line", ".debug_line_str", ".debug_loc", ".debug_loclists", ".debug_ranges", ".debug_rnglists", diff --git a/clang-tools-extra/clang-doc/BitcodeWriter.cpp b/clang-tools-extra/clang-doc/BitcodeWriter.cpp index e23511b..3a7ac6e 100644 --- a/clang-tools-extra/clang-doc/BitcodeWriter.cpp +++ b/clang-tools-extra/clang-doc/BitcodeWriter.cpp @@ -303,8 +303,6 @@ static const std::vector<std::pair<BlockId, std::vector<RecordId>>> // AbbreviationMap -constexpr unsigned char BitCodeConstants::Signature[]; - void ClangDocBitcodeWriter::AbbreviationMap::add(RecordId RID, unsigned AbbrevID) { assert(RecordIdNameMap[RID] && "Unknown RecordId."); diff --git a/clang-tools-extra/clang-doc/HTMLMustacheGenerator.cpp b/clang-tools-extra/clang-doc/HTMLMustacheGenerator.cpp index b4b9322..3650f66 100644 --- a/clang-tools-extra/clang-doc/HTMLMustacheGenerator.cpp +++ b/clang-tools-extra/clang-doc/HTMLMustacheGenerator.cpp @@ -29,7 +29,8 @@ namespace clang { namespace doc { static Error generateDocForJSON(json::Value &JSON, StringRef Filename, StringRef Path, raw_fd_ostream &OS, - const ClangDocContext &CDCtx); + const ClangDocContext &CDCtx, + StringRef HTMLRootPath); static Error createFileOpenError(StringRef FileName, std::error_code EC) { return createFileError("cannot open file " + FileName, EC); @@ -159,16 +160,24 @@ Error MustacheHTMLGenerator::generateDocs( { llvm::TimeTraceScope TS("Iterate JSON files"); std::error_code EC; - sys::fs::directory_iterator JSONIter(JSONPath, EC); + sys::fs::recursive_directory_iterator JSONIter(JSONPath, EC); std::vector<json::Value> JSONFiles; JSONFiles.reserve(Infos.size()); if (EC) return createStringError("Failed to create directory iterator."); - SmallString<128> HTMLDirPath(RootDir.str() + "/html/"); + SmallString<128> HTMLDirPath(RootDir.str() + "/html"); if (auto EC = sys::fs::create_directories(HTMLDirPath)) return createFileError(HTMLDirPath, EC); - while (JSONIter != sys::fs::directory_iterator()) { + while (JSONIter != sys::fs::recursive_directory_iterator()) { + // create the same directory structure in the HTML dir + if (JSONIter->type() == sys::fs::file_type::directory_file) { + SmallString<128> HTMLClonedPath(JSONIter->path()); + sys::path::replace_path_prefix(HTMLClonedPath, JSONPath, HTMLDirPath); + if (auto EC = sys::fs::create_directories(HTMLClonedPath)) + return createFileError(HTMLClonedPath, EC); + } + if (EC) return createFileError("Failed to iterate: " + JSONIter->path(), EC); @@ -190,15 +199,16 @@ Error MustacheHTMLGenerator::generateDocs( return Parsed.takeError(); std::error_code FileErr; - SmallString<128> HTMLFilePath(HTMLDirPath); - sys::path::append(HTMLFilePath, sys::path::filename(Path)); + SmallString<128> HTMLFilePath(JSONIter->path()); + sys::path::replace_path_prefix(HTMLFilePath, JSONPath, HTMLDirPath); sys::path::replace_extension(HTMLFilePath, "html"); raw_fd_ostream InfoOS(HTMLFilePath, FileErr, sys::fs::OF_None); if (FileErr) return createFileOpenError(Path, FileErr); - if (Error Err = generateDocForJSON(*Parsed, sys::path::stem(HTMLFilePath), - HTMLFilePath, InfoOS, CDCtx)) + if (Error Err = + generateDocForJSON(*Parsed, sys::path::stem(HTMLFilePath), + HTMLFilePath, InfoOS, CDCtx, HTMLDirPath)) return Err; JSONIter.increment(EC); } @@ -207,16 +217,16 @@ Error MustacheHTMLGenerator::generateDocs( return Error::success(); } -static Error setupTemplateValue(const ClangDocContext &CDCtx, json::Value &V) { +static Error setupTemplateValue(const ClangDocContext &CDCtx, json::Value &V, + SmallString<128> RelativeHTMLPath) { V.getAsObject()->insert({"ProjectName", CDCtx.ProjectName}); json::Value StylesheetArr = Array(); - SmallString<128> RelativePath("./"); - sys::path::native(RelativePath, sys::path::Style::posix); + sys::path::native(RelativeHTMLPath, sys::path::Style::posix); auto *SSA = StylesheetArr.getAsArray(); SSA->reserve(CDCtx.UserStylesheets.size()); for (const auto &FilePath : CDCtx.UserStylesheets) { - SmallString<128> StylesheetPath = RelativePath; + SmallString<128> StylesheetPath = RelativeHTMLPath; sys::path::append(StylesheetPath, sys::path::Style::posix, sys::path::filename(FilePath)); SSA->emplace_back(StylesheetPath); @@ -227,7 +237,7 @@ static Error setupTemplateValue(const ClangDocContext &CDCtx, json::Value &V) { auto *SCA = ScriptArr.getAsArray(); SCA->reserve(CDCtx.JsScripts.size()); for (auto Script : CDCtx.JsScripts) { - SmallString<128> JsPath = RelativePath; + SmallString<128> JsPath = RelativeHTMLPath; sys::path::append(JsPath, sys::path::Style::posix, sys::path::filename(Script)); SCA->emplace_back(JsPath); @@ -238,7 +248,8 @@ static Error setupTemplateValue(const ClangDocContext &CDCtx, json::Value &V) { static Error generateDocForJSON(json::Value &JSON, StringRef Filename, StringRef Path, raw_fd_ostream &OS, - const ClangDocContext &CDCtx) { + const ClangDocContext &CDCtx, + StringRef HTMLRootPath) { auto StrValue = (*JSON.getAsObject())["InfoType"]; if (StrValue.kind() != json::Value::Kind::String) return createStringError("JSON file '%s' does not contain key: 'InfoType'.", @@ -249,13 +260,17 @@ static Error generateDocForJSON(json::Value &JSON, StringRef Filename, "JSON file '%s' does not contain 'InfoType' field as a string.", Filename.str().c_str()); + SmallString<128> PathVec(Path); + // Remove filename, or else the relative path will have an extra "../" + sys::path::remove_filename(PathVec); + auto RelativeHTMLPath = computeRelativePath(HTMLRootPath, PathVec); if (ObjTypeStr.value() == "namespace") { - if (auto Err = setupTemplateValue(CDCtx, JSON)) + if (auto Err = setupTemplateValue(CDCtx, JSON, RelativeHTMLPath)) return Err; assert(NamespaceTemplate && "NamespaceTemplate is nullptr."); NamespaceTemplate->render(JSON, OS); } else if (ObjTypeStr.value() == "record") { - if (auto Err = setupTemplateValue(CDCtx, JSON)) + if (auto Err = setupTemplateValue(CDCtx, JSON, RelativeHTMLPath)) return Err; assert(RecordTemplate && "RecordTemplate is nullptr."); RecordTemplate->render(JSON, OS); diff --git a/clang-tools-extra/clang-doc/JSONGenerator.cpp b/clang-tools-extra/clang-doc/JSONGenerator.cpp index 93ec90e..9a770d7 100644 --- a/clang-tools-extra/clang-doc/JSONGenerator.cpp +++ b/clang-tools-extra/clang-doc/JSONGenerator.cpp @@ -581,22 +581,14 @@ static SmallString<16> determineFileName(Info *I, SmallString<128> &Path) { if (I->IT == InfoType::IT_record) { auto *RecordSymbolInfo = static_cast<SymbolInfo *>(I); FileName = RecordSymbolInfo->MangledName; - } else if (I->USR == GlobalNamespaceID) + } else if (I->IT == InfoType::IT_namespace) { FileName = "index"; - else if (I->IT == InfoType::IT_namespace) { - for (const auto &NS : I->Namespace) { - FileName += NS.Name; - FileName += "_"; - } - FileName += I->Name; } else FileName = I->Name; sys::path::append(Path, FileName + ".json"); return FileName; } -// FIXME: Revert back to creating nested directories for namespaces instead of -// putting everything in a flat directory structure. Error JSONGenerator::generateDocs( StringRef RootDir, llvm::StringMap<std::unique_ptr<doc::Info>> Infos, const ClangDocContext &CDCtx) { @@ -609,6 +601,7 @@ Error JSONGenerator::generateDocs( auto RootDirStr = RootDir.str() + "/json"; StringRef JSONDir = StringRef(RootDirStr); sys::path::native(JSONDir, Path); + sys::path::append(Path, Info->getRelativeFilePath("")); if (!CreatedDirs.contains(Path)) { if (std::error_code Err = sys::fs::create_directories(Path); Err != std::error_code()) diff --git a/clang-tools-extra/clang-doc/assets/class-template.mustache b/clang-tools-extra/clang-doc/assets/class-template.mustache index b1a7470..a320a93 100644 --- a/clang-tools-extra/clang-doc/assets/class-template.mustache +++ b/clang-tools-extra/clang-doc/assets/class-template.mustache @@ -141,9 +141,7 @@ <div> {{#PublicMembers}} <div id="{{Name}}" class="delimiter-container"> - <pre> - <code class="language-cpp code-clang-doc" >{{Type}} {{Name}}</code> - </pre> + <pre><code class="language-cpp code-clang-doc" >{{Type}} {{Name}}</code></pre> {{#MemberComments}} <div> {{>Comments}} @@ -160,9 +158,7 @@ <div> {{#Obj}} <div id="{{Name}}" class="delimiter-container"> - <pre> -<code class="language-cpp code-clang-doc" >{{Type}} {{Name}}</code> - </pre> + <pre><code class="language-cpp code-clang-doc" >{{Type}} {{Name}}</code></pre> {{#MemberComments}} <div> {{>Comments}} diff --git a/clang-tools-extra/clang-doc/assets/namespace-template.mustache b/clang-tools-extra/clang-doc/assets/namespace-template.mustache index d96bc5c..f4a35cf 100644 --- a/clang-tools-extra/clang-doc/assets/namespace-template.mustache +++ b/clang-tools-extra/clang-doc/assets/namespace-template.mustache @@ -92,9 +92,7 @@ {{#Records}} <li id="{{USR}}" style="max-height: 40px;"> <a href="{{DocumentationFileName}}.html"> - <pre> - <code class="language-cpp code-clang-doc">class {{Name}}</code> - </pre> + <pre><code class="language-cpp code-clang-doc">class {{Name}}</code></pre> </a> </li> {{/Records}} diff --git a/clang-tools-extra/clang-tidy/ClangTidy.cpp b/clang-tools-extra/clang-tidy/ClangTidy.cpp index 7e18f38..76243e9 100644 --- a/clang-tools-extra/clang-tidy/ClangTidy.cpp +++ b/clang-tools-extra/clang-tidy/ClangTidy.cpp @@ -117,7 +117,8 @@ public: void reportDiagnostic(const ClangTidyError &Error) { const tooling::DiagnosticMessage &Message = Error.Message; - SourceLocation Loc = getLocation(Message.FilePath, Message.FileOffset); + const SourceLocation Loc = + getLocation(Message.FilePath, Message.FileOffset); // Contains a pair for each attempted fix: location and whether the fix was // applied successfully. SmallVector<std::pair<SourceLocation, bool>, 4> FixLocations; @@ -157,11 +158,11 @@ public: // FIXME: Implement better conflict handling. llvm::errs() << "Trying to resolve conflict: " << llvm::toString(std::move(Err)) << "\n"; - unsigned NewOffset = + const unsigned NewOffset = Replacements.getShiftedCodePosition(R.getOffset()); - unsigned NewLength = Replacements.getShiftedCodePosition( - R.getOffset() + R.getLength()) - - NewOffset; + const unsigned NewLength = Replacements.getShiftedCodePosition( + R.getOffset() + R.getLength()) - + NewOffset; if (NewLength == R.getLength()) { R = Replacement(R.getFilePath(), NewOffset, NewLength, R.getReplacementText()); @@ -200,7 +201,7 @@ public: for (const auto &FileAndReplacements : FileReplacements) { Rewriter Rewrite(SourceMgr, LangOpts); - StringRef File = FileAndReplacements.first(); + const StringRef File = FileAndReplacements.first(); VFS.setCurrentWorkingDirectory(FileAndReplacements.second.BuildDir); llvm::ErrorOr<std::unique_ptr<MemoryBuffer>> Buffer = SourceMgr.getFileManager().getBufferForFile(File); @@ -210,7 +211,7 @@ public: // FIXME: Maybe don't apply fixes for other files as well. continue; } - StringRef Code = Buffer.get()->getBuffer(); + const StringRef Code = Buffer.get()->getBuffer(); auto Style = format::getStyle( *Context.getOptionsForFile(File).FormatStyle, File, "none"); if (!Style) { @@ -262,7 +263,7 @@ private: if (!File) return {}; - FileID ID = SourceMgr.getOrCreateFileID(*File, SrcMgr::C_User); + const FileID ID = SourceMgr.getOrCreateFileID(*File, SrcMgr::C_User); return SourceMgr.getLocForStartOfFile(ID).getLocWithOffset(Offset); } @@ -284,7 +285,8 @@ private: } void reportNote(const tooling::DiagnosticMessage &Message) { - SourceLocation Loc = getLocation(Message.FilePath, Message.FileOffset); + const SourceLocation Loc = + getLocation(Message.FilePath, Message.FileOffset); auto Diag = Diags.Report(Loc, Diags.getCustomDiagID(DiagnosticsEngine::Note, "%0")) << Message.Message; @@ -296,8 +298,9 @@ private: CharSourceRange getRange(const FileByteRange &Range) { SmallString<128> AbsoluteFilePath{Range.FilePath}; Files.makeAbsolutePath(AbsoluteFilePath); - SourceLocation BeginLoc = getLocation(AbsoluteFilePath, Range.FileOffset); - SourceLocation EndLoc = BeginLoc.getLocWithOffset(Range.Length); + const SourceLocation BeginLoc = + getLocation(AbsoluteFilePath, Range.FileOffset); + const SourceLocation EndLoc = BeginLoc.getLocWithOffset(Range.Length); // Retrieve the source range for applicable highlights and fixes. Macro // definition on the command line have locations in a virtual buffer and // don't have valid file paths and are therefore not applicable. @@ -353,7 +356,8 @@ ClangTidyASTConsumerFactory::ClangTidyASTConsumerFactory( if (Context.canExperimentalCustomChecks() && custom::RegisterCustomChecks) custom::RegisterCustomChecks(Context.getOptions(), *CheckFactories); #endif - for (ClangTidyModuleRegistry::entry E : ClangTidyModuleRegistry::entries()) { + for (ClangTidyModuleRegistry::entry const E : + ClangTidyModuleRegistry::entries()) { std::unique_ptr<ClangTidyModule> Module = E.instantiate(); Module->addCheckFactories(*CheckFactories); } @@ -394,8 +398,9 @@ static CheckersList getAnalyzerCheckersAndPackages(ClangTidyContext &Context, // Always add all core checkers if any other static analyzer check is enabled. // This is currently necessary, as other path sensitive checks rely on the // core checkers. - for (StringRef CheckName : RegisteredCheckers) { - std::string ClangTidyCheckName((AnalyzerCheckNamePrefix + CheckName).str()); + for (const StringRef CheckName : RegisteredCheckers) { + const std::string ClangTidyCheckName( + (AnalyzerCheckNamePrefix + CheckName).str()); if (CheckName.starts_with("core") || Context.isCheckEnabled(ClangTidyCheckName)) { @@ -504,7 +509,7 @@ std::vector<std::string> ClangTidyASTConsumerFactory::getCheckNames() { ClangTidyOptions::OptionMap ClangTidyASTConsumerFactory::getCheckOptions() { ClangTidyOptions::OptionMap Options; - std::vector<std::unique_ptr<ClangTidyCheck>> Checks = + const std::vector<std::unique_ptr<ClangTidyCheck>> Checks = CheckFactories->createChecks(&Context); for (const auto &Check : Checks) Check->storeOptions(Options); @@ -564,7 +569,7 @@ runClangTidy(clang::tidy::ClangTidyContext &Context, std::make_shared<PCHContainerOperations>(), BaseFS); // Add extra arguments passed by the clang-tidy command-line. - ArgumentsAdjuster PerFileExtraArgumentsInserter = + const ArgumentsAdjuster PerFileExtraArgumentsInserter = [&Context](const CommandLineArguments &Args, StringRef Filename) { ClangTidyOptions Opts = Context.getOptionsForFile(Filename); CommandLineArguments AdjustedArgs = Args; @@ -703,7 +708,7 @@ ChecksAndOptions getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers, #if CLANG_TIDY_ENABLE_STATIC_ANALYZER SmallString<64> Buffer(AnalyzerCheckNamePrefix); - size_t DefSize = Buffer.size(); + const size_t DefSize = Buffer.size(); for (const auto &AnalyzerCheck : AnalyzerOptions::getRegisteredCheckers( AllowEnablingAnalyzerAlphaCheckers)) { Buffer.truncate(DefSize); diff --git a/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp b/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp index 6e0c252..b747657 100644 --- a/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp +++ b/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp @@ -161,7 +161,7 @@ ClangTidyCheck::OptionsView::getEnumInt(StringRef LocalName, if (Iter == CheckOptions.end()) return std::nullopt; - StringRef Value = Iter->getValue().Value; + const StringRef Value = Iter->getValue().Value; StringRef Closest; unsigned EditDistance = 3; for (const auto &NameAndEnum : Mapping) { @@ -173,7 +173,7 @@ ClangTidyCheck::OptionsView::getEnumInt(StringRef LocalName, EditDistance = 0; continue; } - unsigned Distance = + const unsigned Distance = Value.edit_distance(NameAndEnum.second, true, EditDistance); if (Distance < EditDistance) { EditDistance = Distance; diff --git a/clang-tools-extra/clang-tidy/ClangTidyCheck.h b/clang-tools-extra/clang-tidy/ClangTidyCheck.h index e53ae53..905e419 100644 --- a/clang-tools-extra/clang-tidy/ClangTidyCheck.h +++ b/clang-tools-extra/clang-tidy/ClangTidyCheck.h @@ -458,7 +458,7 @@ public: template <typename T> std::enable_if_t<std::is_enum_v<T>, std::vector<NameAndValue>> typeEraseMapping() const { - ArrayRef<std::pair<T, StringRef>> Mapping = + const ArrayRef<std::pair<T, StringRef>> Mapping = OptionEnumMapping<T>::getEnumMapping(); std::vector<NameAndValue> Result; Result.reserve(Mapping.size()); diff --git a/clang-tools-extra/clang-tidy/ClangTidyDiagnosticConsumer.cpp b/clang-tools-extra/clang-tidy/ClangTidyDiagnosticConsumer.cpp index 65fd09f..81a9f93 100644 --- a/clang-tools-extra/clang-tidy/ClangTidyDiagnosticConsumer.cpp +++ b/clang-tools-extra/clang-tidy/ClangTidyDiagnosticConsumer.cpp @@ -61,7 +61,7 @@ protected: // FIXME: Remove this once there's a better way to pass check names than // appending the check name to the message in ClangTidyContext::diag and // using getCustomDiagID. - std::string CheckNameInMessage = " [" + Error.DiagnosticName + "]"; + const std::string CheckNameInMessage = " [" + Error.DiagnosticName + "]"; Message.consume_back(CheckNameInMessage); auto TidyMessage = @@ -77,7 +77,7 @@ protected: if (SourceRange.isCharRange()) return SourceRange; assert(SourceRange.isTokenRange()); - SourceLocation End = Lexer::getLocForEndOfToken( + const SourceLocation End = Lexer::getLocForEndOfToken( SourceRange.getEnd(), 0, Loc.getManager(), LangOpts); return CharSourceRange::getCharRange(SourceRange.getBegin(), End); }; @@ -114,14 +114,14 @@ protected: Level == DiagnosticsEngine::Note ? &Error.Notes.back() : &Error.Message; for (const auto &FixIt : Hints) { - CharSourceRange Range = FixIt.RemoveRange; + const CharSourceRange Range = FixIt.RemoveRange; assert(Range.getBegin().isValid() && Range.getEnd().isValid() && "Invalid range in the fix-it hint."); assert(Range.getBegin().isFileID() && Range.getEnd().isFileID() && "Only file locations supported in fix-it hints."); - tooling::Replacement Replacement(Loc.getManager(), Range, - FixIt.CodeToInsert); + const tooling::Replacement Replacement(Loc.getManager(), Range, + FixIt.CodeToInsert); llvm::Error Err = DiagWithFix->Fix[Replacement.getFilePath()].add(Replacement); // FIXME: better error handling (at least, don't let other replacements be @@ -177,7 +177,7 @@ DiagnosticBuilder ClangTidyContext::diag( StringRef CheckName, SourceLocation Loc, StringRef Description, DiagnosticIDs::Level Level /* = DiagnosticIDs::Warning*/) { assert(Loc.isValid()); - unsigned ID = DiagEngine->getDiagnosticIDs()->getCustomDiagID( + const unsigned ID = DiagEngine->getDiagnosticIDs()->getCustomDiagID( Level, (Description + " [" + CheckName + "]").str()); CheckNamesByDiagnosticID.try_emplace(ID, CheckName); return DiagEngine->Report(Loc, ID); @@ -186,7 +186,7 @@ DiagnosticBuilder ClangTidyContext::diag( DiagnosticBuilder ClangTidyContext::diag( StringRef CheckName, StringRef Description, DiagnosticIDs::Level Level /* = DiagnosticIDs::Warning*/) { - unsigned ID = DiagEngine->getDiagnosticIDs()->getCustomDiagID( + const unsigned ID = DiagEngine->getDiagnosticIDs()->getCustomDiagID( Level, (Description + " [" + CheckName + "]").str()); CheckNamesByDiagnosticID.try_emplace(ID, CheckName); return DiagEngine->Report(ID); @@ -195,10 +195,11 @@ DiagnosticBuilder ClangTidyContext::diag( DiagnosticBuilder ClangTidyContext::diag(const tooling::Diagnostic &Error) { SourceManager &SM = DiagEngine->getSourceManager(); FileManager &FM = SM.getFileManager(); - FileEntryRef File = llvm::cantFail(FM.getFileRef(Error.Message.FilePath)); - FileID ID = SM.getOrCreateFileID(File, SrcMgr::C_User); - SourceLocation FileStartLoc = SM.getLocForStartOfFile(ID); - SourceLocation Loc = FileStartLoc.getLocWithOffset( + const FileEntryRef File = + llvm::cantFail(FM.getFileRef(Error.Message.FilePath)); + const FileID ID = SM.getOrCreateFileID(File, SrcMgr::C_User); + const SourceLocation FileStartLoc = SM.getLocForStartOfFile(ID); + const SourceLocation Loc = FileStartLoc.getLocWithOffset( static_cast<SourceLocation::IntTy>(Error.Message.FileOffset)); return diag(Error.DiagnosticName, Loc, Error.Message.Message, static_cast<DiagnosticIDs::Level>(Error.DiagLevel)); @@ -214,7 +215,7 @@ bool ClangTidyContext::shouldSuppressDiagnostic( DiagnosticsEngine::Level DiagLevel, const Diagnostic &Info, SmallVectorImpl<tooling::Diagnostic> &NoLintErrors, bool AllowIO, bool EnableNoLintBlocks) { - std::string CheckName = getCheckName(Info.getID()); + const std::string CheckName = getCheckName(Info.getID()); return NoLintHandler.shouldSuppress(DiagLevel, Info, CheckName, NoLintErrors, AllowIO, EnableNoLintBlocks); } @@ -226,7 +227,7 @@ void ClangTidyContext::setSourceManager(SourceManager *SourceMgr) { static bool parseFileExtensions(llvm::ArrayRef<std::string> AllFileExtensions, FileExtensionsSet &FileExtensions) { FileExtensions.clear(); - for (StringRef Suffix : AllFileExtensions) { + for (const StringRef Suffix : AllFileExtensions) { StringRef Extension = Suffix.trim(); if (!llvm::all_of(Extension, isAlphanumeric)) return false; @@ -294,11 +295,11 @@ bool ClangTidyContext::treatAsError(StringRef CheckName) const { } std::string ClangTidyContext::getCheckName(unsigned DiagnosticID) const { - std::string ClangWarningOption = std::string( + const std::string ClangWarningOption = std::string( DiagEngine->getDiagnosticIDs()->getWarningOptionForDiag(DiagnosticID)); if (!ClangWarningOption.empty()) return "clang-diagnostic-" + ClangWarningOption; - llvm::DenseMap<unsigned, std::string>::const_iterator I = + const llvm::DenseMap<unsigned, std::string>::const_iterator I = CheckNamesByDiagnosticID.find(DiagnosticID); if (I != CheckNamesByDiagnosticID.end()) return I->second; @@ -316,7 +317,7 @@ ClangTidyDiagnosticConsumer::ClangTidyDiagnosticConsumer( void ClangTidyDiagnosticConsumer::finalizeLastError() { if (!Errors.empty()) { - ClangTidyError &Error = Errors.back(); + const ClangTidyError &Error = Errors.back(); if (Error.DiagnosticName == "clang-tidy-config") { // Never ignore these. } else if (!Context.isCheckEnabled(Error.DiagnosticName) && @@ -436,8 +437,8 @@ void ClangTidyDiagnosticConsumer::HandleDiagnostic( Level = ClangTidyError::Remark; } - bool IsWarningAsError = DiagLevel == DiagnosticsEngine::Warning && - Context.treatAsError(CheckName); + const bool IsWarningAsError = DiagLevel == DiagnosticsEngine::Warning && + Context.treatAsError(CheckName); Errors.emplace_back(CheckName, Level, Context.getCurrentBuildDirectory(), IsWarningAsError); } @@ -491,8 +492,9 @@ void ClangTidyDiagnosticConsumer::forwardDiagnostic(const Diagnostic &Info) { // Acquire a diagnostic ID also in the external diagnostics engine. auto DiagLevelAndFormatString = Context.getDiagLevelAndFormatString(Info.getID(), Info.getLocation()); - unsigned ExternalID = ExternalDiagEngine->getDiagnosticIDs()->getCustomDiagID( - DiagLevelAndFormatString.first, DiagLevelAndFormatString.second); + const unsigned ExternalID = + ExternalDiagEngine->getDiagnosticIDs()->getCustomDiagID( + DiagLevelAndFormatString.first, DiagLevelAndFormatString.second); // Forward the details. auto Builder = ExternalDiagEngine->Report(Info.getLocation(), ExternalID); @@ -501,7 +503,7 @@ void ClangTidyDiagnosticConsumer::forwardDiagnostic(const Diagnostic &Info) { for (auto Range : Info.getRanges()) Builder << Range; for (unsigned Index = 0; Index < Info.getNumArgs(); ++Index) { - DiagnosticsEngine::ArgumentKind Kind = Info.getArgKind(Index); + const DiagnosticsEngine::ArgumentKind Kind = Info.getArgKind(Index); switch (Kind) { case clang::DiagnosticsEngine::ak_std_string: Builder << Info.getArgStdStr(Index); @@ -574,7 +576,7 @@ void ClangTidyDiagnosticConsumer::checkFilters(SourceLocation Location, // FIXME: We start with a conservative approach here, but the actual type of // location needed depends on the check (in particular, where this check wants // to apply fixes). - FileID FID = Sources.getDecomposedExpansionLoc(Location).first; + const FileID FID = Sources.getDecomposedExpansionLoc(Location).first; OptionalFileEntryRef File = Sources.getFileEntryRefForID(FID); // -DMACRO definitions on the command line have locations in a virtual buffer @@ -585,13 +587,13 @@ void ClangTidyDiagnosticConsumer::checkFilters(SourceLocation Location, return; } - StringRef FileName(File->getName()); + const StringRef FileName(File->getName()); LastErrorRelatesToUserCode = LastErrorRelatesToUserCode || Sources.isInMainFile(Location) || (getHeaderFilter()->match(FileName) && !getExcludeHeaderFilter()->match(FileName)); - unsigned LineNumber = Sources.getExpansionLineNumber(Location); + const unsigned LineNumber = Sources.getExpansionLineNumber(Location); LastErrorPassesLineFilter = LastErrorPassesLineFilter || passesLineFilter(FileName, LineNumber); } @@ -707,8 +709,8 @@ void ClangTidyDiagnosticConsumer::removeIncompatibleErrors() { for (unsigned I = 0; I < ErrorFixes.size(); ++I) { for (const auto &FileAndReplace : *ErrorFixes[I].second) { for (const auto &Replace : FileAndReplace.second) { - unsigned Begin = Replace.getOffset(); - unsigned End = Begin + Replace.getLength(); + const unsigned Begin = Replace.getOffset(); + const unsigned End = Begin + Replace.getLength(); auto &Events = FileEvents[Replace.getFilePath()]; if (Begin == End) { Events.emplace_back(Begin, End, Event::ET_Insert, I, Sizes[I]); @@ -767,7 +769,7 @@ struct LessClangTidyError { }; struct EqualClangTidyError { bool operator()(const ClangTidyError &LHS, const ClangTidyError &RHS) const { - LessClangTidyError Less; + const LessClangTidyError Less; return !Less(LHS, RHS) && !Less(RHS, LHS); } }; @@ -803,7 +805,7 @@ void ClangTidyDiagnosticConsumer::removeDuplicatedDiagnosticsOfAliasCheckers() { auto IT = Errors.begin(); while (IT != Errors.end()) { ClangTidyError &Error = *IT; - std::pair<UniqueErrorSet::iterator, bool> Inserted = + const std::pair<UniqueErrorSet::iterator, bool> Inserted = UniqueErrors.insert(&Error); // Unique error, we keep it and move along. diff --git a/clang-tools-extra/clang-tidy/ClangTidyOptions.cpp b/clang-tools-extra/clang-tidy/ClangTidyOptions.cpp index c4b47a4..550f780 100644 --- a/clang-tools-extra/clang-tidy/ClangTidyOptions.cpp +++ b/clang-tools-extra/clang-tidy/ClangTidyOptions.cpp @@ -119,7 +119,7 @@ void yamlize(IO &IO, ClangTidyOptions::OptionMap &Val, bool, yamlize(IO, NOpts->Options, true, Ctx); } else if (isa<MappingNode>(I.getCurrentNode())) { IO.beginMapping(); - for (StringRef Key : IO.keys()) { + for (const StringRef Key : IO.keys()) { // Requires 'llvm::yaml::IO' to accept 'StringRef' // NOLINTNEXTLINE(bugprone-suspicious-stringview-data-usage) IO.mapRequired(Key.data(), Val[Key].Value); @@ -392,7 +392,7 @@ llvm::ErrorOr<llvm::SmallString<128>> FileOptionsBaseProvider::getNormalizedAbsolutePath(llvm::StringRef Path) { assert(FS && "FS must be set."); llvm::SmallString<128> NormalizedAbsolutePath = {Path}; - std::error_code Err = FS->makeAbsolute(NormalizedAbsolutePath); + const std::error_code Err = FS->makeAbsolute(NormalizedAbsolutePath); if (Err) return Err; llvm::sys::path::remove_dots(NormalizedAbsolutePath, /*remove_dot_dot=*/true); @@ -463,7 +463,7 @@ FileOptionsProvider::getRawOptions(StringRef FileName) { LLVM_DEBUG(llvm::dbgs() << "Getting options for file " << FileName << "...\n"); - llvm::ErrorOr<llvm::SmallString<128>> AbsoluteFilePath = + const llvm::ErrorOr<llvm::SmallString<128>> AbsoluteFilePath = getNormalizedAbsolutePath(FileName); if (!AbsoluteFilePath) return {}; @@ -471,8 +471,8 @@ FileOptionsProvider::getRawOptions(StringRef FileName) { std::vector<OptionsSource> RawOptions = DefaultOptionsProvider::getRawOptions(AbsoluteFilePath->str()); addRawFileOptions(AbsoluteFilePath->str(), RawOptions); - OptionsSource CommandLineOptions(OverrideOptions, - OptionsSourceTypeCheckCommandLineOption); + const OptionsSource CommandLineOptions( + OverrideOptions, OptionsSourceTypeCheckCommandLineOption); RawOptions.push_back(CommandLineOptions); return RawOptions; @@ -502,7 +502,7 @@ FileOptionsBaseProvider::tryReadConfigFile(StringRef Directory) { llvm::ErrorOr<std::unique_ptr<llvm::MemoryBuffer>> Text = FS->getBufferForFile(ConfigFile); - if (std::error_code EC = Text.getError()) { + if (const std::error_code EC = Text.getError()) { llvm::errs() << "Can't read " << ConfigFile << ": " << EC.message() << "\n"; continue; diff --git a/clang-tools-extra/clang-tidy/ClangTidyProfiling.cpp b/clang-tools-extra/clang-tidy/ClangTidyProfiling.cpp index 8ea6b76..6fee154 100644 --- a/clang-tools-extra/clang-tidy/ClangTidyProfiling.cpp +++ b/clang-tools-extra/clang-tidy/ClangTidyProfiling.cpp @@ -59,7 +59,8 @@ void ClangTidyProfiling::storeProfileData(llvm::TimerGroup &TG) { llvm::SmallString<256> OutputDirectory(Storage->StoreFilename); llvm::sys::path::remove_filename(OutputDirectory); - if (std::error_code EC = llvm::sys::fs::create_directories(OutputDirectory)) { + if (const std::error_code EC = + llvm::sys::fs::create_directories(OutputDirectory)) { llvm::errs() << "Unable to create output directory '" << OutputDirectory << "': " << EC.message() << "\n"; return; diff --git a/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.h b/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.h index aaa0410..60cb01f 100644 --- a/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.h +++ b/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLING_EXPANDMODULARHEADERSPPCALLBACKS_H_ -#define LLVM_CLANG_TOOLING_EXPANDMODULARHEADERSPPCALLBACKS_H_ +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_EXPANDMODULARHEADERSPPCALLBACKS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_EXPANDMODULARHEADERSPPCALLBACKS_H #include "clang/Lex/HeaderSearchOptions.h" #include "clang/Lex/PPCallbacks.h" @@ -144,4 +144,4 @@ private: } // namespace tooling } // namespace clang -#endif // LLVM_CLANG_TOOLING_EXPANDMODULARHEADERSPPCALLBACKS_H_ +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_EXPANDMODULARHEADERSPPCALLBACKS_H diff --git a/clang-tools-extra/clang-tidy/FileExtensionsSet.h b/clang-tools-extra/clang-tidy/FileExtensionsSet.h index 95c221c..f97bb64 100644 --- a/clang-tools-extra/clang-tidy/FileExtensionsSet.h +++ b/clang-tools-extra/clang-tidy/FileExtensionsSet.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FILE_EXTENSIONS_SET_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FILE_EXTENSIONS_SET_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FILEEXTENSIONSSET_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FILEEXTENSIONSSET_H #include "llvm/ADT/SmallSet.h" #include "llvm/ADT/StringRef.h" @@ -16,4 +16,4 @@ namespace clang::tidy { using FileExtensionsSet = llvm::SmallSet<llvm::StringRef, 5>; } // namespace clang::tidy -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FILE_EXTENSIONS_SET_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FILEEXTENSIONSSET_H diff --git a/clang-tools-extra/clang-tidy/GlobList.cpp b/clang-tools-extra/clang-tidy/GlobList.cpp index 667a256..5d5a5f2 100644 --- a/clang-tools-extra/clang-tidy/GlobList.cpp +++ b/clang-tools-extra/clang-tidy/GlobList.cpp @@ -23,16 +23,17 @@ static bool consumeNegativeIndicator(StringRef &GlobList) { // removes it and the trailing comma from the GlobList and // returns the extracted glob. static llvm::StringRef extractNextGlob(StringRef &GlobList) { - StringRef UntrimmedGlob = GlobList.substr(0, GlobList.find_first_of(",\n")); - StringRef Glob = UntrimmedGlob.trim(); + const StringRef UntrimmedGlob = + GlobList.substr(0, GlobList.find_first_of(",\n")); + const StringRef Glob = UntrimmedGlob.trim(); GlobList = GlobList.substr(UntrimmedGlob.size() + 1); return Glob; } static llvm::Regex createRegexFromGlob(StringRef &Glob) { SmallString<128> RegexText("^"); - StringRef MetaChars("()^$|*+?.[]\\{}"); - for (char C : Glob) { + const StringRef MetaChars("()^$|*+?.[]\\{}"); + for (const char C : Glob) { if (C == '*') RegexText.push_back('.'); else if (MetaChars.contains(C)) diff --git a/clang-tools-extra/clang-tidy/NoLintDirectiveHandler.cpp b/clang-tools-extra/clang-tidy/NoLintDirectiveHandler.cpp index ef20ee1..a74ab00 100644 --- a/clang-tools-extra/clang-tidy/NoLintDirectiveHandler.cpp +++ b/clang-tools-extra/clang-tidy/NoLintDirectiveHandler.cpp @@ -134,7 +134,7 @@ static SmallVector<NoLintToken> getNoLints(StringRef Buffer) { // Get checks, if specified. std::optional<StringRef> Checks; if (Pos < Buffer.size() && Buffer[Pos] == '(') { - size_t ClosingBracket = Buffer.find_first_of("\n)", ++Pos); + const size_t ClosingBracket = Buffer.find_first_of("\n)", ++Pos); if (ClosingBracket != StringRef::npos && Buffer[ClosingBracket] == ')') { Checks = Buffer.slice(Pos, ClosingBracket); Pos = ClosingBracket + 1; @@ -183,13 +183,13 @@ static tooling::Diagnostic makeNoLintError(const SourceManager &SrcMgr, tooling::Diagnostic Error; Error.DiagLevel = tooling::Diagnostic::Error; Error.DiagnosticName = "clang-tidy-nolint"; - StringRef Message = + const StringRef Message = (NoLint.Type == NoLintType::NoLintBegin) ? ("unmatched 'NOLINTBEGIN' comment without a subsequent 'NOLINT" "END' comment") : ("unmatched 'NOLINTEND' comment without a previous 'NOLINT" "BEGIN' comment"); - SourceLocation Loc = SrcMgr.getComposedLoc(File, NoLint.Pos); + const SourceLocation Loc = SrcMgr.getComposedLoc(File, NoLint.Pos); Error.Message = tooling::DiagnosticMessage(Message, SrcMgr, Loc); return Error; } @@ -294,8 +294,8 @@ bool NoLintDirectiveHandler::Impl::diagHasNoLintInMacro( // this line. static std::pair<size_t, size_t> getLineStartAndEnd(StringRef Buffer, size_t From) { - size_t StartPos = Buffer.find_last_of('\n', From) + 1; - size_t EndPos = std::min(Buffer.find('\n', From), Buffer.size()); + const size_t StartPos = Buffer.find_last_of('\n', From) + 1; + const size_t EndPos = std::min(Buffer.find('\n', From), Buffer.size()); return {StartPos, EndPos}; } diff --git a/clang-tools-extra/clang-tidy/abseil/AbseilMatcher.h b/clang-tools-extra/clang-tidy/abseil/AbseilMatcher.h index 2ae3c00..982774c 100644 --- a/clang-tools-extra/clang-tidy/abseil/AbseilMatcher.h +++ b/clang-tools-extra/clang-tidy/abseil/AbseilMatcher.h @@ -6,6 +6,9 @@ // //===----------------------------------------------------------------------===// +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_ABSEILMATCHER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_ABSEILMATCHER_H + #include "clang/AST/ASTContext.h" #include "clang/ASTMatchers/ASTMatchFinder.h" #include <algorithm> @@ -31,7 +34,7 @@ AST_POLYMORPHIC_MATCHER( isInAbseilFile, AST_POLYMORPHIC_SUPPORTED_TYPES(Decl, Stmt, TypeLoc, NestedNameSpecifierLoc)) { auto &SourceManager = Finder->getASTContext().getSourceManager(); - SourceLocation Loc = SourceManager.getSpellingLoc(Node.getBeginLoc()); + const SourceLocation Loc = SourceManager.getSpellingLoc(Node.getBeginLoc()); if (Loc.isInvalid()) return false; OptionalFileEntryRef FileEntry = @@ -42,7 +45,7 @@ AST_POLYMORPHIC_MATCHER( // [absl-library] is AbseilLibraries list entry. StringRef Path = FileEntry->getName(); static constexpr llvm::StringLiteral AbslPrefix("absl/"); - size_t PrefixPosition = Path.find(AbslPrefix); + const size_t PrefixPosition = Path.find(AbslPrefix); if (PrefixPosition == StringRef::npos) return false; Path = Path.drop_front(PrefixPosition + AbslPrefix.size()); @@ -57,3 +60,5 @@ AST_POLYMORPHIC_MATCHER( } } // namespace clang::ast_matchers + +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_ABSEILMATCHER_H diff --git a/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.cpp b/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.cpp index 03f78f1..421e5973 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.cpp @@ -41,7 +41,7 @@ void DurationAdditionCheck::check(const MatchFinder::MatchResult &Result) { if (!Scale) return; - llvm::StringRef TimeFactory = getTimeInverseForScale(*Scale); + const llvm::StringRef TimeFactory = getTimeInverseForScale(*Scale); FixItHint Hint; if (Call == Binop->getLHS()->IgnoreParenImpCasts()) { diff --git a/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.h b/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.h index b728118..f5bab53 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.h +++ b/clang-tools-extra/clang-tidy/abseil/DurationAdditionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMEADDITIONCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMEADDITIONCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_DURATIONADDITIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_DURATIONADDITIONCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::abseil -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMEADDITIONCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_DURATIONADDITIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/abseil/DurationComparisonCheck.cpp b/clang-tools-extra/clang-tidy/abseil/DurationComparisonCheck.cpp index 16a244b..f008777 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationComparisonCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationComparisonCheck.cpp @@ -38,9 +38,9 @@ void DurationComparisonCheck::check(const MatchFinder::MatchResult &Result) { // if nothing needs to be done. if (isInMacro(Result, Binop->getLHS()) || isInMacro(Result, Binop->getRHS())) return; - std::string LhsReplacement = + const std::string LhsReplacement = rewriteExprFromNumberToDuration(Result, *Scale, Binop->getLHS()); - std::string RhsReplacement = + const std::string RhsReplacement = rewriteExprFromNumberToDuration(Result, *Scale, Binop->getRHS()); diag(Binop->getBeginLoc(), "perform comparison in the duration domain") diff --git a/clang-tools-extra/clang-tidy/abseil/DurationConversionCastCheck.cpp b/clang-tools-extra/clang-tidy/abseil/DurationConversionCastCheck.cpp index 11d6017..ef06a9e 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationConversionCastCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationConversionCastCheck.cpp @@ -41,7 +41,7 @@ void DurationConversionCastCheck::check( const auto *FuncDecl = Result.Nodes.getNodeAs<FunctionDecl>("func_decl"); const auto *Arg = Result.Nodes.getNodeAs<Expr>("arg"); - StringRef ConversionFuncName = FuncDecl->getName(); + const StringRef ConversionFuncName = FuncDecl->getName(); std::optional<DurationScale> Scale = getScaleForDurationInverse(ConversionFuncName); @@ -51,7 +51,8 @@ void DurationConversionCastCheck::check( // Casting a double to an integer. if (MatchedCast->getTypeAsWritten()->isIntegerType() && ConversionFuncName.contains("Double")) { - llvm::StringRef NewFuncName = getDurationInverseForScale(*Scale).second; + const llvm::StringRef NewFuncName = + getDurationInverseForScale(*Scale).second; diag(MatchedCast->getBeginLoc(), "duration should be converted directly to an integer rather than " @@ -66,7 +67,8 @@ void DurationConversionCastCheck::check( // Casting an integer to a double. if (MatchedCast->getTypeAsWritten()->isRealFloatingType() && ConversionFuncName.contains("Int64")) { - llvm::StringRef NewFuncName = getDurationInverseForScale(*Scale).first; + const llvm::StringRef NewFuncName = + getDurationInverseForScale(*Scale).first; diag(MatchedCast->getBeginLoc(), "duration should be converted directly to " "a floating-point number rather than " diff --git a/clang-tools-extra/clang-tidy/abseil/DurationFactoryScaleCheck.cpp b/clang-tools-extra/clang-tidy/abseil/DurationFactoryScaleCheck.cpp index 3346297..9e403fb8 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationFactoryScaleCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationFactoryScaleCheck.cpp @@ -158,7 +158,7 @@ void DurationFactoryScaleCheck::check(const MatchFinder::MatchResult &Result) { if (!MaybeScale) return; - DurationScale Scale = *MaybeScale; + const DurationScale Scale = *MaybeScale; const Expr *Remainder = nullptr; std::optional<DurationScale> NewScale; diff --git a/clang-tools-extra/clang-tidy/abseil/DurationRewriter.cpp b/clang-tools-extra/clang-tidy/abseil/DurationRewriter.cpp index ee19796..a78d07d 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationRewriter.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationRewriter.cpp @@ -20,7 +20,7 @@ namespace clang::tidy::abseil { /// Returns an integer if the fractional part of a `FloatingLiteral` is `0`. static std::optional<llvm::APSInt> truncateIfIntegral(const FloatingLiteral &FloatLiteral) { - double Value = FloatLiteral.getValueAsApproximateDouble(); + const double Value = FloatLiteral.getValueAsApproximateDouble(); if (std::fmod(Value, 1) == 0) { if (Value >= static_cast<double>(1U << 31)) return std::nullopt; @@ -69,7 +69,7 @@ rewriteInverseDurationCall(const MatchFinder::MatchResult &Result, static std::optional<std::string> rewriteInverseTimeCall(const MatchFinder::MatchResult &Result, DurationScale Scale, const Expr &Node) { - llvm::StringRef InverseFunction = getTimeInverseForScale(Scale); + const llvm::StringRef InverseFunction = getTimeInverseForScale(Scale); if (const auto *MaybeCallArg = selectFirst<const Expr>( "e", match(callExpr(callee(functionDecl(hasName(InverseFunction))), hasArgument(0, expr().bind("e"))), diff --git a/clang-tools-extra/clang-tidy/abseil/DurationSubtractionCheck.cpp b/clang-tools-extra/clang-tidy/abseil/DurationSubtractionCheck.cpp index c5d93ad..42a7df4 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationSubtractionCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationSubtractionCheck.cpp @@ -41,7 +41,7 @@ void DurationSubtractionCheck::check(const MatchFinder::MatchResult &Result) { if (!Scale) return; - std::string RhsReplacement = + const std::string RhsReplacement = rewriteExprFromNumberToDuration(Result, *Scale, Binop->getRHS()); const Expr *LhsArg = Result.Nodes.getNodeAs<Expr>("lhs_arg"); diff --git a/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.cpp b/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.cpp index 805d7da..5867fb6 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.cpp @@ -19,10 +19,10 @@ namespace clang::tidy::abseil { void DurationUnnecessaryConversionCheck::registerMatchers(MatchFinder *Finder) { for (const auto &Scale : {"Hours", "Minutes", "Seconds", "Milliseconds", "Microseconds", "Nanoseconds"}) { - std::string DurationFactory = (llvm::Twine("::absl::") + Scale).str(); - std::string FloatConversion = + const std::string DurationFactory = (llvm::Twine("::absl::") + Scale).str(); + const std::string FloatConversion = (llvm::Twine("::absl::ToDouble") + Scale).str(); - std::string IntegerConversion = + const std::string IntegerConversion = (llvm::Twine("::absl::ToInt64") + Scale).str(); // Matcher which matches the current scale's factory with a `1` argument, diff --git a/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.h b/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.h index 59af8968..f5d2511 100644 --- a/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.h +++ b/clang-tools-extra/clang-tidy/abseil/DurationUnnecessaryConversionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMEDOUBLECONVERSIONCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMEDOUBLECONVERSIONCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_DURATIONUNNECESSARYCONVERSIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_DURATIONUNNECESSARYCONVERSIONCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::abseil -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMEDOUBLECONVERSIONCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_DURATIONUNNECESSARYCONVERSIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/abseil/FasterStrsplitDelimiterCheck.cpp b/clang-tools-extra/clang-tidy/abseil/FasterStrsplitDelimiterCheck.cpp index d9f6551..0827526 100644 --- a/clang-tools-extra/clang-tidy/abseil/FasterStrsplitDelimiterCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/FasterStrsplitDelimiterCheck.cpp @@ -29,7 +29,7 @@ makeCharacterLiteral(const StringLiteral *Literal, const ASTContext &Context) { assert(Literal->getCharByteWidth() == 1 && "StrSplit doesn't support wide char"); std::string Result = clang::tooling::fixit::getText(*Literal, Context).str(); - bool IsRawStringLiteral = StringRef(Result).starts_with(R"(R")"); + const bool IsRawStringLiteral = StringRef(Result).starts_with(R"(R")"); // Since raw string literal might contain unescaped non-printable characters, // we normalize them using `StringLiteral::outputString`. if (IsRawStringLiteral) { diff --git a/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.cpp b/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.cpp index c090e5a..5f4cb66 100644 --- a/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.cpp @@ -32,7 +32,7 @@ void NoInternalDependenciesCheck::check( const auto *InternalDependency = Result.Nodes.getNodeAs<NestedNameSpecifierLoc>("InternalDep"); - SourceLocation LocAtFault = + const SourceLocation LocAtFault = Result.SourceManager->getSpellingLoc(InternalDependency->getBeginLoc()); if (!LocAtFault.isValid()) diff --git a/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.h b/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.h index 2911a1a..2291831 100644 --- a/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.h +++ b/clang-tools-extra/clang-tidy/abseil/NoInternalDependenciesCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_NOINTERNALDEPSCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_NOINTERNALDEPSCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_NOINTERNALDEPENDENCIESCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_NOINTERNALDEPENDENCIESCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::abseil -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_NOINTERNALDEPSCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_NOINTERNALDEPENDENCIESCHECK_H diff --git a/clang-tools-extra/clang-tidy/abseil/StringFindStartswithCheck.cpp b/clang-tools-extra/clang-tidy/abseil/StringFindStartswithCheck.cpp index 92d6305..e1063c4f 100644 --- a/clang-tools-extra/clang-tidy/abseil/StringFindStartswithCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/StringFindStartswithCheck.cpp @@ -92,7 +92,7 @@ void StringFindStartswithCheck::check(const MatchFinder::MatchResult &Result) { const auto *FindFun = Result.Nodes.getNodeAs<CXXMethodDecl>("findfun"); assert(FindFun != nullptr); - bool Rev = FindFun->getName().contains("rfind"); + const bool Rev = FindFun->getName().contains("rfind"); if (ComparisonExpr->getBeginLoc().isMacroID()) return; @@ -107,7 +107,7 @@ void StringFindStartswithCheck::check(const MatchFinder::MatchResult &Result) { Context.getLangOpts()); // Create the StartsWith string, negating if comparison was "!=". - bool Neg = ComparisonExpr->getOpcode() == BO_NE; + const bool Neg = ComparisonExpr->getOpcode() == BO_NE; // Create the warning message and a FixIt hint replacing the original expr. auto Diagnostic = diff --git a/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.cpp b/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.cpp index 7a97a18..5d80b16 100644 --- a/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.cpp @@ -39,9 +39,9 @@ void TimeComparisonCheck::check(const MatchFinder::MatchResult &Result) { // want to handle the case of rewriting both sides. This is much simpler if // we unconditionally try and rewrite both, and let the rewriter determine // if nothing needs to be done. - std::string LhsReplacement = + const std::string LhsReplacement = rewriteExprFromNumberToTime(Result, *Scale, Binop->getLHS()); - std::string RhsReplacement = + const std::string RhsReplacement = rewriteExprFromNumberToTime(Result, *Scale, Binop->getRHS()); diag(Binop->getBeginLoc(), "perform comparison in the time domain") diff --git a/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.h b/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.h index 703d951..74a877a 100644 --- a/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.h +++ b/clang-tools-extra/clang-tidy/abseil/TimeComparisonCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMECOMPARECHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMECOMPARECHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMECOMPARISONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMECOMPARISONCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::abseil -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMECOMPARECHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ABSEIL_TIMECOMPARISONCHECK_H diff --git a/clang-tools-extra/clang-tidy/abseil/TimeSubtractionCheck.cpp b/clang-tools-extra/clang-tidy/abseil/TimeSubtractionCheck.cpp index 228d974..4ae49d2 100644 --- a/clang-tools-extra/clang-tidy/abseil/TimeSubtractionCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/TimeSubtractionCheck.cpp @@ -93,7 +93,7 @@ void TimeSubtractionCheck::emitDiagnostic(const Expr *Node, void TimeSubtractionCheck::registerMatchers(MatchFinder *Finder) { for (const char *ScaleName : {"Hours", "Minutes", "Seconds", "Millis", "Micros", "Nanos"}) { - std::string TimeInverse = (llvm::Twine("ToUnix") + ScaleName).str(); + const std::string TimeInverse = (llvm::Twine("ToUnix") + ScaleName).str(); std::optional<DurationScale> Scale = getScaleForTimeInverse(TimeInverse); assert(Scale && "Unknown scale encountered"); @@ -127,7 +127,7 @@ void TimeSubtractionCheck::registerMatchers(MatchFinder *Finder) { void TimeSubtractionCheck::check(const MatchFinder::MatchResult &Result) { const auto *BinOp = Result.Nodes.getNodeAs<BinaryOperator>("binop"); - std::string InverseName = + const std::string InverseName = Result.Nodes.getNodeAs<FunctionDecl>("func_decl")->getNameAsString(); if (insideMacroDefinition(Result, BinOp->getSourceRange())) return; @@ -144,7 +144,7 @@ void TimeSubtractionCheck::check(const MatchFinder::MatchResult &Result) { // We're working with the first case of matcher, and need to replace the // entire 'Duration' factory call. (Which also means being careful about // our order-of-operations and optionally putting in some parenthesis. - bool NeedParens = parensRequired(Result, OuterCall); + const bool NeedParens = parensRequired(Result, OuterCall); emitDiagnostic( OuterCall, @@ -169,7 +169,7 @@ void TimeSubtractionCheck::check(const MatchFinder::MatchResult &Result) { // converts it from the inverse to a Duration. In this case, we replace // the outer with just the subtraction expression, which gives the right // type and scale, taking care again about parenthesis. - bool NeedParens = parensRequired(Result, MaybeCallArg); + const bool NeedParens = parensRequired(Result, MaybeCallArg); emitDiagnostic( MaybeCallArg, diff --git a/clang-tools-extra/clang-tidy/abseil/UpgradeDurationConversionsCheck.cpp b/clang-tools-extra/clang-tidy/abseil/UpgradeDurationConversionsCheck.cpp index 8b197e5b..1a6ff30 100644 --- a/clang-tools-extra/clang-tidy/abseil/UpgradeDurationConversionsCheck.cpp +++ b/clang-tools-extra/clang-tidy/abseil/UpgradeDurationConversionsCheck.cpp @@ -117,10 +117,10 @@ void UpgradeDurationConversionsCheck::check( "implicit conversion to 'int64_t' is deprecated in this context; use an " "explicit cast instead"; - TraversalKindScope RAII(*Result.Context, TK_AsIs); + const TraversalKindScope RAII(*Result.Context, TK_AsIs); const auto *ArgExpr = Result.Nodes.getNodeAs<Expr>("arg"); - SourceLocation Loc = ArgExpr->getBeginLoc(); + const SourceLocation Loc = ArgExpr->getBeginLoc(); const auto *OuterExpr = Result.Nodes.getNodeAs<Expr>("OuterExpr"); @@ -139,13 +139,13 @@ void UpgradeDurationConversionsCheck::check( // We gather source locations from template matches not in template // instantiations for future matches. - internal::Matcher<Stmt> IsInsideTemplate = + const internal::Matcher<Stmt> IsInsideTemplate = hasAncestor(decl(anyOf(classTemplateDecl(), functionTemplateDecl()))); if (!match(IsInsideTemplate, *ArgExpr, *Result.Context).empty()) MatchedTemplateLocations.insert(Loc); - DiagnosticBuilder Diag = diag(Loc, Message); - CharSourceRange SourceRange = Lexer::makeFileCharRange( + const DiagnosticBuilder Diag = diag(Loc, Message); + const CharSourceRange SourceRange = Lexer::makeFileCharRange( CharSourceRange::getTokenRange(ArgExpr->getSourceRange()), *Result.SourceManager, Result.Context->getLangOpts()); if (SourceRange.isInvalid()) diff --git a/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.cpp b/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.cpp index 49ba17c..519d909 100644 --- a/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.cpp +++ b/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.cpp @@ -76,7 +76,7 @@ void IdDependentBackwardBranchCheck::registerMatchers(MatchFinder *Finder) { this); } -IdDependentBackwardBranchCheck::IdDependencyRecord * +const IdDependentBackwardBranchCheck::IdDependencyRecord * IdDependentBackwardBranchCheck::hasIdDepVar(const Expr *Expression) { if (!Expression) return nullptr; @@ -94,12 +94,12 @@ IdDependentBackwardBranchCheck::hasIdDepVar(const Expr *Expression) { } for (const auto *Child : Expression->children()) if (const auto *ChildExpression = dyn_cast_if_present<Expr>(Child)) - if (IdDependencyRecord *Result = hasIdDepVar(ChildExpression)) + if (const IdDependencyRecord *Result = hasIdDepVar(ChildExpression)) return Result; return nullptr; } -IdDependentBackwardBranchCheck::IdDependencyRecord * +const IdDependentBackwardBranchCheck::IdDependencyRecord * IdDependentBackwardBranchCheck::hasIdDepField(const Expr *Expression) { if (!Expression) return nullptr; @@ -116,7 +116,7 @@ IdDependentBackwardBranchCheck::hasIdDepField(const Expr *Expression) { } for (const auto *Child : Expression->children()) if (const auto *ChildExpression = dyn_cast_if_present<Expr>(Child)) - if (IdDependencyRecord *Result = hasIdDepField(ChildExpression)) + if (const IdDependencyRecord *Result = hasIdDepField(ChildExpression)) return Result; return nullptr; } @@ -239,7 +239,7 @@ void IdDependentBackwardBranchCheck::check( const auto *Loop = Result.Nodes.getNodeAs<Stmt>("backward_branch"); if (!Loop) return; - LoopType Type = getLoopType(Loop); + const LoopType Type = getLoopType(Loop); if (CondExpr) { if (IDCall) { // Conditional expression calls an ID function directly. diag(CondExpr->getBeginLoc(), @@ -249,8 +249,8 @@ void IdDependentBackwardBranchCheck::check( return; } // Conditional expression has DeclRefExpr(s), check ID-dependency. - IdDependencyRecord *IdDepVar = hasIdDepVar(CondExpr); - IdDependencyRecord *IdDepField = hasIdDepField(CondExpr); + const IdDependencyRecord *IdDepVar = hasIdDepVar(CondExpr); + const IdDependencyRecord *IdDepField = hasIdDepField(CondExpr); if (IdDepVar) { diag(CondExpr->getBeginLoc(), "backward branch (%select{do|while|for}0 loop) is ID-dependent due " diff --git a/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.h b/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.h index b777918..297e775 100644 --- a/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.h +++ b/clang-tools-extra/clang-tidy/altera/IdDependentBackwardBranchCheck.h @@ -44,10 +44,10 @@ private: std::map<const FieldDecl *, IdDependencyRecord> IdDepFieldsMap; /// Returns an IdDependencyRecord if the Expression contains an ID-dependent /// variable, returns a nullptr otherwise. - IdDependencyRecord *hasIdDepVar(const Expr *Expression); + const IdDependencyRecord *hasIdDepVar(const Expr *Expression); /// Returns an IdDependencyRecord if the Expression contains an ID-dependent /// field, returns a nullptr otherwise. - IdDependencyRecord *hasIdDepField(const Expr *Expression); + const IdDependencyRecord *hasIdDepField(const Expr *Expression); /// Stores the location an ID-dependent variable is created from a call to /// an ID function in IdDepVarsMap. void saveIdDepVar(const Stmt *Statement, const VarDecl *Variable); diff --git a/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.cpp b/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.cpp index 4c740e3..8aa23b8 100644 --- a/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.cpp +++ b/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.cpp @@ -77,7 +77,7 @@ void KernelNameRestrictionPPCallbacks::EndOfMainFile() { // Check main file for restricted names. OptionalFileEntryRef Entry = SM.getFileEntryRefForID(SM.getMainFileID()); - StringRef FileName = llvm::sys::path::filename(Entry->getName()); + const StringRef FileName = llvm::sys::path::filename(Entry->getName()); if (fileNameIsRestricted(FileName)) Check.diag(SM.getLocForStartOfFile(SM.getMainFileID()), "compiling '%0' may cause additional compilation errors due " @@ -90,7 +90,7 @@ void KernelNameRestrictionPPCallbacks::EndOfMainFile() { // Check included files for restricted names. for (const IncludeDirective &ID : IncludeDirectives) { - StringRef FileName = llvm::sys::path::filename(ID.FileName); + const StringRef FileName = llvm::sys::path::filename(ID.FileName); if (fileNameIsRestricted(FileName)) Check.diag(ID.Loc, "including '%0' may cause additional compilation errors due " diff --git a/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.h b/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.h index 182d10b..441cf36 100644 --- a/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.h +++ b/clang-tools-extra/clang-tidy/altera/KernelNameRestrictionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_KERNEL_NAME_RESTRICTION_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_KERNEL_NAME_RESTRICTION_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_KERNELNAMERESTRICTIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_KERNELNAMERESTRICTIONCHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::altera -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_KERNEL_NAME_RESTRICTION_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_KERNELNAMERESTRICTIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/altera/SingleWorkItemBarrierCheck.h b/clang-tools-extra/clang-tidy/altera/SingleWorkItemBarrierCheck.h index dab3dbc..dcfefcb 100644 --- a/clang-tools-extra/clang-tidy/altera/SingleWorkItemBarrierCheck.h +++ b/clang-tools-extra/clang-tidy/altera/SingleWorkItemBarrierCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_SINGLE_WORK_ITEM_BARRIER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_SINGLE_WORK_ITEM_BARRIER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_SINGLEWORKITEMBARRIERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_SINGLEWORKITEMBARRIERCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ public: } // namespace clang::tidy::altera -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_SINGLE_WORK_ITEM_BARRIER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ALTERA_SINGLEWORKITEMBARRIERCHECK_H diff --git a/clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp b/clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp index 0a19378..2b1312c 100644 --- a/clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp +++ b/clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp @@ -60,10 +60,10 @@ void StructPackAlignCheck::check(const MatchFinder::MatchResult &Result) { // For each StructField, record how big it is (in bits). // Would be good to use a pair of <offset, size> to advise a better // packing order. - QualType StructFieldTy = StructField->getType(); + const QualType StructFieldTy = StructField->getType(); if (StructFieldTy->isIncompleteType()) return; - unsigned int StructFieldWidth = + const unsigned int StructFieldWidth = (unsigned int)Result.Context->getTypeInfo(StructFieldTy.getTypePtr()) .Width; FieldSizes.emplace_back(StructFieldWidth, StructField->getFieldIndex()); @@ -72,21 +72,22 @@ void StructPackAlignCheck::check(const MatchFinder::MatchResult &Result) { TotalBitSize += StructFieldWidth; } - uint64_t CharSize = Result.Context->getCharWidth(); - CharUnits CurrSize = Result.Context->getASTRecordLayout(Struct).getSize(); - CharUnits MinByteSize = + const uint64_t CharSize = Result.Context->getCharWidth(); + const CharUnits CurrSize = + Result.Context->getASTRecordLayout(Struct).getSize(); + const CharUnits MinByteSize = CharUnits::fromQuantity(std::max<clang::CharUnits::QuantityType>( std::ceil(static_cast<float>(TotalBitSize) / CharSize), 1)); - CharUnits MaxAlign = CharUnits::fromQuantity( + const CharUnits MaxAlign = CharUnits::fromQuantity( std::ceil((float)Struct->getMaxAlignment() / CharSize)); - CharUnits CurrAlign = + const CharUnits CurrAlign = Result.Context->getASTRecordLayout(Struct).getAlignment(); - CharUnits NewAlign = computeRecommendedAlignment(MinByteSize); + const CharUnits NewAlign = computeRecommendedAlignment(MinByteSize); - bool IsPacked = Struct->hasAttr<PackedAttr>(); - bool NeedsPacking = (MinByteSize < CurrSize) && (MaxAlign != NewAlign) && - (CurrSize != NewAlign); - bool NeedsAlignment = CurrAlign.getQuantity() != NewAlign.getQuantity(); + const bool IsPacked = Struct->hasAttr<PackedAttr>(); + const bool NeedsPacking = (MinByteSize < CurrSize) && + (MaxAlign != NewAlign) && (CurrSize != NewAlign); + const bool NeedsAlignment = CurrAlign.getQuantity() != NewAlign.getQuantity(); if (!NeedsAlignment && !NeedsPacking) return; @@ -111,7 +112,8 @@ void StructPackAlignCheck::check(const MatchFinder::MatchResult &Result) { FixItHint FixIt; auto *Attribute = Struct->getAttr<AlignedAttr>(); - std::string NewAlignQuantity = std::to_string((int)NewAlign.getQuantity()); + const std::string NewAlignQuantity = + std::to_string((int)NewAlign.getQuantity()); if (Attribute) { FixIt = FixItHint::CreateReplacement( Attribute->getRange(), diff --git a/clang-tools-extra/clang-tidy/altera/UnrollLoopsCheck.cpp b/clang-tools-extra/clang-tidy/altera/UnrollLoopsCheck.cpp index e90cdd0..b0cd4cd 100644 --- a/clang-tools-extra/clang-tidy/altera/UnrollLoopsCheck.cpp +++ b/clang-tools-extra/clang-tidy/altera/UnrollLoopsCheck.cpp @@ -127,7 +127,7 @@ bool UnrollLoopsCheck::hasKnownBounds(const Stmt *Statement, if (const auto *InitDeclStatement = dyn_cast<DeclStmt>(Initializer)) { if (const auto *VariableDecl = dyn_cast<VarDecl>(InitDeclStatement->getSingleDecl())) { - APValue *Evaluation = VariableDecl->evaluateValue(); + const APValue *Evaluation = VariableDecl->evaluateValue(); if (!Evaluation || !Evaluation->hasValue()) return false; } diff --git a/clang-tools-extra/clang-tidy/android/CloexecAccept4Check.h b/clang-tools-extra/clang-tidy/android/CloexecAccept4Check.h index 02c4e00..5637fc8 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecAccept4Check.h +++ b/clang-tools-extra/clang-tidy/android/CloexecAccept4Check.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_ACCEPT4_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_ACCEPT4_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECACCEPT4CHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECACCEPT4CHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_ACCEPT4_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECACCEPT4CHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.cpp b/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.cpp index 9cd888c..a624523 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.cpp +++ b/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.cpp @@ -26,10 +26,11 @@ void CloexecAcceptCheck::registerMatchers(MatchFinder *Finder) { } void CloexecAcceptCheck::check(const MatchFinder::MatchResult &Result) { - std::string ReplacementText = (Twine("accept4(") + getSpellingArg(Result, 0) + - ", " + getSpellingArg(Result, 1) + ", " + - getSpellingArg(Result, 2) + ", SOCK_CLOEXEC)") - .str(); + const std::string ReplacementText = + (Twine("accept4(") + getSpellingArg(Result, 0) + ", " + + getSpellingArg(Result, 1) + ", " + getSpellingArg(Result, 2) + + ", SOCK_CLOEXEC)") + .str(); replaceFunc( Result, diff --git a/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.h b/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.h index 4540f93..332a97a 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecAcceptCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_ACCEPT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_ACCEPT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECACCEPTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECACCEPTCHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_ACCEPT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECACCEPTCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecCheck.cpp b/clang-tools-extra/clang-tidy/android/CloexecCheck.cpp index 48c54c0..c046578 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecCheck.cpp +++ b/clang-tools-extra/clang-tidy/android/CloexecCheck.cpp @@ -30,7 +30,8 @@ static std::string buildFixMsgForStringFlag(const Expr *Arg, " \"" + Twine(Mode) + "\"") .str(); - StringRef SR = cast<StringLiteral>(Arg->IgnoreParenCasts())->getString(); + const StringRef SR = + cast<StringLiteral>(Arg->IgnoreParenCasts())->getString(); return ("\"" + SR + Twine(Mode) + "\"").str(); } @@ -49,14 +50,14 @@ void CloexecCheck::insertMacroFlag(const MatchFinder::MatchResult &Result, const auto *MatchedCall = Result.Nodes.getNodeAs<CallExpr>(FuncBindingStr); const auto *FlagArg = MatchedCall->getArg(ArgPos); const auto *FD = Result.Nodes.getNodeAs<FunctionDecl>(FuncDeclBindingStr); - SourceManager &SM = *Result.SourceManager; + const SourceManager &SM = *Result.SourceManager; if (utils::exprHasBitFlagWithSpelling(FlagArg->IgnoreParenCasts(), SM, Result.Context->getLangOpts(), MacroFlag)) return; - SourceLocation EndLoc = + const SourceLocation EndLoc = Lexer::getLocForEndOfToken(SM.getFileLoc(FlagArg->getEndLoc()), 0, SM, Result.Context->getLangOpts()); @@ -84,7 +85,7 @@ void CloexecCheck::insertStringFlag( if (!ModeStr || ModeStr->getString().contains(Mode)) return; - std::string ReplacementText = buildFixMsgForStringFlag( + const std::string ReplacementText = buildFixMsgForStringFlag( ModeArg, *Result.SourceManager, Result.Context->getLangOpts(), Mode); diag(ModeArg->getBeginLoc(), "use %0 mode '%1' to set O_CLOEXEC") diff --git a/clang-tools-extra/clang-tidy/android/CloexecCheck.h b/clang-tools-extra/clang-tidy/android/CloexecCheck.h index 858d96a..fc1accd 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecCheck.h @@ -12,8 +12,8 @@ /// //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECCHECK_H #include "../ClangTidyCheck.h" @@ -97,4 +97,4 @@ protected: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecCreatCheck.h b/clang-tools-extra/clang-tidy/android/CloexecCreatCheck.h index ee2f51a..d7d2b42 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecCreatCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecCreatCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_CREAT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_CREAT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECCREATCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECCREATCHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_CREAT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECCREATCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecDupCheck.cpp b/clang-tools-extra/clang-tidy/android/CloexecDupCheck.cpp index 5ac1b6f..5db5746 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecDupCheck.cpp +++ b/clang-tools-extra/clang-tidy/android/CloexecDupCheck.cpp @@ -20,7 +20,7 @@ void CloexecDupCheck::registerMatchers(MatchFinder *Finder) { } void CloexecDupCheck::check(const MatchFinder::MatchResult &Result) { - std::string ReplacementText = + const std::string ReplacementText = (Twine("fcntl(") + getSpellingArg(Result, 0) + ", F_DUPFD_CLOEXEC)") .str(); diff --git a/clang-tools-extra/clang-tidy/android/CloexecDupCheck.h b/clang-tools-extra/clang-tidy/android/CloexecDupCheck.h index f569968..4eae507 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecDupCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecDupCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_DUP_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_DUP_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECDUPCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECDUPCHECK_H #include "CloexecCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_DUP_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECDUPCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecEpollCreate1Check.h b/clang-tools-extra/clang-tidy/android/CloexecEpollCreate1Check.h index f467b87..03a529f 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecEpollCreate1Check.h +++ b/clang-tools-extra/clang-tidy/android/CloexecEpollCreate1Check.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_EPOLL_CREATE1_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_EPOLL_CREATE1_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECEPOLLCREATE1CHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECEPOLLCREATE1CHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_EPOLL_CREATE1_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECEPOLLCREATE1CHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecEpollCreateCheck.h b/clang-tools-extra/clang-tidy/android/CloexecEpollCreateCheck.h index a8d17c8..243b9bd 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecEpollCreateCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecEpollCreateCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_EPOLL_CREATE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_EPOLL_CREATE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECEPOLLCREATECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECEPOLLCREATECHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_EPOLL_CREATE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECEPOLLCREATECHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecFopenCheck.h b/clang-tools-extra/clang-tidy/android/CloexecFopenCheck.h index 646b237..a018fc5 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecFopenCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecFopenCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_FOPEN_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_FOPEN_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECFOPENCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECFOPENCHECK_H #include "CloexecCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_FOPEN_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECFOPENCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecInotifyInit1Check.h b/clang-tools-extra/clang-tidy/android/CloexecInotifyInit1Check.h index 3960d05..c2e4533 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecInotifyInit1Check.h +++ b/clang-tools-extra/clang-tidy/android/CloexecInotifyInit1Check.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_INOTIFY_INIT1_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_INOTIFY_INIT1_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECINOTIFYINIT1CHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECINOTIFYINIT1CHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_INOTIFY_INIT1_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECINOTIFYINIT1CHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecInotifyInitCheck.h b/clang-tools-extra/clang-tidy/android/CloexecInotifyInitCheck.h index cb9e682..cd202c2 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecInotifyInitCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecInotifyInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_INOTIFY_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_INOTIFY_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECINOTIFYINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECINOTIFYINITCHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_INOTIFY_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECINOTIFYINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecMemfdCreateCheck.h b/clang-tools-extra/clang-tidy/android/CloexecMemfdCreateCheck.h index dd96ee9..1a77c7f 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecMemfdCreateCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecMemfdCreateCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_MEMFD_CREATE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_MEMFD_CREATE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECMEMFDCREATECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECMEMFDCREATECHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_MEMFD_CREATE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECMEMFDCREATECHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.cpp b/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.cpp index 8c24482..9938027 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.cpp +++ b/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.cpp @@ -30,7 +30,7 @@ void CloexecOpenCheck::registerMatchers(MatchFinder *Finder) { void CloexecOpenCheck::check(const MatchFinder::MatchResult &Result) { const auto *FD = Result.Nodes.getNodeAs<FunctionDecl>(FuncDeclBindingStr); assert(FD->param_size() > 1); - int ArgPos = (FD->param_size() > 2) ? 2 : 1; + const int ArgPos = (FD->param_size() > 2) ? 2 : 1; insertMacroFlag(Result, /*MacroFlag=*/"O_CLOEXEC", ArgPos); } diff --git a/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.h b/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.h index d95fe21..d30b456 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecOpenCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_OPEN_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_OPEN_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECOPENCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECOPENCHECK_H #include "CloexecCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_OPEN_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECOPENCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecPipe2Check.h b/clang-tools-extra/clang-tidy/android/CloexecPipe2Check.h index 496bd6b..3165308 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecPipe2Check.h +++ b/clang-tools-extra/clang-tidy/android/CloexecPipe2Check.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_PIPE2_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_PIPE2_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECPIPE2CHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECPIPE2CHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_PIPE2_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECPIPE2CHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.cpp b/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.cpp index a475dff..37e3c56 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.cpp +++ b/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.cpp @@ -20,7 +20,7 @@ void CloexecPipeCheck::registerMatchers(MatchFinder *Finder) { } void CloexecPipeCheck::check(const MatchFinder::MatchResult &Result) { - std::string ReplacementText = + const std::string ReplacementText = (Twine("pipe2(") + getSpellingArg(Result, 0) + ", O_CLOEXEC)").str(); replaceFunc(Result, diff --git a/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.h b/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.h index f0145e1..721a688 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecPipeCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_PIPE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_PIPE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECPIPECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECPIPECHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_PIPE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECPIPECHECK_H diff --git a/clang-tools-extra/clang-tidy/android/CloexecSocketCheck.h b/clang-tools-extra/clang-tidy/android/CloexecSocketCheck.h index 0a29d72..8865db3a 100644 --- a/clang-tools-extra/clang-tidy/android/CloexecSocketCheck.h +++ b/clang-tools-extra/clang-tidy/android/CloexecSocketCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_SOCKET_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_SOCKET_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECSOCKETCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECSOCKETCHECK_H #include "CloexecCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::android -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXEC_SOCKET_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ANDROID_CLOEXECSOCKETCHECK_H diff --git a/clang-tools-extra/clang-tidy/android/ComparisonInTempFailureRetryCheck.cpp b/clang-tools-extra/clang-tidy/android/ComparisonInTempFailureRetryCheck.cpp index 36ac9a4..c42f069 100644 --- a/clang-tools-extra/clang-tidy/android/ComparisonInTempFailureRetryCheck.cpp +++ b/clang-tools-extra/clang-tidy/android/ComparisonInTempFailureRetryCheck.cpp @@ -64,7 +64,7 @@ void ComparisonInTempFailureRetryCheck::check( const LangOptions &Opts = Result.Context->getLangOpts(); SourceLocation LocStart = Node.getBeginLoc(); while (LocStart.isMacroID()) { - SourceLocation Invocation = SM.getImmediateMacroCallerLoc(LocStart); + const SourceLocation Invocation = SM.getImmediateMacroCallerLoc(LocStart); Token Tok; if (!Lexer::getRawToken(SM.getSpellingLoc(Invocation), Tok, SM, Opts, /*IgnoreWhiteSpace=*/true)) { diff --git a/clang-tools-extra/clang-tidy/boost/UseRangesCheck.cpp b/clang-tools-extra/clang-tidy/boost/UseRangesCheck.cpp index 34ecee5..02fc8d8 100644 --- a/clang-tools-extra/clang-tidy/boost/UseRangesCheck.cpp +++ b/clang-tools-extra/clang-tidy/boost/UseRangesCheck.cpp @@ -341,7 +341,7 @@ void UseRangesCheck::storeOptions(ClangTidyOptions::OptionMap &Opts) { } DiagnosticBuilder UseRangesCheck::createDiag(const CallExpr &Call) { - DiagnosticBuilder D = + const DiagnosticBuilder D = diag(Call.getBeginLoc(), "use a %0 version of this algorithm"); D << (Call.getDirectCallee()->isInStdNamespace() ? "boost" : "ranged"); return D; diff --git a/clang-tools-extra/clang-tidy/boost/UseToStringCheck.h b/clang-tools-extra/clang-tidy/boost/UseToStringCheck.h index af87f15..dae3f7c 100644 --- a/clang-tools-extra/clang-tidy/boost/UseToStringCheck.h +++ b/clang-tools-extra/clang-tidy/boost/UseToStringCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BOOST_USE_TO_STRING_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BOOST_USE_TO_STRING_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BOOST_USETOSTRINGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BOOST_USETOSTRINGCHECK_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::boost -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BOOST_USE_TO_STRING_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BOOST_USETOSTRINGCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp b/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp index e1856ff..baea231 100644 --- a/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp +++ b/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp @@ -22,6 +22,7 @@ #include "CommandProcessorCheck.h" #include "ComparePointerToMemberVirtualFunctionCheck.h" #include "CopyConstructorInitCheck.h" +#include "CopyConstructorMutatesArgumentCheck.h" #include "CrtpConstructorAccessibilityCheck.h" #include "DanglingHandleCheck.h" #include "DefaultOperatorNewOnOveralignedTypeCheck.h" @@ -141,6 +142,8 @@ public: "bugprone-compare-pointer-to-member-virtual-function"); CheckFactories.registerCheck<CopyConstructorInitCheck>( "bugprone-copy-constructor-init"); + CheckFactories.registerCheck<CopyConstructorMutatesArgumentCheck>( + "bugprone-copy-constructor-mutates-argument"); CheckFactories.registerCheck<DanglingHandleCheck>( "bugprone-dangling-handle"); CheckFactories.registerCheck<DefaultOperatorNewOnOveralignedTypeCheck>( diff --git a/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt b/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt index 7d2e108..aacaa61 100644 --- a/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt +++ b/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt @@ -18,6 +18,7 @@ add_clang_library(clangTidyBugproneModule STATIC CommandProcessorCheck.cpp ComparePointerToMemberVirtualFunctionCheck.cpp CopyConstructorInitCheck.cpp + CopyConstructorMutatesArgumentCheck.cpp CrtpConstructorAccessibilityCheck.cpp DanglingHandleCheck.cpp DefaultOperatorNewOnOveralignedTypeCheck.cpp diff --git a/clang-tools-extra/clang-tidy/bugprone/CopyConstructorInitCheck.h b/clang-tools-extra/clang-tidy/bugprone/CopyConstructorInitCheck.h index cba1a25..e977bc2 100644 --- a/clang-tools-extra/clang-tidy/bugprone/CopyConstructorInitCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/CopyConstructorInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPY_CONSTRUCTOR_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPY_CONSTRUCTOR_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPYCONSTRUCTORINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPYCONSTRUCTORINITCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPY_CONSTRUCTOR_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPYCONSTRUCTORINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/cert/MutatingCopyCheck.cpp b/clang-tools-extra/clang-tidy/bugprone/CopyConstructorMutatesArgumentCheck.cpp index fb9d72c..cbbb1a0 100644 --- a/clang-tools-extra/clang-tidy/cert/MutatingCopyCheck.cpp +++ b/clang-tools-extra/clang-tidy/bugprone/CopyConstructorMutatesArgumentCheck.cpp @@ -6,19 +6,20 @@ // //===----------------------------------------------------------------------===// -#include "MutatingCopyCheck.h" +#include "CopyConstructorMutatesArgumentCheck.h" #include "clang/AST/ASTContext.h" #include "clang/ASTMatchers/ASTMatchFinder.h" using namespace clang::ast_matchers; -namespace clang::tidy::cert { +namespace clang::tidy::bugprone { static constexpr llvm::StringLiteral SourceDeclName = "ChangedPVD"; static constexpr llvm::StringLiteral MutatingOperatorName = "MutatingOp"; static constexpr llvm::StringLiteral MutatingCallName = "MutatingCall"; -void MutatingCopyCheck::registerMatchers(MatchFinder *Finder) { +void CopyConstructorMutatesArgumentCheck::registerMatchers( + MatchFinder *Finder) { const auto MemberExprOrSourceObject = anyOf( memberExpr(), declRefExpr(to(decl(equalsBoundNode(std::string(SourceDeclName)))))); @@ -60,7 +61,8 @@ void MutatingCopyCheck::registerMatchers(MatchFinder *Finder) { this); } -void MutatingCopyCheck::check(const MatchFinder::MatchResult &Result) { +void CopyConstructorMutatesArgumentCheck::check( + const MatchFinder::MatchResult &Result) { if (const auto *MemberCall = Result.Nodes.getNodeAs<CXXMemberCallExpr>(MutatingCallName)) diag(MemberCall->getBeginLoc(), "call mutates copied object"); @@ -69,4 +71,4 @@ void MutatingCopyCheck::check(const MatchFinder::MatchResult &Result) { diag(Assignment->getBeginLoc(), "mutating copied object"); } -} // namespace clang::tidy::cert +} // namespace clang::tidy::bugprone diff --git a/clang-tools-extra/clang-tidy/cert/MutatingCopyCheck.h b/clang-tools-extra/clang-tidy/bugprone/CopyConstructorMutatesArgumentCheck.h index c211fa0..0fed572 100644 --- a/clang-tools-extra/clang-tidy/cert/MutatingCopyCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/CopyConstructorMutatesArgumentCheck.h @@ -6,21 +6,21 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_MUTATINGCOPYCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_MUTATINGCOPYCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPYCONSTRUCTORMUTATESARGUMENTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPYCONSTRUCTORMUTATESARGUMENTCHECK_H #include "../ClangTidyCheck.h" -namespace clang::tidy::cert { +namespace clang::tidy::bugprone { /// Finds assignments to the copied object and its direct or indirect members /// in copy constructors and copy assignment operators. /// /// For the user-facing documentation see: -/// https://clang.llvm.org/extra/clang-tidy/checks/cert/oop58-cpp.html -class MutatingCopyCheck : public ClangTidyCheck { +/// https://clang.llvm.org/extra/clang-tidy/checks/bugprone/copy-constructor-mutates-argument.html +class CopyConstructorMutatesArgumentCheck : public ClangTidyCheck { public: - MutatingCopyCheck(StringRef Name, ClangTidyContext *Context) + CopyConstructorMutatesArgumentCheck(StringRef Name, ClangTidyContext *Context) : ClangTidyCheck(Name, Context) {} bool isLanguageVersionSupported(const LangOptions &LangOpts) const override { return LangOpts.CPlusPlus; @@ -29,6 +29,6 @@ public: void check(const ast_matchers::MatchFinder::MatchResult &Result) override; }; -} // namespace clang::tidy::cert +} // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_MUTATINGCOPYCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_COPYCONSTRUCTORMUTATESARGUMENTCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/DanglingHandleCheck.h b/clang-tools-extra/clang-tidy/bugprone/DanglingHandleCheck.h index 486562c..0b71bc4 100644 --- a/clang-tools-extra/clang-tidy/bugprone/DanglingHandleCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/DanglingHandleCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DANGLING_HANDLE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DANGLING_HANDLE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DANGLINGHANDLECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DANGLINGHANDLECHECK_H #include "../ClangTidyCheck.h" @@ -37,4 +37,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DANGLING_HANDLE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DANGLINGHANDLECHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/DynamicStaticInitializersCheck.h b/clang-tools-extra/clang-tidy/bugprone/DynamicStaticInitializersCheck.h index e02c62a..00e4bb1 100644 --- a/clang-tools-extra/clang-tidy/bugprone/DynamicStaticInitializersCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/DynamicStaticInitializersCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DYNAMIC_STATIC_INITIALIZERS_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DYNAMIC_STATIC_INITIALIZERS_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DYNAMICSTATICINITIALIZERSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DYNAMICSTATICINITIALIZERSCHECK_H #include "../ClangTidyCheck.h" #include "../FileExtensionsSet.h" @@ -30,4 +30,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DYNAMIC_STATIC_INITIALIZERS_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_DYNAMICSTATICINITIALIZERSCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/ExceptionEscapeCheck.h b/clang-tools-extra/clang-tidy/bugprone/ExceptionEscapeCheck.h index bd1e7ba..31d9e85 100644 --- a/clang-tools-extra/clang-tidy/bugprone/ExceptionEscapeCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/ExceptionEscapeCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_EXCEPTION_ESCAPE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_EXCEPTION_ESCAPE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_EXCEPTIONESCAPECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_EXCEPTIONESCAPECHECK_H #include "../ClangTidyCheck.h" #include "../utils/ExceptionAnalyzer.h" @@ -42,4 +42,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_EXCEPTION_ESCAPE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_EXCEPTIONESCAPECHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/FoldInitTypeCheck.h b/clang-tools-extra/clang-tidy/bugprone/FoldInitTypeCheck.h index 119728d..ef8b4d1 100644 --- a/clang-tools-extra/clang-tidy/bugprone/FoldInitTypeCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/FoldInitTypeCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_FOLD_INIT_TYPE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_FOLD_INIT_TYPE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_FOLDINITTYPECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_FOLDINITTYPECHECK_H #include "../ClangTidyCheck.h" @@ -39,4 +39,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_FOLD_INIT_TYPE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_FOLDINITTYPECHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/IntegerDivisionCheck.h b/clang-tools-extra/clang-tidy/bugprone/IntegerDivisionCheck.h index 777e3186..acab7be 100644 --- a/clang-tools-extra/clang-tidy/bugprone/IntegerDivisionCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/IntegerDivisionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_INTEGER_DIVISION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_INTEGER_DIVISION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_INTEGERDIVISIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_INTEGERDIVISIONCHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_INTEGER_DIVISION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_INTEGERDIVISIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/MisplacedOperatorInStrlenInAllocCheck.h b/clang-tools-extra/clang-tidy/bugprone/MisplacedOperatorInStrlenInAllocCheck.h index f650145..c40aef3 100644 --- a/clang-tools-extra/clang-tidy/bugprone/MisplacedOperatorInStrlenInAllocCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/MisplacedOperatorInStrlenInAllocCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACED_OPERATOR_IN_STRLEN_IN_ALLOC_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACED_OPERATOR_IN_STRLEN_IN_ALLOC_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACEDOPERATORINSTRLENINALLOCCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACEDOPERATORINSTRLENINALLOCCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACED_OPERATOR_IN_STRLEN_IN_ALLOC_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACEDOPERATORINSTRLENINALLOCCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/MisplacedPointerArithmeticInAllocCheck.h b/clang-tools-extra/clang-tidy/bugprone/MisplacedPointerArithmeticInAllocCheck.h index e78c30cb..9f6504f 100644 --- a/clang-tools-extra/clang-tidy/bugprone/MisplacedPointerArithmeticInAllocCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/MisplacedPointerArithmeticInAllocCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACED_OPERATOR_IN_ALLOC_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACED_OPERATOR_IN_ALLOC_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACEDPOINTERARITHMETICINALLOCCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACEDPOINTERARITHMETICINALLOCCHECK_H #include "../ClangTidyCheck.h" @@ -29,4 +29,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACED_OPERATOR_IN_ALLOC_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MISPLACEDPOINTERARITHMETICINALLOCCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/MultipleStatementMacroCheck.h b/clang-tools-extra/clang-tidy/bugprone/MultipleStatementMacroCheck.h index 1a2d4a4..1c3679a 100644 --- a/clang-tools-extra/clang-tidy/bugprone/MultipleStatementMacroCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/MultipleStatementMacroCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MULTIPLE_STATEMENT_MACRO_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MULTIPLE_STATEMENT_MACRO_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MULTIPLESTATEMENTMACROCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MULTIPLESTATEMENTMACROCHECK_H #include "../ClangTidyCheck.h" @@ -29,4 +29,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MULTIPLE_STATEMENT_MACRO_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_MULTIPLESTATEMENTMACROCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/NarrowingConversionsCheck.h b/clang-tools-extra/clang-tidy/bugprone/NarrowingConversionsCheck.h index 9631c71..e506e5b 100644 --- a/clang-tools-extra/clang-tidy/bugprone/NarrowingConversionsCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/NarrowingConversionsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NARROWING_CONVERSIONS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NARROWING_CONVERSIONS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NARROWINGCONVERSIONSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NARROWINGCONVERSIONSCHECK_H #include "../ClangTidyCheck.h" @@ -108,4 +108,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NARROWING_CONVERSIONS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NARROWINGCONVERSIONSCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/NondeterministicPointerIterationOrderCheck.h b/clang-tools-extra/clang-tidy/bugprone/NondeterministicPointerIterationOrderCheck.h index 054d580..46b4e12 100644 --- a/clang-tools-extra/clang-tidy/bugprone/NondeterministicPointerIterationOrderCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/NondeterministicPointerIterationOrderCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NONDETERMINISTIC_POINTER_ITERATION_ORDER_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NONDETERMINISTIC_POINTER_ITERATION_ORDER_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NONDETERMINISTICPOINTERITERATIONORDERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NONDETERMINISTICPOINTERITERATIONORDERCHECK_H #include "../ClangTidyCheck.h" @@ -36,4 +36,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NONDETERMINISTIC_POINTER_ITERATION_ORDER_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NONDETERMINISTICPOINTERITERATIONORDERCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/NotNullTerminatedResultCheck.h b/clang-tools-extra/clang-tidy/bugprone/NotNullTerminatedResultCheck.h index a8f4ca3..cf61eb5 100644 --- a/clang-tools-extra/clang-tidy/bugprone/NotNullTerminatedResultCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/NotNullTerminatedResultCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NOT_NULL_TERMINATED_RESULT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NOT_NULL_TERMINATED_RESULT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NOTNULLTERMINATEDRESULTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NOTNULLTERMINATEDRESULTCHECK_H #include "../ClangTidyCheck.h" @@ -60,4 +60,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NOT_NULL_TERMINATED_RESULT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_NOTNULLTERMINATEDRESULTCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/PosixReturnCheck.h b/clang-tools-extra/clang-tidy/bugprone/PosixReturnCheck.h index d72c86c..a9cb7a6 100644 --- a/clang-tools-extra/clang-tidy/bugprone/PosixReturnCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/PosixReturnCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_POSIX_RETURN_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_POSIX_RETURN_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_POSIXRETURNCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_POSIXRETURNCHECK_H #include "../ClangTidyCheck.h" @@ -23,4 +23,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_POSIX_RETURN_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_POSIXRETURNCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/StringConstructorCheck.h b/clang-tools-extra/clang-tidy/bugprone/StringConstructorCheck.h index 0d7a203..9c08e4b 100644 --- a/clang-tools-extra/clang-tidy/bugprone/StringConstructorCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/StringConstructorCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_STRING_CONSTRUCTOR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_STRING_CONSTRUCTOR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_STRINGCONSTRUCTORCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_STRINGCONSTRUCTORCHECK_H #include "../ClangTidyCheck.h" @@ -36,4 +36,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_STRING_CONSTRUCTOR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_STRINGCONSTRUCTORCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/SuspiciousMemsetUsageCheck.h b/clang-tools-extra/clang-tidy/bugprone/SuspiciousMemsetUsageCheck.h index a1f5f2b..c45f332 100644 --- a/clang-tools-extra/clang-tidy/bugprone/SuspiciousMemsetUsageCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/SuspiciousMemsetUsageCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_SUSPICIOUS_MEMSET_USAGE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_SUSPICIOUS_MEMSET_USAGE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_SUSPICIOUSMEMSETUSAGECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_SUSPICIOUSMEMSETUSAGECHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_SUSPICIOUS_MEMSET_USAGE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_SUSPICIOUSMEMSETUSAGECHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/UndefinedMemoryManipulationCheck.h b/clang-tools-extra/clang-tidy/bugprone/UndefinedMemoryManipulationCheck.h index c9a232a..409122f 100644 --- a/clang-tools-extra/clang-tidy/bugprone/UndefinedMemoryManipulationCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/UndefinedMemoryManipulationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDEFINED_MEMORY_MANIPULATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDEFINED_MEMORY_MANIPULATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDEFINEDMEMORYMANIPULATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDEFINEDMEMORYMANIPULATIONCHECK_H #include "../ClangTidyCheck.h" @@ -29,4 +29,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDEFINED_MEMORY_MANIPULATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDEFINEDMEMORYMANIPULATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/UndelegatedConstructorCheck.h b/clang-tools-extra/clang-tidy/bugprone/UndelegatedConstructorCheck.h index 18465f7..c7cadbf6 100644 --- a/clang-tools-extra/clang-tidy/bugprone/UndelegatedConstructorCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/UndelegatedConstructorCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNDELEGATEDCONSTRUCTOR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNDELEGATEDCONSTRUCTOR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDELEGATEDCONSTRUCTORCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDELEGATEDCONSTRUCTORCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNDELEGATEDCONSTRUCTOR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_UNDELEGATEDCONSTRUCTORCHECK_H diff --git a/clang-tools-extra/clang-tidy/bugprone/VirtualNearMissCheck.h b/clang-tools-extra/clang-tidy/bugprone/VirtualNearMissCheck.h index 71d302f4..2278817 100644 --- a/clang-tools-extra/clang-tidy/bugprone/VirtualNearMissCheck.h +++ b/clang-tools-extra/clang-tidy/bugprone/VirtualNearMissCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_VIRTUAL_NEAR_MISS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_VIRTUAL_NEAR_MISS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_VIRTUALNEARMISSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_VIRTUALNEARMISSCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/DenseMap.h" @@ -60,4 +60,4 @@ private: } // namespace clang::tidy::bugprone -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_VIRTUAL_NEAR_MISS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_VIRTUALNEARMISSCHECK_H diff --git a/clang-tools-extra/clang-tidy/cert/CERTTidyModule.cpp b/clang-tools-extra/clang-tidy/cert/CERTTidyModule.cpp index a1f6270..b0799da 100644 --- a/clang-tools-extra/clang-tidy/cert/CERTTidyModule.cpp +++ b/clang-tools-extra/clang-tidy/cert/CERTTidyModule.cpp @@ -11,6 +11,7 @@ #include "../ClangTidyModuleRegistry.h" #include "../bugprone/BadSignalToKillThreadCheck.h" #include "../bugprone/CommandProcessorCheck.h" +#include "../bugprone/CopyConstructorMutatesArgumentCheck.h" #include "../bugprone/DefaultOperatorNewOnOveralignedTypeCheck.h" #include "../bugprone/FloatLoopCounterCheck.h" #include "../bugprone/PointerArithmeticOnPolymorphicObjectCheck.h" @@ -39,7 +40,6 @@ #include "../readability/EnumInitialValueCheck.h" #include "../readability/UppercaseLiteralSuffixCheck.h" #include "LimitedRandomnessCheck.h" -#include "MutatingCopyCheck.h" #include "ProperlySeededRandomGeneratorCheck.h" #include "ThrownExceptionTypeCheck.h" @@ -282,7 +282,8 @@ public: "cert-oop54-cpp"); CheckFactories.registerCheck<bugprone::RawMemoryCallOnNonTrivialTypeCheck>( "cert-oop57-cpp"); - CheckFactories.registerCheck<MutatingCopyCheck>("cert-oop58-cpp"); + CheckFactories.registerCheck<bugprone::CopyConstructorMutatesArgumentCheck>( + "cert-oop58-cpp"); // C checkers // ARR diff --git a/clang-tools-extra/clang-tidy/cert/CMakeLists.txt b/clang-tools-extra/clang-tidy/cert/CMakeLists.txt index b25576a3..5abb472 100644 --- a/clang-tools-extra/clang-tidy/cert/CMakeLists.txt +++ b/clang-tools-extra/clang-tidy/cert/CMakeLists.txt @@ -6,7 +6,6 @@ set(LLVM_LINK_COMPONENTS add_clang_library(clangTidyCERTModule STATIC CERTTidyModule.cpp LimitedRandomnessCheck.cpp - MutatingCopyCheck.cpp ProperlySeededRandomGeneratorCheck.cpp ThrownExceptionTypeCheck.cpp diff --git a/clang-tools-extra/clang-tidy/cert/LimitedRandomnessCheck.h b/clang-tools-extra/clang-tidy/cert/LimitedRandomnessCheck.h index a9d6076..a806cd3 100644 --- a/clang-tools-extra/clang-tidy/cert/LimitedRandomnessCheck.h +++ b/clang-tools-extra/clang-tidy/cert/LimitedRandomnessCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_LIMITED_RANDOMNESS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_LIMITED_RANDOMNESS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_LIMITEDRANDOMNESSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_LIMITEDRANDOMNESSCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::cert -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_LIMITED_RANDOMNESS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_LIMITEDRANDOMNESSCHECK_H diff --git a/clang-tools-extra/clang-tidy/cert/ProperlySeededRandomGeneratorCheck.h b/clang-tools-extra/clang-tidy/cert/ProperlySeededRandomGeneratorCheck.h index 7da01cc..8cb2e62 100644 --- a/clang-tools-extra/clang-tidy/cert/ProperlySeededRandomGeneratorCheck.h +++ b/clang-tools-extra/clang-tidy/cert/ProperlySeededRandomGeneratorCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_PROPERLY_SEEDED_RANDOM_GENERATOR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_PROPERLY_SEEDED_RANDOM_GENERATOR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_PROPERLYSEEDEDRANDOMGENERATORCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_PROPERLYSEEDEDRANDOMGENERATORCHECK_H #include "../ClangTidyCheck.h" #include <string> @@ -39,4 +39,4 @@ private: } // namespace clang::tidy::cert -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_PROPERLY_SEEDED_RANDOM_GENERATOR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CERT_PROPERLYSEEDEDRANDOMGENERATORCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/InitVariablesCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/InitVariablesCheck.cpp index 2545548..93b5b96 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/InitVariablesCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/InitVariablesCheck.cpp @@ -37,7 +37,7 @@ void InitVariablesCheck::storeOptions(ClangTidyOptions::OptionMap &Opts) { } void InitVariablesCheck::registerMatchers(MatchFinder *Finder) { - std::string BadDecl = "badDecl"; + const std::string BadDecl = "badDecl"; Finder->addMatcher( varDecl(unless(hasInitializer(anything())), unless(isInstantiated()), isLocalVarDecl(), unless(isStaticLocal()), isDefinition(), @@ -82,7 +82,7 @@ void InitVariablesCheck::check(const MatchFinder::MatchResult &Result) { if (MatchedDecl->getEndLoc().isMacroID()) return; - QualType TypePtr = MatchedDecl->getType(); + const QualType TypePtr = MatchedDecl->getType(); std::optional<const char *> InitializationString; bool AddMathInclude = false; diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/InterfacesGlobalInitCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/InterfacesGlobalInitCheck.h index 780b4b3..dc91854 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/InterfacesGlobalInitCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/InterfacesGlobalInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_INTERFACES_GLOBAL_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_INTERFACES_GLOBAL_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_INTERFACESGLOBALINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_INTERFACESGLOBALINITCHECK_H #include "../ClangTidyCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_INTERFACES_GLOBAL_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_INTERFACESGLOBALINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/MacroUsageCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/MacroUsageCheck.cpp index 0836a5c..b301a2b 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/MacroUsageCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/MacroUsageCheck.cpp @@ -47,7 +47,7 @@ public: SM.isWrittenInCommandLineFile(MD->getLocation())) return; - StringRef MacroName = MacroNameTok.getIdentifierInfo()->getName(); + const StringRef MacroName = MacroNameTok.getIdentifierInfo()->getName(); if (MacroName == "__GCC_HAVE_DWARF2_CFI_ASM") return; if (!CheckCapsOnly && !RegExp.match(MacroName)) diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/MisleadingCaptureDefaultByValueCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/MisleadingCaptureDefaultByValueCheck.cpp index 57d98ee..366bd12 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/MisleadingCaptureDefaultByValueCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/MisleadingCaptureDefaultByValueCheck.cpp @@ -81,7 +81,7 @@ void MisleadingCaptureDefaultByValueCheck::check( return; if (Lambda->getCaptureDefault() == LCD_ByCopy) { - bool IsThisImplicitlyCaptured = std::any_of( + const bool IsThisImplicitlyCaptured = std::any_of( Lambda->implicit_capture_begin(), Lambda->implicit_capture_end(), [](const LambdaCapture &Capture) { return Capture.capturesThis(); }); auto Diag = diag(Lambda->getCaptureDefaultLoc(), @@ -89,8 +89,8 @@ void MisleadingCaptureDefaultByValueCheck::check( "should not specify a by-value capture default") << IsThisImplicitlyCaptured; - std::string ReplacementText = createReplacementText(Lambda); - SourceLocation DefaultCaptureEnd = + const std::string ReplacementText = createReplacementText(Lambda); + const SourceLocation DefaultCaptureEnd = findDefaultCaptureEnd(Lambda, *Result.Context); Diag << FixItHint::CreateReplacement( CharSourceRange::getCharRange(Lambda->getCaptureDefaultLoc(), diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/MissingStdForwardCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/MissingStdForwardCheck.cpp index 090ab2f..d1d81d5 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/MissingStdForwardCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/MissingStdForwardCheck.cpp @@ -26,11 +26,12 @@ AST_MATCHER_P(QualType, possiblyPackExpansionOf, } AST_MATCHER(ParmVarDecl, isTemplateTypeParameter) { - ast_matchers::internal::Matcher<QualType> Inner = possiblyPackExpansionOf( - qualType(rValueReferenceType(), - references(templateTypeParmType( - hasDeclaration(templateTypeParmDecl()))), - unless(references(qualType(isConstQualified()))))); + const ast_matchers::internal::Matcher<QualType> Inner = + possiblyPackExpansionOf( + qualType(rValueReferenceType(), + references(templateTypeParmType( + hasDeclaration(templateTypeParmDecl()))), + unless(references(qualType(isConstQualified()))))); if (!Inner.matches(Node.getType(), Finder, Builder)) return false; @@ -43,7 +44,7 @@ AST_MATCHER(ParmVarDecl, isTemplateTypeParameter) { if (!FuncTemplate) return false; - QualType ParamType = + const QualType ParamType = Node.getType().getNonPackExpansionType()->getPointeeType(); const auto *TemplateType = ParamType->getAsCanonical<TemplateTypeParmType>(); if (!TemplateType) @@ -54,10 +55,10 @@ AST_MATCHER(ParmVarDecl, isTemplateTypeParameter) { } AST_MATCHER_P(NamedDecl, hasSameNameAsBoundNode, std::string, BindingID) { - IdentifierInfo *II = Node.getIdentifier(); + const IdentifierInfo *II = Node.getIdentifier(); if (nullptr == II) return false; - StringRef Name = II->getName(); + const StringRef Name = II->getName(); return Builder->removeBindings( [this, Name](const ast_matchers::internal::BoundNodesMap &Nodes) { diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/NoMallocCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/NoMallocCheck.h index e4dece6..da35b53 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/NoMallocCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/NoMallocCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_NO_MALLOC_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_NO_MALLOC_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_NOMALLOCCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_NOMALLOCCHECK_H #include "../ClangTidyCheck.h" @@ -56,4 +56,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_NO_MALLOC_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_NOMALLOCCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/NoSuspendWithLockCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/NoSuspendWithLockCheck.cpp index 43df277..8ecbccd 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/NoSuspendWithLockCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/NoSuspendWithLockCheck.cpp @@ -27,7 +27,7 @@ void NoSuspendWithLockCheck::registerMatchers(MatchFinder *Finder) { hasDeclaration(namedDecl(matchers::matchesAnyListedName( utils::options::parseStringList(LockGuards))))); - StatementMatcher Lock = + const StatementMatcher Lock = declStmt(has(varDecl(hasType(LockType)).bind("lock-decl"))) .bind("lock-decl-stmt"); Finder->addMatcher( @@ -55,12 +55,12 @@ void NoSuspendWithLockCheck::check(const MatchFinder::MatchResult &Result) { Options.AddImplicitDtors = true; Options.AddTemporaryDtors = true; - std::unique_ptr<CFG> TheCFG = CFG::buildCFG( + const std::unique_ptr<CFG> TheCFG = CFG::buildCFG( nullptr, const_cast<clang::CompoundStmt *>(Block), &Context, Options); if (!TheCFG) return; - utils::ExprSequence Sequence(TheCFG.get(), Block, &Context); + const utils::ExprSequence Sequence(TheCFG.get(), Block, &Context); const Stmt *LastBlockStmt = Block->body_back(); if (Sequence.inSequence(LockStmt, Suspend) && (Suspend == LastBlockStmt || diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/OwningMemoryCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/OwningMemoryCheck.h index 462e986..248b5c2 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/OwningMemoryCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/OwningMemoryCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_OWNING_MEMORY_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_OWNING_MEMORY_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_OWNINGMEMORYCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_OWNINGMEMORYCHECK_H #include "../ClangTidyCheck.h" @@ -61,4 +61,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_OWNING_MEMORY_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_OWNINGMEMORYCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/PreferMemberInitializerCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/PreferMemberInitializerCheck.cpp index 9913671..51a1468 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/PreferMemberInitializerCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/PreferMemberInitializerCheck.cpp @@ -164,12 +164,12 @@ void PreferMemberInitializerCheck::check( llvm::DenseMap<const FieldDecl *, AssignedLevel> AssignedFields{}; for (const CXXCtorInitializer *Init : Ctor->inits()) - if (FieldDecl *Field = Init->getMember()) + if (const FieldDecl *Field = Init->getMember()) updateAssignmentLevel(Field, Init->getInit(), Ctor, AssignedFields); for (const Stmt *S : Body->body()) { if (S->getBeginLoc().isMacroID()) { - StringRef MacroName = Lexer::getImmediateMacroName( + const StringRef MacroName = Lexer::getImmediateMacroName( S->getBeginLoc(), *Result.SourceManager, getLangOpts()); if (MacroName.contains_insensitive("assert")) return; @@ -206,7 +206,7 @@ void PreferMemberInitializerCheck::check( bool AddComma = false; bool AddBrace = false; bool InvalidFix = false; - unsigned Index = Field->getFieldIndex(); + const unsigned Index = Field->getFieldIndex(); const CXXCtorInitializer *LastInListInit = nullptr; for (const CXXCtorInitializer *Init : Ctor->inits()) { if (!Init->isWritten() || Init->isInClassMemberInitializer()) @@ -276,7 +276,7 @@ void PreferMemberInitializerCheck::check( << Field; if (InvalidFix) continue; - StringRef NewInit = Lexer::getSourceText( + const StringRef NewInit = Lexer::getSourceText( Result.SourceManager->getExpansionRange(InitValue->getSourceRange()), *Result.SourceManager, getLangOpts()); if (HasInitAlready) { @@ -288,8 +288,8 @@ void PreferMemberInitializerCheck::check( else Diag << FixItHint::CreateReplacement(ReplaceRange, NewInit); } else { - SmallString<128> Insertion({InsertPrefix, Field->getName(), "(", NewInit, - AddComma ? "), " : ")"}); + const SmallString<128> Insertion({InsertPrefix, Field->getName(), "(", + NewInit, AddComma ? "), " : ")"}); Diag << FixItHint::CreateInsertion(InsertPos, Insertion, FirstToCtorInits); FirstToCtorInits = areDiagsSelfContained(); diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.cpp index f3237f4..d0f8652 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.cpp @@ -35,7 +35,7 @@ AST_MATCHER_P(Expr, hasParentIgnoringImpCasts, ast_matchers::internal::Matcher<Expr>, InnerMatcher) { const Expr *E = &Node; do { - DynTypedNodeList Parents = Finder->getASTContext().getParents(*E); + const DynTypedNodeList Parents = Finder->getASTContext().getParents(*E); if (Parents.size() != 1) return false; E = Parents[0].get<Expr>(); diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.h index cea4bfa..2d4b40b 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsArrayToPointerDecayCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_ARRAY_TO_POINTER_DECAY_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_ARRAY_TO_POINTER_DECAY_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSARRAYTOPOINTERDECAYCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSARRAYTOPOINTERDECAYCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_ARRAY_TO_POINTER_DECAY_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSARRAYTOPOINTERDECAYCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsAvoidUncheckedContainerAccess.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsAvoidUncheckedContainerAccess.h index 0755da7..2462e7a 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsAvoidUncheckedContainerAccess.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsAvoidUncheckedContainerAccess.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_AVOID_UNCHECKED_CONTAINER_ACCESS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_AVOID_UNCHECKED_CONTAINER_ACCESS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSAVOIDUNCHECKEDCONTAINERACCESS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSAVOIDUNCHECKEDCONTAINERACCESS_H #include "../ClangTidyCheck.h" @@ -53,4 +53,4 @@ struct OptionEnumMapping< getEnumMapping(); }; } // namespace clang::tidy -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_AVOID_UNCHECKED_CONTAINER_ACCESS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSAVOIDUNCHECKEDCONTAINERACCESS_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.cpp index 634ec18..82fc9f2 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.cpp @@ -78,7 +78,7 @@ void ProBoundsConstantArrayIndexCheck::check( else BaseRange = cast<CXXOperatorCallExpr>(Matched)->getArg(0)->getSourceRange(); - SourceRange IndexRange = IndexExpr->getSourceRange(); + const SourceRange IndexRange = IndexExpr->getSourceRange(); auto Diag = diag(Matched->getExprLoc(), "do not use array subscript when the index is " @@ -115,7 +115,7 @@ void ProBoundsConstantArrayIndexCheck::check( const auto &SizeArg = TemplateArgs[1]; if (SizeArg.getKind() != TemplateArgument::Integral) return; - llvm::APInt ArraySize = SizeArg.getAsIntegral(); + const llvm::APInt ArraySize = SizeArg.getAsIntegral(); // Get uint64_t values, because different bitwidths would lead to an assertion // in APInt::uge. diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.h index 73f1855..7c8fec0 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsConstantArrayIndexCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_CONSTANT_ARRAY_INDEX_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_CONSTANT_ARRAY_INDEX_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSCONSTANTARRAYINDEXCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSCONSTANTARRAYINDEXCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -37,4 +37,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_CONSTANT_ARRAY_INDEX_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSCONSTANTARRAYINDEXCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsPointerArithmeticCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsPointerArithmeticCheck.h index 45b7985..4f6b17f 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsPointerArithmeticCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProBoundsPointerArithmeticCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_POINTER_ARITHMETIC_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_POINTER_ARITHMETIC_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSPOINTERARITHMETICCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSPOINTERARITHMETICCHECK_H #include "../ClangTidyCheck.h" @@ -35,4 +35,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_BOUNDS_POINTER_ARITHMETIC_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROBOUNDSPOINTERARITHMETICCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeConstCastCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeConstCastCheck.h index 0b8cfc8..a0a368c 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeConstCastCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeConstCastCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_CONST_CAST_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_CONST_CAST_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPECONSTCASTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPECONSTCASTCHECK_H #include "../ClangTidyCheck.h" @@ -36,4 +36,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_CONST_CAST_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPECONSTCASTCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.cpp index b9867c2..fcd9c6d 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.cpp @@ -45,7 +45,7 @@ void ProTypeCstyleCastCheck::check(const MatchFinder::MatchResult &Result) { return; } - QualType SourceType = MatchedCast->getSubExpr()->getType(); + const QualType SourceType = MatchedCast->getSubExpr()->getType(); if (MatchedCast->getCastKind() == CK_BaseToDerived) { const auto *SourceDecl = SourceType->getPointeeCXXRecordDecl(); @@ -58,7 +58,7 @@ void ProTypeCstyleCastCheck::check(const MatchFinder::MatchResult &Result) { // Leave type spelling exactly as it was (unlike // getTypeAsWritten().getAsString() which would spell enum types 'enum // X'). - StringRef DestTypeString = Lexer::getSourceText( + const StringRef DestTypeString = Lexer::getSourceText( CharSourceRange::getTokenRange( MatchedCast->getLParenLoc().getLocWithOffset(1), MatchedCast->getRParenLoc().getLocWithOffset(-1)), diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.h index f8e1d5a..5fd0208 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeCstyleCastCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_CSTYLE_CAST_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_CSTYLE_CAST_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPECSTYLECASTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPECSTYLECASTCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_CSTYLE_CAST_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPECSTYLECASTCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.cpp index 1ac9b8b..111c621 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.cpp @@ -173,7 +173,7 @@ struct InitializerInsertion { assert(!Initializers.empty() && "No initializers to insert"); std::string Code; llvm::raw_string_ostream Stream(Code); - std::string Joined = + const std::string Joined = llvm::join(Initializers.begin(), Initializers.end(), "(), "); switch (Placement) { case InitializerPlacement::New: @@ -434,7 +434,7 @@ static llvm::StringLiteral getInitializer(QualType QT, bool UseAssignment) { void ProTypeMemberInitCheck::checkMissingMemberInitializer( ASTContext &Context, const CXXRecordDecl &ClassDecl, const CXXConstructorDecl *Ctor) { - bool IsUnion = ClassDecl.isUnion(); + const bool IsUnion = ClassDecl.isUnion(); if (IsUnion && ClassDecl.hasInClassInitializer()) return; @@ -583,7 +583,7 @@ void ProTypeMemberInitCheck::checkMissingBaseClassInitializer( void ProTypeMemberInitCheck::checkUninitializedTrivialType( const ASTContext &Context, const VarDecl *Var) { - DiagnosticBuilder Diag = + const DiagnosticBuilder Diag = diag(Var->getBeginLoc(), "uninitialized record type: %0") << Var; Diag << FixItHint::CreateInsertion( diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.h index 8beaab3..89d3074 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeMemberInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_MEMBER_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_MEMBER_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEMEMBERINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEMEMBERINITCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/DenseSet.h" @@ -79,4 +79,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_MEMBER_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEMEMBERINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeReinterpretCastCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeReinterpretCastCheck.h index 4948d0a..566944d 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeReinterpretCastCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeReinterpretCastCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_REINTERPRETCAST_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_REINTERPRETCAST_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEREINTERPRETCASTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEREINTERPRETCASTCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_REINTERPRETCAST_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEREINTERPRETCASTCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.cpp index c200a79..e7b92fc 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.cpp @@ -33,7 +33,7 @@ void ProTypeStaticCastDowncastCheck::check( const MatchFinder::MatchResult &Result) { const auto *MatchedCast = Result.Nodes.getNodeAs<CXXStaticCastExpr>("cast"); - QualType SourceType = MatchedCast->getSubExpr()->getType(); + const QualType SourceType = MatchedCast->getSubExpr()->getType(); const auto *SourceDecl = SourceType->getPointeeCXXRecordDecl(); if (!SourceDecl) // The cast is from object to reference SourceDecl = SourceType->getAsCXXRecordDecl(); diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.h index 3d01fb9..02d54a5 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeStaticCastDowncastCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_STATIC_CAST_DOWNCAST_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_STATIC_CAST_DOWNCAST_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPESTATICCASTDOWNCASTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPESTATICCASTDOWNCASTCHECK_H #include "../ClangTidyCheck.h" @@ -37,4 +37,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_STATIC_CAST_DOWNCAST_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPESTATICCASTDOWNCASTCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeUnionAccessCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeUnionAccessCheck.h index fe82ce96..41154e8 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeUnionAccessCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeUnionAccessCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_UNION_ACCESS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_UNION_ACCESS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEUNIONACCESSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEUNIONACCESSCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_UNION_ACCESS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEUNIONACCESSCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.cpp index 431b2a7..c223ed1 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.cpp @@ -65,9 +65,9 @@ static constexpr StringRef VaArgWarningMessage = namespace { AST_MATCHER(QualType, isVAList) { - ASTContext &Context = Finder->getASTContext(); - QualType Desugar = Node.getDesugaredType(Context); - QualType NodeTy = Node.getUnqualifiedType(); + const ASTContext &Context = Finder->getASTContext(); + const QualType Desugar = Node.getDesugaredType(Context); + const QualType NodeTy = Node.getUnqualifiedType(); auto CheckVaList = [](QualType NodeTy, QualType Expected, const ASTContext &Context) { @@ -88,7 +88,8 @@ AST_MATCHER(QualType, isVAList) { // type. Some targets implements va_list as 'char *' or 'void *'. // In these cases we need to remove all typedefs one by one to check this. using BuiltinVaListKind = TargetInfo::BuiltinVaListKind; - BuiltinVaListKind VaListKind = Context.getTargetInfo().getBuiltinVaListKind(); + const BuiltinVaListKind VaListKind = + Context.getTargetInfo().getBuiltinVaListKind(); if (VaListKind == BuiltinVaListKind::CharPtrBuiltinVaList || VaListKind == BuiltinVaListKind::VoidPtrBuiltinVaList) { if (CheckVaList(NodeTy, Context.getBuiltinVaListType(), Context)) diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.h index b28d365..5be6163 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/ProTypeVarargCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_VARARG_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_VARARG_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEVARARGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEVARARGCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ public: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PRO_TYPE_VARARG_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_PROTYPEVARARGCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/RvalueReferenceParamNotMovedCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/RvalueReferenceParamNotMovedCheck.cpp index c40ac7a..28bfe57 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/RvalueReferenceParamNotMovedCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/RvalueReferenceParamNotMovedCheck.cpp @@ -39,7 +39,7 @@ AST_MATCHER_P2(Stmt, argumentOf, bool, AllowPartialMove, StatementMatcher, void RvalueReferenceParamNotMovedCheck::registerMatchers(MatchFinder *Finder) { auto ToParam = hasAnyParameter(parmVarDecl(equalsBoundNode("param"))); - StatementMatcher MoveCallMatcher = + const StatementMatcher MoveCallMatcher = callExpr( argumentCountIs(1), anyOf(callee(functionDecl(hasName(MoveFunction))), diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/SlicingCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/SlicingCheck.h index 61990e6..520a763 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/SlicingCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/SlicingCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SLICING_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SLICING_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SLICINGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SLICINGCHECK_H #include "../ClangTidyCheck.h" @@ -36,4 +36,4 @@ private: } // namespace clang::tidy::cppcoreguidelines -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SLICING_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SLICINGCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.cpp index b38a0c6..77a7b2b 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.cpp @@ -109,7 +109,7 @@ join(ArrayRef<SpecialMemberFunctionsCheck::SpecialMemberFunctionKind> SMFS, llvm::raw_string_ostream Stream(Buffer); Stream << toString(SMFS[0]); - size_t LastIndex = SMFS.size() - 1; + const size_t LastIndex = SMFS.size() - 1; for (size_t I = 1; I < LastIndex; ++I) { Stream << ", " << toString(SMFS[I]); } @@ -146,7 +146,7 @@ void SpecialMemberFunctionsCheck::check( StoreMember({DestructorType, Dtor->isDeleted()}); } - std::initializer_list<std::pair<std::string, SpecialMemberFunctionKind>> + const std::initializer_list<std::pair<std::string, SpecialMemberFunctionKind>> Matchers = {{"copy-ctor", SpecialMemberFunctionKind::CopyConstructor}, {"copy-assign", SpecialMemberFunctionKind::CopyAssignment}, {"move-ctor", SpecialMemberFunctionKind::MoveConstructor}, @@ -202,7 +202,7 @@ void SpecialMemberFunctionsCheck::checkForMissingMembers( MissingMembers.push_back(Kind2); }; - bool RequireThree = + const bool RequireThree = HasMember(SpecialMemberFunctionKind::NonDefaultDestructor) || (!AllowSoleDefaultDtor && (HasMember(SpecialMemberFunctionKind::Destructor) || @@ -212,10 +212,11 @@ void SpecialMemberFunctionsCheck::checkForMissingMembers( HasMember(SpecialMemberFunctionKind::MoveConstructor) || HasMember(SpecialMemberFunctionKind::MoveAssignment); - bool RequireFive = (!AllowMissingMoveFunctions && RequireThree && - getLangOpts().CPlusPlus11) || - HasMember(SpecialMemberFunctionKind::MoveConstructor) || - HasMember(SpecialMemberFunctionKind::MoveAssignment); + const bool RequireFive = + (!AllowMissingMoveFunctions && RequireThree && + getLangOpts().CPlusPlus11) || + HasMember(SpecialMemberFunctionKind::MoveConstructor) || + HasMember(SpecialMemberFunctionKind::MoveAssignment); if (RequireThree) { if (!HasMember(SpecialMemberFunctionKind::Destructor) && diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.h b/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.h index 507aaa1..1d41b57 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.h +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/SpecialMemberFunctionsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SPECIAL_MEMBER_FUNCTIONS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SPECIAL_MEMBER_FUNCTIONS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SPECIALMEMBERFUNCTIONSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SPECIALMEMBERFUNCTIONSCHECK_H #include "../ClangTidyCheck.h" @@ -97,7 +97,7 @@ struct DenseMapInfo< assert(Val != getEmptyKey() && "Cannot hash the empty key!"); assert(Val != getTombstoneKey() && "Cannot hash the tombstone key!"); - std::hash<ClassDefId::second_type> SecondHash; + const std::hash<ClassDefId::second_type> SecondHash; return Val.first.getHashValue() + SecondHash(Val.second); } @@ -112,4 +112,4 @@ struct DenseMapInfo< } // namespace llvm -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SPECIAL_MEMBER_FUNCTIONS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_CPPCOREGUIDELINES_SPECIALMEMBERFUNCTIONSCHECK_H diff --git a/clang-tools-extra/clang-tidy/cppcoreguidelines/VirtualClassDestructorCheck.cpp b/clang-tools-extra/clang-tidy/cppcoreguidelines/VirtualClassDestructorCheck.cpp index 7700889..89f1bf1 100644 --- a/clang-tools-extra/clang-tidy/cppcoreguidelines/VirtualClassDestructorCheck.cpp +++ b/clang-tools-extra/clang-tidy/cppcoreguidelines/VirtualClassDestructorCheck.cpp @@ -56,10 +56,10 @@ getVirtualKeywordRange(const CXXDestructorDecl &Destructor, if (Destructor.getLocation().isMacroID()) return std::nullopt; - SourceLocation VirtualBeginLoc = Destructor.getBeginLoc(); - SourceLocation VirtualBeginSpellingLoc = + const SourceLocation VirtualBeginLoc = Destructor.getBeginLoc(); + const SourceLocation VirtualBeginSpellingLoc = SM.getSpellingLoc(Destructor.getBeginLoc()); - SourceLocation VirtualEndLoc = VirtualBeginSpellingLoc.getLocWithOffset( + const SourceLocation VirtualEndLoc = VirtualBeginSpellingLoc.getLocWithOffset( Lexer::MeasureTokenLength(VirtualBeginSpellingLoc, SM, LangOpts)); /// Range ends with \c StartOfNextToken so that any whitespace after \c @@ -68,7 +68,7 @@ getVirtualKeywordRange(const CXXDestructorDecl &Destructor, Lexer::findNextToken(VirtualEndLoc, SM, LangOpts); if (!NextToken) return std::nullopt; - SourceLocation StartOfNextToken = NextToken->getLocation(); + const SourceLocation StartOfNextToken = NextToken->getLocation(); return CharSourceRange::getCharRange(VirtualBeginLoc, StartOfNextToken); } @@ -79,7 +79,7 @@ getPublicASDecl(const CXXRecordDecl &StructOrClass) { AS{StructOrClass.decls_begin()}, ASEnd{StructOrClass.decls_end()}; AS != ASEnd; ++AS) { - AccessSpecDecl *ASDecl = *AS; + const AccessSpecDecl *ASDecl = *AS; if (ASDecl->getAccess() == AccessSpecifier::AS_public) return ASDecl; } @@ -125,7 +125,7 @@ static std::string getSourceText(const CXXDestructorDecl &Destructor) { static std::string eraseKeyword(std::string &DestructorString, const std::string &Keyword) { - size_t KeywordIndex = DestructorString.find(Keyword); + const size_t KeywordIndex = DestructorString.find(Keyword); if (KeywordIndex != std::string::npos) DestructorString.erase(KeywordIndex, Keyword.length()); return DestructorString; diff --git a/clang-tools-extra/clang-tidy/custom/QueryCheck.cpp b/clang-tools-extra/clang-tidy/custom/QueryCheck.cpp index f83c138f..315ce58 100644 --- a/clang-tools-extra/clang-tidy/custom/QueryCheck.cpp +++ b/clang-tools-extra/clang-tidy/custom/QueryCheck.cpp @@ -33,7 +33,7 @@ parseQuery(const ClangTidyOptions::CustomCheckValue &V, clang::query::QuerySession QS({}); llvm::StringRef QueryStringRef{V.Query}; while (!QueryStringRef.empty()) { - query::QueryRef Q = query::QueryParser::parse(QueryStringRef, QS); + const query::QueryRef Q = query::QueryParser::parse(QueryStringRef, QS); switch (Q->Kind) { case query::QK_Match: { const auto &MatchQuery = llvm::cast<query::MatchQuery>(*Q); @@ -126,11 +126,11 @@ void QueryCheck::registerMatchers(MatchFinder *Finder) { void QueryCheck::check(const MatchFinder::MatchResult &Result) { auto Emit = [this](const DiagMaps &DiagMaps, const std::string &BindName, const DynTypedNode &Node, DiagnosticIDs::Level Level) { - DiagMaps::const_iterator DiagMapIt = DiagMaps.find(Level); + const DiagMaps::const_iterator DiagMapIt = DiagMaps.find(Level); if (DiagMapIt == DiagMaps.end()) return; const BindNameMapToDiagMessage &BindNameMap = DiagMapIt->second; - BindNameMapToDiagMessage::const_iterator BindNameMapIt = + const BindNameMapToDiagMessage::const_iterator BindNameMapIt = BindNameMap.find(BindName); if (BindNameMapIt == BindNameMap.end()) return; diff --git a/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsCallsCheck.h b/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsCallsCheck.h index 51bb153..ee08b76 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsCallsCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsCallsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULT_ARGUMENTS_CALLS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULT_ARGUMENTS_CALLS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULTARGUMENTSCALLSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULTARGUMENTSCALLSCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULT_ARGUMENTS_CALLS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULTARGUMENTSCALLSCHECK_H diff --git a/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.cpp b/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.cpp index d80511e..b22aff1 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.cpp +++ b/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.cpp @@ -24,7 +24,7 @@ void DefaultArgumentsDeclarationsCheck::check( if (!D) return; - SourceRange DefaultArgRange = D->getDefaultArgRange(); + const SourceRange DefaultArgRange = D->getDefaultArgRange(); if (DefaultArgRange.getEnd() != D->getEndLoc()) return; @@ -35,10 +35,10 @@ void DefaultArgumentsDeclarationsCheck::check( return; } - SourceLocation StartLocation = + const SourceLocation StartLocation = D->getName().empty() ? D->getBeginLoc() : D->getLocation(); - SourceRange RemovalRange( + const SourceRange RemovalRange( Lexer::getLocForEndOfToken(StartLocation, 0, *Result.SourceManager, Result.Context->getLangOpts()), DefaultArgRange.getEnd()); diff --git a/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.h b/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.h index 1b0e3dd..aa991f8 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/DefaultArgumentsDeclarationsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULT_ARGUMENTS_DECLARATIONS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULT_ARGUMENTS_DECLARATIONS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULTARGUMENTSDECLARATIONSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULTARGUMENTSDECLARATIONSCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULT_ARGUMENTS_DECLARATIONS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_DEFAULTARGUMENTSDECLARATIONSCHECK_H diff --git a/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.cpp b/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.cpp index 80de028..652dec9 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.cpp +++ b/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.cpp @@ -28,7 +28,7 @@ AST_MATCHER(CXXRecordDecl, hasBases) { void MultipleInheritanceCheck::addNodeToInterfaceMap(const CXXRecordDecl *Node, bool IsInterface) { assert(Node->getIdentifier()); - StringRef Name = Node->getIdentifier()->getName(); + const StringRef Name = Node->getIdentifier()->getName(); InterfaceMap.insert(std::make_pair(Name, IsInterface)); } @@ -38,7 +38,7 @@ void MultipleInheritanceCheck::addNodeToInterfaceMap(const CXXRecordDecl *Node, bool MultipleInheritanceCheck::getInterfaceStatus(const CXXRecordDecl *Node, bool &IsInterface) const { assert(Node->getIdentifier()); - StringRef Name = Node->getIdentifier()->getName(); + const StringRef Name = Node->getIdentifier()->getName(); auto Pair = InterfaceMap.find(Name); if (Pair == InterfaceMap.end()) return false; @@ -81,7 +81,7 @@ bool MultipleInheritanceCheck::isInterface(const CXXRecordDecl *Node) { } } - bool CurrentClassIsInterface = isCurrentClassInterface(Node); + const bool CurrentClassIsInterface = isCurrentClassInterface(Node); addNodeToInterfaceMap(Node, CurrentClassIsInterface); return CurrentClassIsInterface; } diff --git a/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.h b/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.h index 66be182..2e26843 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/MultipleInheritanceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_MULTIPLE_INHERITANCE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_MULTIPLE_INHERITANCE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_MULTIPLEINHERITANCECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_MULTIPLEINHERITANCECHECK_H #include "../ClangTidyCheck.h" @@ -43,4 +43,4 @@ private: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_MULTIPLE_INHERITANCE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_MULTIPLEINHERITANCECHECK_H diff --git a/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.cpp b/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.cpp index e202c288..4a498fb 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.cpp +++ b/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.cpp @@ -34,7 +34,7 @@ void OverloadedOperatorCheck::check(const MatchFinder::MatchResult &Result) { const auto *D = Result.Nodes.getNodeAs<FunctionDecl>("decl"); assert(D && "No FunctionDecl captured!"); - SourceLocation Loc = D->getBeginLoc(); + const SourceLocation Loc = D->getBeginLoc(); if (Loc.isValid()) diag(Loc, "overloading %0 is disallowed") << D; } diff --git a/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.h b/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.h index d91ecf8..4945ad2 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/OverloadedOperatorCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_OVERLOADED_OPERATOR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_OVERLOADED_OPERATOR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_OVERLOADEDOPERATORCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_OVERLOADEDOPERATORCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_OVERLOADED_OPERATOR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_OVERLOADEDOPERATORCHECK_H diff --git a/clang-tools-extra/clang-tidy/fuchsia/StaticallyConstructedObjectsCheck.h b/clang-tools-extra/clang-tidy/fuchsia/StaticallyConstructedObjectsCheck.h index 42d643e..d2403b0 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/StaticallyConstructedObjectsCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/StaticallyConstructedObjectsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_STATICALLY_CONSTRUCTED_OBJECTS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_STATICALLY_CONSTRUCTED_OBJECTS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_STATICALLYCONSTRUCTEDOBJECTSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_STATICALLYCONSTRUCTEDOBJECTSCHECK_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_STATICALLY_CONSTRUCTED_OBJECTS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_STATICALLYCONSTRUCTEDOBJECTSCHECK_H diff --git a/clang-tools-extra/clang-tidy/fuchsia/TemporaryObjectsCheck.cpp b/clang-tools-extra/clang-tidy/fuchsia/TemporaryObjectsCheck.cpp index 7b910b1..3acd5fb 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/TemporaryObjectsCheck.cpp +++ b/clang-tools-extra/clang-tidy/fuchsia/TemporaryObjectsCheck.cpp @@ -20,7 +20,7 @@ namespace clang::tidy::fuchsia { namespace { AST_MATCHER_P(CXXRecordDecl, matchesAnyName, ArrayRef<StringRef>, Names) { - std::string QualifiedName = Node.getQualifiedNameAsString(); + const std::string QualifiedName = Node.getQualifiedNameAsString(); return llvm::is_contained(Names, QualifiedName); } diff --git a/clang-tools-extra/clang-tidy/fuchsia/TrailingReturnCheck.h b/clang-tools-extra/clang-tidy/fuchsia/TrailingReturnCheck.h index c644e87..ba1dbeb 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/TrailingReturnCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/TrailingReturnCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_TRAILING_RETURN_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_TRAILING_RETURN_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_TRAILINGRETURNCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_TRAILINGRETURNCHECK_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_TRAILING_RETURN_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_TRAILINGRETURNCHECK_H diff --git a/clang-tools-extra/clang-tidy/fuchsia/VirtualInheritanceCheck.h b/clang-tools-extra/clang-tidy/fuchsia/VirtualInheritanceCheck.h index 45c6019..e940602 100644 --- a/clang-tools-extra/clang-tidy/fuchsia/VirtualInheritanceCheck.h +++ b/clang-tools-extra/clang-tidy/fuchsia/VirtualInheritanceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_VIRTUAL_INHERITANCE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_VIRTUAL_INHERITANCE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_VIRTUALINHERITANCECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_VIRTUALINHERITANCECHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::fuchsia -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_VIRTUAL_INHERITANCE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_FUCHSIA_VIRTUALINHERITANCECHECK_H diff --git a/clang-tools-extra/clang-tidy/google/AvoidCStyleCastsCheck.cpp b/clang-tools-extra/clang-tidy/google/AvoidCStyleCastsCheck.cpp index 174ecb0..3a5a9cd 100644 --- a/clang-tools-extra/clang-tidy/google/AvoidCStyleCastsCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/AvoidCStyleCastsCheck.cpp @@ -140,7 +140,7 @@ void AvoidCStyleCastsCheck::check(const MatchFinder::MatchResult &Result) { CharSourceRange ReplaceRange = getReplaceRange(CastExpr); - bool FnToFnCast = + const bool FnToFnCast = IsFunction(SourceTypeAsWritten) && IsFunction(DestTypeAsWritten); const bool ConstructorCast = !CastExpr->getTypeAsWritten().hasQualifiers() && @@ -239,8 +239,8 @@ void AvoidCStyleCastsCheck::check(const MatchFinder::MatchResult &Result) { return; } if (DestType->isReferenceType()) { - QualType Dest = DestType.getNonReferenceType(); - QualType Source = SourceType.getNonReferenceType(); + const QualType Dest = DestType.getNonReferenceType(); + const QualType Source = SourceType.getNonReferenceType(); if (Source == Dest.withConst() || SourceType.getNonReferenceType() == DestType.getNonReferenceType()) { ReplaceWithNamedCast("const_cast"); diff --git a/clang-tools-extra/clang-tidy/google/AvoidNSObjectNewCheck.cpp b/clang-tools-extra/clang-tidy/google/AvoidNSObjectNewCheck.cpp index daf4948..5221e4b 100644 --- a/clang-tools-extra/clang-tidy/google/AvoidNSObjectNewCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/AvoidNSObjectNewCheck.cpp @@ -22,11 +22,11 @@ using namespace clang::ast_matchers; namespace clang::tidy::google::objc { static bool isMessageExpressionInsideMacro(const ObjCMessageExpr *Expr) { - SourceLocation ReceiverLocation = Expr->getReceiverRange().getBegin(); + const SourceLocation ReceiverLocation = Expr->getReceiverRange().getBegin(); if (ReceiverLocation.isMacroID()) return true; - SourceLocation SelectorLocation = Expr->getSelectorStartLoc(); + const SourceLocation SelectorLocation = Expr->getSelectorStartLoc(); if (SelectorLocation.isMacroID()) return true; @@ -58,7 +58,7 @@ static bool isInitMethodAvailable(const ObjCInterfaceDecl *ClassDecl) { static StringRef getReceiverString(SourceRange ReceiverRange, const SourceManager &SM, const LangOptions &LangOpts) { - CharSourceRange CharRange = Lexer::makeFileCharRange( + const CharSourceRange CharRange = Lexer::makeFileCharRange( CharSourceRange::getTokenRange(ReceiverRange), SM, LangOpts); return Lexer::getSourceText(CharRange, SM, LangOpts); } @@ -77,13 +77,13 @@ static FixItHint getCallFixItHint(const ObjCMessageExpr *Expr, if (FoundClassFactory != ClassToFactoryMethodMap.end()) { StringRef ClassName = FoundClassFactory->first; StringRef FactorySelector = FoundClassFactory->second; - std::string NewCall = + const std::string NewCall = std::string(llvm::formatv("[{0} {1}]", ClassName, FactorySelector)); return FixItHint::CreateReplacement(Expr->getSourceRange(), NewCall); } if (isInitMethodAvailable(Expr->getReceiverInterface())) { - std::string NewCall = + const std::string NewCall = std::string(llvm::formatv("[[{0} alloc] init]", Receiver)); return FixItHint::CreateReplacement(Expr->getSourceRange(), NewCall); } diff --git a/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.cpp b/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.cpp index 7347657..6b96f71 100644 --- a/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.cpp @@ -40,7 +40,7 @@ void AvoidThrowingObjCExceptionCheck::check( // If the match location was in a macro, check if the macro was in a system // header. if (SourceLoc.isMacroID()) { - SourceManager &SM = *Result.SourceManager; + const SourceManager &SM = *Result.SourceManager; auto MacroLoc = SM.getImmediateMacroCallerLoc(SourceLoc); // Matches in system header macros should be ignored. diff --git a/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.h b/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.h index 26a0465..417bb8f 100644 --- a/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.h +++ b/clang-tools-extra/clang-tidy/google/AvoidThrowingObjCExceptionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_AVOID_THROWING_EXCEPTION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_AVOID_THROWING_EXCEPTION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_AVOIDTHROWINGOBJCEXCEPTIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_AVOIDTHROWINGOBJCEXCEPTIONCHECK_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::google::objc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_AVOID_THROWING_EXCEPTION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_AVOIDTHROWINGOBJCEXCEPTIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/google/AvoidUnderscoreInGoogletestNameCheck.cpp b/clang-tools-extra/clang-tidy/google/AvoidUnderscoreInGoogletestNameCheck.cpp index b335463..b156d75 100644 --- a/clang-tools-extra/clang-tidy/google/AvoidUnderscoreInGoogletestNameCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/AvoidUnderscoreInGoogletestNameCheck.cpp @@ -39,10 +39,11 @@ public: void MacroExpands(const Token &MacroNameToken, const MacroDefinition &MacroDefinition, SourceRange Range, const MacroArgs *Args) override { - IdentifierInfo *NameIdentifierInfo = MacroNameToken.getIdentifierInfo(); + const IdentifierInfo *NameIdentifierInfo = + MacroNameToken.getIdentifierInfo(); if (!NameIdentifierInfo) return; - StringRef MacroName = NameIdentifierInfo->getName(); + const StringRef MacroName = NameIdentifierInfo->getName(); if (!isGoogletestTestMacro(MacroName) || !Args || Args->getNumMacroArguments() < 2) return; @@ -50,7 +51,7 @@ public: const Token *TestNameToken = Args->getUnexpArgument(1); if (!TestSuiteNameToken || !TestNameToken) return; - std::string TestSuiteNameMaybeDisabled = + const std::string TestSuiteNameMaybeDisabled = PP->getSpelling(*TestSuiteNameToken); StringRef TestSuiteName = TestSuiteNameMaybeDisabled; TestSuiteName.consume_front(KDisabledTestPrefix); @@ -60,7 +61,7 @@ public: "Googletest FAQ") << TestSuiteName; - std::string TestNameMaybeDisabled = PP->getSpelling(*TestNameToken); + const std::string TestNameMaybeDisabled = PP->getSpelling(*TestNameToken); StringRef TestName = TestNameMaybeDisabled; TestName.consume_front(KDisabledTestPrefix); if (TestName.contains('_')) diff --git a/clang-tools-extra/clang-tidy/google/DefaultArgumentsCheck.h b/clang-tools-extra/clang-tidy/google/DefaultArgumentsCheck.h index 1d1e4e3..0f397b4 100644 --- a/clang-tools-extra/clang-tidy/google/DefaultArgumentsCheck.h +++ b/clang-tools-extra/clang-tidy/google/DefaultArgumentsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_DEFAULT_ARGUMENTS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_DEFAULT_ARGUMENTS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_DEFAULTARGUMENTSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_DEFAULTARGUMENTSCHECK_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::google -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_DEFAULT_ARGUMENTS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_DEFAULTARGUMENTSCHECK_H diff --git a/clang-tools-extra/clang-tidy/google/ExplicitConstructorCheck.cpp b/clang-tools-extra/clang-tidy/google/ExplicitConstructorCheck.cpp index 6d5182d..ac604b7 100644 --- a/clang-tools-extra/clang-tidy/google/ExplicitConstructorCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/ExplicitConstructorCheck.cpp @@ -39,8 +39,8 @@ static SourceRange findToken(const SourceManager &Sources, bool (*Pred)(const Token &)) { if (StartLoc.isMacroID() || EndLoc.isMacroID()) return {}; - FileID File = Sources.getFileID(Sources.getSpellingLoc(StartLoc)); - StringRef Buf = Sources.getBufferData(File); + const FileID File = Sources.getFileID(Sources.getSpellingLoc(StartLoc)); + const StringRef Buf = Sources.getBufferData(File); const char *StartChar = Sources.getCharacterData(StartLoc); Lexer Lex(StartLoc, LangOpts, StartChar, StartChar, Buf.end()); Lex.SetCommentRetentionState(true); @@ -88,7 +88,7 @@ void ExplicitConstructorCheck::check(const MatchFinder::MatchResult &Result) { Result.Nodes.getNodeAs<CXXConversionDecl>("conversion")) { if (Conversion->isOutOfLine()) return; - SourceLocation Loc = Conversion->getLocation(); + const SourceLocation Loc = Conversion->getLocation(); // Ignore all macros until we learn to ignore specific ones (e.g. used in // gmock to define matchers). if (Loc.isMacroID()) @@ -105,7 +105,7 @@ void ExplicitConstructorCheck::check(const MatchFinder::MatchResult &Result) { const ExplicitSpecifier ExplicitSpec = Ctor->getExplicitSpecifier(); - bool TakesInitializerList = isStdInitializerList( + const bool TakesInitializerList = isStdInitializerList( Ctor->getParamDecl(0)->getType().getNonReferenceType()); if (ExplicitSpec.isExplicit() && (Ctor->isCopyOrMoveConstructor() || TakesInitializerList)) { @@ -113,7 +113,7 @@ void ExplicitConstructorCheck::check(const MatchFinder::MatchResult &Result) { return Tok.is(tok::raw_identifier) && Tok.getRawIdentifier() == "explicit"; }; - SourceRange ExplicitTokenRange = + const SourceRange ExplicitTokenRange = findToken(*Result.SourceManager, getLangOpts(), Ctor->getOuterLocStart(), Ctor->getEndLoc(), IsKwExplicit); StringRef ConstructorDescription; @@ -149,7 +149,7 @@ void ExplicitConstructorCheck::check(const MatchFinder::MatchResult &Result) { const bool SingleArgument = Ctor->getNumParams() == 1 && !Ctor->getParamDecl(0)->isParameterPack(); - SourceLocation Loc = Ctor->getLocation(); + const SourceLocation Loc = Ctor->getLocation(); auto Diag = diag(Loc, ExplicitExpr ? WithExpressionWarningMessage : NoExpressionWarningMessage) diff --git a/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.cpp b/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.cpp index ce0e4e6..2b9183c 100644 --- a/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.cpp @@ -33,7 +33,7 @@ static std::string validFunctionNameRegex(bool RequirePrefix) { // If a prefix is required, the regex checks for a capital letter followed by // another capital letter or number that is part of the prefix and another // capital letter or number that begins the name following the prefix. - std::string FunctionNameMatcher = + const std::string FunctionNameMatcher = std::string(RequirePrefix ? "[A-Z][A-Z0-9]+" : "") + "[A-Z][a-zA-Z0-9]*"; return std::string("::(") + FunctionNameMatcher + ")$"; } @@ -48,13 +48,13 @@ static FixItHint generateFixItHint(const FunctionDecl *Decl) { if (Decl->getStorageClass() != SC_Static) return {}; - StringRef Name = Decl->getName(); + const StringRef Name = Decl->getName(); std::string NewName = Decl->getName().str(); size_t Index = 0; bool AtWordBoundary = true; while (Index < NewName.size()) { - char Ch = NewName[Index]; + const char Ch = NewName[Index]; if (isalnum(Ch)) { // Capitalize the first letter after every word boundary. if (AtWordBoundary) { @@ -101,7 +101,7 @@ void FunctionNamingCheck::registerMatchers(MatchFinder *Finder) { void FunctionNamingCheck::check(const MatchFinder::MatchResult &Result) { const auto *MatchedDecl = Result.Nodes.getNodeAs<FunctionDecl>("function"); - bool IsGlobal = MatchedDecl->getStorageClass() != SC_Static; + const bool IsGlobal = MatchedDecl->getStorageClass() != SC_Static; diag(MatchedDecl->getLocation(), "%select{static function|function in global namespace}1 named %0 must " "%select{be in|have an appropriate prefix followed by}1 Pascal case as " diff --git a/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.h b/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.h index e4efadf..6acc184 100644 --- a/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.h +++ b/clang-tools-extra/clang-tidy/google/FunctionNamingCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_FUNCTION_NAMING_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_FUNCTION_NAMING_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_FUNCTIONNAMINGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_FUNCTIONNAMINGCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/StringRef.h" @@ -36,4 +36,4 @@ public: } // namespace clang::tidy::google::objc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_FUNCTION_NAMING_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_FUNCTIONNAMINGCHECK_H diff --git a/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.cpp b/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.cpp index a4c76be9..7470b1eb 100644 --- a/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.cpp @@ -30,7 +30,7 @@ static FixItHint generateFixItHint(const VarDecl *Decl, bool IsConst) { return {}; } - char FC = Decl->getName()[0]; + const char FC = Decl->getName()[0]; if (!llvm::isAlpha(FC) || Decl->getName().size() == 1) { // No fix available if first character is not alphabetical character, or it // is a single-character variable, since it is difficult to determine the @@ -38,7 +38,7 @@ static FixItHint generateFixItHint(const VarDecl *Decl, bool IsConst) { // their own. return {}; } - char SC = Decl->getName()[1]; + const char SC = Decl->getName()[1]; if ((FC == 'k' || FC == 'g') && !llvm::isAlpha(SC)) { // No fix available if the prefix is correct but the second character is // not alphabetical, since it is difficult to determine the proper fix in diff --git a/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.h b/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.h index 9b55855..e0693d4 100644 --- a/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.h +++ b/clang-tools-extra/clang-tidy/google/GlobalVariableDeclarationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_GLOBAL_VARIABLE_DECLARATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_GLOBAL_VARIABLE_DECLARATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_GLOBALVARIABLEDECLARATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_GLOBALVARIABLEDECLARATIONCHECK_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::google::objc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_OBJC_GLOBAL_VARIABLE_DECLARATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_GOOGLE_GLOBALVARIABLEDECLARATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/google/IntegerTypesCheck.cpp b/clang-tools-extra/clang-tidy/google/IntegerTypesCheck.cpp index 52777fa..52bcf1b 100644 --- a/clang-tools-extra/clang-tidy/google/IntegerTypesCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/IntegerTypesCheck.cpp @@ -103,7 +103,7 @@ void IntegerTypesCheck::registerMatchers(MatchFinder *Finder) { void IntegerTypesCheck::check(const MatchFinder::MatchResult &Result) { auto TL = *Result.Nodes.getNodeAs<TypeLoc>("tl"); - SourceLocation Loc = TL.getBeginLoc(); + const SourceLocation Loc = TL.getBeginLoc(); // Look through qualification. if (auto QualLoc = TL.getAs<QualifiedTypeLoc>()) @@ -113,7 +113,7 @@ void IntegerTypesCheck::check(const MatchFinder::MatchResult &Result) { if (!BuiltinLoc) return; - Token Tok = getTokenAtLoc(Loc, Result, *IdentTable); + const Token Tok = getTokenAtLoc(Loc, Result, *IdentTable); // Ensure the location actually points to one of the builting integral type // names we're interested in. Otherwise, we might be getting this match from // implicit code (e.g. an implicit assignment operator of a class containing @@ -164,7 +164,7 @@ void IntegerTypesCheck::check(const MatchFinder::MatchResult &Result) { !isAsciiIdentifierContinue(Data[Port.size()])) return; - std::string Replacement = + const std::string Replacement = ((IsSigned ? SignedTypePrefix : UnsignedTypePrefix) + Twine(Width) + TypeSuffix) .str(); diff --git a/clang-tools-extra/clang-tidy/google/TodoCommentCheck.cpp b/clang-tools-extra/clang-tidy/google/TodoCommentCheck.cpp index 8554870..7331b36 100644 --- a/clang-tools-extra/clang-tidy/google/TodoCommentCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/TodoCommentCheck.cpp @@ -20,7 +20,7 @@ public: TodoMatch("^// *TODO *(\\(.*\\))?:?( )?(.*)$") {} bool HandleComment(Preprocessor &PP, SourceRange Range) override { - StringRef Text = + const StringRef Text = Lexer::getSourceText(CharSourceRange::getCharRange(Range), PP.getSourceManager(), PP.getLangOpts()); @@ -28,13 +28,14 @@ public: if (!TodoMatch.match(Text, &Matches)) return false; - StringRef Username = Matches[1]; - StringRef Comment = Matches[3]; + const StringRef Username = Matches[1]; + const StringRef Comment = Matches[3]; if (!Username.empty()) return false; - std::string NewText = ("// TODO(" + Twine(User) + "): " + Comment).str(); + const std::string NewText = + ("// TODO(" + Twine(User) + "): " + Comment).str(); Check.diag(Range.getBegin(), "missing username/bug in TODO") << FixItHint::CreateReplacement(CharSourceRange::getCharRange(Range), diff --git a/clang-tools-extra/clang-tidy/google/UnnamedNamespaceInHeaderCheck.cpp b/clang-tools-extra/clang-tidy/google/UnnamedNamespaceInHeaderCheck.cpp index 3066dd0..054bdc8 100644 --- a/clang-tools-extra/clang-tidy/google/UnnamedNamespaceInHeaderCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/UnnamedNamespaceInHeaderCheck.cpp @@ -28,7 +28,7 @@ void UnnamedNamespaceInHeaderCheck::registerMatchers( void UnnamedNamespaceInHeaderCheck::check( const MatchFinder::MatchResult &Result) { const auto *N = Result.Nodes.getNodeAs<NamespaceDecl>("anonymousNamespace"); - SourceLocation Loc = N->getBeginLoc(); + const SourceLocation Loc = N->getBeginLoc(); if (!Loc.isValid()) return; diff --git a/clang-tools-extra/clang-tidy/google/UpgradeGoogletestCaseCheck.cpp b/clang-tools-extra/clang-tidy/google/UpgradeGoogletestCaseCheck.cpp index 9da1915..87fd046 100644 --- a/clang-tools-extra/clang-tidy/google/UpgradeGoogletestCaseCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/UpgradeGoogletestCaseCheck.cpp @@ -64,7 +64,7 @@ public: // We check if the newly defined macro is one of the target replacements. // This ensures that the check creates warnings only if it is including a // recent enough version of Google Test. - llvm::StringRef FileName = PP->getSourceManager().getFilename( + const llvm::StringRef FileName = PP->getSourceManager().getFilename( MD->getMacroInfo()->getDefinitionLoc()); ReplacementFound = FileName.ends_with("gtest/gtest-typed-test.h") && PP->getSpelling(MacroNameTok) == "TYPED_TEST_SUITE"; @@ -94,18 +94,18 @@ private: if (!ReplacementFound) return; - std::string Name = PP->getSpelling(MacroNameTok); + const std::string Name = PP->getSpelling(MacroNameTok); std::optional<llvm::StringRef> Replacement = getNewMacroName(Name); if (!Replacement) return; - llvm::StringRef FileName = PP->getSourceManager().getFilename( + const llvm::StringRef FileName = PP->getSourceManager().getFilename( MD.getMacroInfo()->getDefinitionLoc()); if (!FileName.ends_with("gtest/gtest-typed-test.h")) return; - DiagnosticBuilder Diag = Check->diag(Loc, RenameCaseToSuiteMessage); + const DiagnosticBuilder Diag = Check->diag(Loc, RenameCaseToSuiteMessage); if (Action == CheckAction::Rename) Diag << FixItHint::CreateReplacement( @@ -234,7 +234,7 @@ static bool isInInstantiation(const NodeType &Node, template <typename NodeType> static bool isInTemplate(const NodeType &Node, const MatchFinder::MatchResult &Result) { - internal::Matcher<NodeType> IsInsideTemplate = + const internal::Matcher<NodeType> IsInsideTemplate = hasAncestor(decl(anyOf(classTemplateDecl(), functionTemplateDecl()))); return !match(IsInsideTemplate, Node, *Result.Context).empty(); } @@ -340,7 +340,7 @@ void UpgradeGoogletestCaseCheck::check(const MatchFinder::MatchResult &Result) { // will only be instantiated with the true type name, `TestSuite`. } - DiagnosticBuilder Diag = + const DiagnosticBuilder Diag = diag(ReplacementRange.getBegin(), RenameCaseToSuiteMessage); ReplacementRange = Lexer::makeFileCharRange( diff --git a/clang-tools-extra/clang-tidy/google/UsingNamespaceDirectiveCheck.cpp b/clang-tools-extra/clang-tidy/google/UsingNamespaceDirectiveCheck.cpp index fbfd5d34..00446dc 100644 --- a/clang-tools-extra/clang-tidy/google/UsingNamespaceDirectiveCheck.cpp +++ b/clang-tools-extra/clang-tidy/google/UsingNamespaceDirectiveCheck.cpp @@ -22,7 +22,7 @@ void UsingNamespaceDirectiveCheck::registerMatchers( void UsingNamespaceDirectiveCheck::check( const MatchFinder::MatchResult &Result) { const auto *U = Result.Nodes.getNodeAs<UsingDirectiveDecl>("usingNamespace"); - SourceLocation Loc = U->getBeginLoc(); + const SourceLocation Loc = U->getBeginLoc(); if (U->isImplicit() || !Loc.isValid()) return; diff --git a/clang-tools-extra/clang-tidy/hicpp/ExceptionBaseclassCheck.h b/clang-tools-extra/clang-tidy/hicpp/ExceptionBaseclassCheck.h index 12fe7f7..800e7ac 100644 --- a/clang-tools-extra/clang-tidy/hicpp/ExceptionBaseclassCheck.h +++ b/clang-tools-extra/clang-tidy/hicpp/ExceptionBaseclassCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_EXCEPTION_BASECLASS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_EXCEPTION_BASECLASS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_EXCEPTIONBASECLASSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_EXCEPTIONBASECLASSCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::hicpp -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_EXCEPTION_BASECLASS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_EXCEPTIONBASECLASSCHECK_H diff --git a/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.cpp b/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.cpp index e610d99..e9a5819 100644 --- a/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.cpp +++ b/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.cpp @@ -152,7 +152,7 @@ void MultiwayPathsCoveredCheck::handleSwitchWithoutDefault( assert(CaseCount > 0 && "Switch statement without any case found. This case " "should be excluded by the matcher and is handled " "separately."); - std::size_t MaxPathsPossible = [&]() { + const std::size_t MaxPathsPossible = [&]() { if (const auto *GeneralCondition = Result.Nodes.getNodeAs<DeclRefExpr>("non-enum-condition")) { return getNumberOfPossibleValues(GeneralCondition->getType(), diff --git a/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.h b/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.h index 902be2d..e22e31a 100644 --- a/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.h +++ b/clang-tools-extra/clang-tidy/hicpp/MultiwayPathsCoveredCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_MULTIWAY_PATHS_COVERED_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_MULTIWAY_PATHS_COVERED_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_MULTIWAYPATHSCOVEREDCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_MULTIWAYPATHSCOVEREDCHECK_H #include "../ClangTidyCheck.h" @@ -41,4 +41,4 @@ private: } // namespace clang::tidy::hicpp -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_MULTIWAY_PATHS_COVERED_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_MULTIWAYPATHSCOVEREDCHECK_H diff --git a/clang-tools-extra/clang-tidy/hicpp/NoAssemblerCheck.h b/clang-tools-extra/clang-tidy/hicpp/NoAssemblerCheck.h index 1ff40ea..15d646f 100644 --- a/clang-tools-extra/clang-tidy/hicpp/NoAssemblerCheck.h +++ b/clang-tools-extra/clang-tidy/hicpp/NoAssemblerCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_NO_ASSEMBLER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_NO_ASSEMBLER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_NOASSEMBLERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_NOASSEMBLERCHECK_H #include "../ClangTidyCheck.h" @@ -27,4 +27,4 @@ public: } // namespace clang::tidy::hicpp -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_NO_ASSEMBLER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_NOASSEMBLERCHECK_H diff --git a/clang-tools-extra/clang-tidy/hicpp/SignedBitwiseCheck.h b/clang-tools-extra/clang-tidy/hicpp/SignedBitwiseCheck.h index 499a4e7..ef92a4d 100644 --- a/clang-tools-extra/clang-tidy/hicpp/SignedBitwiseCheck.h +++ b/clang-tools-extra/clang-tidy/hicpp/SignedBitwiseCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_SIGNED_BITWISE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_SIGNED_BITWISE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_SIGNEDBITWISECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_SIGNEDBITWISECHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ private: } // namespace clang::tidy::hicpp -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_SIGNED_BITWISE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_HICPP_SIGNEDBITWISECHECK_H diff --git a/clang-tools-extra/clang-tidy/llvm/.clang-tidy b/clang-tools-extra/clang-tidy/llvm/.clang-tidy new file mode 100644 index 0000000..08c2c1d --- /dev/null +++ b/clang-tools-extra/clang-tidy/llvm/.clang-tidy @@ -0,0 +1,5 @@ +InheritParentConfig: true +# FIXME(vbvictor) enable this check when https://github.com/llvm/llvm-project/issues/166750 is fixed +# and github runners are updated to include the fix +Checks: > + -llvm-header-guard diff --git a/clang-tools-extra/clang-tidy/llvm/HeaderGuardCheck.cpp b/clang-tools-extra/clang-tidy/llvm/HeaderGuardCheck.cpp index 8737c1e..ef8b6b1 100644 --- a/clang-tools-extra/clang-tidy/llvm/HeaderGuardCheck.cpp +++ b/clang-tools-extra/clang-tidy/llvm/HeaderGuardCheck.cpp @@ -28,24 +28,24 @@ std::string LLVMHeaderGuardCheck::getHeaderGuard(StringRef Filename, // style in include/llvm and include/clang which we want to preserve. // We don't want _INCLUDE_ in our guards. - size_t PosInclude = Guard.rfind("include/"); + const size_t PosInclude = Guard.rfind("include/"); if (PosInclude != StringRef::npos) Guard = Guard.substr(PosInclude + std::strlen("include/")); // For clang we drop the _TOOLS_. - size_t PosToolsClang = Guard.rfind("tools/clang/"); + const size_t PosToolsClang = Guard.rfind("tools/clang/"); if (PosToolsClang != StringRef::npos) Guard = Guard.substr(PosToolsClang + std::strlen("tools/")); // Unlike LLVM svn, LLVM git monorepo is named llvm-project, so we replace // "/llvm-project/" with the canonical "/llvm/". const static StringRef LLVMProject = "/llvm-project/"; - size_t PosLLVMProject = Guard.rfind(LLVMProject); + const size_t PosLLVMProject = Guard.rfind(LLVMProject); if (PosLLVMProject != StringRef::npos) Guard = Guard.replace(PosLLVMProject, LLVMProject.size(), "/llvm/"); // The remainder is LLVM_FULL_PATH_TO_HEADER_H - size_t PosLLVM = Guard.rfind("llvm/"); + const size_t PosLLVM = Guard.rfind("llvm/"); if (PosLLVM != StringRef::npos) Guard = Guard.substr(PosLLVM); diff --git a/clang-tools-extra/clang-tidy/llvm/IncludeOrderCheck.cpp b/clang-tools-extra/clang-tidy/llvm/IncludeOrderCheck.cpp index f34e3a6..416aca1 100644 --- a/clang-tools-extra/clang-tidy/llvm/IncludeOrderCheck.cpp +++ b/clang-tools-extra/clang-tidy/llvm/IncludeOrderCheck.cpp @@ -162,15 +162,15 @@ void IncludeOrderPPCallbacks::EndOfMainFile() { continue; const IncludeDirective &CopyFrom = FileDirectives[IncludeIndices[I]]; - SourceLocation FromLoc = CopyFrom.Range.getBegin(); + const SourceLocation FromLoc = CopyFrom.Range.getBegin(); const char *FromData = SM.getCharacterData(FromLoc); - unsigned FromLen = std::strcspn(FromData, "\n"); + const unsigned FromLen = std::strcspn(FromData, "\n"); - StringRef FixedName(FromData, FromLen); + const StringRef FixedName(FromData, FromLen); - SourceLocation ToLoc = FileDirectives[I].Range.getBegin(); + const SourceLocation ToLoc = FileDirectives[I].Range.getBegin(); const char *ToData = SM.getCharacterData(ToLoc); - unsigned ToLen = std::strcspn(ToData, "\n"); + const unsigned ToLen = std::strcspn(ToData, "\n"); auto ToRange = CharSourceRange::getCharRange(ToLoc, ToLoc.getLocWithOffset(ToLen)); diff --git a/clang-tools-extra/clang-tidy/llvm/PreferIsaOrDynCastInConditionalsCheck.cpp b/clang-tools-extra/clang-tidy/llvm/PreferIsaOrDynCastInConditionalsCheck.cpp index f4f3543..8966745 100644 --- a/clang-tools-extra/clang-tidy/llvm/PreferIsaOrDynCastInConditionalsCheck.cpp +++ b/clang-tools-extra/clang-tidy/llvm/PreferIsaOrDynCastInConditionalsCheck.cpp @@ -68,8 +68,8 @@ void PreferIsaOrDynCastInConditionalsCheck::check( // llvm::cast<T>(x) // ^ ^ // StartLoc EndLoc - SourceLocation StartLoc = Callee->getLocation(); - SourceLocation EndLoc = Callee->getNameInfo().getEndLoc(); + const SourceLocation StartLoc = Callee->getLocation(); + const SourceLocation EndLoc = Callee->getNameInfo().getEndLoc(); if (Result.Nodes.getNodeAs<VarDecl>("var")) { diag(StartLoc, diff --git a/clang-tools-extra/clang-tidy/llvm/TwineLocalCheck.cpp b/clang-tools-extra/clang-tidy/llvm/TwineLocalCheck.cpp index b8b7c41..7dea845 100644 --- a/clang-tools-extra/clang-tidy/llvm/TwineLocalCheck.cpp +++ b/clang-tools-extra/clang-tidy/llvm/TwineLocalCheck.cpp @@ -40,13 +40,13 @@ void TwineLocalCheck::check(const MatchFinder::MatchResult &Result) { C = cast<CXXConstructExpr>(C)->getArg(0)->IgnoreParenImpCasts(); } - SourceRange TypeRange = + const SourceRange TypeRange = VD->getTypeSourceInfo()->getTypeLoc().getSourceRange(); // A real Twine, turn it into a std::string. if (VD->getType()->getCanonicalTypeUnqualified() == C->getType()->getCanonicalTypeUnqualified()) { - SourceLocation EndLoc = Lexer::getLocForEndOfToken( + const SourceLocation EndLoc = Lexer::getLocForEndOfToken( VD->getInit()->getEndLoc(), 0, *Result.SourceManager, getLangOpts()); Diag << FixItHint::CreateReplacement(TypeRange, "std::string") << FixItHint::CreateInsertion(VD->getInit()->getBeginLoc(), "(") diff --git a/clang-tools-extra/clang-tidy/llvm/UseNewMLIROpBuilderCheck.cpp b/clang-tools-extra/clang-tidy/llvm/UseNewMLIROpBuilderCheck.cpp index 0014153..55b383e 100644 --- a/clang-tools-extra/clang-tidy/llvm/UseNewMLIROpBuilderCheck.cpp +++ b/clang-tools-extra/clang-tidy/llvm/UseNewMLIROpBuilderCheck.cpp @@ -36,7 +36,7 @@ static EditGenerator rewrite(RangeSelector Call, RangeSelector Builder) { SourceLocation Begin = CallRange->getBegin(); // This will result in just a warning and no edit. - bool InMacro = CallRange->getBegin().isMacroID(); + const bool InMacro = CallRange->getBegin().isMacroID(); if (InMacro) { while (SM.isMacroArgExpansion(Begin)) Begin = SM.getImmediateExpansionRange(Begin).getBegin(); @@ -119,11 +119,11 @@ static EditGenerator rewrite(RangeSelector Call, RangeSelector Builder) { } static RewriteRuleWith<std::string> useNewMlirOpBuilderCheckRule() { - Stencil Message = cat("use 'OpType::create(builder, ...)' instead of " - "'builder.create<OpType>(...)'"); + const Stencil Message = cat("use 'OpType::create(builder, ...)' instead of " + "'builder.create<OpType>(...)'"); // Match a create call on an OpBuilder. auto BuilderType = cxxRecordDecl(isSameOrDerivedFrom("::mlir::OpBuilder")); - ast_matchers::internal::Matcher<Stmt> Base = + const ast_matchers::internal::Matcher<Stmt> Base = cxxMemberCallExpr( on(expr(anyOf(hasType(BuilderType), hasType(pointsTo(BuilderType)))) .bind("builder")), diff --git a/clang-tools-extra/clang-tidy/llvmlibc/InlineFunctionDeclCheck.cpp b/clang-tools-extra/clang-tidy/llvmlibc/InlineFunctionDeclCheck.cpp index 091e4fe..9f09947 100644 --- a/clang-tools-extra/clang-tidy/llvmlibc/InlineFunctionDeclCheck.cpp +++ b/clang-tools-extra/clang-tidy/llvmlibc/InlineFunctionDeclCheck.cpp @@ -78,7 +78,8 @@ void InlineFunctionDeclCheck::check(const MatchFinder::MatchResult &Result) { // Check if decl starts with LIBC_INLINE auto Loc = FullSourceLoc(Result.SourceManager->getFileLoc(SrcBegin), *Result.SourceManager); - llvm::StringRef SrcText = Loc.getBufferData().drop_front(Loc.getFileOffset()); + const llvm::StringRef SrcText = + Loc.getBufferData().drop_front(Loc.getFileOffset()); if (SrcText.starts_with("LIBC_INLINE")) return; diff --git a/clang-tools-extra/clang-tidy/llvmlibc/NamespaceConstants.h b/clang-tools-extra/clang-tidy/llvmlibc/NamespaceConstants.h index 50669dc..8b8b719 100644 --- a/clang-tools-extra/clang-tidy/llvmlibc/NamespaceConstants.h +++ b/clang-tools-extra/clang-tidy/llvmlibc/NamespaceConstants.h @@ -6,6 +6,9 @@ // //===----------------------------------------------------------------------===// +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_LLVMLIBC_NAMESPACECONSTANTS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_LLVMLIBC_NAMESPACECONSTANTS_H + #include "llvm/ADT/StringRef.h" namespace clang::tidy::llvm_libc { @@ -18,3 +21,5 @@ const static llvm::StringRef RequiredNamespaceDeclMacroName = "LIBC_NAMESPACE_DECL"; } // namespace clang::tidy::llvm_libc + +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_LLVMLIBC_NAMESPACECONSTANTS_H diff --git a/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.cpp b/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.cpp index b05fd04..61d5477 100644 --- a/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.cpp @@ -56,7 +56,7 @@ static llvm::SmallString<64U> skeleton(StringRef Name) { const char *Prev = Curr; UTF32 CodePoint = 0; - ConversionResult Result = convertUTF8Sequence( + const ConversionResult Result = convertUTF8Sequence( reinterpret_cast<const UTF8 **>(&Curr), reinterpret_cast<const UTF8 *>(End), &CodePoint, strictConversion); if (Result != conversionOK) { @@ -64,7 +64,7 @@ static llvm::SmallString<64U> skeleton(StringRef Name) { break; } - StringRef Key(Prev, Curr - Prev); + const StringRef Key(Prev, Curr - Prev); auto *Where = llvm::lower_bound(ConfusableEntries, CodePoint, [](decltype(ConfusableEntries[0]) X, UTF32 Y) { return X.codepoint < Y; }); @@ -183,7 +183,7 @@ void ConfusableIdentifierCheck::addDeclToCheck(const NamedDecl *ND, if (!NDII) return; - StringRef NDName = NDII->getName(); + const StringRef NDName = NDII->getName(); if (NDName.empty()) return; @@ -216,7 +216,7 @@ void ConfusableIdentifierCheck::onEndOfTranslationUnit() { // the same skeleton. for (const IdentifierInfo *II : Idents) { for (auto [OuterND, OuterParent] : NameToDecls[II]) { - for (Entry Inner : DeclsWithinContext[OuterParent]) { + for (const Entry Inner : DeclsWithinContext[OuterParent]) { // Don't complain if the identifiers are the same. if (OuterND->getIdentifier() == Inner.ND->getIdentifier()) continue; diff --git a/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.h b/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.h index b341d03..5b98d48 100644 --- a/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.h +++ b/clang-tools-extra/clang-tidy/misc/ConfusableIdentifierCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_CONFUSABLE_IDENTIFIER_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_CONFUSABLE_IDENTIFIER_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_CONFUSABLEIDENTIFIERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_CONFUSABLEIDENTIFIERCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/DenseMap.h" @@ -41,4 +41,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_CONFUSABLE_IDENTIFIER_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_CONFUSABLEIDENTIFIERCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/ConfusableTable/BuildConfusableTable.cpp b/clang-tools-extra/clang-tidy/misc/ConfusableTable/BuildConfusableTable.cpp index 6a07902..f5dcc7f 100644 --- a/clang-tools-extra/clang-tidy/misc/ConfusableTable/BuildConfusableTable.cpp +++ b/clang-tools-extra/clang-tidy/misc/ConfusableTable/BuildConfusableTable.cpp @@ -26,7 +26,7 @@ int main(int argc, char *argv[]) { std::vector<std::pair<llvm::UTF32, SmallVector<llvm::UTF32>>> Entries; SmallVector<StringRef> Values; - for (StringRef Line : Lines) { + for (const StringRef Line : Lines) { if (Line.starts_with("#")) continue; @@ -37,14 +37,14 @@ int main(int argc, char *argv[]) { return 2; } - llvm::StringRef From = Values[0].trim(); + const llvm::StringRef From = Values[0].trim(); llvm::UTF32 CodePoint = 0; From.getAsInteger(16, CodePoint); SmallVector<llvm::UTF32> To; SmallVector<StringRef> ToN; Values[1].split(ToN, ' ', -1, false); - for (StringRef ToI : ToN) { + for (const StringRef ToI : ToN) { llvm::UTF32 ToCodePoint = 0; ToI.trim().getAsInteger(16, ToCodePoint); To.push_back(ToCodePoint); @@ -56,7 +56,7 @@ int main(int argc, char *argv[]) { } llvm::sort(Entries); - unsigned LargestValue = + const unsigned LargestValue = llvm::max_element(Entries, [](const auto &Entry0, const auto &Entry1) { return Entry0.second.size() < Entry1.second.size(); })->second.size(); diff --git a/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.cpp b/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.cpp index 3b9b8e0..a2d3d3f 100644 --- a/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.cpp @@ -56,7 +56,7 @@ AST_MATCHER_P(Stmt, forEachPrevStmt, ast_matchers::internal::Matcher<Stmt>, // Matches the expression awaited by the `co_await`. AST_MATCHER_P(CoawaitExpr, awaitable, ast_matchers::internal::Matcher<Expr>, InnerMatcher) { - if (Expr *E = Node.getOperand()) + if (const Expr *E = Node.getOperand()) return InnerMatcher.matches(*E, Finder, Builder); return false; } diff --git a/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.h b/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.h index e100509..768b62e 100644 --- a/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.h +++ b/clang-tools-extra/clang-tidy/misc/CoroutineHostileRAIICheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_COROUTINESHOSTILERAIICHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_COROUTINESHOSTILERAIICHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_COROUTINEHOSTILERAIICHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_COROUTINEHOSTILERAIICHECK_H #include "../ClangTidyCheck.h" #include "clang/AST/ASTTypeTraits.h" @@ -50,4 +50,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_COROUTINESHOSTILERAIICHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_COROUTINEHOSTILERAIICHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.cpp b/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.cpp index 714af11..c10ee1d 100644 --- a/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.cpp @@ -93,7 +93,8 @@ void DefinitionsInHeadersCheck::check(const MatchFinder::MatchResult &Result) { } } - bool IsFullSpec = FD->getTemplateSpecializationKind() != TSK_Undeclared; + const bool IsFullSpec = + FD->getTemplateSpecializationKind() != TSK_Undeclared; diag(FD->getLocation(), "%select{function|full function template specialization}0 %1 defined " "in a header file; function definitions in header files can lead to " diff --git a/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.h b/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.h index 0c162cc..e52fa20 100644 --- a/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.h +++ b/clang-tools-extra/clang-tidy/misc/DefinitionsInHeadersCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_DEFINITIONS_IN_HEADERS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_DEFINITIONS_IN_HEADERS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_DEFINITIONSINHEADERSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_DEFINITIONSINHEADERSCHECK_H #include "../ClangTidyCheck.h" #include "../utils/FileExtensionsUtils.h" @@ -38,4 +38,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_DEFINITIONS_IN_HEADERS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_DEFINITIONSINHEADERSCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.cpp b/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.cpp index 1a5aa4b..558c368 100644 --- a/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.cpp @@ -200,7 +200,7 @@ void IncludeCleanerCheck::check(const MatchFinder::MatchResult &Result) { Unused.push_back(&I); } - llvm::StringRef Code = SM->getBufferData(SM->getMainFileID()); + const llvm::StringRef Code = SM->getBufferData(SM->getMainFileID()); auto FileStyle = format::getStyle(format::DefaultFormatStyle, getCurrentMainFile(), format::DefaultFallbackStyle, Code, @@ -220,14 +220,14 @@ void IncludeCleanerCheck::check(const MatchFinder::MatchResult &Result) { } if (MissingIncludes) { - tooling::HeaderIncludes HeaderIncludes(getCurrentMainFile(), Code, - FileStyle->IncludeStyle); + const tooling::HeaderIncludes HeaderIncludes(getCurrentMainFile(), Code, + FileStyle->IncludeStyle); // Deduplicate insertions when running in bulk fix mode. llvm::StringSet<> InsertedHeaders{}; for (const auto &Inc : Missing) { - std::string Spelling = include_cleaner::spellHeader( + const std::string Spelling = include_cleaner::spellHeader( {Inc.Missing, PP->getHeaderSearchInfo(), MainFile}); - bool Angled = llvm::StringRef{Spelling}.starts_with("<"); + const bool Angled = llvm::StringRef{Spelling}.starts_with("<"); // We might suggest insertion of an existing include in edge cases, e.g., // include is present in a PP-disabled region, or spelling of the header // turns out to be the same as one of the unresolved includes in the @@ -235,7 +235,7 @@ void IncludeCleanerCheck::check(const MatchFinder::MatchResult &Result) { if (auto Replacement = HeaderIncludes.insert( llvm::StringRef{Spelling}.trim("\"<>"), Angled, tooling::IncludeDirective::Include)) { - DiagnosticBuilder DB = + const DiagnosticBuilder DB = diag(SM->getSpellingLoc(Inc.SymRef.RefLocation), "no header providing \"%0\" is directly included") << Inc.SymRef.Target.name(); diff --git a/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.h b/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.h index 43e1ed8..619d819 100644 --- a/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.h +++ b/clang-tools-extra/clang-tidy/misc/IncludeCleanerCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_INCLUDECLEANER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_INCLUDECLEANER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_INCLUDECLEANERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_INCLUDECLEANERCHECK_H #include "../ClangTidyCheck.h" #include "../ClangTidyDiagnosticConsumer.h" @@ -57,4 +57,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_INCLUDECLEANER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_INCLUDECLEANERCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp b/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp index f89c539..db5bc61 100644 --- a/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp +++ b/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp @@ -40,18 +40,18 @@ static bool containsMisleadingBidi(StringRef Buffer, // // Warn if we end up with an unclosed context. while (CurPtr < Buffer.end()) { - unsigned char C = *CurPtr; + const unsigned char C = *CurPtr; if (isASCII(C)) { ++CurPtr; - bool IsParagrapSep = + const bool IsParagrapSep = (C == 0xA || C == 0xD || (0x1C <= C && C <= 0x1E) || C == 0x85); - bool IsSegmentSep = (C == 0x9 || C == 0xB || C == 0x1F); + const bool IsSegmentSep = (C == 0x9 || C == 0xB || C == 0x1F); if (IsParagrapSep || IsSegmentSep) BidiContexts.clear(); continue; } llvm::UTF32 CodePoint = 0; - llvm::ConversionResult Result = llvm::convertUTF8Sequence( + const llvm::ConversionResult Result = llvm::convertUTF8Sequence( (const llvm::UTF8 **)&CurPtr, (const llvm::UTF8 *)Buffer.end(), &CodePoint, llvm::strictConversion); @@ -94,7 +94,7 @@ public: bool HandleComment(Preprocessor &PP, SourceRange Range) override { // FIXME: check that we are in a /* */ comment - StringRef Text = + const StringRef Text = Lexer::getSourceText(CharSourceRange::getCharRange(Range), PP.getSourceManager(), PP.getLangOpts()); @@ -124,7 +124,7 @@ void MisleadingBidirectionalCheck::registerPPCallbacks( void MisleadingBidirectionalCheck::check( const ast_matchers::MatchFinder::MatchResult &Result) { if (const auto *SL = Result.Nodes.getNodeAs<StringLiteral>("strlit")) { - StringRef Literal = SL->getBytes(); + const StringRef Literal = SL->getBytes(); if (containsMisleadingBidi(Literal, false)) diag(SL->getBeginLoc(), "string literal contains misleading " "bidirectional Unicode characters"); diff --git a/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.h b/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.h index ba895b9..50e70cb 100644 --- a/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.h +++ b/clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGBIDIRECTIONALCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGBIDIRECTIONALCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGBIDIRECTIONAL_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGBIDIRECTIONAL_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGBIDIRECTIONALCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGBIDIRECTIONAL_H diff --git a/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.cpp b/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.cpp index ce04fb6..3c8a03a 100644 --- a/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.cpp +++ b/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.cpp @@ -124,7 +124,7 @@ static bool hasRTLCharacters(StringRef Buffer) { const char *EndPtr = Buffer.end(); while (CurPtr < EndPtr) { llvm::UTF32 CodePoint = 0; - llvm::ConversionResult Result = llvm::convertUTF8Sequence( + const llvm::ConversionResult Result = llvm::convertUTF8Sequence( (const llvm::UTF8 **)&CurPtr, (const llvm::UTF8 *)EndPtr, &CodePoint, llvm::strictConversion); if (Result != llvm::conversionOK) @@ -144,9 +144,9 @@ MisleadingIdentifierCheck::~MisleadingIdentifierCheck() = default; void MisleadingIdentifierCheck::check( const ast_matchers::MatchFinder::MatchResult &Result) { if (const auto *ND = Result.Nodes.getNodeAs<NamedDecl>("nameddecl")) { - IdentifierInfo *II = ND->getIdentifier(); + const IdentifierInfo *II = ND->getIdentifier(); if (II) { - StringRef NDName = II->getName(); + const StringRef NDName = II->getName(); if (hasRTLCharacters(NDName)) diag(ND->getBeginLoc(), "identifier has right-to-left codepoints"); } diff --git a/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.h b/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.h index 06b83d5..5c184f8 100644 --- a/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.h +++ b/clang-tools-extra/clang-tidy/misc/MisleadingIdentifier.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGIDENTIFIERCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGIDENTIFIERCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGIDENTIFIER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGIDENTIFIER_H #include "../ClangTidyCheck.h" @@ -24,4 +24,4 @@ public: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGIDENTIFIERCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISLEADINGIDENTIFIER_H diff --git a/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.cpp b/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.cpp index afa59f3..c8c0cfd 100644 --- a/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.cpp @@ -40,7 +40,7 @@ static QualType guessAlternateQualification(ASTContext &Context, QualType QT) { Qualifiers Quals = QT.getLocalQualifiers(); Quals.removeConst(); - QualType NewQT = Context.getPointerType( + const QualType NewQT = Context.getPointerType( QualType(QT->getPointeeType().getTypePtr(), Qualifiers::Const)); return NewQT.withCVRQualifiers(Quals.getCVRQualifiers()); } @@ -48,7 +48,7 @@ static QualType guessAlternateQualification(ASTContext &Context, QualType QT) { void MisplacedConstCheck::check(const MatchFinder::MatchResult &Result) { const auto *Var = Result.Nodes.getNodeAs<ValueDecl>("decl"); ASTContext &Ctx = *Result.Context; - QualType CanQT = Var->getType().getCanonicalType(); + const QualType CanQT = Var->getType().getCanonicalType(); SourceLocation AliasLoc; const char *AliasType = nullptr; diff --git a/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.h b/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.h index 2b8a05d..5f5a4cfd 100644 --- a/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.h +++ b/clang-tools-extra/clang-tidy/misc/MisplacedConstCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISPLACED_CONST_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISPLACED_CONST_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISPLACEDCONSTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISPLACEDCONSTCHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISPLACED_CONST_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MISPLACEDCONSTCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.cpp b/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.cpp index 9801c9e..a44e9b3 100644 --- a/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.cpp @@ -51,7 +51,7 @@ AST_MATCHER(FunctionDecl, isPlacementOverload) { return true; const auto *FPT = Node.getType()->castAs<FunctionProtoType>(); - ASTContext &Ctx = Node.getASTContext(); + const ASTContext &Ctx = Node.getASTContext(); if (Ctx.getLangOpts().SizedDeallocation && ASTContext::hasSameType(FPT->getParamType(1), Ctx.getSizeType())) return false; diff --git a/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.h b/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.h index 93c39fc..9c7aff0 100644 --- a/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.h +++ b/clang-tools-extra/clang-tidy/misc/NewDeleteOverloadsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_NEWDELETEOVERLOADS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_NEWDELETEOVERLOADS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_NEWDELETEOVERLOADSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_NEWDELETEOVERLOADSCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/SmallVector.h" @@ -33,4 +33,4 @@ public: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_NEWDELETEOVERLOADS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_NEWDELETEOVERLOADSCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/NoRecursionCheck.cpp b/clang-tools-extra/clang-tidy/misc/NoRecursionCheck.cpp index 035598d..8bcbb61 100644 --- a/clang-tools-extra/clang-tidy/misc/NoRecursionCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/NoRecursionCheck.cpp @@ -122,7 +122,7 @@ private: } // Set time! // Note that this must be after `populateSet()` might have been called. - bool SetInsertionSucceeded = Set.insert(V).second; + const bool SetInsertionSucceeded = Set.insert(V).second; (void)SetInsertionSucceeded; assert(SetInsertionSucceeded && "We did check that no such value existed"); return true; @@ -132,7 +132,7 @@ public: /// Insert a new element into the SmartSmallSetVector. /// \returns true if the element was inserted into the SmartSmallSetVector. bool insert(const T &X) { - bool Result = setInsert(X); + const bool Result = setInsert(X); if (Result) Vector.push_back(X); return Result; @@ -200,8 +200,8 @@ void NoRecursionCheck::handleSCC(ArrayRef<CallGraphNode *> SCC) { assert(!SCC.empty() && "Empty SCC does not make sense."); // First of all, call out every strongly connected function. - for (CallGraphNode *N : SCC) { - FunctionDecl *D = N->getDefinition(); + for (const CallGraphNode *N : SCC) { + const FunctionDecl *D = N->getDefinition(); diag(D->getLocation(), "function %0 is within a recursive call chain") << D; } @@ -224,7 +224,8 @@ void NoRecursionCheck::handleSCC(ArrayRef<CallGraphNode *> SCC) { assert(CyclicCallStack.size() >= 2 && "Cycle requires at least 2 frames"); // Which function we decided to be the entry point that lead to the recursion? - FunctionDecl *CycleEntryFn = CyclicCallStack.front().Callee->getDefinition(); + const FunctionDecl *CycleEntryFn = + CyclicCallStack.front().Callee->getDefinition(); // And now, for ease of understanding, let's print the call sequence that // forms the cycle in question. diag(CycleEntryFn->getLocation(), @@ -233,8 +234,8 @@ void NoRecursionCheck::handleSCC(ArrayRef<CallGraphNode *> SCC) { << CycleEntryFn; for (int CurFrame = 1, NumFrames = CyclicCallStack.size(); CurFrame != NumFrames; ++CurFrame) { - CallGraphNode::CallRecord PrevNode = CyclicCallStack[CurFrame - 1]; - CallGraphNode::CallRecord CurrNode = CyclicCallStack[CurFrame]; + const CallGraphNode::CallRecord PrevNode = CyclicCallStack[CurFrame - 1]; + const CallGraphNode::CallRecord CurrNode = CyclicCallStack[CurFrame]; Decl *PrevDecl = PrevNode.Callee->getDecl(); Decl *CurrDecl = CurrNode.Callee->getDecl(); diff --git a/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.cpp b/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.cpp index 6baa12a..c8ae41c 100644 --- a/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.cpp @@ -77,8 +77,8 @@ static bool areEquivalentExpr(const Expr *Left, const Expr *Right) { return cast<CharacterLiteral>(Left)->getValue() == cast<CharacterLiteral>(Right)->getValue(); case Stmt::IntegerLiteralClass: { - llvm::APInt LeftLit = cast<IntegerLiteral>(Left)->getValue(); - llvm::APInt RightLit = cast<IntegerLiteral>(Right)->getValue(); + const llvm::APInt LeftLit = cast<IntegerLiteral>(Left)->getValue(); + const llvm::APInt RightLit = cast<IntegerLiteral>(Right)->getValue(); return LeftLit.getBitWidth() == RightLit.getBitWidth() && LeftLit == RightLit; } @@ -256,7 +256,7 @@ static bool rangeSubsumesRange(BinaryOperatorKind OpcodeLHS, const APSInt &ValueLHS, BinaryOperatorKind OpcodeRHS, const APSInt &ValueRHS) { - int Comparison = APSInt::compareValues(ValueLHS, ValueRHS); + const int Comparison = APSInt::compareValues(ValueLHS, ValueRHS); switch (OpcodeLHS) { case BO_EQ: return OpcodeRHS == BO_EQ && Comparison == 0; @@ -352,11 +352,11 @@ static bool hasSameOperatorParent(const Expr *TheExpr, ASTContext &Context) { // IgnoreParenImpCasts logic in reverse: skip surrounding uninteresting nodes const DynTypedNodeList Parents = Context.getParents(*TheExpr); - for (DynTypedNode DynParent : Parents) { + for (const DynTypedNode DynParent : Parents) { if (const auto *Parent = DynParent.get<Expr>()) { - bool Skip = isa<ParenExpr>(Parent) || isa<ImplicitCastExpr>(Parent) || - isa<FullExpr>(Parent) || - isa<MaterializeTemporaryExpr>(Parent); + const bool Skip = + isa<ParenExpr>(Parent) || isa<ImplicitCastExpr>(Parent) || + isa<FullExpr>(Parent) || isa<MaterializeTemporaryExpr>(Parent); if (Skip && hasSameOperatorParent<TExpr>(Parent, OpKind, Context)) return true; if (checkOpKind<TExpr>(Parent, OpKind)) @@ -392,7 +392,7 @@ markDuplicateOperands(const TExpr *TheExpr, return false; if (collectOperands<TExpr>(Operands.second, AllOperands, OpKind)) return false; - size_t NumOperands = AllOperands.size(); + const size_t NumOperands = AllOperands.size(); llvm::SmallBitVector Duplicates(NumOperands); for (size_t I = 0; I < NumOperands; I++) { if (Duplicates[I]) @@ -463,7 +463,7 @@ AST_MATCHER_P(Expr, expandedByMacro, ArrayRef<llvm::StringLiteral>, Names) { const LangOptions &LO = Finder->getASTContext().getLangOpts(); SourceLocation Loc = Node.getExprLoc(); while (Loc.isMacroID()) { - StringRef MacroName = Lexer::getImmediateMacroName(Loc, SM, LO); + const StringRef MacroName = Lexer::getImmediateMacroName(Loc, SM, LO); if (llvm::is_contained(Names, MacroName)) return true; Loc = SM.getImmediateMacroCallerLoc(Loc); @@ -476,7 +476,7 @@ AST_MATCHER_P(Expr, expandedByMacro, ArrayRef<llvm::StringLiteral>, Names) { // Returns a matcher for integer constant expressions. static ast_matchers::internal::Matcher<Expr> matchIntegerConstantExpr(StringRef Id) { - std::string CstId = (Id + "-const").str(); + const std::string CstId = (Id + "-const").str(); return expr(isIntegerConstantExpr()).bind(CstId); } @@ -486,7 +486,7 @@ matchIntegerConstantExpr(StringRef Id) { static bool retrieveIntegerConstantExpr(const MatchFinder::MatchResult &Result, StringRef Id, APSInt &Value, const Expr *&ConstExpr) { - std::string CstId = (Id + "-const").str(); + const std::string CstId = (Id + "-const").str(); ConstExpr = Result.Nodes.getNodeAs<Expr>(CstId); if (!ConstExpr) return false; @@ -508,7 +508,7 @@ static bool retrieveIntegerConstantExpr(const MatchFinder::MatchResult &Result, // Returns a matcher for symbolic expressions (matches every expression except // ingeter constant expressions). static ast_matchers::internal::Matcher<Expr> matchSymbolicExpr(StringRef Id) { - std::string SymId = (Id + "-sym").str(); + const std::string SymId = (Id + "-sym").str(); return ignoringParenImpCasts( expr(unless(isIntegerConstantExpr())).bind(SymId)); } @@ -517,7 +517,7 @@ static ast_matchers::internal::Matcher<Expr> matchSymbolicExpr(StringRef Id) { // stores it into 'SymExpr'. static bool retrieveSymbolicExpr(const MatchFinder::MatchResult &Result, StringRef Id, const Expr *&SymExpr) { - std::string SymId = (Id + "-sym").str(); + const std::string SymId = (Id + "-sym").str(); if (const auto *Node = Result.Nodes.getNodeAs<Expr>(SymId)) { SymExpr = Node; return true; @@ -557,11 +557,11 @@ retrieveBinOpIntegerConstantExpr(const MatchFinder::MatchResult &Result, // Matches relational expressions: 'Expr <op> k' (i.e. x < 2, x != 3, 12 <= x). static ast_matchers::internal::Matcher<Expr> matchRelationalIntegerConstantExpr(StringRef Id) { - std::string CastId = (Id + "-cast").str(); - std::string SwapId = (Id + "-swap").str(); - std::string NegateId = (Id + "-negate").str(); - std::string OverloadId = (Id + "-overload").str(); - std::string ConstId = (Id + "-const").str(); + const std::string CastId = (Id + "-cast").str(); + const std::string SwapId = (Id + "-swap").str(); + const std::string NegateId = (Id + "-negate").str(); + const std::string OverloadId = (Id + "-overload").str(); + const std::string ConstId = (Id + "-const").str(); const auto RelationalExpr = ignoringParenImpCasts(binaryOperator( isComparisonOperator(), expr().bind(Id), @@ -625,7 +625,7 @@ canOverloadedOperatorArgsBeModified(const CXXOperatorCallExpr *OperatorCall, if (!OperatorDecl) return true; - unsigned ParamCount = OperatorDecl->getNumParams(); + const unsigned ParamCount = OperatorDecl->getNumParams(); // Overloaded operators declared inside a class have only one param. // These functions must be declared const in order to not be able to modify @@ -647,10 +647,10 @@ static bool retrieveRelationalIntegerConstantExpr( const MatchFinder::MatchResult &Result, StringRef Id, const Expr *&OperandExpr, BinaryOperatorKind &Opcode, const Expr *&Symbol, APSInt &Value, const Expr *&ConstExpr) { - std::string CastId = (Id + "-cast").str(); - std::string SwapId = (Id + "-swap").str(); - std::string NegateId = (Id + "-negate").str(); - std::string OverloadId = (Id + "-overload").str(); + const std::string CastId = (Id + "-cast").str(); + const std::string SwapId = (Id + "-swap").str(); + const std::string NegateId = (Id + "-negate").str(); + const std::string OverloadId = (Id + "-overload").str(); if (const auto *Bin = Result.Nodes.getNodeAs<BinaryOperator>(Id)) { // Operand received with explicit comparator. @@ -829,11 +829,11 @@ static bool areExprsFromDifferentMacros(const Expr *LhsExpr, const SourceManager &SM = AstCtx->getSourceManager(); const LangOptions &LO = AstCtx->getLangOpts(); - std::pair<FileID, unsigned> LsrLocInfo = + const std::pair<FileID, unsigned> LsrLocInfo = SM.getDecomposedLoc(SM.getExpansionLoc(Lsr.getBegin())); - std::pair<FileID, unsigned> RsrLocInfo = + const std::pair<FileID, unsigned> RsrLocInfo = SM.getDecomposedLoc(SM.getExpansionLoc(Rsr.getBegin())); - llvm::MemoryBufferRef MB = SM.getBufferOrFake(LsrLocInfo.first); + const llvm::MemoryBufferRef MB = SM.getBufferOrFake(LsrLocInfo.first); const char *LTokenPos = MB.getBufferStart() + LsrLocInfo.second; const char *RTokenPos = MB.getBufferStart() + RsrLocInfo.second; @@ -1097,7 +1097,7 @@ void RedundantExpressionCheck::checkArithmeticExpr( if (const auto *ComparisonOperator = Result.Nodes.getNodeAs<BinaryOperator>( "binop-const-compare-to-sym")) { - BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); + const BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); if (!retrieveBinOpIntegerConstantExpr(Result, "lhs", LhsOpcode, LhsSymbol, LhsValue) || !retrieveSymbolicExpr(Result, "rhs", RhsSymbol) || @@ -1118,7 +1118,7 @@ void RedundantExpressionCheck::checkArithmeticExpr( } else if (const auto *ComparisonOperator = Result.Nodes.getNodeAs<BinaryOperator>( "binop-const-compare-to-binop-const")) { - BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); + const BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); if (!retrieveBinOpIntegerConstantExpr(Result, "lhs", LhsOpcode, LhsSymbol, LhsValue) || @@ -1165,7 +1165,7 @@ void RedundantExpressionCheck::checkBitwiseExpr( const MatchFinder::MatchResult &Result) { if (const auto *ComparisonOperator = Result.Nodes.getNodeAs<BinaryOperator>( "binop-const-compare-to-const")) { - BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); + const BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); APSInt LhsValue, RhsValue; const Expr *LhsSymbol = nullptr; @@ -1175,9 +1175,9 @@ void RedundantExpressionCheck::checkBitwiseExpr( !retrieveIntegerConstantExpr(Result, "rhs", RhsValue)) return; - uint64_t LhsConstant = LhsValue.getZExtValue(); - uint64_t RhsConstant = RhsValue.getZExtValue(); - SourceLocation Loc = ComparisonOperator->getOperatorLoc(); + const uint64_t LhsConstant = LhsValue.getZExtValue(); + const uint64_t RhsConstant = RhsValue.getZExtValue(); + const SourceLocation Loc = ComparisonOperator->getOperatorLoc(); // Check expression: x & k1 == k2 (i.e. x & 0xFF == 0xF00) if (LhsOpcode == BO_And && (LhsConstant & RhsConstant) != RhsConstant) { @@ -1208,24 +1208,24 @@ void RedundantExpressionCheck::checkBitwiseExpr( if ((Value != 0 && ~Value != 0) || Sym->getExprLoc().isMacroID()) return; - SourceLocation Loc = IneffectiveOperator->getOperatorLoc(); + const SourceLocation Loc = IneffectiveOperator->getOperatorLoc(); - BinaryOperatorKind Opcode = IneffectiveOperator->getOpcode(); + const BinaryOperatorKind Opcode = IneffectiveOperator->getOpcode(); if (exprEvaluatesToZero(Opcode, Value)) { diag(Loc, "expression always evaluates to 0"); } else if (exprEvaluatesToBitwiseNegatedZero(Opcode, Value)) { - SourceRange ConstExprRange(ConstExpr->getBeginLoc(), - ConstExpr->getEndLoc()); - StringRef ConstExprText = Lexer::getSourceText( + const SourceRange ConstExprRange(ConstExpr->getBeginLoc(), + ConstExpr->getEndLoc()); + const StringRef ConstExprText = Lexer::getSourceText( CharSourceRange::getTokenRange(ConstExprRange), *Result.SourceManager, Result.Context->getLangOpts()); diag(Loc, "expression always evaluates to '%0'") << ConstExprText; } else if (exprEvaluatesToSymbolic(Opcode, Value)) { - SourceRange SymExprRange(Sym->getBeginLoc(), Sym->getEndLoc()); + const SourceRange SymExprRange(Sym->getBeginLoc(), Sym->getEndLoc()); - StringRef ExprText = Lexer::getSourceText( + const StringRef ExprText = Lexer::getSourceText( CharSourceRange::getTokenRange(SymExprRange), *Result.SourceManager, Result.Context->getLangOpts()); @@ -1240,7 +1240,7 @@ void RedundantExpressionCheck::checkRelationalExpr( "comparisons-of-symbol-and-const")) { // Matched expressions are: (x <op> k1) <REL> (x <op> k2). // E.g.: (X < 2) && (X > 4) - BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); + const BinaryOperatorKind Opcode = ComparisonOperator->getOpcode(); const Expr *LhsExpr = nullptr, *RhsExpr = nullptr; const Expr *LhsSymbol = nullptr, *RhsSymbol = nullptr; @@ -1392,7 +1392,7 @@ void RedundantExpressionCheck::check(const MatchFinder::MatchResult &Result) { if (Call && canOverloadedOperatorArgsBeModified(Call, true)) return; - StringRef Message = + const StringRef Message = Call ? "overloaded operator has equivalent nested operands" : "operator has equivalent nested operands"; @@ -1405,12 +1405,12 @@ void RedundantExpressionCheck::check(const MatchFinder::MatchResult &Result) { if (const auto *NegateOperator = Result.Nodes.getNodeAs<UnaryOperator>("logical-bitwise-confusion")) { - SourceLocation OperatorLoc = NegateOperator->getOperatorLoc(); + const SourceLocation OperatorLoc = NegateOperator->getOperatorLoc(); auto Diag = diag(OperatorLoc, "ineffective logical negation operator used; did you mean '~'?"); - SourceLocation LogicalNotLocation = OperatorLoc.getLocWithOffset(1); + const SourceLocation LogicalNotLocation = OperatorLoc.getLocWithOffset(1); if (!LogicalNotLocation.isMacroID()) Diag << FixItHint::CreateReplacement( diff --git a/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.h b/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.h index 57289c39..f127007 100644 --- a/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.h +++ b/clang-tools-extra/clang-tidy/misc/RedundantExpressionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_REDUNDANT_EXPRESSION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_REDUNDANT_EXPRESSION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_REDUNDANTEXPRESSIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_REDUNDANTEXPRESSIONCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_REDUNDANT_EXPRESSION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_REDUNDANTEXPRESSIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/StaticAssertCheck.cpp b/clang-tools-extra/clang-tidy/misc/StaticAssertCheck.cpp index 5ac5300..ff866e9 100644 --- a/clang-tools-extra/clang-tidy/misc/StaticAssertCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/StaticAssertCheck.cpp @@ -84,12 +84,12 @@ void StaticAssertCheck::check(const MatchFinder::MatchResult &Result) { const auto *AssertExprRoot = Result.Nodes.getNodeAs<BinaryOperator>("assertExprRoot"); const auto *CastExpr = Result.Nodes.getNodeAs<CStyleCastExpr>("castExpr"); - SourceLocation AssertExpansionLoc = CondStmt->getBeginLoc(); + const SourceLocation AssertExpansionLoc = CondStmt->getBeginLoc(); if (!AssertExpansionLoc.isValid() || !AssertExpansionLoc.isMacroID()) return; - StringRef MacroName = + const StringRef MacroName = Lexer::getImmediateMacroName(AssertExpansionLoc, SM, Opts); if (MacroName != "assert" || Condition->isValueDependent() || @@ -99,19 +99,20 @@ void StaticAssertCheck::check(const MatchFinder::MatchResult &Result) { // False literal is not the result of macro expansion. if (IsAlwaysFalse && (!CastExpr || CastExpr->getType()->isPointerType())) { - SourceLocation FalseLiteralLoc = + const SourceLocation FalseLiteralLoc = SM.getImmediateSpellingLoc(IsAlwaysFalse->getExprLoc()); if (!FalseLiteralLoc.isMacroID()) return; - StringRef FalseMacroName = + const StringRef FalseMacroName = Lexer::getImmediateMacroName(FalseLiteralLoc, SM, Opts); if (FalseMacroName.compare_insensitive("false") == 0 || FalseMacroName.compare_insensitive("null") == 0) return; } - SourceLocation AssertLoc = SM.getImmediateMacroCallerLoc(AssertExpansionLoc); + const SourceLocation AssertLoc = + SM.getImmediateMacroCallerLoc(AssertExpansionLoc); SmallVector<FixItHint, 4> FixItHints; SourceLocation LastParenLoc; diff --git a/clang-tools-extra/clang-tidy/misc/ThrowByValueCatchByReferenceCheck.h b/clang-tools-extra/clang-tidy/misc/ThrowByValueCatchByReferenceCheck.h index 15c17e7..56e4c12 100644 --- a/clang-tools-extra/clang-tidy/misc/ThrowByValueCatchByReferenceCheck.h +++ b/clang-tools-extra/clang-tidy/misc/ThrowByValueCatchByReferenceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_THROW_BY_VALUE_CATCH_BY_REFERENCE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_THROW_BY_VALUE_CATCH_BY_REFERENCE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_THROWBYVALUECATCHBYREFERENCECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_THROWBYVALUECATCHBYREFERENCECHECK_H #include "../ClangTidyCheck.h" @@ -49,4 +49,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_THROW_BY_VALUE_CATCH_BY_REFERENCE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_THROWBYVALUECATCHBYREFERENCECHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.cpp b/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.cpp index 8a85e79..a14a559 100644 --- a/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.cpp @@ -18,7 +18,7 @@ namespace { AST_MATCHER_P(CXXMethodDecl, firstParameter, ast_matchers::internal::Matcher<ParmVarDecl>, InnerMatcher) { - unsigned N = Node.isExplicitObjectMemberFunction() ? 1 : 0; + const unsigned N = Node.isExplicitObjectMemberFunction() ? 1 : 0; return (N < Node.parameters().size() && InnerMatcher.matches(*Node.parameters()[N], Finder, Builder)); } diff --git a/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.h b/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.h index be9e7b9..941fe72 100644 --- a/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.h +++ b/clang-tools-extra/clang-tidy/misc/UnconventionalAssignOperatorCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_ASSIGNOPERATORSIGNATURECHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_ASSIGNOPERATORSIGNATURECHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNCONVENTIONALASSIGNOPERATORCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNCONVENTIONALASSIGNOPERATORCHECK_H #include "../ClangTidyCheck.h" @@ -37,4 +37,4 @@ public: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_ASSIGNOPERATORSIGNATURECHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNCONVENTIONALASSIGNOPERATORCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/UnusedAliasDeclsCheck.h b/clang-tools-extra/clang-tidy/misc/UnusedAliasDeclsCheck.h index ffe82ca..b9d85c1 100644 --- a/clang-tools-extra/clang-tidy/misc/UnusedAliasDeclsCheck.h +++ b/clang-tools-extra/clang-tidy/misc/UnusedAliasDeclsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_ALIAS_DECLS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_ALIAS_DECLS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDALIASDECLSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDALIASDECLSCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/DenseMap.h" @@ -32,4 +32,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_ALIAS_DECLS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDALIASDECLSCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.cpp b/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.cpp index f2189f5..ae08096 100644 --- a/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.cpp @@ -165,7 +165,7 @@ void UnusedParametersCheck::warnOnUnusedParameter( if (!Result.Context->getLangOpts().CPlusPlus) return; - SourceRange RemovalRange(Param->getLocation()); + const SourceRange RemovalRange(Param->getLocation()); // Note: We always add a space before the '/*' to not accidentally create // a '*/*' for pointer types, which doesn't start a comment. clang-format // will clean this up afterwards. diff --git a/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.h b/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.h index 877fc4d..fe2cc6e 100644 --- a/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.h +++ b/clang-tools-extra/clang-tidy/misc/UnusedParametersCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_PARAMETERS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_PARAMETERS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDPARAMETERSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDPARAMETERSCHECK_H #include "../ClangTidyCheck.h" @@ -36,4 +36,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_PARAMETERS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDPARAMETERSCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/UnusedUsingDeclsCheck.h b/clang-tools-extra/clang-tidy/misc/UnusedUsingDeclsCheck.h index 96d8d9d..986bf37 100644 --- a/clang-tools-extra/clang-tidy/misc/UnusedUsingDeclsCheck.h +++ b/clang-tools-extra/clang-tidy/misc/UnusedUsingDeclsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_USING_DECLS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_USING_DECLS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDUSINGDECLSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDUSINGDECLSCHECK_H #include "../ClangTidyCheck.h" #include "../utils/FileExtensionsUtils.h" @@ -56,4 +56,4 @@ private: } // namespace clang::tidy::misc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSED_USING_DECLS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_UNUSEDUSINGDECLSCHECK_H diff --git a/clang-tools-extra/clang-tidy/misc/UseAnonymousNamespaceCheck.cpp b/clang-tools-extra/clang-tidy/misc/UseAnonymousNamespaceCheck.cpp index aa0cc1e..ed36d48 100644 --- a/clang-tools-extra/clang-tidy/misc/UseAnonymousNamespaceCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/UseAnonymousNamespaceCheck.cpp @@ -52,7 +52,8 @@ void UseAnonymousNamespaceCheck::registerMatchers(MatchFinder *Finder) { void UseAnonymousNamespaceCheck::check(const MatchFinder::MatchResult &Result) { if (const auto *MatchedDecl = Result.Nodes.getNodeAs<NamedDecl>("x")) { - StringRef Type = llvm::isa<VarDecl>(MatchedDecl) ? "variable" : "function"; + const StringRef Type = + llvm::isa<VarDecl>(MatchedDecl) ? "variable" : "function"; diag(MatchedDecl->getLocation(), "%0 %1 declared 'static', move to anonymous namespace instead") << Type << MatchedDecl; diff --git a/clang-tools-extra/clang-tidy/misc/UseInternalLinkageCheck.cpp b/clang-tools-extra/clang-tidy/misc/UseInternalLinkageCheck.cpp index 1d27064..bad51c60 100644 --- a/clang-tools-extra/clang-tidy/misc/UseInternalLinkageCheck.cpp +++ b/clang-tools-extra/clang-tidy/misc/UseInternalLinkageCheck.cpp @@ -142,7 +142,8 @@ static constexpr StringRef Message = void UseInternalLinkageCheck::check(const MatchFinder::MatchResult &Result) { if (const auto *FD = Result.Nodes.getNodeAs<FunctionDecl>("fn")) { - DiagnosticBuilder DB = diag(FD->getLocation(), Message) << "function" << FD; + const DiagnosticBuilder DB = diag(FD->getLocation(), Message) + << "function" << FD; const SourceLocation FixLoc = FD->getInnerLocStart(); if (FixLoc.isInvalid() || FixLoc.isMacroID()) return; @@ -157,7 +158,8 @@ void UseInternalLinkageCheck::check(const MatchFinder::MatchResult &Result) { if (getLangOpts().CPlusPlus && VD->getType().isConstQualified()) return; - DiagnosticBuilder DB = diag(VD->getLocation(), Message) << "variable" << VD; + const DiagnosticBuilder DB = diag(VD->getLocation(), Message) + << "variable" << VD; const SourceLocation FixLoc = VD->getInnerLocStart(); if (FixLoc.isInvalid() || FixLoc.isMacroID()) return; diff --git a/clang-tools-extra/clang-tidy/modernize/AvoidBindCheck.h b/clang-tools-extra/clang-tidy/modernize/AvoidBindCheck.h index 94838cb..22e629f 100644 --- a/clang-tools-extra/clang-tidy/modernize/AvoidBindCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/AvoidBindCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_AVOID_BIND_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_AVOID_BIND_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_AVOIDBINDCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_AVOIDBINDCHECK_H #include "../ClangTidyCheck.h" @@ -34,4 +34,4 @@ private: }; } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_AVOID_BIND_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_AVOIDBINDCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/DeprecatedHeadersCheck.h b/clang-tools-extra/clang-tidy/modernize/DeprecatedHeadersCheck.h index badb2b4..015404e 100644 --- a/clang-tools-extra/clang-tidy/modernize/DeprecatedHeadersCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/DeprecatedHeadersCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_C_HEADERS_TO_CXX_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_C_HEADERS_TO_CXX_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_DEPRECATEDHEADERSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_DEPRECATEDHEADERSCHECK_H #include "../ClangTidyCheck.h" @@ -57,4 +57,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_C_HEADERS_TO_CXX_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_DEPRECATEDHEADERSCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/IntegralLiteralExpressionMatcher.h b/clang-tools-extra/clang-tidy/modernize/IntegralLiteralExpressionMatcher.h index d495087..ce0d7c0 100644 --- a/clang-tools-extra/clang-tidy/modernize/IntegralLiteralExpressionMatcher.h +++ b/clang-tools-extra/clang-tidy/modernize/IntegralLiteralExpressionMatcher.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_INTEGRAL_LITERAL_EXPRESSION_MATCHER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_INTEGRAL_LITERAL_EXPRESSION_MATCHER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_INTEGRALLITERALEXPRESSIONMATCHER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_INTEGRALLITERALEXPRESSIONMATCHER_H #include <clang/Lex/Token.h> #include <llvm/ADT/ArrayRef.h> @@ -73,4 +73,4 @@ private: } // namespace clang::tidy::modernize -#endif +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_INTEGRALLITERALEXPRESSIONMATCHER_H diff --git a/clang-tools-extra/clang-tidy/modernize/LoopConvertCheck.h b/clang-tools-extra/clang-tidy/modernize/LoopConvertCheck.h index 5548782..958b4eb 100644 --- a/clang-tools-extra/clang-tidy/modernize/LoopConvertCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/LoopConvertCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOP_CONVERT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOP_CONVERT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOPCONVERTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOPCONVERTCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -85,4 +85,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOP_CONVERT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOPCONVERTCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/LoopConvertUtils.h b/clang-tools-extra/clang-tidy/modernize/LoopConvertUtils.h index 0a0db5e..5d0800d 100644 --- a/clang-tools-extra/clang-tidy/modernize/LoopConvertUtils.h +++ b/clang-tools-extra/clang-tidy/modernize/LoopConvertUtils.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOP_CONVERT_UTILS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOP_CONVERT_UTILS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOPCONVERTUTILS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOPCONVERTUTILS_H #include "clang/AST/ASTContext.h" #include "clang/AST/RecursiveASTVisitor.h" @@ -466,4 +466,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOP_CONVERT_UTILS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_LOOPCONVERTUTILS_H diff --git a/clang-tools-extra/clang-tidy/modernize/MakeSharedCheck.h b/clang-tools-extra/clang-tidy/modernize/MakeSharedCheck.h index 063b35fc..4b7f625 100644 --- a/clang-tools-extra/clang-tidy/modernize/MakeSharedCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/MakeSharedCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_SHARED_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_SHARED_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKESHAREDCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKESHAREDCHECK_H #include "MakeSmartPtrCheck.h" @@ -35,4 +35,4 @@ protected: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_SHARED_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKESHAREDCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/MakeSmartPtrCheck.h b/clang-tools-extra/clang-tidy/modernize/MakeSmartPtrCheck.h index 28d5b45..1d70f62 100644 --- a/clang-tools-extra/clang-tidy/modernize/MakeSmartPtrCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/MakeSmartPtrCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_SMART_PTR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_SMART_PTR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKESMARTPTRCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKESMARTPTRCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -64,4 +64,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_SMART_PTR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKESMARTPTRCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/MakeUniqueCheck.h b/clang-tools-extra/clang-tidy/modernize/MakeUniqueCheck.h index 9c4f6bc..170343b9 100644 --- a/clang-tools-extra/clang-tidy/modernize/MakeUniqueCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/MakeUniqueCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_UNIQUE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_UNIQUE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKEUNIQUECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKEUNIQUECHECK_H #include "MakeSmartPtrCheck.h" @@ -37,4 +37,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKE_UNIQUE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MAKEUNIQUECHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/PassByValueCheck.h b/clang-tools-extra/clang-tidy/modernize/PassByValueCheck.h index f27871c..eb51f4a 100644 --- a/clang-tools-extra/clang-tidy/modernize/PassByValueCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/PassByValueCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_PASS_BY_VALUE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_PASS_BY_VALUE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_PASSBYVALUECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_PASSBYVALUECHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -33,4 +33,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_PASS_BY_VALUE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_PASSBYVALUECHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/RawStringLiteralCheck.h b/clang-tools-extra/clang-tidy/modernize/RawStringLiteralCheck.h index 8ce6ec0..5be38dd 100644 --- a/clang-tools-extra/clang-tidy/modernize/RawStringLiteralCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/RawStringLiteralCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RAW_STRING_LITERAL_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RAW_STRING_LITERAL_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RAWSTRINGLITERALCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RAWSTRINGLITERALCHECK_H #include "../ClangTidyCheck.h" #include <bitset> @@ -40,4 +40,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RAW_STRING_LITERAL_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RAWSTRINGLITERALCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/RedundantVoidArgCheck.h b/clang-tools-extra/clang-tidy/modernize/RedundantVoidArgCheck.h index 53de74b..d6edd99 100644 --- a/clang-tools-extra/clang-tidy/modernize/RedundantVoidArgCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/RedundantVoidArgCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REDUNDANT_VOID_ARG_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REDUNDANT_VOID_ARG_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REDUNDANTVOIDARGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REDUNDANTVOIDARGCHECK_H #include "../ClangTidyCheck.h" #include "clang/Lex/Token.h" @@ -73,4 +73,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REDUNDANT_VOID_ARG_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REDUNDANTVOIDARGCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/ReplaceAutoPtrCheck.h b/clang-tools-extra/clang-tidy/modernize/ReplaceAutoPtrCheck.h index 9a6e2bb..18f4740 100644 --- a/clang-tools-extra/clang-tidy/modernize/ReplaceAutoPtrCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/ReplaceAutoPtrCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACE_AUTO_PTR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACE_AUTO_PTR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACEAUTOPTRCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACEAUTOPTRCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -56,4 +56,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACE_AUTO_PTR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACEAUTOPTRCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/ReplaceRandomShuffleCheck.h b/clang-tools-extra/clang-tidy/modernize/ReplaceRandomShuffleCheck.h index 5f2be10..3ffa387 100644 --- a/clang-tools-extra/clang-tidy/modernize/ReplaceRandomShuffleCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/ReplaceRandomShuffleCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACE_RANDOM_SHUFFLE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACE_RANDOM_SHUFFLE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACERANDOMSHUFFLECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACERANDOMSHUFFLECHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -37,4 +37,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACE_RANDOM_SHUFFLE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_REPLACERANDOMSHUFFLECHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/ReturnBracedInitListCheck.h b/clang-tools-extra/clang-tidy/modernize/ReturnBracedInitListCheck.h index ef465ea..be78571 100644 --- a/clang-tools-extra/clang-tidy/modernize/ReturnBracedInitListCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/ReturnBracedInitListCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RETURN_BRACED_INIT_LIST_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RETURN_BRACED_INIT_LIST_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RETURNBRACEDINITLISTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RETURNBRACEDINITLISTCHECK_H #include "../ClangTidyCheck.h" @@ -34,4 +34,4 @@ public: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RETURN_BRACED_INIT_LIST_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_RETURNBRACEDINITLISTCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UnaryStaticAssertCheck.h b/clang-tools-extra/clang-tidy/modernize/UnaryStaticAssertCheck.h index 95611c9..ebe77b9 100644 --- a/clang-tools-extra/clang-tidy/modernize/UnaryStaticAssertCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UnaryStaticAssertCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_UNARY_STATIC_ASSERT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_UNARY_STATIC_ASSERT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_UNARYSTATICASSERTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_UNARYSTATICASSERTCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ public: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_UNARY_STATIC_ASSERT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_UNARYSTATICASSERTCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseAutoCheck.h b/clang-tools-extra/clang-tidy/modernize/UseAutoCheck.h index dc39077..85e87fe 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseAutoCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseAutoCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_AUTO_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_AUTO_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEAUTOCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEAUTOCHECK_H #include "../ClangTidyCheck.h" @@ -35,4 +35,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_AUTO_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEAUTOCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseBoolLiteralsCheck.h b/clang-tools-extra/clang-tidy/modernize/UseBoolLiteralsCheck.h index 64aff84..95bce07 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseBoolLiteralsCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseBoolLiteralsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_BOOL_LITERALS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_BOOL_LITERALS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEBOOLLITERALSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEBOOLLITERALSCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_BOOL_LITERALS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEBOOLLITERALSCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseDefaultMemberInitCheck.h b/clang-tools-extra/clang-tidy/modernize/UseDefaultMemberInitCheck.h index be6a18a..f379214 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseDefaultMemberInitCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseDefaultMemberInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_DEFAULT_MEMBER_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_DEFAULT_MEMBER_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEDEFAULTMEMBERINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEDEFAULTMEMBERINITCHECK_H #include "../ClangTidyCheck.h" @@ -44,4 +44,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_DEFAULT_MEMBER_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEDEFAULTMEMBERINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseEmplaceCheck.h b/clang-tools-extra/clang-tidy/modernize/UseEmplaceCheck.h index 87ebf6f..a7ad5bb 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseEmplaceCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseEmplaceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EMPLACE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EMPLACE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEMPLACECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEMPLACECHECK_H #include "../ClangTidyCheck.h" #include <string> @@ -45,4 +45,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EMPLACE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEMPLACECHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseEqualsDefaultCheck.h b/clang-tools-extra/clang-tidy/modernize/UseEqualsDefaultCheck.h index 519f189..a17d3d8 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseEqualsDefaultCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseEqualsDefaultCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EQUALS_DEFAULT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EQUALS_DEFAULT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEQUALSDEFAULTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEQUALSDEFAULTCHECK_H #include "../ClangTidyCheck.h" @@ -48,4 +48,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EQUALS_DEFAULT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEQUALSDEFAULTCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseEqualsDeleteCheck.h b/clang-tools-extra/clang-tidy/modernize/UseEqualsDeleteCheck.h index 31a956b..17155fe 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseEqualsDeleteCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseEqualsDeleteCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EQUALS_DELETE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EQUALS_DELETE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEQUALSDELETECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEQUALSDELETECHECK_H #include "../ClangTidyCheck.h" @@ -38,4 +38,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_EQUALS_DELETE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEEQUALSDELETECHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseNoexceptCheck.h b/clang-tools-extra/clang-tidy/modernize/UseNoexceptCheck.h index 30b5d4e..a97b39b 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseNoexceptCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseNoexceptCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_NOEXCEPT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_NOEXCEPT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USENOEXCEPTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USENOEXCEPTCHECK_H #include "../ClangTidyCheck.h" @@ -44,4 +44,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_NOEXCEPT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USENOEXCEPTCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h b/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h index 7c7b5ae..1caa07a 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_NULLPTR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_NULLPTR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USENULLPTRCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USENULLPTRCHECK_H #include "../ClangTidyCheck.h" @@ -31,4 +31,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_NULLPTR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USENULLPTRCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseStdPrintCheck.h b/clang-tools-extra/clang-tidy/modernize/UseStdPrintCheck.h index f5b3f71..18cff9a 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseStdPrintCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseStdPrintCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_USESTDPRINTCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_USESTDPRINTCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USESTDPRINTCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USESTDPRINTCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -48,4 +48,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_USESTDPRINTCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USESTDPRINTCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseTransparentFunctorsCheck.h b/clang-tools-extra/clang-tidy/modernize/UseTransparentFunctorsCheck.h index 0af729b..936eaf1 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseTransparentFunctorsCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseTransparentFunctorsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_TRANSPARENT_FUNCTORS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_TRANSPARENT_FUNCTORS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USETRANSPARENTFUNCTORSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USETRANSPARENTFUNCTORSCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ private: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_TRANSPARENT_FUNCTORS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USETRANSPARENTFUNCTORSCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseUncaughtExceptionsCheck.h b/clang-tools-extra/clang-tidy/modernize/UseUncaughtExceptionsCheck.h index 772133d..09b0ba5 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseUncaughtExceptionsCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseUncaughtExceptionsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_UNCAUGHT_EXCEPTIONS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_UNCAUGHT_EXCEPTIONS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEUNCAUGHTEXCEPTIONSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEUNCAUGHTEXCEPTIONSCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ public: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_UNCAUGHT_EXCEPTIONS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEUNCAUGHTEXCEPTIONSCHECK_H diff --git a/clang-tools-extra/clang-tidy/modernize/UseUsingCheck.h b/clang-tools-extra/clang-tidy/modernize/UseUsingCheck.h index 5ecabc7..60813cd 100644 --- a/clang-tools-extra/clang-tidy/modernize/UseUsingCheck.h +++ b/clang-tools-extra/clang-tidy/modernize/UseUsingCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_USING_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_USING_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEUSINGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEUSINGCHECK_H #include "../ClangTidyCheck.h" @@ -40,4 +40,4 @@ public: } // namespace clang::tidy::modernize -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USE_USING_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_USEUSINGCHECK_H diff --git a/clang-tools-extra/clang-tidy/mpi/BufferDerefCheck.h b/clang-tools-extra/clang-tidy/mpi/BufferDerefCheck.h index 07ee68a..a44ef31 100644 --- a/clang-tools-extra/clang-tidy/mpi/BufferDerefCheck.h +++ b/clang-tools-extra/clang-tidy/mpi/BufferDerefCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_BUFFER_DEREF_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_BUFFER_DEREF_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_BUFFERDEREFCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_BUFFERDEREFCHECK_H #include "../ClangTidyCheck.h" #include "clang/StaticAnalyzer/Checkers/MPIFunctionClassifier.h" @@ -48,4 +48,4 @@ private: } // namespace clang::tidy::mpi -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_BUFFER_DEREF_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_BUFFERDEREFCHECK_H diff --git a/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.cpp b/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.cpp index 17c1283..370a54d 100644 --- a/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.cpp +++ b/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.cpp @@ -41,39 +41,39 @@ isMPITypeMatching(const std::multimap<BuiltinType::Kind, StringRef> &MultiMap, /// /// \returns true if the type is a standard type static bool isStandardMPIDatatype(StringRef MPIDatatype) { - static llvm::StringSet<> AllTypes = {"MPI_C_BOOL", - "MPI_CHAR", - "MPI_SIGNED_CHAR", - "MPI_UNSIGNED_CHAR", - "MPI_WCHAR", - "MPI_INT", - "MPI_LONG", - "MPI_SHORT", - "MPI_LONG_LONG", - "MPI_LONG_LONG_INT", - "MPI_UNSIGNED", - "MPI_UNSIGNED_SHORT", - "MPI_UNSIGNED_LONG", - "MPI_UNSIGNED_LONG_LONG", - "MPI_FLOAT", - "MPI_DOUBLE", - "MPI_LONG_DOUBLE", - "MPI_C_COMPLEX", - "MPI_C_FLOAT_COMPLEX", - "MPI_C_DOUBLE_COMPLEX", - "MPI_C_LONG_DOUBLE_COMPLEX", - "MPI_INT8_T", - "MPI_INT16_T", - "MPI_INT32_T", - "MPI_INT64_T", - "MPI_UINT8_T", - "MPI_UINT16_T", - "MPI_UINT32_T", - "MPI_UINT64_T", - "MPI_CXX_BOOL", - "MPI_CXX_FLOAT_COMPLEX", - "MPI_CXX_DOUBLE_COMPLEX", - "MPI_CXX_LONG_DOUBLE_COMPLEX"}; + static const llvm::StringSet<> AllTypes = {"MPI_C_BOOL", + "MPI_CHAR", + "MPI_SIGNED_CHAR", + "MPI_UNSIGNED_CHAR", + "MPI_WCHAR", + "MPI_INT", + "MPI_LONG", + "MPI_SHORT", + "MPI_LONG_LONG", + "MPI_LONG_LONG_INT", + "MPI_UNSIGNED", + "MPI_UNSIGNED_SHORT", + "MPI_UNSIGNED_LONG", + "MPI_UNSIGNED_LONG_LONG", + "MPI_FLOAT", + "MPI_DOUBLE", + "MPI_LONG_DOUBLE", + "MPI_C_COMPLEX", + "MPI_C_FLOAT_COMPLEX", + "MPI_C_DOUBLE_COMPLEX", + "MPI_C_LONG_DOUBLE_COMPLEX", + "MPI_INT8_T", + "MPI_INT16_T", + "MPI_INT32_T", + "MPI_INT64_T", + "MPI_UINT8_T", + "MPI_UINT16_T", + "MPI_UINT32_T", + "MPI_UINT64_T", + "MPI_CXX_BOOL", + "MPI_CXX_FLOAT_COMPLEX", + "MPI_CXX_DOUBLE_COMPLEX", + "MPI_CXX_LONG_DOUBLE_COMPLEX"}; return AllTypes.contains(MPIDatatype); } @@ -90,7 +90,7 @@ static bool isBuiltinTypeMatching(const BuiltinType *Builtin, std::string &BufferTypeName, StringRef MPIDatatype, const LangOptions &LO) { - static std::multimap<BuiltinType::Kind, StringRef> BuiltinMatches = { + static const std::multimap<BuiltinType::Kind, StringRef> BuiltinMatches = { // On some systems like PPC or ARM, 'char' is unsigned by default which is // why distinct signedness for the buffer and MPI type is tolerated. {BuiltinType::SChar, "MPI_CHAR"}, @@ -143,7 +143,7 @@ static bool isCComplexTypeMatching(const ComplexType *const Complex, std::string &BufferTypeName, StringRef MPIDatatype, const LangOptions &LO) { - static std::multimap<BuiltinType::Kind, StringRef> ComplexCMatches = { + static const std::multimap<BuiltinType::Kind, StringRef> ComplexCMatches = { {BuiltinType::Float, "MPI_C_COMPLEX"}, {BuiltinType::Float, "MPI_C_FLOAT_COMPLEX"}, {BuiltinType::Double, "MPI_C_DOUBLE_COMPLEX"}, @@ -173,7 +173,7 @@ static bool isCXXComplexTypeMatching(const TemplateSpecializationType *const Template, std::string &BufferTypeName, StringRef MPIDatatype, const LangOptions &LO) { - static std::multimap<BuiltinType::Kind, StringRef> ComplexCXXMatches = { + static const std::multimap<BuiltinType::Kind, StringRef> ComplexCXXMatches = { {BuiltinType::Float, "MPI_CXX_FLOAT_COMPLEX"}, {BuiltinType::Double, "MPI_CXX_DOUBLE_COMPLEX"}, {BuiltinType::LongDouble, "MPI_CXX_LONG_DOUBLE_COMPLEX"}}; @@ -264,7 +264,7 @@ void TypeMismatchCheck::check(const MatchFinder::MatchResult &Result) { "MPI_IN_PLACE") return; - StringRef MPIDatatype = + const StringRef MPIDatatype = tooling::fixit::getText(*CE->getArg(DatatypeIdx), *Result.Context); const Type *ArgType = argumentType(CE, BufferIdx); diff --git a/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.h b/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.h index 5a7db17..043f99e 100644 --- a/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.h +++ b/clang-tools-extra/clang-tidy/mpi/TypeMismatchCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_TYPE_MISMATCH_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_TYPE_MISMATCH_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_TYPEMISMATCHCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_TYPEMISMATCHCHECK_H #include "../ClangTidyCheck.h" #include "clang/ASTMatchers/ASTMatchFinder.h" @@ -49,4 +49,4 @@ private: } // namespace clang::tidy::mpi -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_TYPE_MISMATCH_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MPI_TYPEMISMATCHCHECK_H diff --git a/clang-tools-extra/clang-tidy/objc/AssertEquals.cpp b/clang-tools-extra/clang-tidy/objc/AssertEquals.cpp index 3f1bc17..9d274ee 100644 --- a/clang-tools-extra/clang-tidy/objc/AssertEquals.cpp +++ b/clang-tools-extra/clang-tidy/objc/AssertEquals.cpp @@ -7,8 +7,8 @@ //===----------------------------------------------------------------------===// #include "AssertEquals.h" +#include "llvm/ADT/StringMap.h" -#include <map> #include <string> using namespace clang::ast_matchers; @@ -16,44 +16,39 @@ using namespace clang::ast_matchers; namespace clang::tidy::objc { // Mapping from `XCTAssert*Equal` to `XCTAssert*EqualObjects` name. -static const std::map<std::string, std::string> &nameMap() { - static std::map<std::string, std::string> Map{ - {"XCTAssertEqual", "XCTAssertEqualObjects"}, - {"XCTAssertNotEqual", "XCTAssertNotEqualObjects"}, - - }; - return Map; -} +static const llvm::StringMap<StringRef> NameMap{ + {"XCTAssertEqual", "XCTAssertEqualObjects"}, + {"XCTAssertNotEqual", "XCTAssertNotEqualObjects"}, +}; void AssertEquals::registerMatchers(MatchFinder *Finder) { - for (const auto &Pair : nameMap()) { + for (const auto &[CurrName, _] : NameMap) { Finder->addMatcher( binaryOperator(anyOf(hasOperatorName("!="), hasOperatorName("==")), - isExpandedFromMacro(Pair.first), + isExpandedFromMacro(std::string(CurrName)), anyOf(hasLHS(hasType(qualType( hasCanonicalType(asString("NSString *"))))), hasRHS(hasType(qualType( - hasCanonicalType(asString("NSString *")))))) - - ) - .bind(Pair.first), + hasCanonicalType(asString("NSString *"))))))) + .bind(CurrName), this); } } void AssertEquals::check(const ast_matchers::MatchFinder::MatchResult &Result) { - for (const auto &Pair : nameMap()) { - if (const auto *Root = Result.Nodes.getNodeAs<BinaryOperator>(Pair.first)) { - SourceManager *Sm = Result.SourceManager; + for (const auto &[CurrName, TargetName] : NameMap) { + if (const auto *Root = Result.Nodes.getNodeAs<BinaryOperator>(CurrName)) { + const SourceManager *Sm = Result.SourceManager; // The macros are nested two levels, so going up twice. auto MacroCallsite = Sm->getImmediateMacroCallerLoc( Sm->getImmediateMacroCallerLoc(Root->getBeginLoc())); - diag(MacroCallsite, "use " + Pair.second + " for comparing objects") + diag(MacroCallsite, + (Twine("use ") + TargetName + " for comparing objects").str()) << FixItHint::CreateReplacement( clang::CharSourceRange::getCharRange( MacroCallsite, - MacroCallsite.getLocWithOffset(Pair.first.length())), - Pair.second); + MacroCallsite.getLocWithOffset(CurrName.size())), + TargetName); } } } diff --git a/clang-tools-extra/clang-tidy/objc/AssertEquals.h b/clang-tools-extra/clang-tidy/objc/AssertEquals.h index 8c21f9b..1405065 100644 --- a/clang-tools-extra/clang-tidy/objc/AssertEquals.h +++ b/clang-tools-extra/clang-tidy/objc/AssertEquals.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef THIRD_PARTY_LLVM_LLVM_PROJECT_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_OBJCASSERTEQUALS_H_ -#define THIRD_PARTY_LLVM_LLVM_PROJECT_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_OBJCASSERTEQUALS_H_ +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_ASSERTEQUALS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_ASSERTEQUALS_H #include "../ClangTidyCheck.h" #include "clang/ASTMatchers/ASTMatchFinder.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::objc -#endif // THIRD_PARTY_LLVM_LLVM_PROJECT_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_OBJCASSERTEQUALS_H_ +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_ASSERTEQUALS_H diff --git a/clang-tools-extra/clang-tidy/objc/ForbiddenSubclassingCheck.h b/clang-tools-extra/clang-tidy/objc/ForbiddenSubclassingCheck.h index 2d23869..6e0a12a 100644 --- a/clang-tools-extra/clang-tidy/objc/ForbiddenSubclassingCheck.h +++ b/clang-tools-extra/clang-tidy/objc/ForbiddenSubclassingCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_FORBIDDEN_SUBCLASSING_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_FORBIDDEN_SUBCLASSING_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_FORBIDDENSUBCLASSINGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_FORBIDDENSUBCLASSINGCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/StringRef.h" @@ -36,4 +36,4 @@ private: } // namespace clang::tidy::objc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_FORBIDDEN_SUBCLASSING_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_FORBIDDENSUBCLASSINGCHECK_H diff --git a/clang-tools-extra/clang-tidy/objc/NSDateFormatterCheck.cpp b/clang-tools-extra/clang-tidy/objc/NSDateFormatterCheck.cpp index 1481b2b..31d098e 100644 --- a/clang-tools-extra/clang-tidy/objc/NSDateFormatterCheck.cpp +++ b/clang-tools-extra/clang-tidy/objc/NSDateFormatterCheck.cpp @@ -49,7 +49,7 @@ void NSDateFormatterCheck::check(const MatchFinder::MatchResult &Result) { // Callback implementation. const auto *StrExpr = Result.Nodes.getNodeAs<ObjCStringLiteral>("str_lit"); const StringLiteral *SL = cast<ObjCStringLiteral>(StrExpr)->getString(); - StringRef SR = SL->getString(); + const StringRef SR = SL->getString(); if (!isValidDatePattern(SR)) { diag(StrExpr->getExprLoc(), "invalid date format specifier"); diff --git a/clang-tools-extra/clang-tidy/objc/NSInvocationArgumentLifetimeCheck.cpp b/clang-tools-extra/clang-tidy/objc/NSInvocationArgumentLifetimeCheck.cpp index 8a32c38..69caaed 100644 --- a/clang-tools-extra/clang-tidy/objc/NSInvocationArgumentLifetimeCheck.cpp +++ b/clang-tools-extra/clang-tidy/objc/NSInvocationArgumentLifetimeCheck.cpp @@ -43,7 +43,7 @@ AST_POLYMORPHIC_MATCHER(isObjCManagedLifetime, AST_POLYMORPHIC_SUPPORTED_TYPES(ObjCIvarRefExpr, DeclRefExpr, MemberExpr)) { - QualType QT = Node.getType(); + const QualType QT = Node.getType(); return QT->isScalarType() && (QT->getScalarTypeKind() == Type::STK_ObjCObjectPointer || QT->getScalarTypeKind() == Type::STK_BlockPointer) && @@ -55,12 +55,12 @@ AST_POLYMORPHIC_MATCHER(isObjCManagedLifetime, static std::optional<FixItHint> fixItHintReplacementForOwnershipString(StringRef Text, CharSourceRange Range, StringRef Ownership) { - size_t Index = Text.find(Ownership); + const size_t Index = Text.find(Ownership); if (Index == StringRef::npos) return std::nullopt; - SourceLocation Begin = Range.getBegin().getLocWithOffset(Index); - SourceLocation End = Begin.getLocWithOffset(Ownership.size()); + const SourceLocation Begin = Range.getBegin().getLocWithOffset(Index); + const SourceLocation End = Begin.getLocWithOffset(Ownership.size()); return FixItHint::CreateReplacement(SourceRange(Begin, End), UnsafeUnretainedText); } @@ -76,7 +76,7 @@ fixItHintForVarDecl(const VarDecl *VD, const SourceManager &SM, // Currently there is no way to directly get the source range for the // __weak/__strong ObjC lifetime qualifiers, so it's necessary to string // search in the source code. - CharSourceRange Range = Lexer::makeFileCharRange( + const CharSourceRange Range = Lexer::makeFileCharRange( CharSourceRange::getTokenRange(VD->getSourceRange()), SM, LangOpts); if (Range.isInvalid()) { // An invalid range likely means inside a macro, in which case don't supply @@ -84,7 +84,7 @@ fixItHintForVarDecl(const VarDecl *VD, const SourceManager &SM, return std::nullopt; } - StringRef VarDeclText = Lexer::getSourceText(Range, SM, LangOpts); + const StringRef VarDeclText = Lexer::getSourceText(Range, SM, LangOpts); if (std::optional<FixItHint> Hint = fixItHintReplacementForOwnershipString(VarDeclText, Range, WeakText)) return Hint; diff --git a/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.cpp b/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.cpp index 4a586c8..690572f 100644 --- a/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.cpp +++ b/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.cpp @@ -39,7 +39,7 @@ static FixItHint generateFixItHint(const ObjCPropertyDecl *Decl, auto NewName = Decl->getName().str(); size_t Index = 0; if (Style == CategoryProperty) { - size_t UnderScorePos = Name.find_first_of('_'); + const size_t UnderScorePos = Name.find_first_of('_'); if (UnderScorePos != llvm::StringRef::npos) { Index = UnderScorePos + 1; NewName.replace(0, Index - 1, Name.substr(0, Index - 1).lower()); @@ -74,7 +74,7 @@ static std::string validPropertyNameRegex(bool UsedInMatcher) { // // aRbITRaRyCapS is allowed to avoid generating false positives for names // like isVitaminBSupplement, CProgrammingLanguage, and isBeforeM. - std::string StartMatcher = UsedInMatcher ? "::" : "^"; + const std::string StartMatcher = UsedInMatcher ? "::" : "^"; return StartMatcher + "([a-z]|[A-Z][A-Z0-9])[a-z0-9A-Z]*$"; } @@ -85,7 +85,7 @@ static bool hasCategoryPropertyPrefix(llvm::StringRef PropertyName) { } static bool prefixedPropertyNameValid(llvm::StringRef PropertyName) { - size_t Start = PropertyName.find_first_of('_'); + const size_t Start = PropertyName.find_first_of('_'); assert(Start != llvm::StringRef::npos && Start + 1 < PropertyName.size()); auto Prefix = PropertyName.substr(0, Start); if (Prefix.lower() != Prefix) { @@ -115,8 +115,9 @@ void PropertyDeclarationCheck::check(const MatchFinder::MatchResult &Result) { hasCategoryPropertyPrefix(MatchedDecl->getName())) { if (!prefixedPropertyNameValid(MatchedDecl->getName()) || CategoryDecl->IsClassExtension()) { - NamingStyle Style = CategoryDecl->IsClassExtension() ? StandardProperty - : CategoryProperty; + const NamingStyle Style = CategoryDecl->IsClassExtension() + ? StandardProperty + : CategoryProperty; diag(MatchedDecl->getLocation(), "property name '%0' not using lowerCamelCase style or not prefixed " "in a category, according to the Apple Coding Guidelines") diff --git a/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.h b/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.h index daaebb1..1e185b9 100644 --- a/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.h +++ b/clang-tools-extra/clang-tidy/objc/PropertyDeclarationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_PROPERTY_DECLARATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_PROPERTY_DECLARATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_PROPERTYDECLARATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_PROPERTYDECLARATIONCHECK_H #include "../ClangTidyCheck.h" @@ -34,4 +34,4 @@ public: } // namespace clang::tidy::objc -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_PROPERTY_DECLARATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_OBJC_PROPERTYDECLARATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/objc/SuperSelfCheck.cpp b/clang-tools-extra/clang-tidy/objc/SuperSelfCheck.cpp index 3c133ad..3887afe7 100644 --- a/clang-tools-extra/clang-tidy/objc/SuperSelfCheck.cpp +++ b/clang-tools-extra/clang-tidy/objc/SuperSelfCheck.cpp @@ -90,11 +90,11 @@ void SuperSelfCheck::check(const MatchFinder::MatchResult &Result) { "invoke a superclass initializer?") << Message->getMethodDecl(); - SourceLocation ReceiverLoc = Message->getReceiverRange().getBegin(); + const SourceLocation ReceiverLoc = Message->getReceiverRange().getBegin(); if (ReceiverLoc.isMacroID() || ReceiverLoc.isInvalid()) return; - SourceLocation SelectorLoc = Message->getSelectorStartLoc(); + const SourceLocation SelectorLoc = Message->getSelectorStartLoc(); if (SelectorLoc.isMacroID() || SelectorLoc.isInvalid()) return; diff --git a/clang-tools-extra/clang-tidy/performance/FasterStringFindCheck.h b/clang-tools-extra/clang-tidy/performance/FasterStringFindCheck.h index 2452d2e..74067c1 100644 --- a/clang-tools-extra/clang-tidy/performance/FasterStringFindCheck.h +++ b/clang-tools-extra/clang-tidy/performance/FasterStringFindCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_FASTER_STRING_FIND_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_FASTER_STRING_FIND_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_FASTERSTRINGFINDCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_FASTERSTRINGFINDCHECK_H #include "../ClangTidyCheck.h" @@ -38,4 +38,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_FASTER_STRING_FIND_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_FASTERSTRINGFINDCHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.cpp b/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.cpp index a558954b..74a76fa 100644 --- a/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.cpp @@ -87,8 +87,8 @@ void ImplicitConversionInLoopCheck::reportAndFix(const ASTContext *Context, const Expr *OperatorCall) { // We only match on const ref, so we should print a const ref version of the // type. - QualType ConstType = OperatorCall->getType().withConst(); - QualType ConstRefType = Context->getLValueReferenceType(ConstType); + const QualType ConstType = OperatorCall->getType().withConst(); + const QualType ConstRefType = Context->getLValueReferenceType(ConstType); const char Message[] = "the type of the loop variable %0 is different from the one returned " "by the iterator and generates an implicit conversion; you can either " diff --git a/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.h b/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.h index 786081a..4690caa 100644 --- a/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.h +++ b/clang-tools-extra/clang-tidy/performance/ImplicitConversionInLoopCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_IMPLICIT_CONVERSION_IN_LOOP_CHECK_H_ -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_IMPLICIT_CONVERSION_IN_LOOP_CHECK_H_ +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_IMPLICITCONVERSIONINLOOPCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_IMPLICITCONVERSIONINLOOPCHECK_H #include "../ClangTidyCheck.h" @@ -33,4 +33,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_IMPLICIT_CONVERSION_IN_LOOP_CHECK_H_ +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_IMPLICITCONVERSIONINLOOPCHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/InefficientAlgorithmCheck.cpp b/clang-tools-extra/clang-tidy/performance/InefficientAlgorithmCheck.cpp index cd128c3..b57fdb2 100644 --- a/clang-tools-extra/clang-tidy/performance/InefficientAlgorithmCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/InefficientAlgorithmCheck.cpp @@ -71,8 +71,8 @@ void InefficientAlgorithmCheck::check(const MatchFinder::MatchResult &Result) { // Store if the key type of the container is compatible with the value // that is searched for. - QualType ValueType = AlgCall->getArg(2)->getType(); - QualType KeyType = + const QualType ValueType = AlgCall->getArg(2)->getType(); + const QualType KeyType = IneffCont->getTemplateArgs()[0].getAsType().getCanonicalType(); const bool CompatibleTypes = areTypesCompatible(KeyType, ValueType); @@ -104,8 +104,8 @@ void InefficientAlgorithmCheck::check(const MatchFinder::MatchResult &Result) { const auto *IneffContExpr = Result.Nodes.getNodeAs<Expr>("IneffContExpr"); FixItHint Hint; - SourceManager &SM = *Result.SourceManager; - LangOptions LangOpts = getLangOpts(); + const SourceManager &SM = *Result.SourceManager; + const LangOptions LangOpts = getLangOpts(); CharSourceRange CallRange = CharSourceRange::getTokenRange(AlgCall->getSourceRange()); @@ -128,13 +128,13 @@ void InefficientAlgorithmCheck::check(const MatchFinder::MatchResult &Result) { } if (!CallRange.getBegin().isMacroID() && !Maplike && CompatibleTypes) { - StringRef ContainerText = Lexer::getSourceText( + const StringRef ContainerText = Lexer::getSourceText( CharSourceRange::getTokenRange(IneffContExpr->getSourceRange()), SM, LangOpts); - StringRef ParamText = Lexer::getSourceText( + const StringRef ParamText = Lexer::getSourceText( CharSourceRange::getTokenRange(AlgParam->getSourceRange()), SM, LangOpts); - std::string ReplacementText = + const std::string ReplacementText = (llvm::Twine(ContainerText) + (PtrToContainer ? "->" : ".") + AlgDecl->getName() + "(" + ParamText + ")") .str(); diff --git a/clang-tools-extra/clang-tidy/performance/InefficientStringConcatenationCheck.h b/clang-tools-extra/clang-tidy/performance/InefficientStringConcatenationCheck.h index b82a838..0fae10f 100644 --- a/clang-tools-extra/clang-tidy/performance/InefficientStringConcatenationCheck.h +++ b/clang-tools-extra/clang-tidy/performance/InefficientStringConcatenationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTSTRINGCONCATENATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTSTRINGCONCATENATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTSTRINGCONCATENATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTSTRINGCONCATENATIONCHECK_H #include "../ClangTidyCheck.h" @@ -35,4 +35,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTSTRINGCONCATENATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTSTRINGCONCATENATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.cpp b/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.cpp index 4a8f292..a59ab33 100644 --- a/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.cpp @@ -208,7 +208,7 @@ void InefficientVectorOperationCheck::check( if (!TargetVarDecl) TargetVarDecl = ProtoVarDecl; - llvm::SmallPtrSet<const DeclRefExpr *, 16> AllVarRefs = + const llvm::SmallPtrSet<const DeclRefExpr *, 16> AllVarRefs = utils::decl_ref_expr::allDeclRefExprs(*TargetVarDecl, *LoopParent, *Context); for (const auto *Ref : AllVarRefs) { @@ -231,12 +231,12 @@ void InefficientVectorOperationCheck::check( } else { llvm::StringRef FieldName = ProtoAddFieldCall->getMethodDecl()->getName(); FieldName.consume_front("add_"); - std::string MutableFieldName = ("mutable_" + FieldName).str(); + const std::string MutableFieldName = ("mutable_" + FieldName).str(); PartialReserveStmt = "." + MutableFieldName + "()->Reserve"; // e.g., ".mutable_xxx()->Reserve" } - llvm::StringRef VarName = Lexer::getSourceText( + const llvm::StringRef VarName = Lexer::getSourceText( CharSourceRange::getTokenRange( AppendCall->getImplicitObjectArgument()->getSourceRange()), SM, Context->getLangOpts()); @@ -246,14 +246,14 @@ void InefficientVectorOperationCheck::check( if (RangeLoop) { // Get the range-expression in a for-range statement represented as // `for (range-declarator: range-expression)`. - StringRef RangeInitExpName = + const StringRef RangeInitExpName = Lexer::getSourceText(CharSourceRange::getTokenRange( RangeLoop->getRangeInit()->getSourceRange()), SM, Context->getLangOpts()); ReserveSize = (RangeInitExpName + ".size()").str(); } else if (ForLoop) { // Handle counter-based loop cases. - StringRef LoopEndSource = Lexer::getSourceText( + const StringRef LoopEndSource = Lexer::getSourceText( CharSourceRange::getTokenRange(LoopEndExpr->getSourceRange()), SM, Context->getLangOpts()); ReserveSize = std::string(LoopEndSource); @@ -264,7 +264,7 @@ void InefficientVectorOperationCheck::check( "container capacity before the loop") << AppendCall->getMethodDecl()->getDeclName(); if (!ReserveSize.empty()) { - std::string ReserveStmt = + const std::string ReserveStmt = (VarName + PartialReserveStmt + "(" + ReserveSize + ");\n").str(); Diag << FixItHint::CreateInsertion(LoopStmt->getBeginLoc(), ReserveStmt); } diff --git a/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.h b/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.h index 18f7c19..5f3b88f 100644 --- a/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.h +++ b/clang-tools-extra/clang-tidy/performance/InefficientVectorOperationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENT_VECTOR_OPERATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENT_VECTOR_OPERATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTVECTOROPERATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTVECTOROPERATIONCHECK_H #include "../ClangTidyCheck.h" @@ -44,4 +44,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENT_VECTOR_OPERATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_INEFFICIENTVECTOROPERATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.cpp b/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.cpp index 854f09a..4d26c39 100644 --- a/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.cpp @@ -19,10 +19,10 @@ static void replaceCallWithArg(const CallExpr *Call, DiagnosticBuilder &Diag, const LangOptions &LangOpts) { const Expr *Arg = Call->getArg(0); - CharSourceRange BeforeArgumentsRange = Lexer::makeFileCharRange( + const CharSourceRange BeforeArgumentsRange = Lexer::makeFileCharRange( CharSourceRange::getCharRange(Call->getBeginLoc(), Arg->getBeginLoc()), SM, LangOpts); - CharSourceRange AfterArgumentsRange = Lexer::makeFileCharRange( + const CharSourceRange AfterArgumentsRange = Lexer::makeFileCharRange( CharSourceRange::getCharRange(Call->getEndLoc(), Call->getEndLoc().getLocWithOffset(1)), SM, LangOpts); @@ -114,17 +114,18 @@ void MoveConstArgCheck::check(const MatchFinder::MatchResult &Result) { const Expr *Arg = CallMove->getArg(0); const QualType ArgType = Arg->getType().getCanonicalType(); - SourceManager &SM = Result.Context->getSourceManager(); + const SourceManager &SM = Result.Context->getSourceManager(); - CharSourceRange MoveRange = + const CharSourceRange MoveRange = CharSourceRange::getCharRange(CallMove->getSourceRange()); - CharSourceRange FileMoveRange = + const CharSourceRange FileMoveRange = Lexer::makeFileCharRange(MoveRange, SM, getLangOpts()); if (!FileMoveRange.isValid()) return; - bool IsConstArg = ArgType.isConstQualified(); - bool IsTriviallyCopyable = ArgType.isTriviallyCopyableType(*Result.Context); + const bool IsConstArg = ArgType.isConstQualified(); + const bool IsTriviallyCopyable = + ArgType.isTriviallyCopyableType(*Result.Context); if (IsConstArg || IsTriviallyCopyable) { if (const CXXRecordDecl *R = ArgType->getAsCXXRecordDecl()) { @@ -143,10 +144,10 @@ void MoveConstArgCheck::check(const MatchFinder::MatchResult &Result) { if (!IsConstArg && IsTriviallyCopyable && !CheckTriviallyCopyableMove) return; - bool IsVariable = isa<DeclRefExpr>(Arg); + const bool IsVariable = isa<DeclRefExpr>(Arg); // std::move shouldn't be removed when an lvalue wrapped by std::move is // passed to the function with an rvalue reference parameter. - bool IsRVRefParam = + const bool IsRVRefParam = isRValueReferenceParam(ReceivingExpr, InvocationParmType, Arg); const auto *Var = IsVariable ? dyn_cast<DeclRefExpr>(Arg)->getDecl() : nullptr; diff --git a/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.h b/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.h index 9f67f64..ff1d67b 100644 --- a/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.h +++ b/clang-tools-extra/clang-tidy/performance/MoveConstArgCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MOVECONSTANTARGUMENTCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MOVECONSTANTARGUMENTCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_MOVECONSTARGCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_MOVECONSTARGCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/DenseSet.h" @@ -43,4 +43,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_MOVECONSTANTARGUMENTCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_MOVECONSTARGCHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/MoveConstructorInitCheck.cpp b/clang-tools-extra/clang-tidy/performance/MoveConstructorInitCheck.cpp index 44f6d20..d8cdc0d 100644 --- a/clang-tools-extra/clang-tidy/performance/MoveConstructorInitCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/MoveConstructorInitCheck.cpp @@ -39,7 +39,7 @@ void MoveConstructorInitCheck::check(const MatchFinder::MatchResult &Result) { // Do not diagnose if the expression used to perform the initialization is a // trivially-copyable type. - QualType QT = Initializer->getInit()->getType(); + const QualType QT = Initializer->getInit()->getType(); if (QT.isTriviallyCopyableType(*Result.Context)) return; diff --git a/clang-tools-extra/clang-tidy/performance/NoexceptFunctionBaseCheck.h b/clang-tools-extra/clang-tidy/performance/NoexceptFunctionBaseCheck.h index 56a1e4a..6ed3025 100644 --- a/clang-tools-extra/clang-tidy/performance/NoexceptFunctionBaseCheck.h +++ b/clang-tools-extra/clang-tidy/performance/NoexceptFunctionBaseCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_NOEXCEPTFUNCTIONCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_NOEXCEPTFUNCTIONCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_NOEXCEPTFUNCTIONBASECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_NOEXCEPTFUNCTIONBASECHECK_H #include "../ClangTidyCheck.h" #include "../utils/ExceptionSpecAnalyzer.h" @@ -46,4 +46,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_NOEXCEPTFUNCTIONCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_NOEXCEPTFUNCTIONBASECHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/TriviallyDestructibleCheck.cpp b/clang-tools-extra/clang-tidy/performance/TriviallyDestructibleCheck.cpp index 0db66c0..2f54b17 100644 --- a/clang-tools-extra/clang-tidy/performance/TriviallyDestructibleCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/TriviallyDestructibleCheck.cpp @@ -24,7 +24,7 @@ AST_MATCHER(Decl, isFirstDecl) { return Node.isFirstDecl(); } AST_MATCHER_P(CXXRecordDecl, hasBase, Matcher<QualType>, InnerMatcher) { for (const CXXBaseSpecifier &BaseSpec : Node.bases()) { - QualType BaseType = BaseSpec.getType(); + const QualType BaseType = BaseSpec.getType(); if (InnerMatcher.matches(BaseType, Finder, Builder)) return true; } @@ -50,7 +50,7 @@ void TriviallyDestructibleCheck::check(const MatchFinder::MatchResult &Result) { const auto *MatchedDecl = Result.Nodes.getNodeAs<CXXDestructorDecl>("decl"); // Get locations of both first and out-of-line declarations. - SourceManager &SM = *Result.SourceManager; + const SourceManager &SM = *Result.SourceManager; const auto *FirstDecl = cast<CXXMethodDecl>(MatchedDecl->getFirstDecl()); const SourceLocation FirstDeclEnd = utils::lexer::findNextTerminator( FirstDecl->getEndLoc(), SM, getLangOpts()); diff --git a/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.cpp b/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.cpp index 096ca57e..a462c55 100644 --- a/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.cpp @@ -155,18 +155,18 @@ void TypePromotionInMathFnCheck::check(const MatchFinder::MatchResult &Result) { const auto *Call = Result.Nodes.getNodeAs<CallExpr>("call"); assert(Call != nullptr); - StringRef OldFnName = Call->getDirectCallee()->getName(); + const StringRef OldFnName = Call->getDirectCallee()->getName(); // In C++ mode, we prefer std::foo to ::foof. But some of these suggestions // are only valid in C++11 and newer. - static llvm::StringSet<> Cpp11OnlyFns = { + static const llvm::StringSet<> Cpp11OnlyFns = { "acosh", "asinh", "atanh", "cbrt", "copysign", "erf", "erfc", "exp2", "expm1", "fdim", "fma", "fmax", "fmin", "hypot", "ilogb", "lgamma", "llrint", "llround", "log1p", "log2", "logb", "lrint", "lround", "nearbyint", "nextafter", "nexttoward", "remainder", "remquo", "rint", "round", "scalbln", "scalbn", "tgamma", "trunc"}; - bool StdFnRequiresCpp11 = Cpp11OnlyFns.contains(OldFnName); + const bool StdFnRequiresCpp11 = Cpp11OnlyFns.contains(OldFnName); std::string NewFnName; bool FnInCmath = false; diff --git a/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.h b/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.h index cf74f80..21a7f4d 100644 --- a/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.h +++ b/clang-tools-extra/clang-tidy/performance/TypePromotionInMathFnCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_TYPE_PROMOTION_IN_MATH_FN_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_TYPE_PROMOTION_IN_MATH_FN_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_TYPEPROMOTIONINMATHFNCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_TYPEPROMOTIONINMATHFNCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -39,4 +39,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_TYPE_PROMOTION_IN_MATH_FN_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_TYPEPROMOTIONINMATHFNCHECK_H diff --git a/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp b/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp index 5918366..fec4c14 100644 --- a/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp +++ b/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp @@ -46,7 +46,7 @@ static std::optional<SourceLocation> firstLocAfterNewLine(SourceLocation Loc, if (Invalid) { return std::nullopt; } - size_t Offset = std::strcspn(TextAfter, "\n"); + const size_t Offset = std::strcspn(TextAfter, "\n"); return Loc.getLocWithOffset(TextAfter[Offset] == '\0' ? Offset : Offset + 1); } @@ -148,7 +148,7 @@ AST_MATCHER_FUNCTION_P(StatementMatcher, initializerReturnsReferenceToConst, static bool isInitializingVariableImmutable( const VarDecl &InitializingVar, const Stmt &BlockStmt, ASTContext &Context, const std::vector<StringRef> &ExcludedContainerTypes) { - QualType T = InitializingVar.getType().getCanonicalType(); + const QualType T = InitializingVar.getType().getCanonicalType(); if (!isOnlyUsedAsConst(InitializingVar, BlockStmt, Context, T->isPointerType() ? 1 : 0)) return false; @@ -297,7 +297,7 @@ void UnnecessaryCopyInitialization::check( const auto *ObjectArg = Result.Nodes.getNodeAs<VarDecl>(ObjectArgId); const auto *CtorCall = Result.Nodes.getNodeAs<CXXConstructExpr>("ctorCall"); - TraversalKindScope RAII(*Result.Context, TK_AsIs); + const TraversalKindScope RAII(*Result.Context, TK_AsIs); // A constructor that looks like T(const T& t, bool arg = false) counts as a // copy only when it is called with default arguments for the arguments after @@ -327,7 +327,7 @@ void UnnecessaryCopyInitialization::check( void UnnecessaryCopyInitialization::handleCopyFromMethodReturn( const CheckContext &Ctx, const VarDecl *ObjectArg) { - bool IsConstQualified = Ctx.Var.getType().isConstQualified(); + const bool IsConstQualified = Ctx.Var.getType().isConstQualified(); if (!IsConstQualified && !Ctx.IsVarOnlyUsedAsConst) return; if (ObjectArg != nullptr && diff --git a/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.h b/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.h index 6623188..5edc744 100644 --- a/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.h +++ b/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARY_COPY_INITIALIZATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARY_COPY_INITIALIZATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARYCOPYINITIALIZATION_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARYCOPYINITIALIZATION_H #include "../ClangTidyCheck.h" #include "clang/AST/Decl.h" @@ -64,4 +64,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARY_COPY_INITIALIZATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARYCOPYINITIALIZATION_H diff --git a/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.cpp b/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.cpp index 3f5b43f..d626297 100644 --- a/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.cpp +++ b/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.cpp @@ -73,7 +73,7 @@ void UnnecessaryValueParamCheck::check(const MatchFinder::MatchResult &Result) { const auto *Param = Result.Nodes.getNodeAs<ParmVarDecl>("param"); const auto *Function = Result.Nodes.getNodeAs<FunctionDecl>("functionDecl"); - TraversalKindScope RAII(*Result.Context, TK_AsIs); + const TraversalKindScope RAII(*Result.Context, TK_AsIs); FunctionParmMutationAnalyzer *Analyzer = FunctionParmMutationAnalyzer::getFunctionParmMutationAnalyzer( diff --git a/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.h b/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.h index d59fb41..22df689 100644 --- a/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.h +++ b/clang-tools-extra/clang-tidy/performance/UnnecessaryValueParamCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARY_VALUE_PARAM_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARY_VALUE_PARAM_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARYVALUEPARAMCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARYVALUEPARAMCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -51,4 +51,4 @@ private: } // namespace clang::tidy::performance -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARY_VALUE_PARAM_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PERFORMANCE_UNNECESSARYVALUEPARAMCHECK_H diff --git a/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp b/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp index 195418d..172b918 100644 --- a/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp +++ b/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp @@ -55,13 +55,13 @@ public: bool ParseArgs(const CompilerInstance &, const std::vector<std::string> &Args) override { - ClangTidyGlobalOptions GlobalOptions; - ClangTidyOptions DefaultOptions; + const ClangTidyGlobalOptions GlobalOptions; + const ClangTidyOptions DefaultOptions; ClangTidyOptions OverrideOptions; // Parse the extra command line args. // FIXME: This is very limited at the moment. - for (StringRef Arg : Args) + for (const StringRef Arg : Args) if (Arg.starts_with("-checks=")) OverrideOptions.Checks = std::string(Arg.substr(strlen("-checks="))); diff --git a/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.cpp b/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.cpp index 5174f56..4225c3e 100644 --- a/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.cpp +++ b/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.cpp @@ -43,8 +43,9 @@ void RestrictedIncludesPPCallbacks::EndOfMainFile() { for (const auto &Include : FileDirectives) { // Fetch the length of the include statement from the start to just after // the newline, for finding the end (including the newline). - unsigned ToLen = std::strcspn(SM.getCharacterData(Include.Loc), "\n") + 1; - CharSourceRange ToRange = CharSourceRange::getCharRange( + const unsigned ToLen = + std::strcspn(SM.getCharacterData(Include.Loc), "\n") + 1; + const CharSourceRange ToRange = CharSourceRange::getCharRange( Include.Loc, Include.Loc.getLocWithOffset(ToLen)); if (!Include.IsInMainFile) { diff --git a/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.h b/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.h index e37f8933..0dd6402 100644 --- a/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.h +++ b/clang-tools-extra/clang-tidy/portability/RestrictSystemIncludesCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_RESTRICTINCLUDESSCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_RESTRICTINCLUDESSCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_RESTRICTSYSTEMINCLUDESCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_RESTRICTSYSTEMINCLUDESCHECK_H #include "../ClangTidyCheck.h" #include "../GlobList.h" @@ -79,4 +79,4 @@ private: } // namespace clang::tidy::portability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_RESTRICTINCLUDESSCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_RESTRICTSYSTEMINCLUDESCHECK_H diff --git a/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.cpp b/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.cpp index d90b09a..fb4b22c 100644 --- a/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.cpp +++ b/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.cpp @@ -104,9 +104,9 @@ void SIMDIntrinsicsCheck::check(const MatchFinder::MatchResult &Result) { if (!Callee) return; - StringRef Old = Callee->getName(); + const StringRef Old = Callee->getName(); StringRef New; - llvm::Triple::ArchType Arch = + const llvm::Triple::ArchType Arch = Result.Context->getTargetInfo().getTriple().getArch(); // We warn or suggest if this SIMD intrinsic function has a std::simd diff --git a/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.h b/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.h index db2d230..addcecbc 100644 --- a/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.h +++ b/clang-tools-extra/clang-tidy/portability/SIMDIntrinsicsCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_SIMD_INTRINSICS_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_SIMD_INTRINSICS_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_SIMDINTRINSICSCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_SIMDINTRINSICSCHECK_H #include "../ClangTidyCheck.h" @@ -37,4 +37,4 @@ private: } // namespace clang::tidy::portability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_SIMD_INTRINSICS_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_PORTABILITY_SIMDINTRINSICSCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/AvoidConstParamsInDecls.h b/clang-tools-extra/clang-tidy/readability/AvoidConstParamsInDecls.h index 1dd28fd..6490508 100644 --- a/clang-tools-extra/clang-tidy/readability/AvoidConstParamsInDecls.h +++ b/clang-tools-extra/clang-tidy/readability/AvoidConstParamsInDecls.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOID_CONST_PARAMS_IN_DECLS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOID_CONST_PARAMS_IN_DECLS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOIDCONSTPARAMSINDECLS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOIDCONSTPARAMSINDECLS_H #include "../ClangTidyCheck.h" @@ -34,4 +34,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOID_CONST_PARAMS_IN_DECLS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOIDCONSTPARAMSINDECLS_H diff --git a/clang-tools-extra/clang-tidy/readability/AvoidNestedConditionalOperatorCheck.h b/clang-tools-extra/clang-tidy/readability/AvoidNestedConditionalOperatorCheck.h index 260c843..0e729ec 100644 --- a/clang-tools-extra/clang-tidy/readability/AvoidNestedConditionalOperatorCheck.h +++ b/clang-tools-extra/clang-tidy/readability/AvoidNestedConditionalOperatorCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOID_NESTED_CONDITIONAL_OPERATOR_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOID_NESTED_CONDITIONAL_OPERATOR_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOIDNESTEDCONDITIONALOPERATORCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOIDNESTEDCONDITIONALOPERATORCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOID_NESTED_CONDITIONAL_OPERATOR_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_AVOIDNESTEDCONDITIONALOPERATORCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/ConvertMemberFunctionsToStatic.h b/clang-tools-extra/clang-tidy/readability/ConvertMemberFunctionsToStatic.h index 2aab03f1..20cb7f2 100644 --- a/clang-tools-extra/clang-tidy/readability/ConvertMemberFunctionsToStatic.h +++ b/clang-tools-extra/clang-tidy/readability/ConvertMemberFunctionsToStatic.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_CONVERTMEMFUNCTOSTATIC_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_CONVERTMEMFUNCTOSTATIC_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_CONVERTMEMBERFUNCTIONSTOSTATIC_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_CONVERTMEMBERFUNCTIONSTOSTATIC_H #include "../ClangTidyCheck.h" @@ -32,4 +32,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_CONVERTMEMFUNCTOSTATIC_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_CONVERTMEMBERFUNCTIONSTOSTATIC_H diff --git a/clang-tools-extra/clang-tidy/readability/DeleteNullPointerCheck.h b/clang-tools-extra/clang-tidy/readability/DeleteNullPointerCheck.h index 52b1b26..b346f68 100644 --- a/clang-tools-extra/clang-tidy/readability/DeleteNullPointerCheck.h +++ b/clang-tools-extra/clang-tidy/readability/DeleteNullPointerCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DELETE_NULL_POINTER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DELETE_NULL_POINTER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DELETENULLPOINTERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DELETENULLPOINTERCHECK_H #include "../ClangTidyCheck.h" @@ -34,4 +34,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DELETE_NULL_POINTER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DELETENULLPOINTERCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/DuplicateIncludeCheck.h b/clang-tools-extra/clang-tidy/readability/DuplicateIncludeCheck.h index 297999c..ca36791 100644 --- a/clang-tools-extra/clang-tidy/readability/DuplicateIncludeCheck.h +++ b/clang-tools-extra/clang-tidy/readability/DuplicateIncludeCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DUPLICATE_INCLUDE_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DUPLICATE_INCLUDE_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DUPLICATEINCLUDECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DUPLICATEINCLUDECHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DUPLICATE_INCLUDE_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_DUPLICATEINCLUDECHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/ImplicitBoolConversionCheck.h b/clang-tools-extra/clang-tidy/readability/ImplicitBoolConversionCheck.h index f88ceb1..101089c 100644 --- a/clang-tools-extra/clang-tidy/readability/ImplicitBoolConversionCheck.h +++ b/clang-tools-extra/clang-tidy/readability/ImplicitBoolConversionCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_IMPLICIT_BOOL_CONVERSION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_IMPLICIT_BOOL_CONVERSION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_IMPLICITBOOLCONVERSIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_IMPLICITBOOLCONVERSIONCHECK_H #include "../ClangTidyCheck.h" @@ -41,4 +41,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_IMPLICIT_BOOL_CONVERSION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_IMPLICITBOOLCONVERSIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/InconsistentDeclarationParameterNameCheck.h b/clang-tools-extra/clang-tidy/readability/InconsistentDeclarationParameterNameCheck.h index 289e131d..32218e1 100644 --- a/clang-tools-extra/clang-tidy/readability/InconsistentDeclarationParameterNameCheck.h +++ b/clang-tools-extra/clang-tidy/readability/InconsistentDeclarationParameterNameCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_INCONSISTENT_DECLARATION_PARAMETER_NAME_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_INCONSISTENT_DECLARATION_PARAMETER_NAME_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_INCONSISTENTDECLARATIONPARAMETERNAMECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_INCONSISTENTDECLARATIONPARAMETERNAMECHECK_H #include "../ClangTidyCheck.h" @@ -45,4 +45,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_INCONSISTENT_DECLARATION_PARAMETER_NAME_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_INCONSISTENTDECLARATIONPARAMETERNAMECHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/IsolateDeclarationCheck.h b/clang-tools-extra/clang-tidy/readability/IsolateDeclarationCheck.h index 750b4d8..0bf22e5 100644 --- a/clang-tools-extra/clang-tidy/readability/IsolateDeclarationCheck.h +++ b/clang-tools-extra/clang-tidy/readability/IsolateDeclarationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_ISOLATEDECLCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_ISOLATEDECLCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_ISOLATEDECLARATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_ISOLATEDECLARATIONCHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_ISOLATEDECLCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_ISOLATEDECLARATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/MisleadingIndentationCheck.h b/clang-tools-extra/clang-tidy/readability/MisleadingIndentationCheck.h index 8347f1a..edd2b1a 100644 --- a/clang-tools-extra/clang-tidy/readability/MisleadingIndentationCheck.h +++ b/clang-tools-extra/clang-tidy/readability/MisleadingIndentationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISLEADING_INDENTATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISLEADING_INDENTATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISLEADINGINDENTATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISLEADINGINDENTATIONCHECK_H #include "../ClangTidyCheck.h" @@ -38,4 +38,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISLEADING_INDENTATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISLEADINGINDENTATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/MisplacedArrayIndexCheck.h b/clang-tools-extra/clang-tidy/readability/MisplacedArrayIndexCheck.h index 0a6e0c8..f0c565b 100644 --- a/clang-tools-extra/clang-tidy/readability/MisplacedArrayIndexCheck.h +++ b/clang-tools-extra/clang-tidy/readability/MisplacedArrayIndexCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISPLACED_ARRAY_INDEX_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISPLACED_ARRAY_INDEX_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISPLACEDARRAYINDEXCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISPLACEDARRAYINDEXCHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISPLACED_ARRAY_INDEX_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MISPLACEDARRAYINDEXCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/NonConstParameterCheck.h b/clang-tools-extra/clang-tidy/readability/NonConstParameterCheck.h index b015618..7dcb16e 100644 --- a/clang-tools-extra/clang-tidy/readability/NonConstParameterCheck.h +++ b/clang-tools-extra/clang-tidy/readability/NonConstParameterCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_NON_CONST_PARAMETER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_NON_CONST_PARAMETER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_NONCONSTPARAMETERCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_NONCONSTPARAMETERCHECK_H #include "../ClangTidyCheck.h" @@ -59,4 +59,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_NON_CONST_PARAMETER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_NONCONSTPARAMETERCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/RedundantControlFlowCheck.h b/clang-tools-extra/clang-tidy/readability/RedundantControlFlowCheck.h index 3018b1f..fde3050 100644 --- a/clang-tools-extra/clang-tidy/readability/RedundantControlFlowCheck.h +++ b/clang-tools-extra/clang-tidy/readability/RedundantControlFlowCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_CONTROL_FLOW_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_CONTROL_FLOW_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTCONTROLFLOWCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTCONTROLFLOWCHECK_H #include "../ClangTidyCheck.h" @@ -47,4 +47,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_CONTROL_FLOW_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTCONTROLFLOWCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/RedundantDeclarationCheck.h b/clang-tools-extra/clang-tidy/readability/RedundantDeclarationCheck.h index b22cef9..9b1b09f 100644 --- a/clang-tools-extra/clang-tidy/readability/RedundantDeclarationCheck.h +++ b/clang-tools-extra/clang-tidy/readability/RedundantDeclarationCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_DECLARATION_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_DECLARATION_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTDECLARATIONCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTDECLARATIONCHECK_H #include "../ClangTidyCheck.h" @@ -30,4 +30,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_DECLARATION_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTDECLARATIONCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/RedundantFunctionPtrDereferenceCheck.h b/clang-tools-extra/clang-tidy/readability/RedundantFunctionPtrDereferenceCheck.h index 5c82a5e..49cbf69 100644 --- a/clang-tools-extra/clang-tidy/readability/RedundantFunctionPtrDereferenceCheck.h +++ b/clang-tools-extra/clang-tidy/readability/RedundantFunctionPtrDereferenceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_FUNCTION_PTR_DEREFERENCE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_FUNCTION_PTR_DEREFERENCE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTFUNCTIONPTRDEREFERENCECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTFUNCTIONPTRDEREFERENCECHECK_H #include "../ClangTidyCheck.h" @@ -28,4 +28,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_FUNCTION_PTR_DEREFERENCE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTFUNCTIONPTRDEREFERENCECHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/RedundantMemberInitCheck.h b/clang-tools-extra/clang-tidy/readability/RedundantMemberInitCheck.h index 64d365d..ff8b02d 100644 --- a/clang-tools-extra/clang-tidy/readability/RedundantMemberInitCheck.h +++ b/clang-tools-extra/clang-tidy/readability/RedundantMemberInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_MEMBER_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_MEMBER_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTMEMBERINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTMEMBERINITCHECK_H #include "../ClangTidyCheck.h" @@ -40,4 +40,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_MEMBER_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTMEMBERINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/RedundantStringInitCheck.h b/clang-tools-extra/clang-tidy/readability/RedundantStringInitCheck.h index 853ea2f..5c4b744 100644 --- a/clang-tools-extra/clang-tidy/readability/RedundantStringInitCheck.h +++ b/clang-tools-extra/clang-tidy/readability/RedundantStringInitCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_STRING_INIT_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_STRING_INIT_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTSTRINGINITCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTSTRINGINITCHECK_H #include "../ClangTidyCheck.h" #include <string> @@ -32,4 +32,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANT_STRING_INIT_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_REDUNDANTSTRINGINITCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/SimplifyBooleanExprCheck.h b/clang-tools-extra/clang-tidy/readability/SimplifyBooleanExprCheck.h index 466bc41..99520d7 100644 --- a/clang-tools-extra/clang-tidy/readability/SimplifyBooleanExprCheck.h +++ b/clang-tools-extra/clang-tidy/readability/SimplifyBooleanExprCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_SIMPLIFY_BOOLEAN_EXPR_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_SIMPLIFY_BOOLEAN_EXPR_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_SIMPLIFYBOOLEANEXPRCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_SIMPLIFYBOOLEANEXPRCHECK_H #include "../ClangTidyCheck.h" @@ -75,4 +75,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_SIMPLIFY_BOOLEAN_EXPR_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_SIMPLIFYBOOLEANEXPRCHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/StaticAccessedThroughInstanceCheck.h b/clang-tools-extra/clang-tidy/readability/StaticAccessedThroughInstanceCheck.h index c376806..38a2ea6 100644 --- a/clang-tools-extra/clang-tidy/readability/StaticAccessedThroughInstanceCheck.h +++ b/clang-tools-extra/clang-tidy/readability/StaticAccessedThroughInstanceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATIC_ACCESSED_THROUGH_INSTANCE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATIC_ACCESSED_THROUGH_INSTANCE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATICACCESSEDTHROUGHINSTANCECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATICACCESSEDTHROUGHINSTANCECHECK_H #include "../ClangTidyCheck.h" @@ -41,4 +41,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATIC_ACCESSED_THROUGH_INSTANCE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATICACCESSEDTHROUGHINSTANCECHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/StaticDefinitionInAnonymousNamespaceCheck.h b/clang-tools-extra/clang-tidy/readability/StaticDefinitionInAnonymousNamespaceCheck.h index 5530655..e096682 100644 --- a/clang-tools-extra/clang-tidy/readability/StaticDefinitionInAnonymousNamespaceCheck.h +++ b/clang-tools-extra/clang-tidy/readability/StaticDefinitionInAnonymousNamespaceCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATIC_DEFINITION_IN_ANONYMOUS_NAMESPACE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATIC_DEFINITION_IN_ANONYMOUS_NAMESPACE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATICDEFINITIONINANONYMOUSNAMESPACECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATICDEFINITIONINANONYMOUSNAMESPACECHECK_H #include "../ClangTidyCheck.h" @@ -34,4 +34,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATIC_DEFINITION_IN_ANONYMOUS_NAMESPACE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_STATICDEFINITIONINANONYMOUSNAMESPACECHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/UniqueptrDeleteReleaseCheck.h b/clang-tools-extra/clang-tidy/readability/UniqueptrDeleteReleaseCheck.h index ab6449e..2874208 100644 --- a/clang-tools-extra/clang-tidy/readability/UniqueptrDeleteReleaseCheck.h +++ b/clang-tools-extra/clang-tidy/readability/UniqueptrDeleteReleaseCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_UNIQUEPTR_DELETE_RELEASE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_UNIQUEPTR_DELETE_RELEASE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_UNIQUEPTRDELETERELEASECHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_UNIQUEPTRDELETERELEASECHECK_H #include "../ClangTidyCheck.h" @@ -37,4 +37,4 @@ private: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_UNIQUEPTR_DELETE_RELEASE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_UNIQUEPTRDELETERELEASECHECK_H diff --git a/clang-tools-extra/clang-tidy/readability/UseAnyOfAllOfCheck.h b/clang-tools-extra/clang-tidy/readability/UseAnyOfAllOfCheck.h index f431311b..32983e4 100644 --- a/clang-tools-extra/clang-tidy/readability/UseAnyOfAllOfCheck.h +++ b/clang-tools-extra/clang-tidy/readability/UseAnyOfAllOfCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_USEALGORITHMCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_USEALGORITHMCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_USEANYOFALLOFCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_USEANYOFALLOFCHECK_H #include "../ClangTidyCheck.h" #include "../utils/IncludeInserter.h" @@ -33,4 +33,4 @@ public: } // namespace clang::tidy::readability -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_USEALGORITHMCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_USEANYOFALLOFCHECK_H diff --git a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp index 1ae8756..f5e4bf0 100644 --- a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp +++ b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp @@ -390,7 +390,7 @@ static void printStats(const ClangTidyStats &Stats) { static std::unique_ptr<ClangTidyOptionsProvider> createOptionsProvider(llvm::IntrusiveRefCntPtr<vfs::FileSystem> FS) { ClangTidyGlobalOptions GlobalOptions; - if (std::error_code Err = parseLineFilter(LineFilter, GlobalOptions)) { + if (const std::error_code Err = parseLineFilter(LineFilter, GlobalOptions)) { llvm::errs() << "Invalid LineFilter: " << Err.message() << "\n\nUsage:\n"; llvm::cl::PrintHelpMessage(/*Hidden=*/false, /*Categorized=*/true); return nullptr; @@ -448,7 +448,7 @@ createOptionsProvider(llvm::IntrusiveRefCntPtr<vfs::FileSystem> FS) { llvm::ErrorOr<std::unique_ptr<llvm::MemoryBuffer>> Text = llvm::MemoryBuffer::getFile(ConfigFile); - if (std::error_code EC = Text.getError()) { + if (const std::error_code EC = Text.getError()) { llvm::errs() << "Error: can't read config-file '" << ConfigFile << "': " << EC.message() << "\n"; return nullptr; @@ -491,7 +491,7 @@ static StringRef closest(StringRef Value, const StringSet<> &Allowed) { unsigned MaxEdit = 5U; StringRef Closest; for (auto Item : Allowed.keys()) { - unsigned Cur = Value.edit_distance_insensitive(Item, true, MaxEdit); + const unsigned Cur = Value.edit_distance_insensitive(Item, true, MaxEdit); if (Cur < MaxEdit) { Closest = Item; MaxEdit = Cur; @@ -504,7 +504,7 @@ static constexpr StringLiteral VerifyConfigWarningEnd = " [-verify-config]\n"; static bool verifyChecks(const StringSet<> &AllChecks, StringRef CheckGlob, StringRef Source) { - GlobList Globs(CheckGlob); + const GlobList Globs(CheckGlob); bool AnyInvalid = false; for (const auto &Item : Globs.getItems()) { if (Item.Text.starts_with("clang-diagnostic")) @@ -520,7 +520,7 @@ static bool verifyChecks(const StringSet<> &AllChecks, StringRef CheckGlob, llvm::raw_ostream &Output = llvm::WithColor::warning(llvm::errs(), Source) << "unknown check '" << Item.Text << '\''; - llvm::StringRef Closest = closest(Item.Text, AllChecks); + const llvm::StringRef Closest = closest(Item.Text, AllChecks); if (!Closest.empty()) Output << "; did you mean '" << Closest << '\''; Output << VerifyConfigWarningEnd; @@ -560,7 +560,7 @@ static bool verifyOptions(const llvm::StringSet<> &ValidOptions, AnyInvalid = true; auto &Output = llvm::WithColor::warning(llvm::errs(), Source) << "unknown check option '" << Key << '\''; - llvm::StringRef Closest = closest(Key, ValidOptions); + const llvm::StringRef Closest = closest(Key, ValidOptions); if (!Closest.empty()) Output << "; did you mean '" << Closest << '\''; Output << VerifyConfigWarningEnd; @@ -572,7 +572,7 @@ static SmallString<256> makeAbsolute(llvm::StringRef Input) { if (Input.empty()) return {}; SmallString<256> AbsolutePath(Input); - if (std::error_code EC = llvm::sys::fs::make_absolute(AbsolutePath)) { + if (const std::error_code EC = llvm::sys::fs::make_absolute(AbsolutePath)) { llvm::errs() << "Can't make absolute path from " << Input << ": " << EC.message() << "\n"; } @@ -594,7 +594,7 @@ static llvm::IntrusiveRefCntPtr<vfs::OverlayFileSystem> createBaseFS() { } int clangTidyMain(int argc, const char **argv) { - llvm::InitLLVM X(argc, argv); + const llvm::InitLLVM X(argc, argv); SmallVector<const char *> Args{argv, argv + argc}; // expand parameters file to argc and argv. @@ -623,7 +623,8 @@ int clangTidyMain(int argc, const char **argv) { return 1; } - llvm::IntrusiveRefCntPtr<vfs::OverlayFileSystem> BaseFS = createBaseFS(); + const llvm::IntrusiveRefCntPtr<vfs::OverlayFileSystem> BaseFS = + createBaseFS(); if (!BaseFS) return 1; @@ -632,7 +633,7 @@ int clangTidyMain(int argc, const char **argv) { if (!OptionsProvider) return 1; - SmallString<256> ProfilePrefix = makeAbsolute(StoreCheckProfile); + const SmallString<256> ProfilePrefix = makeAbsolute(StoreCheckProfile); StringRef FileName("dummy"); auto PathList = OptionsParser->getSourcePathList(); @@ -640,10 +641,10 @@ int clangTidyMain(int argc, const char **argv) { FileName = PathList.front(); } - SmallString<256> FilePath = makeAbsolute(FileName); + const SmallString<256> FilePath = makeAbsolute(FileName); ClangTidyOptions EffectiveOptions = OptionsProvider->getOptions(FilePath); - std::vector<std::string> EnabledChecks = + const std::vector<std::string> EnabledChecks = getCheckNames(EffectiveOptions, AllowEnablingAnalyzerAlphaCheckers, ExperimentalCustomChecks); @@ -687,9 +688,9 @@ int clangTidyMain(int argc, const char **argv) { } if (VerifyConfig) { - std::vector<ClangTidyOptionsProvider::OptionsSource> RawOptions = + const std::vector<ClangTidyOptionsProvider::OptionsSource> RawOptions = OptionsProvider->getRawOptions(FileName); - ChecksAndOptions Valid = getAllChecksAndOptions( + const ChecksAndOptions Valid = getAllChecksAndOptions( AllowEnablingAnalyzerAlphaCheckers, ExperimentalCustomChecks); bool AnyInvalid = false; for (const auto &[Opts, Source] : RawOptions) { @@ -733,14 +734,14 @@ int clangTidyMain(int argc, const char **argv) { std::vector<ClangTidyError> Errors = runClangTidy(Context, OptionsParser->getCompilations(), PathList, BaseFS, FixNotes, EnableCheckProfile, ProfilePrefix, Quiet); - bool FoundErrors = llvm::any_of(Errors, [](const ClangTidyError &E) { + const bool FoundErrors = llvm::any_of(Errors, [](const ClangTidyError &E) { return E.DiagLevel == ClangTidyError::Error; }); // --fix-errors and --fix-notes imply --fix. - FixBehaviour Behaviour = FixNotes ? FB_FixNotes - : (Fix || FixErrors) ? FB_Fix - : FB_NoFix; + const FixBehaviour Behaviour = FixNotes ? FB_FixNotes + : (Fix || FixErrors) ? FB_Fix + : FB_NoFix; const bool DisableFixes = FoundErrors && !FixErrors; @@ -769,7 +770,7 @@ int clangTidyMain(int argc, const char **argv) { if (WErrorCount) { if (!Quiet) { - StringRef Plural = WErrorCount == 1 ? "" : "s"; + const StringRef Plural = WErrorCount == 1 ? "" : "s"; llvm::errs() << WErrorCount << " warning" << Plural << " treated as error" << Plural << "\n"; } diff --git a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.h b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.h index f86828e..44b7a37 100644 --- a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.h +++ b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.h @@ -14,8 +14,13 @@ /// //===----------------------------------------------------------------------===// +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_TOOL_CLANGTIDYMAIN_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_TOOL_CLANGTIDYMAIN_H + namespace clang::tidy { int clangTidyMain(int argc, const char **argv); } // namespace clang::tidy + +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_TOOL_CLANGTIDYMAIN_H diff --git a/clang-tools-extra/clang-tidy/utils/ASTUtils.cpp b/clang-tools-extra/clang-tidy/utils/ASTUtils.cpp index d5deb99..2627813 100644 --- a/clang-tools-extra/clang-tidy/utils/ASTUtils.cpp +++ b/clang-tools-extra/clang-tidy/utils/ASTUtils.cpp @@ -67,7 +67,7 @@ bool rangeIsEntirelyWithinMacroArgument(SourceRange Range, // Check if the range is entirely contained within a macro argument. SourceLocation MacroArgExpansionStartForRangeBegin; SourceLocation MacroArgExpansionStartForRangeEnd; - bool RangeIsEntirelyWithinMacroArgument = + const bool RangeIsEntirelyWithinMacroArgument = SM && SM->isMacroArgExpansion(Range.getBegin(), &MacroArgExpansionStartForRangeBegin) && diff --git a/clang-tools-extra/clang-tidy/utils/ASTUtils.h b/clang-tools-extra/clang-tidy/utils/ASTUtils.h index c2127f0..808cd4a 100644 --- a/clang-tools-extra/clang-tidy/utils/ASTUtils.h +++ b/clang-tools-extra/clang-tidy/utils/ASTUtils.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ASTUTILS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ASTUTILS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_ASTUTILS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_ASTUTILS_H #include "clang/AST/AST.h" @@ -47,4 +47,4 @@ findOutermostIndirectFieldDeclForField(const FieldDecl *FD); } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_ASTUTILS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_ASTUTILS_H diff --git a/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.cpp b/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.cpp index 14770c4..aacb4e3 100644 --- a/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.cpp +++ b/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.cpp @@ -48,7 +48,8 @@ FixItHint BraceInsertionHints::closingBraceFixIt() const { static tok::TokenKind getTokenKind(SourceLocation Loc, const SourceManager &SM, const LangOptions &LangOpts) { Token Tok; - SourceLocation Beginning = Lexer::GetBeginningOfToken(Loc, SM, LangOpts); + const SourceLocation Beginning = + Lexer::GetBeginningOfToken(Loc, SM, LangOpts); const bool Invalid = Lexer::getRawToken(Beginning, Tok, SM, LangOpts); assert(!Invalid && "Expected a valid token."); @@ -77,15 +78,16 @@ static SourceLocation findEndLocation(const Stmt &S, const SourceManager &SM, // EOL, insert brace before. break; } - tok::TokenKind TokKind = getTokenKind(Loc, SM, LangOpts); + const tok::TokenKind TokKind = getTokenKind(Loc, SM, LangOpts); if (TokKind != tok::comment) { // Non-comment token, insert brace before. break; } - SourceLocation TokEndLoc = Lexer::getLocForEndOfToken(Loc, 0, SM, LangOpts); - SourceRange TokRange(Loc, TokEndLoc); - StringRef Comment = Lexer::getSourceText( + const SourceLocation TokEndLoc = + Lexer::getLocForEndOfToken(Loc, 0, SM, LangOpts); + const SourceRange TokRange(Loc, TokEndLoc); + const StringRef Comment = Lexer::getSourceText( CharSourceRange::getTokenRange(TokRange), SM, LangOpts); if (Comment.starts_with("/*") && Comment.contains('\n')) { // Multi-line block comment, insert brace before. diff --git a/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.h b/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.h index 53ce2e0..879c84d 100644 --- a/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.h +++ b/clang-tools-extra/clang-tidy/utils/BracesAroundStatement.h @@ -11,6 +11,9 @@ /// //===----------------------------------------------------------------------===// +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_BRACESAROUNDSTATEMENT_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_BRACESAROUNDSTATEMENT_H + #include "clang/AST/Stmt.h" #include "clang/Basic/Diagnostic.h" #include "clang/Basic/SourceLocation.h" @@ -73,3 +76,5 @@ getBraceInsertionsHints(const Stmt *S, const LangOptions &LangOpts, SourceLocation EndLocHint = SourceLocation()); } // namespace clang::tidy::utils + +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_BRACESAROUNDSTATEMENT_H diff --git a/clang-tools-extra/clang-tidy/utils/DeclRefExprUtils.cpp b/clang-tools-extra/clang-tidy/utils/DeclRefExprUtils.cpp index a5b0883..75a6daf 100644 --- a/clang-tools-extra/clang-tidy/utils/DeclRefExprUtils.cpp +++ b/clang-tools-extra/clang-tidy/utils/DeclRefExprUtils.cpp @@ -63,7 +63,7 @@ static bool hasSameParameterTypes(const CXXMethodDecl &D, static const CXXMethodDecl *findConstOverload(const CXXMethodDecl &D) { assert(!D.isConst()); - DeclContext::lookup_result LookupResult = + const DeclContext::lookup_result LookupResult = D.getParent()->lookup(D.getNameInfo().getName()); if (LookupResult.isSingleResult()) { // No overload. diff --git a/clang-tools-extra/clang-tidy/utils/DesignatedInitializers.h b/clang-tools-extra/clang-tidy/utils/DesignatedInitializers.h index 9109601..1960eab 100644 --- a/clang-tools-extra/clang-tidy/utils/DesignatedInitializers.h +++ b/clang-tools-extra/clang-tidy/utils/DesignatedInitializers.h @@ -11,6 +11,9 @@ /// //===----------------------------------------------------------------------===// +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_DESIGNATEDINITIALIZERS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_DESIGNATEDINITIALIZERS_H + #include "clang/AST/Expr.h" #include "clang/Basic/SourceLocation.h" #include "llvm/ADT/DenseMap.h" @@ -40,3 +43,5 @@ llvm::DenseMap<clang::SourceLocation, std::string> getUnwrittenDesignators(const clang::InitListExpr *Syn); } // namespace clang::tidy::utils + +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_DESIGNATEDINITIALIZERS_H diff --git a/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.cpp b/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.cpp index 706dd67..5fd1b73 100644 --- a/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.cpp +++ b/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.cpp @@ -360,8 +360,9 @@ ExceptionAnalyzer::ExceptionInfo::filterByCatch(const Type *HandlerTy, llvm::SmallVector<const Type *, 8> TypesToDelete; for (const auto &ThrownException : ThrownExceptions) { const Type *ExceptionTy = ThrownException.getFirst(); - CanQualType ExceptionCanTy = ExceptionTy->getCanonicalTypeUnqualified(); - CanQualType HandlerCanTy = HandlerTy->getCanonicalTypeUnqualified(); + const CanQualType ExceptionCanTy = + ExceptionTy->getCanonicalTypeUnqualified(); + const CanQualType HandlerCanTy = HandlerTy->getCanonicalTypeUnqualified(); // The handler is of type cv T or cv T& and E and T are the same type // (ignoring the top-level cv-qualifiers) ... @@ -476,7 +477,7 @@ ExceptionAnalyzer::ExceptionInfo ExceptionAnalyzer::throwsException( // For a constructor, we also have to check the initializers. if (const auto *Ctor = dyn_cast<CXXConstructorDecl>(Func)) { for (const CXXCtorInitializer *Init : Ctor->inits()) { - ExceptionInfo Excs = + const ExceptionInfo Excs = throwsException(Init->getInit(), Caught, CallStack); Result.merge(Excs); } @@ -533,7 +534,7 @@ ExceptionAnalyzer::throwsException(const Stmt *St, // Everything is caught through 'catch(...)'. if (!Catch->getExceptionDecl()) { - ExceptionInfo Rethrown = throwsException( + const ExceptionInfo Rethrown = throwsException( Catch->getHandlerBlock(), Uncaught.getExceptions(), CallStack); Results.merge(Rethrown); Uncaught.clear(); @@ -554,7 +555,7 @@ ExceptionAnalyzer::throwsException(const Stmt *St, Uncaught.filterByCatch(CaughtType, Catch->getExceptionDecl()->getASTContext()); if (!FilteredExceptions.empty()) { - ExceptionInfo Rethrown = throwsException( + const ExceptionInfo Rethrown = throwsException( Catch->getHandlerBlock(), FilteredExceptions, CallStack); Results.merge(Rethrown); } @@ -563,44 +564,46 @@ ExceptionAnalyzer::throwsException(const Stmt *St, Results.merge(Uncaught); } else if (const auto *Call = dyn_cast<CallExpr>(St)) { if (const FunctionDecl *Func = Call->getDirectCallee()) { - ExceptionInfo Excs = + const ExceptionInfo Excs = throwsException(Func, Caught, CallStack, Call->getBeginLoc()); Results.merge(Excs); } } else if (const auto *Construct = dyn_cast<CXXConstructExpr>(St)) { - ExceptionInfo Excs = throwsException(Construct->getConstructor(), Caught, - CallStack, Construct->getBeginLoc()); + const ExceptionInfo Excs = + throwsException(Construct->getConstructor(), Caught, CallStack, + Construct->getBeginLoc()); Results.merge(Excs); } else if (const auto *DefaultInit = dyn_cast<CXXDefaultInitExpr>(St)) { - ExceptionInfo Excs = + const ExceptionInfo Excs = throwsException(DefaultInit->getExpr(), Caught, CallStack); Results.merge(Excs); } else if (const auto *Coro = dyn_cast<CoroutineBodyStmt>(St)) { for (const Stmt *Child : Coro->childrenExclBody()) { if (Child != Coro->getExceptionHandler()) { - ExceptionInfo Excs = throwsException(Child, Caught, CallStack); + const ExceptionInfo Excs = throwsException(Child, Caught, CallStack); Results.merge(Excs); } } - ExceptionInfo Excs = throwsException(Coro->getBody(), Caught, CallStack); + const ExceptionInfo Excs = + throwsException(Coro->getBody(), Caught, CallStack); Results.merge(throwsException(Coro->getExceptionHandler(), Excs.getExceptions(), CallStack)); for (const auto &Exception : Excs.getExceptions()) { const Type *ExcType = Exception.getFirst(); if (const CXXRecordDecl *ThrowableRec = ExcType->getAsCXXRecordDecl()) { - ExceptionInfo DestructorExcs = throwsException( + const ExceptionInfo DestructorExcs = throwsException( ThrowableRec->getDestructor(), Caught, CallStack, SourceLocation{}); Results.merge(DestructorExcs); } } } else if (const auto *Lambda = dyn_cast<LambdaExpr>(St)) { for (const Stmt *Init : Lambda->capture_inits()) { - ExceptionInfo Excs = throwsException(Init, Caught, CallStack); + const ExceptionInfo Excs = throwsException(Init, Caught, CallStack); Results.merge(Excs); } } else { for (const Stmt *Child : St->children()) { - ExceptionInfo Excs = throwsException(Child, Caught, CallStack); + const ExceptionInfo Excs = throwsException(Child, Caught, CallStack); Results.merge(Excs); } } diff --git a/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.h b/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.h index 1ab6dcb..1a277c8 100644 --- a/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.h +++ b/clang-tools-extra/clang-tidy/utils/ExceptionAnalyzer.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTION_ANALYZER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTION_ANALYZER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTIONANALYZER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTIONANALYZER_H #include "clang/AST/ASTContext.h" #include "clang/ASTMatchers/ASTMatchFinder.h" @@ -158,4 +158,4 @@ private: } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTION_ANALYZER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTIONANALYZER_H diff --git a/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.cpp b/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.cpp index b1d6b19..2da0966 100644 --- a/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.cpp +++ b/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.cpp @@ -20,7 +20,7 @@ ExceptionSpecAnalyzer::analyze(const FunctionDecl *FuncDecl) { const auto [CacheEntry, NotFound] = FunctionCache.try_emplace(FuncDecl, State::NotThrowing); if (NotFound) { - ExceptionSpecAnalyzer::State State = analyzeImpl(FuncDecl); + const ExceptionSpecAnalyzer::State State = analyzeImpl(FuncDecl); // Update result with calculated value FunctionCache[FuncDecl] = State; return State; @@ -87,20 +87,20 @@ ExceptionSpecAnalyzer::analyzeRecord(const CXXRecordDecl *RecordDecl, return analyze(MethodDecl); for (const auto &BaseSpec : RecordDecl->bases()) { - State Result = analyzeBase(BaseSpec, Kind); + const State Result = analyzeBase(BaseSpec, Kind); if (Result == State::Throwing || Result == State::Unknown) return Result; } for (const auto &BaseSpec : RecordDecl->vbases()) { - State Result = analyzeBase(BaseSpec, Kind); + const State Result = analyzeBase(BaseSpec, Kind); if (Result == State::Throwing || Result == State::Unknown) return Result; } for (const auto *FDecl : RecordDecl->fields()) if (!FDecl->isInvalidDecl() && !FDecl->isUnnamedBitField()) { - State Result = analyzeFieldDecl(FDecl, Kind); + const State Result = analyzeFieldDecl(FDecl, Kind); if (Result == State::Throwing || Result == State::Unknown) return Result; } diff --git a/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.h b/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.h index 3fd6fe1..06d11c8 100644 --- a/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.h +++ b/clang-tools-extra/clang-tidy/utils/ExceptionSpecAnalyzer.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTION_SPEC_ANALYZER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTION_SPEC_ANALYZER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTIONSPECANALYZER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTIONSPECANALYZER_H #include "clang/AST/DeclCXX.h" #include "llvm/ADT/DenseMap.h" @@ -86,4 +86,4 @@ private: } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTION_SPEC_ANALYZER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXCEPTIONSPECANALYZER_H diff --git a/clang-tools-extra/clang-tidy/utils/ExprSequence.cpp b/clang-tools-extra/clang-tidy/utils/ExprSequence.cpp index 46eebf4..0375d0f 100644 --- a/clang-tools-extra/clang-tidy/utils/ExprSequence.cpp +++ b/clang-tools-extra/clang-tidy/utils/ExprSequence.cpp @@ -29,13 +29,13 @@ static SmallVector<const Stmt *, 1> getParentStmts(const Stmt *S, ASTContext *Context) { SmallVector<const Stmt *, 1> Result; - TraversalKindScope RAII(*Context, TK_AsIs); + const TraversalKindScope RAII(*Context, TK_AsIs); DynTypedNodeList Parents = Context->getParents(*S); SmallVector<DynTypedNode, 1> NodesToProcess(Parents.begin(), Parents.end()); while (!NodesToProcess.empty()) { - DynTypedNode Node = NodesToProcess.back(); + const DynTypedNode Node = NodesToProcess.back(); NodesToProcess.pop_back(); if (const auto *S = Node.get<Stmt>()) { @@ -95,7 +95,8 @@ bool ExprSequence::inSequence(const Stmt *Before, const Stmt *After) const { return true; } - SmallVector<const Stmt *, 1> BeforeParents = getParentStmts(Before, Context); + const SmallVector<const Stmt *, 1> BeforeParents = + getParentStmts(Before, Context); // Since C++17, the callee of a call expression is guaranteed to be sequenced // before all of the arguments. diff --git a/clang-tools-extra/clang-tidy/utils/ExprSequence.h b/clang-tools-extra/clang-tidy/utils/ExprSequence.h index 9ef94e0..2aea99e 100644 --- a/clang-tools-extra/clang-tidy/utils/ExprSequence.h +++ b/clang-tools-extra/clang-tidy/utils/ExprSequence.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_EXPRSEQUENCE_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_EXPRSEQUENCE_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXPRSEQUENCE_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXPRSEQUENCE_H #include "clang/Analysis/CFG.h" #include "clang/Lex/Lexer.h" @@ -117,4 +117,4 @@ private: } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_EXPRSEQUENCE_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_EXPRSEQUENCE_H diff --git a/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.cpp b/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.cpp index 41d5131..97be36a 100644 --- a/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.cpp +++ b/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.cpp @@ -15,19 +15,19 @@ namespace clang::tidy::utils { bool isExpansionLocInHeaderFile(SourceLocation Loc, const SourceManager &SM, const FileExtensionsSet &HeaderFileExtensions) { - SourceLocation ExpansionLoc = SM.getExpansionLoc(Loc); + const SourceLocation ExpansionLoc = SM.getExpansionLoc(Loc); return isFileExtension(SM.getFilename(ExpansionLoc), HeaderFileExtensions); } bool isPresumedLocInHeaderFile(SourceLocation Loc, SourceManager &SM, const FileExtensionsSet &HeaderFileExtensions) { - PresumedLoc PresumedLocation = SM.getPresumedLoc(Loc); + const PresumedLoc PresumedLocation = SM.getPresumedLoc(Loc); return isFileExtension(PresumedLocation.getFilename(), HeaderFileExtensions); } bool isSpellingLocInHeaderFile(SourceLocation Loc, SourceManager &SM, const FileExtensionsSet &HeaderFileExtensions) { - SourceLocation SpellingLoc = SM.getSpellingLoc(Loc); + const SourceLocation SpellingLoc = SM.getSpellingLoc(Loc); return isFileExtension(SM.getFilename(SpellingLoc), HeaderFileExtensions); } @@ -35,7 +35,7 @@ bool parseFileExtensions(StringRef AllFileExtensions, FileExtensionsSet &FileExtensions, StringRef Delimiters) { SmallVector<StringRef, 5> Suffixes; - for (char Delimiter : Delimiters) { + for (const char Delimiter : Delimiters) { if (AllFileExtensions.contains(Delimiter)) { AllFileExtensions.split(Suffixes, Delimiter); break; @@ -43,7 +43,7 @@ bool parseFileExtensions(StringRef AllFileExtensions, } FileExtensions.clear(); - for (StringRef Suffix : Suffixes) { + for (const StringRef Suffix : Suffixes) { StringRef Extension = Suffix.trim(); if (!llvm::all_of(Extension, isAlphanumeric)) return false; diff --git a/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.h b/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.h index dfab141..425c29c 100644 --- a/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.h +++ b/clang-tools-extra/clang-tidy/utils/FileExtensionsUtils.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_FILE_EXTENSIONS_UTILS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_FILE_EXTENSIONS_UTILS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_FILEEXTENSIONSUTILS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_FILEEXTENSIONSUTILS_H #include "../FileExtensionsSet.h" #include "clang/Basic/SourceLocation.h" @@ -60,4 +60,4 @@ bool isFileExtension(StringRef FileName, } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_FILE_EXTENSIONS_UTILS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_FILEEXTENSIONSUTILS_H diff --git a/clang-tools-extra/clang-tidy/utils/FixItHintUtils.cpp b/clang-tools-extra/clang-tidy/utils/FixItHintUtils.cpp index b30c83e..c4cdf0d 100644 --- a/clang-tools-extra/clang-tidy/utils/FixItHintUtils.cpp +++ b/clang-tools-extra/clang-tidy/utils/FixItHintUtils.cpp @@ -140,7 +140,7 @@ changePointer(const VarDecl &Var, Qualifiers::TQ Qualifier, const Type *Pointee, // the `*` token and placing the `const` left of it. // (`int const* p = nullptr;`) if (QualPolicy == QualifierPolicy::Right) { - SourceLocation BeforeStar = lexer::findPreviousTokenKind( + const SourceLocation BeforeStar = lexer::findPreviousTokenKind( Var.getLocation(), Context.getSourceManager(), Context.getLangOpts(), tok::star); if (locDangerous(BeforeStar)) @@ -161,7 +161,7 @@ changePointer(const VarDecl &Var, Qualifiers::TQ Qualifier, const Type *Pointee, // is the same as 'QualPolicy == Right && isValueType(Pointee)'. // The `const` must be left of the last `*` token. // (`int * const* p = nullptr;`) - SourceLocation BeforeStar = lexer::findPreviousTokenKind( + const SourceLocation BeforeStar = lexer::findPreviousTokenKind( Var.getLocation(), Context.getSourceManager(), Context.getLangOpts(), tok::star); return fixIfNotDangerous(BeforeStar, buildQualifier(Qualifier, true)); @@ -178,7 +178,7 @@ changeReferencee(const VarDecl &Var, Qualifiers::TQ Qualifier, QualType Pointee, return fixIfNotDangerous(Var.getTypeSpecStartLoc(), buildQualifier(Qualifier)); - SourceLocation BeforeRef = lexer::findPreviousAnyTokenKind( + const SourceLocation BeforeRef = lexer::findPreviousAnyTokenKind( Var.getLocation(), Context.getSourceManager(), Context.getLangOpts(), tok::amp, tok::ampamp); std::optional<SourceLocation> IgnoredParens = @@ -201,7 +201,7 @@ std::optional<FixItHint> addQualifierToVarDecl(const VarDecl &Var, QualTarget == QualifierTarget::Value) && "Unexpected Target"); - QualType ParenStrippedType = Var.getType().IgnoreParens(); + const QualType ParenStrippedType = Var.getType().IgnoreParens(); if (isValueType(ParenStrippedType)) return changeValue(Var, Qualifier, QualTarget, QualPolicy, Context); diff --git a/clang-tools-extra/clang-tidy/utils/FormatStringConverter.cpp b/clang-tools-extra/clang-tidy/utils/FormatStringConverter.cpp index f4945b2..127de30 100644 --- a/clang-tools-extra/clang-tidy/utils/FormatStringConverter.cpp +++ b/clang-tools-extra/clang-tidy/utils/FormatStringConverter.cpp @@ -245,7 +245,7 @@ FormatStringConverter::formatStringContainsUnreplaceableMacro( // inhibit conversion. The whole format string will appear to come from that // macro, as will the function call. std::optional<StringRef> MaybeSurroundingMacroName; - if (SourceLocation BeginCallLoc = Call->getBeginLoc(); + if (const SourceLocation BeginCallLoc = Call->getBeginLoc(); BeginCallLoc.isMacroID()) MaybeSurroundingMacroName = Lexer::getImmediateMacroName(BeginCallLoc, SM, PP.getLangOpts()); @@ -283,7 +283,8 @@ FormatStringConverter::formatStringContainsUnreplaceableMacro( void FormatStringConverter::emitAlignment(const PrintfSpecifier &FS, std::string &FormatSpec) { - ConversionSpecifier::Kind ArgKind = FS.getConversionSpecifier().getKind(); + const ConversionSpecifier::Kind ArgKind = + FS.getConversionSpecifier().getKind(); // We only care about alignment if a field width is specified if (FS.getFieldWidth().getHowSpecified() != OptionalAmount::NotSpecified) { @@ -499,7 +500,8 @@ bool FormatStringConverter::emitIntegerArgument( /// @returns true on success, false on failure bool FormatStringConverter::emitType(const PrintfSpecifier &FS, const Expr *Arg, std::string &FormatSpec) { - ConversionSpecifier::Kind ArgKind = FS.getConversionSpecifier().getKind(); + const ConversionSpecifier::Kind ArgKind = + FS.getConversionSpecifier().getKind(); switch (ArgKind) { case ConversionSpecifier::Kind::sArg: emitStringArgument(FS.getArgIndex() + ArgsOffset, Arg); @@ -798,7 +800,7 @@ void FormatStringConverter::applyFixes(DiagnosticBuilder &Diag, } for (const auto &[ArgIndex, Replacement] : ArgFixes) { - SourceLocation AfterOtherSide = + const SourceLocation AfterOtherSide = Lexer::findNextToken(Args[ArgIndex]->getEndLoc(), SM, LangOpts) ->getLocation(); diff --git a/clang-tools-extra/clang-tidy/utils/HeaderGuard.cpp b/clang-tools-extra/clang-tidy/utils/HeaderGuard.cpp index e1d1387..d36b187 100644 --- a/clang-tools-extra/clang-tidy/utils/HeaderGuard.cpp +++ b/clang-tools-extra/clang-tidy/utils/HeaderGuard.cpp @@ -32,11 +32,11 @@ public: FileID PrevFID) override { // Record all files we enter. We'll need them to diagnose headers without // guards. - SourceManager &SM = PP->getSourceManager(); + const SourceManager &SM = PP->getSourceManager(); if (Reason == EnterFile && FileType == SrcMgr::C_User) { if (OptionalFileEntryRef FE = SM.getFileEntryRefForID(SM.getFileID(Loc))) { - std::string FileName = cleanPath(FE->getName()); + const std::string FileName = cleanPath(FE->getName()); Files[FileName] = *FE; } } @@ -66,7 +66,7 @@ public: void EndOfMainFile() override { // Now that we have all this information from the preprocessor, use it! - SourceManager &SM = PP->getSourceManager(); + const SourceManager &SM = PP->getSourceManager(); for (const auto &MacroEntry : Macros) { const MacroInfo *MI = MacroEntry.second; @@ -79,7 +79,7 @@ public: OptionalFileEntryRef FE = SM.getFileEntryRefForID(SM.getFileID(MI->getDefinitionLoc())); - std::string FileName = cleanPath(FE->getName()); + const std::string FileName = cleanPath(FE->getName()); Files.erase(FileName); // See if we should check and fix this header guard. @@ -88,16 +88,16 @@ public: // Look up Locations for this guard. const auto &Locs = Ifndefs[MacroEntry.first.getIdentifierInfo()]; - SourceLocation Ifndef = Locs.second; - SourceLocation Define = MacroEntry.first.getLocation(); - SourceLocation EndIf = EndIfs[Locs.first]; + const SourceLocation Ifndef = Locs.second; + const SourceLocation Define = MacroEntry.first.getLocation(); + const SourceLocation EndIf = EndIfs[Locs.first]; // If the macro Name is not equal to what we can compute, correct it in // the #ifndef and #define. - StringRef CurHeaderGuard = + const StringRef CurHeaderGuard = MacroEntry.first.getIdentifierInfo()->getName(); std::vector<FixItHint> FixIts; - std::string NewGuard = checkHeaderGuardDefinition( + const std::string NewGuard = checkHeaderGuardDefinition( Ifndef, Define, EndIf, FileName, CurHeaderGuard, FixIts); // Now look at the #endif. We want a comment with the header guard. Fix it @@ -129,7 +129,7 @@ public: if (!EndIf.isValid()) return false; const char *EndIfData = PP->getSourceManager().getCharacterData(EndIf); - size_t EndIfLen = std::strcspn(EndIfData, "\r\n"); + const size_t EndIfLen = std::strcspn(EndIfData, "\r\n"); if (EndIfLenPtr) *EndIfLenPtr = EndIfLen; @@ -137,12 +137,12 @@ public: EndIfStr = EndIfStr.substr(EndIfStr.find_first_not_of("#endif \t")); // Give up if there's an escaped newline. - size_t FindEscapedNewline = EndIfStr.find_last_not_of(' '); + const size_t FindEscapedNewline = EndIfStr.find_last_not_of(' '); if (FindEscapedNewline != StringRef::npos && EndIfStr[FindEscapedNewline] == '\\') return false; - bool IsLineComment = + const bool IsLineComment = EndIfStr.consume_front("//") || (EndIfStr.consume_front("/*") && EndIfStr.consume_back("*/")); if (!IsLineComment) @@ -162,7 +162,7 @@ public: std::vector<FixItHint> &FixIts) { std::string CPPVar = Check->getHeaderGuard(FileName, CurHeaderGuard); CPPVar = Check->sanitizeHeaderGuard(CPPVar); - std::string CPPVarUnder = CPPVar + '_'; + const std::string CPPVarUnder = CPPVar + '_'; // Allow a trailing underscore if and only if we don't have to change the // endif comment too. @@ -203,19 +203,20 @@ public: // fix-its to add the guard. // TODO: Insert the guard after top comments. for (const auto &FE : Files) { - StringRef FileName = FE.getKey(); + const StringRef FileName = FE.getKey(); if (!Check->shouldSuggestToAddHeaderGuard(FileName)) continue; - SourceManager &SM = PP->getSourceManager(); - FileID FID = SM.translateFile(FE.getValue()); - SourceLocation StartLoc = SM.getLocForStartOfFile(FID); + const SourceManager &SM = PP->getSourceManager(); + const FileID FID = SM.translateFile(FE.getValue()); + const SourceLocation StartLoc = SM.getLocForStartOfFile(FID); if (StartLoc.isInvalid()) continue; std::string CPPVar = Check->getHeaderGuard(FileName); CPPVar = Check->sanitizeHeaderGuard(CPPVar); - std::string CPPVarUnder = CPPVar + '_'; // Allow a trailing underscore. + const std::string CPPVarUnder = + CPPVar + '_'; // Allow a trailing underscore. // If there's a macro with a name that follows the header guard convention // but was not recognized by the preprocessor as a header guard there must // be code outside of the guarded area. Emit a plain warning without @@ -223,8 +224,8 @@ public: // FIXME: Can we move it into the right spot? bool SeenMacro = false; for (const auto &MacroEntry : Macros) { - StringRef Name = MacroEntry.first.getIdentifierInfo()->getName(); - SourceLocation DefineLoc = MacroEntry.first.getLocation(); + const StringRef Name = MacroEntry.first.getIdentifierInfo()->getName(); + const SourceLocation DefineLoc = MacroEntry.first.getLocation(); if ((Name == CPPVar || Name == CPPVarUnder) && SM.isWrittenInSameFile(StartLoc, DefineLoc)) { Check->diag(DefineLoc, "code/includes outside of area guarded by " diff --git a/clang-tools-extra/clang-tidy/utils/IncludeInserter.cpp b/clang-tools-extra/clang-tidy/utils/IncludeInserter.cpp index 0b67cba..81bb8de 100644 --- a/clang-tools-extra/clang-tidy/utils/IncludeInserter.cpp +++ b/clang-tools-extra/clang-tidy/utils/IncludeInserter.cpp @@ -69,7 +69,7 @@ IncludeSorter &IncludeInserter::getOrCreate(FileID FileID) { std::optional<FixItHint> IncludeInserter::createIncludeInsertion(FileID FileID, llvm::StringRef Header) { - bool IsAngled = Header.consume_front("<"); + const bool IsAngled = Header.consume_front("<"); if (IsAngled != Header.consume_back(">")) return std::nullopt; // We assume the same Header will never be included both angled and not @@ -94,7 +94,7 @@ void IncludeInserter::addInclude(StringRef FileName, bool IsAngled, SourceLocation EndLocation) { assert(SourceMgr && "SourceMgr shouldn't be null; did you remember to call " "registerPreprocessor()?"); - FileID FileID = SourceMgr->getFileID(HashLocation); + const FileID FileID = SourceMgr->getFileID(HashLocation); getOrCreate(FileID).addInclude(FileName, IsAngled, HashLocation, EndLocation); } diff --git a/clang-tools-extra/clang-tidy/utils/IncludeInserter.h b/clang-tools-extra/clang-tidy/utils/IncludeInserter.h index f6ca7d6..9dbf2a7 100644 --- a/clang-tools-extra/clang-tidy/utils/IncludeInserter.h +++ b/clang-tools-extra/clang-tidy/utils/IncludeInserter.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_INCLUDEINSERTER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_INCLUDEINSERTER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_INCLUDEINSERTER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_INCLUDEINSERTER_H #include "IncludeSorter.h" #include "clang/Basic/Diagnostic.h" @@ -100,4 +100,4 @@ private: } // namespace tidy::utils } // namespace clang -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_INCLUDEINSERTER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_INCLUDEINSERTER_H diff --git a/clang-tools-extra/clang-tidy/utils/IncludeSorter.cpp b/clang-tools-extra/clang-tidy/utils/IncludeSorter.cpp index 7e2aad9..f113f8c 100644 --- a/clang-tools-extra/clang-tidy/utils/IncludeSorter.cpp +++ b/clang-tools-extra/clang-tidy/utils/IncludeSorter.cpp @@ -17,7 +17,7 @@ namespace utils { static StringRef removeFirstSuffix(StringRef Str, ArrayRef<const char *> Suffixes) { - for (StringRef Suffix : Suffixes) { + for (const StringRef Suffix : Suffixes) { if (Str.consume_back(Suffix)) return Str; } @@ -37,7 +37,7 @@ static StringRef makeCanonicalName(StringRef Str, removeFirstSuffix(Str, {".cc", ".cpp", ".c", ".h", ".hpp"}), {"Test"}); } if (Style == IncludeSorter::IS_Google_ObjC) { - StringRef Canonical = + const StringRef Canonical = removeFirstSuffix(removeFirstSuffix(Str, {".cc", ".cpp", ".c", ".h", ".hpp", ".mm", ".m"}), {"_unittest", "_regtest", "_test", "Test"}); @@ -57,7 +57,7 @@ static StringRef makeCanonicalName(StringRef Str, // Scan to the end of the line and return the offset of the next line. static size_t findNextLine(const char *Text) { - size_t EOLIndex = std::strcspn(Text, "\n"); + const size_t EOLIndex = std::strcspn(Text, "\n"); return Text[EOLIndex] == '\0' ? EOLIndex : EOLIndex + 1; } @@ -74,14 +74,15 @@ determineIncludeKind(StringRef CanonicalFile, StringRef IncludeFile, return IncludeFile.ends_with(".h") ? IncludeSorter::IK_CSystemInclude : IncludeSorter::IK_CXXSystemInclude; } - StringRef CanonicalInclude = makeCanonicalName(IncludeFile, Style); + const StringRef CanonicalInclude = makeCanonicalName(IncludeFile, Style); if (CanonicalFile.ends_with(CanonicalInclude) || CanonicalInclude.ends_with(CanonicalFile)) { return IncludeSorter::IK_MainTUInclude; } if ((Style == IncludeSorter::IS_Google) || (Style == IncludeSorter::IS_Google_ObjC)) { - std::pair<StringRef, StringRef> Parts = CanonicalInclude.split("/public/"); + const std::pair<StringRef, StringRef> Parts = + CanonicalInclude.split("/public/"); StringRef FileCopy = CanonicalFile; if (FileCopy.consume_front(Parts.first) && FileCopy.consume_back(Parts.second)) { @@ -126,7 +127,7 @@ IncludeSorter::IncludeSorter(const SourceManager *SourceMgr, FileID FileID, void IncludeSorter::addInclude(StringRef FileName, bool IsAngled, SourceLocation HashLocation, SourceLocation EndLocation) { - int Offset = findNextLine(SourceMgr->getCharacterData(EndLocation)); + const int Offset = findNextLine(SourceMgr->getCharacterData(EndLocation)); // Record the relevant location information for this inclusion directive. auto &IncludeLocation = IncludeLocations[FileName]; @@ -139,7 +140,7 @@ void IncludeSorter::addInclude(StringRef FileName, bool IsAngled, return; // Add the included file's name to the appropriate bucket. - IncludeKinds Kind = + const IncludeKinds Kind = determineIncludeKind(CanonicalFile, FileName, IsAngled, Style); if (Kind != IK_InvalidInclude) IncludeBucket[Kind].push_back(FileName.str()); @@ -181,7 +182,8 @@ IncludeSorter::createIncludeInsertion(StringRef FileName, bool IsAngled) { // FileName comes after all include entries in bucket, insert it after // last. const std::string &LastInclude = IncludeBucket[IncludeKind].back(); - SourceRange LastIncludeLocation = IncludeLocations[LastInclude].back(); + const SourceRange LastIncludeLocation = + IncludeLocations[LastInclude].back(); return FixItHint::CreateInsertion(LastIncludeLocation.getEnd(), IncludeStmt); } @@ -205,14 +207,16 @@ IncludeSorter::createIncludeInsertion(StringRef FileName, bool IsAngled) { if (NonEmptyKind < IncludeKind) { // Create a block after. const std::string &LastInclude = IncludeBucket[NonEmptyKind].back(); - SourceRange LastIncludeLocation = IncludeLocations[LastInclude].back(); + const SourceRange LastIncludeLocation = + IncludeLocations[LastInclude].back(); IncludeStmt = '\n' + IncludeStmt; return FixItHint::CreateInsertion(LastIncludeLocation.getEnd(), IncludeStmt); } // Create a block before. const std::string &FirstInclude = IncludeBucket[NonEmptyKind][0]; - SourceRange FirstIncludeLocation = IncludeLocations[FirstInclude].back(); + const SourceRange FirstIncludeLocation = + IncludeLocations[FirstInclude].back(); IncludeStmt.append("\n"); return FixItHint::CreateInsertion(FirstIncludeLocation.getBegin(), IncludeStmt); diff --git a/clang-tools-extra/clang-tidy/utils/IncludeSorter.h b/clang-tools-extra/clang-tidy/utils/IncludeSorter.h index 66830ee..6efec97 100644 --- a/clang-tools-extra/clang-tidy/utils/IncludeSorter.h +++ b/clang-tools-extra/clang-tidy/utils/IncludeSorter.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_INCLUDESORTER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_INCLUDESORTER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_INCLUDESORTER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_INCLUDESORTER_H #include "../ClangTidyCheck.h" #include <optional> @@ -73,4 +73,4 @@ template <> struct OptionEnumMapping<utils::IncludeSorter::IncludeStyle> { getEnumMapping(); }; } // namespace clang::tidy -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_INCLUDESORTER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_INCLUDESORTER_H diff --git a/clang-tools-extra/clang-tidy/utils/LexerUtils.cpp b/clang-tools-extra/clang-tidy/utils/LexerUtils.cpp index 7222f64..0f20fde 100644 --- a/clang-tools-extra/clang-tidy/utils/LexerUtils.cpp +++ b/clang-tools-extra/clang-tidy/utils/LexerUtils.cpp @@ -42,7 +42,7 @@ SourceLocation findPreviousTokenStart(SourceLocation Start, if (Start.isInvalid() || Start.isMacroID()) return {}; - SourceLocation BeforeStart = Start.getLocWithOffset(-1); + const SourceLocation BeforeStart = Start.getLocWithOffset(-1); if (BeforeStart.isInvalid() || BeforeStart.isMacroID()) return {}; @@ -57,7 +57,7 @@ SourceLocation findPreviousTokenKind(SourceLocation Start, return {}; while (true) { - SourceLocation L = findPreviousTokenStart(Start, SM, LangOpts); + const SourceLocation L = findPreviousTokenStart(Start, SM, LangOpts); if (L.isInvalid() || L.isMacroID()) return {}; @@ -123,8 +123,9 @@ std::optional<Token> getQualifyingToken(tok::TokenKind TK, assert((TK == tok::kw_const || TK == tok::kw_volatile || TK == tok::kw_restrict) && "TK is not a qualifier keyword"); - std::pair<FileID, unsigned> LocInfo = SM.getDecomposedLoc(Range.getBegin()); - StringRef File = SM.getBufferData(LocInfo.first); + const std::pair<FileID, unsigned> LocInfo = + SM.getDecomposedLoc(Range.getBegin()); + const StringRef File = SM.getBufferData(LocInfo.first); Lexer RawLexer(SM.getLocForStartOfFile(LocInfo.first), Context.getLangOpts(), File.begin(), File.data() + LocInfo.second, File.end()); std::optional<Token> LastMatchBeforeTemplate; diff --git a/clang-tools-extra/clang-tidy/utils/LexerUtils.h b/clang-tools-extra/clang-tidy/utils/LexerUtils.h index b76a378..c5fb646 100644 --- a/clang-tools-extra/clang-tidy/utils/LexerUtils.h +++ b/clang-tools-extra/clang-tidy/utils/LexerUtils.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_LEXER_UTILS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_LEXER_UTILS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_LEXERUTILS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_LEXERUTILS_H #include "clang/AST/ASTContext.h" #include "clang/Basic/TokenKinds.h" @@ -48,7 +48,7 @@ SourceLocation findPreviousAnyTokenKind(SourceLocation Start, if (Start.isInvalid() || Start.isMacroID()) return {}; while (true) { - SourceLocation L = findPreviousTokenStart(Start, SM, LangOpts); + const SourceLocation L = findPreviousTokenStart(Start, SM, LangOpts); if (L.isInvalid() || L.isMacroID()) return {}; @@ -76,7 +76,7 @@ SourceLocation findNextAnyTokenKind(SourceLocation Start, if (!CurrentToken) return {}; - Token PotentialMatch = *CurrentToken; + const Token PotentialMatch = *CurrentToken; if (PotentialMatch.isOneOf(TK, TKs...)) return PotentialMatch.getLocation(); @@ -130,4 +130,4 @@ SourceLocation getLocationForNoexceptSpecifier(const FunctionDecl *FuncDecl, } // namespace tidy::utils::lexer } // namespace clang -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_LEXER_UTILS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_LEXERUTILS_H diff --git a/clang-tools-extra/clang-tidy/utils/Matchers.h b/clang-tools-extra/clang-tidy/utils/Matchers.h index 4eac065..a444ab62 100644 --- a/clang-tools-extra/clang-tidy/utils/Matchers.h +++ b/clang-tools-extra/clang-tidy/utils/Matchers.h @@ -162,7 +162,7 @@ struct NotIdenticalStatementsPredicate { // Checks if statement is identical (utils::areStatementsIdentical) to one bound // to ID node. AST_MATCHER_P(Stmt, isStatementIdenticalToBoundNode, std::string, ID) { - NotIdenticalStatementsPredicate Predicate{ + const NotIdenticalStatementsPredicate Predicate{ ID, ::clang::DynTypedNode::create(Node), &(Finder->getASTContext())}; return Builder->removeBindings(Predicate); } diff --git a/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.cpp b/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.cpp index 3af7f8d..c862364 100644 --- a/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.cpp +++ b/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.cpp @@ -55,7 +55,7 @@ NamespaceAliaser::createAlias(ASTContext &Context, const Stmt &Statement, } for (const auto &Abbreviation : Abbreviations) { - DeclarationMatcher ConflictMatcher = namedDecl(hasName(Abbreviation)); + const DeclarationMatcher ConflictMatcher = namedDecl(hasName(Abbreviation)); const auto HasConflictingChildren = !match(findAll(ConflictMatcher), *Function, Context).empty(); const auto HasConflictingAncestors = @@ -65,10 +65,10 @@ NamespaceAliaser::createAlias(ASTContext &Context, const Stmt &Statement, if (HasConflictingAncestors || HasConflictingChildren) continue; - std::string Declaration = + const std::string Declaration = (llvm::Twine("\nnamespace ") + Abbreviation + " = " + Namespace + ";") .str(); - SourceLocation Loc = + const SourceLocation Loc = Lexer::getLocForEndOfToken(Function->getBody()->getBeginLoc(), 0, SourceMgr, Context.getLangOpts()); AddedAliases[Function][Namespace.str()] = Abbreviation; diff --git a/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.h b/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.h index 497b67e..fbf0ade 100644 --- a/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.h +++ b/clang-tools-extra/clang-tidy/utils/NamespaceAliaser.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_NAMESPACEALIASER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_NAMESPACEALIASER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_NAMESPACEALIASER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_NAMESPACEALIASER_H #include "clang/AST/ASTContext.h" #include "clang/AST/Stmt.h" @@ -45,4 +45,4 @@ private: } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_NAMESPACEALIASER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_NAMESPACEALIASER_H diff --git a/clang-tools-extra/clang-tidy/utils/OptionsUtils.h b/clang-tools-extra/clang-tidy/utils/OptionsUtils.h index aec24ab..3a12348 100644 --- a/clang-tools-extra/clang-tidy/utils/OptionsUtils.h +++ b/clang-tools-extra/clang-tidy/utils/OptionsUtils.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_OPTIONUTILS_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_OPTIONUTILS_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_OPTIONSUTILS_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_OPTIONSUTILS_H #include "clang/Basic/LLVM.h" #include <string> @@ -26,4 +26,4 @@ std::string serializeStringList(ArrayRef<StringRef> Strings); } // namespace clang::tidy::utils::options -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MISC_OPTIONUTILS_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_OPTIONSUTILS_H diff --git a/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.cpp b/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.cpp index 6bd6d98..bc6de97 100644 --- a/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.cpp +++ b/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.cpp @@ -103,7 +103,7 @@ static const CXXMethodDecl *getOverrideMethod(const CXXMethodDecl *Method) { while (true) { Method = *Method->begin_overridden_methods(); assert(Method && "Overridden method shouldn't be null"); - unsigned NumOverrides = Method->size_overridden_methods(); + const unsigned NumOverrides = Method->size_overridden_methods(); if (NumOverrides == 0) return Method; if (NumOverrides > 1) @@ -148,7 +148,7 @@ static NameLookup findDeclInBases(const CXXRecordDecl &Parent, return NameLookup(InClassRef); const NamedDecl *Found = nullptr; - for (CXXBaseSpecifier Base : Parent.bases()) { + for (const CXXBaseSpecifier Base : Parent.bases()) { const auto *Record = Base.getType()->getAsCXXRecordDecl(); if (!Record && AggressiveTemplateLookup) { if (const auto *TST = @@ -269,7 +269,7 @@ public: } bool VisitNamedDecl(NamedDecl *Decl) { - SourceRange UsageRange = + const SourceRange UsageRange = DeclarationNameInfo(Decl->getDeclName(), Decl->getLocation()) .getSourceRange(); Check->addUsage(Decl, UsageRange, SM); @@ -277,13 +277,13 @@ public: } bool VisitDeclRefExpr(DeclRefExpr *DeclRef) { - SourceRange Range = DeclRef->getNameInfo().getSourceRange(); + const SourceRange Range = DeclRef->getNameInfo().getSourceRange(); Check->addUsage(DeclRef->getDecl(), Range, SM); return true; } bool TraverseNestedNameSpecifierLoc(NestedNameSpecifierLoc Loc) { - if (NestedNameSpecifier Spec = Loc.getNestedNameSpecifier(); + if (const NestedNameSpecifier Spec = Loc.getNestedNameSpecifier(); Spec.getKind() == NestedNameSpecifier::Kind::Namespace) { if (const auto *Decl = dyn_cast<NamespaceDecl>(Spec.getAsNamespaceAndPrefix().Namespace)) @@ -295,27 +295,28 @@ public: } bool VisitMemberExpr(MemberExpr *MemberRef) { - SourceRange Range = MemberRef->getMemberNameInfo().getSourceRange(); + const SourceRange Range = MemberRef->getMemberNameInfo().getSourceRange(); Check->addUsage(MemberRef->getMemberDecl(), Range, SM); return true; } bool VisitCXXDependentScopeMemberExpr(CXXDependentScopeMemberExpr *DepMemberRef) { - QualType BaseType = DepMemberRef->isArrow() - ? DepMemberRef->getBaseType()->getPointeeType() - : DepMemberRef->getBaseType(); + const QualType BaseType = + DepMemberRef->isArrow() ? DepMemberRef->getBaseType()->getPointeeType() + : DepMemberRef->getBaseType(); if (BaseType.isNull()) return true; const CXXRecordDecl *Base = BaseType.getTypePtr()->getAsCXXRecordDecl(); if (!Base) return true; - DeclarationName DeclName = DepMemberRef->getMemberNameInfo().getName(); + const DeclarationName DeclName = + DepMemberRef->getMemberNameInfo().getName(); if (!DeclName.isIdentifier()) return true; - StringRef DependentName = DeclName.getAsIdentifierInfo()->getName(); + const StringRef DependentName = DeclName.getAsIdentifierInfo()->getName(); - if (NameLookup Resolved = findDeclInBases( + if (const NameLookup Resolved = findDeclInBases( *Base, DependentName, AggressiveDependentMemberLookup)) { if (*Resolved) Check->addUsage(*Resolved, @@ -370,7 +371,7 @@ public: const IdentifierInfo *II = FD->getIdentifier(); if (!II) continue; - SourceRange FixLocation{D.getFieldLoc(), D.getFieldLoc()}; + const SourceRange FixLocation{D.getFieldLoc(), D.getFieldLoc()}; Check->addUsage(FD, FixLocation, SM); } @@ -473,7 +474,8 @@ void RenamerClangTidyCheck::addUsage(const NamedDecl *Decl, if (!MaybeFailure) return; - NamingCheckId FailureId(FailureDecl->getLocation(), FailureDecl->getName()); + const NamingCheckId FailureId(FailureDecl->getLocation(), + FailureDecl->getName()); auto [FailureIter, NewFailure] = addUsage(FailureId, UsageRange, SourceMgr); @@ -527,10 +529,10 @@ void RenamerClangTidyCheck::checkMacro(const Token &MacroNameTok, if (!MaybeFailure) return; FailureInfo &Info = *MaybeFailure; - StringRef Name = MacroNameTok.getIdentifierInfo()->getName(); - NamingCheckId ID(MI->getDefinitionLoc(), Name); + const StringRef Name = MacroNameTok.getIdentifierInfo()->getName(); + const NamingCheckId ID(MI->getDefinitionLoc(), Name); NamingCheckFailure &Failure = NamingCheckFailures[ID]; - SourceRange Range(MacroNameTok.getLocation(), MacroNameTok.getEndLoc()); + const SourceRange Range(MacroNameTok.getLocation(), MacroNameTok.getEndLoc()); if (!isValidAsciiIdentifier(Info.Fixup)) Failure.FixStatus = ShouldFixStatus::FixInvalidIdentifier; @@ -542,14 +544,14 @@ void RenamerClangTidyCheck::checkMacro(const Token &MacroNameTok, void RenamerClangTidyCheck::expandMacro(const Token &MacroNameTok, const MacroInfo *MI, const SourceManager &SourceMgr) { - StringRef Name = MacroNameTok.getIdentifierInfo()->getName(); - NamingCheckId ID(MI->getDefinitionLoc(), Name); + const StringRef Name = MacroNameTok.getIdentifierInfo()->getName(); + const NamingCheckId ID(MI->getDefinitionLoc(), Name); auto Failure = NamingCheckFailures.find(ID); if (Failure == NamingCheckFailures.end()) return; - SourceRange Range(MacroNameTok.getLocation(), MacroNameTok.getEndLoc()); + const SourceRange Range(MacroNameTok.getLocation(), MacroNameTok.getEndLoc()); addUsage(ID, Range, SourceMgr); } diff --git a/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.h b/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.h index b38bc08..fe1059d5 100644 --- a/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.h +++ b/clang-tools-extra/clang-tidy/utils/RenamerClangTidyCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_RENAMERCLANGTIDYCHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_RENAMERCLANGTIDYCHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_RENAMERCLANGTIDYCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_RENAMERCLANGTIDYCHECK_H #include "../ClangTidyCheck.h" #include "llvm/ADT/DenseMap.h" @@ -167,4 +167,4 @@ private: } // namespace tidy } // namespace clang -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_RENAMERCLANGTIDYCHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_RENAMERCLANGTIDYCHECK_H diff --git a/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.cpp b/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.cpp index 87602d1..d7dad62 100644 --- a/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.cpp +++ b/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.cpp @@ -105,7 +105,7 @@ void TransformerClangTidyCheck::check( if (Result.Context->getDiagnostics().hasErrorOccurred()) return; - size_t I = transformer::detail::findSelectedCase(Result, Rule); + const size_t I = transformer::detail::findSelectedCase(Result, Rule); Expected<SmallVector<transformer::Edit, 1>> Edits = Rule.Cases[I].Edits(Result); if (!Edits) { @@ -127,7 +127,7 @@ void TransformerClangTidyCheck::check( // Associate the diagnostic with the location of the first change. { - DiagnosticBuilder Diag = + const DiagnosticBuilder Diag = diag((*Edits)[0].Range.getBegin(), escapeForDiagnostic(*Explanation)); for (const auto &T : *Edits) { switch (T.Kind) { diff --git a/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.h b/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.h index ad20fbd..e77f84b 100644 --- a/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.h +++ b/clang-tools-extra/clang-tidy/utils/TransformerClangTidyCheck.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_TRANSFORMER_CLANG_TIDY_CHECK_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_TRANSFORMER_CLANG_TIDY_CHECK_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_TRANSFORMERCLANGTIDYCHECK_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_TRANSFORMERCLANGTIDYCHECK_H #include "../ClangTidyCheck.h" #include "IncludeInserter.h" @@ -83,4 +83,4 @@ private: } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_TRANSFORMER_CLANG_TIDY_CHECK_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_TRANSFORMERCLANGTIDYCHECK_H diff --git a/clang-tools-extra/clang-tidy/utils/TypeTraits.cpp b/clang-tools-extra/clang-tidy/utils/TypeTraits.cpp index d4e079f..98a5d40 100644 --- a/clang-tools-extra/clang-tidy/utils/TypeTraits.cpp +++ b/clang-tools-extra/clang-tidy/utils/TypeTraits.cpp @@ -111,7 +111,7 @@ bool isTriviallyDefaultConstructible(QualType Type, const ASTContext &Context) { } } - QualType CanonicalType = Type.getCanonicalType(); + const QualType CanonicalType = Type.getCanonicalType(); if (CanonicalType->isDependentType()) return false; diff --git a/clang-tools-extra/clang-tidy/utils/UseRangesCheck.cpp b/clang-tools-extra/clang-tidy/utils/UseRangesCheck.cpp index cb14951..8b7019b 100644 --- a/clang-tools-extra/clang-tidy/utils/UseRangesCheck.cpp +++ b/clang-tools-extra/clang-tidy/utils/UseRangesCheck.cpp @@ -73,7 +73,7 @@ makeMatcherPair(StringRef State, const UseRangesCheck::Indexes &Indexes, const std::optional<UseRangesCheck::ReverseIteratorDescriptor> &ReverseDescriptor) { std::string ArgBound = (ArgName + llvm::Twine(Indexes.BeginArg)).str(); - SmallString<64> ID = {BoundCall, State}; + const SmallString<64> ID = {BoundCall, State}; ast_matchers::internal::Matcher<CallExpr> ArgumentMatcher = allOf( hasArgument(Indexes.BeginArg, makeExprMatcher(expr(unless(hasSideEffects())).bind(ArgBound), @@ -84,9 +84,9 @@ makeMatcherPair(StringRef State, const UseRangesCheck::Indexes &Indexes, {"end", "cend"}, EndFreeNames))); if (ReverseDescriptor) { ArgBound.push_back('R'); - SmallVector<StringRef> RBegin{ + const SmallVector<StringRef> RBegin{ llvm::make_first_range(ReverseDescriptor->FreeReverseNames)}; - SmallVector<StringRef> REnd{ + const SmallVector<StringRef> REnd{ llvm::make_second_range(ReverseDescriptor->FreeReverseNames)}; ArgumentMatcher = anyOf( ArgumentMatcher, @@ -110,9 +110,9 @@ void UseRangesCheck::registerMatchers(MatchFinder *Finder) { auto Replaces = getReplacerMap(); ReverseDescriptor = getReverseDescriptor(); auto BeginEndNames = getFreeBeginEndMethods(); - llvm::SmallVector<StringRef, 4> BeginNames{ + const llvm::SmallVector<StringRef, 4> BeginNames{ llvm::make_first_range(BeginEndNames)}; - llvm::SmallVector<StringRef, 4> EndNames{ + const llvm::SmallVector<StringRef, 4> EndNames{ llvm::make_second_range(BeginEndNames)}; Replacers.clear(); llvm::DenseSet<Replacer *> SeenRepl; @@ -169,7 +169,7 @@ static void removeFunctionArgs(DiagnosticBuilder &Diag, const CallExpr &Call, llvm::SmallBitVector Commas(Call.getNumArgs()); // The first comma is actually the '(' which we can't remove Commas[0] = true; - for (unsigned Index : Sorted) { + for (const unsigned Index : Sorted) { const Expr *Arg = Call.getArg(Index); if (Commas[Index]) { if (Index >= Commas.size()) { @@ -192,7 +192,7 @@ static void removeFunctionArgs(DiagnosticBuilder &Diag, const CallExpr &Call, } void UseRangesCheck::check(const MatchFinder::MatchResult &Result) { - Replacer *Replacer = nullptr; + const Replacer *Replacer = nullptr; const FunctionDecl *Function = nullptr; for (const auto &[Node, Value] : Result.Nodes.getMap()) { StringRef NodeStr(Node); @@ -254,7 +254,7 @@ void UseRangesCheck::check(const MatchFinder::MatchResult &Result) { Diag << Inserter.createIncludeInsertion( Result.SourceManager->getFileID(Call->getBeginLoc()), *ReverseDescriptor->ReverseHeader); - StringRef ArgText = Lexer::getSourceText( + const StringRef ArgText = Lexer::getSourceText( CharSourceRange::getTokenRange(ArgExpr->getSourceRange()), Result.Context->getSourceManager(), Result.Context->getLangOpts()); SmallString<128> ReplaceText; diff --git a/clang-tools-extra/clang-tidy/utils/UsingInserter.cpp b/clang-tools-extra/clang-tidy/utils/UsingInserter.cpp index e4c71aa..6a591c1 100644 --- a/clang-tools-extra/clang-tidy/utils/UsingInserter.cpp +++ b/clang-tools-extra/clang-tidy/utils/UsingInserter.cpp @@ -19,7 +19,7 @@ namespace clang::tidy::utils { using namespace ast_matchers; static StringRef getUnqualifiedName(StringRef QualifiedName) { - size_t LastSeparatorPos = QualifiedName.rfind("::"); + const size_t LastSeparatorPos = QualifiedName.rfind("::"); if (LastSeparatorPos == StringRef::npos) return QualifiedName; return QualifiedName.drop_front(LastSeparatorPos + 2); @@ -30,7 +30,7 @@ UsingInserter::UsingInserter(const SourceManager &SourceMgr) std::optional<FixItHint> UsingInserter::createUsingDeclaration( ASTContext &Context, const Stmt &Statement, StringRef QualifiedName) { - StringRef UnqualifiedName = getUnqualifiedName(QualifiedName); + const StringRef UnqualifiedName = getUnqualifiedName(QualifiedName); const FunctionDecl *Function = getSurroundingFunction(Context, Statement); if (!Function) return std::nullopt; @@ -38,7 +38,7 @@ std::optional<FixItHint> UsingInserter::createUsingDeclaration( if (AddedUsing.count(std::make_pair(Function, QualifiedName.str())) != 0) return std::nullopt; - SourceLocation InsertLoc = Lexer::getLocForEndOfToken( + const SourceLocation InsertLoc = Lexer::getLocForEndOfToken( Function->getBody()->getBeginLoc(), 0, SourceMgr, Context.getLangOpts()); // Only use using declarations in the main file, not in includes. @@ -47,7 +47,7 @@ std::optional<FixItHint> UsingInserter::createUsingDeclaration( // FIXME: This declaration could be masked. Investigate if // there is a way to avoid using Sema. - bool AlreadyHasUsingDecl = + const bool AlreadyHasUsingDecl = !match(stmt(hasAncestor(decl(has(usingDecl(hasAnyUsingShadowDecl( hasTargetDecl(hasName(QualifiedName.str())))))))), Statement, Context) @@ -58,15 +58,15 @@ std::optional<FixItHint> UsingInserter::createUsingDeclaration( } // Find conflicting declarations and references. auto ConflictingDecl = namedDecl(hasName(UnqualifiedName)); - bool HasConflictingDeclaration = + const bool HasConflictingDeclaration = !match(findAll(ConflictingDecl), *Function, Context).empty(); - bool HasConflictingDeclRef = + const bool HasConflictingDeclRef = !match(findAll(declRefExpr(to(ConflictingDecl))), *Function, Context) .empty(); if (HasConflictingDeclaration || HasConflictingDeclRef) return std::nullopt; - std::string Declaration = + const std::string Declaration = (llvm::Twine("\nusing ") + QualifiedName + ";").str(); AddedUsing.emplace(Function, QualifiedName.str()); diff --git a/clang-tools-extra/clang-tidy/utils/UsingInserter.h b/clang-tools-extra/clang-tidy/utils/UsingInserter.h index 23c3175..3e94356 100644 --- a/clang-tools-extra/clang-tidy/utils/UsingInserter.h +++ b/clang-tools-extra/clang-tidy/utils/UsingInserter.h @@ -6,8 +6,8 @@ // //===----------------------------------------------------------------------===// -#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_USINGINSERTER_H -#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_USINGINSERTER_H +#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_USINGINSERTER_H +#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_USINGINSERTER_H #include "clang/AST/Decl.h" #include "clang/AST/Stmt.h" @@ -43,4 +43,4 @@ private: }; } // namespace clang::tidy::utils -#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_USINGINSERTER_H +#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_UTILS_USINGINSERTER_H diff --git a/clang-tools-extra/clangd/ClangdLSPServer.cpp b/clang-tools-extra/clangd/ClangdLSPServer.cpp index 0f765e9..f8e6da7 100644 --- a/clang-tools-extra/clangd/ClangdLSPServer.cpp +++ b/clang-tools-extra/clangd/ClangdLSPServer.cpp @@ -456,7 +456,6 @@ private: ClangdLSPServer &Server; }; -constexpr int ClangdLSPServer::MessageHandler::MaxReplayCallbacks; // call(), notify(), and reply() wrap the Transport, adding logging and locking. void ClangdLSPServer::callMethod(StringRef Method, llvm::json::Value Params, diff --git a/clang-tools-extra/clangd/FileDistance.cpp b/clang-tools-extra/clangd/FileDistance.cpp index 06c1a8b..d587c26 100644 --- a/clang-tools-extra/clangd/FileDistance.cpp +++ b/clang-tools-extra/clangd/FileDistance.cpp @@ -54,7 +54,6 @@ static llvm::SmallString<128> canonicalize(llvm::StringRef Path) { return Result; } -constexpr const unsigned FileDistance::Unreachable; const llvm::hash_code FileDistance::RootHash = llvm::hash_value(llvm::StringRef("/")); diff --git a/clang-tools-extra/clangd/FuzzyMatch.cpp b/clang-tools-extra/clangd/FuzzyMatch.cpp index de7280d..cf5182b 100644 --- a/clang-tools-extra/clangd/FuzzyMatch.cpp +++ b/clang-tools-extra/clangd/FuzzyMatch.cpp @@ -62,9 +62,6 @@ namespace clang { namespace clangd { -constexpr int FuzzyMatcher::MaxPat; -constexpr int FuzzyMatcher::MaxWord; - static char lower(char C) { return C >= 'A' && C <= 'Z' ? C + ('a' - 'A') : C; } // A "negative infinity" score that won't overflow. // We use this to mark unreachable states and forbidden solutions. diff --git a/clang-tools-extra/clangd/index/SymbolLocation.cpp b/clang-tools-extra/clangd/index/SymbolLocation.cpp index 61da267..058cb1e 100644 --- a/clang-tools-extra/clangd/index/SymbolLocation.cpp +++ b/clang-tools-extra/clangd/index/SymbolLocation.cpp @@ -11,9 +11,6 @@ namespace clang { namespace clangd { -constexpr uint32_t SymbolLocation::Position::MaxLine; -constexpr uint32_t SymbolLocation::Position::MaxColumn; - void SymbolLocation::Position::setLine(uint32_t L) { if (L > MaxLine) L = MaxLine; diff --git a/clang-tools-extra/docs/ReleaseNotes.rst b/clang-tools-extra/docs/ReleaseNotes.rst index 12435c9..92a2d33 100644 --- a/clang-tools-extra/docs/ReleaseNotes.rst +++ b/clang-tools-extra/docs/ReleaseNotes.rst @@ -284,6 +284,11 @@ New check aliases <clang-tidy/checks/bugprone/raw-memory-call-on-non-trivial-type>` keeping initial check as an alias to the new one. +- Renamed :doc:`cert-oop58-cpp <clang-tidy/checks/cert/oop58-cpp>` to + :doc:`bugprone-copy-constructor-mutates-argument + <clang-tidy/checks/bugprone/copy-constructor-mutates-argument>` + keeping initial check as an alias to the new one. + Changes in existing checks ^^^^^^^^^^^^^^^^^^^^^^^^^^ diff --git a/clang-tools-extra/docs/clang-tidy/checks/bugprone/copy-constructor-mutates-argument.rst b/clang-tools-extra/docs/clang-tidy/checks/bugprone/copy-constructor-mutates-argument.rst new file mode 100644 index 0000000..28e5015 --- /dev/null +++ b/clang-tools-extra/docs/clang-tidy/checks/bugprone/copy-constructor-mutates-argument.rst @@ -0,0 +1,11 @@ +.. title:: clang-tidy - bugprone-copy-constructor-mutates-argument + +bugprone-copy-constructor-mutates-argument +========================================== + +Finds assignments to the copied object and its direct or indirect members +in copy constructors and copy assignment operators. + +This check corresponds to the CERT C Coding Standard rule +`OOP58-CPP. Copy operations must not mutate the source object +<https://wiki.sei.cmu.edu/confluence/display/cplusplus/OOP58-CPP.+Copy+operations+must+not+mutate+the+source+object>`_.
\ No newline at end of file diff --git a/clang-tools-extra/docs/clang-tidy/checks/cert/oop58-cpp.rst b/clang-tools-extra/docs/clang-tidy/checks/cert/oop58-cpp.rst index 399fb1b..e435490 100644 --- a/clang-tools-extra/docs/clang-tidy/checks/cert/oop58-cpp.rst +++ b/clang-tools-extra/docs/clang-tidy/checks/cert/oop58-cpp.rst @@ -1,11 +1,10 @@ -.. title:: clang-tidy - cert-mutating-copy +.. title:: clang-tidy - cert-oop58-cpp +.. meta:: + :http-equiv=refresh: 5;URL=../bugprone/copy-constructor-mutates-argument.html cert-oop58-cpp ============== -Finds assignments to the copied object and its direct or indirect members -in copy constructors and copy assignment operators. - -This check corresponds to the CERT C Coding Standard rule -`OOP58-CPP. Copy operations must not mutate the source object -<https://wiki.sei.cmu.edu/confluence/display/cplusplus/OOP58-CPP.+Copy+operations+must+not+mutate+the+source+object>`_. +The `cert-oop58-cpp` check is an alias, please see +:doc:`bugprone-copy-constructor-mutates-argument <../bugprone/copy-constructor-mutates-argument>` +for more information.
\ No newline at end of file diff --git a/clang-tools-extra/docs/clang-tidy/checks/list.rst b/clang-tools-extra/docs/clang-tidy/checks/list.rst index be509ab..e287560 100644 --- a/clang-tools-extra/docs/clang-tidy/checks/list.rst +++ b/clang-tools-extra/docs/clang-tidy/checks/list.rst @@ -90,6 +90,7 @@ Clang-Tidy Checks :doc:`bugprone-command-processor <bugprone/command-processor>`, :doc:`bugprone-compare-pointer-to-member-virtual-function <bugprone/compare-pointer-to-member-virtual-function>`, :doc:`bugprone-copy-constructor-init <bugprone/copy-constructor-init>`, "Yes" + :doc:`bugprone-copy-constructor-mutates-argument <bugprone/copy-constructor-mutates-argument>`, :doc:`bugprone-crtp-constructor-accessibility <bugprone/crtp-constructor-accessibility>`, "Yes" :doc:`bugprone-dangling-handle <bugprone/dangling-handle>`, :doc:`bugprone-default-operator-new-on-overaligned-type <bugprone/default-operator-new-on-overaligned-type>`, @@ -463,6 +464,7 @@ Check aliases :doc:`cert-oop11-cpp <cert/oop11-cpp>`, :doc:`performance-move-constructor-init <performance/move-constructor-init>`, :doc:`cert-oop54-cpp <cert/oop54-cpp>`, :doc:`bugprone-unhandled-self-assignment <bugprone/unhandled-self-assignment>`, :doc:`cert-oop57-cpp <cert/oop57-cpp>`, :doc:`bugprone-raw-memory-call-on-non-trivial-type <bugprone/raw-memory-call-on-non-trivial-type>`, + :doc:`cert-oop58-cpp <cert/oop58-cpp>`, :doc:`bugprone-copy-constructor-mutates-argument <bugprone/copy-constructor-mutates-argument>`, :doc:`cert-pos44-c <cert/pos44-c>`, :doc:`bugprone-bad-signal-to-kill-thread <bugprone/bad-signal-to-kill-thread>`, :doc:`cert-pos47-c <cert/pos47-c>`, :doc:`concurrency-thread-canceltype-asynchronous <concurrency/thread-canceltype-asynchronous>`, :doc:`cert-sig30-c <cert/sig30-c>`, :doc:`bugprone-signal-handler <bugprone/signal-handler>`, diff --git a/clang-tools-extra/test/clang-doc/basic-project.mustache.test b/clang-tools-extra/test/clang-doc/basic-project.mustache.test index 5509951..5a40a6b 100644 --- a/clang-tools-extra/test/clang-doc/basic-project.mustache.test +++ b/clang-tools-extra/test/clang-doc/basic-project.mustache.test @@ -2,17 +2,17 @@ // RUN: sed 's|$test_dir|%/S|g' %S/Inputs/basic-project/database_template.json > %t/build/compile_commands.json // RUN: clang-doc --format=mustache --output=%t/docs --executor=all-TUs %t/build/compile_commands.json -// RUN: FileCheck %s -input-file=%t/docs/html/_ZTV5Shape.html -check-prefix=HTML-SHAPE -// RUN: FileCheck %s -input-file=%t/docs/html/_ZTV10Calculator.html -check-prefix=HTML-CALC -// RUN: FileCheck %s -input-file=%t/docs/html/_ZTV9Rectangle.html -check-prefix=HTML-RECTANGLE -// RUN: FileCheck %s -input-file=%t/docs/html/_ZTV6Circle.html -check-prefix=HTML-CIRCLE +// RUN: FileCheck %s -input-file=%t/docs/html/GlobalNamespace/_ZTV5Shape.html -check-prefix=HTML-SHAPE +// RUN: FileCheck %s -input-file=%t/docs/html/GlobalNamespace/_ZTV10Calculator.html -check-prefix=HTML-CALC +// RUN: FileCheck %s -input-file=%t/docs/html/GlobalNamespace/_ZTV9Rectangle.html -check-prefix=HTML-RECTANGLE +// RUN: FileCheck %s -input-file=%t/docs/html/GlobalNamespace/_ZTV6Circle.html -check-prefix=HTML-CIRCLE HTML-SHAPE: <html lang="en-US"> HTML-SHAPE: <head> HTML-SHAPE: <meta charset="utf-8"/> HTML-SHAPE: <title>Shape</title> -HTML-SHAPE: <link rel="stylesheet" type="text/css" href="./clang-doc-mustache.css"/> -HTML-SHAPE: <script src="./mustache-index.js"></script> +HTML-SHAPE: <link rel="stylesheet" type="text/css" href="../clang-doc-mustache.css"/> +HTML-SHAPE: <script src="../mustache-index.js"></script> HTML-SHAPE: <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css"> HTML-SHAPE: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js"></script> HTML-SHAPE: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/languages/cpp.min.js"></script> @@ -151,8 +151,8 @@ HTML-CALC: <html lang="en-US"> HTML-CALC: <head> HTML-CALC: <meta charset="utf-8"/> HTML-CALC: <title>Calculator</title> -HTML-CALC: <link rel="stylesheet" type="text/css" href="./clang-doc-mustache.css"/> -HTML-CALC: <script src="./mustache-index.js"></script> +HTML-CALC: <link rel="stylesheet" type="text/css" href="../clang-doc-mustache.css"/> +HTML-CALC: <script src="../mustache-index.js"></script> HTML-CALC: <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css"> HTML-CALC: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js"></script> HTML-CALC: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/languages/cpp.min.js"></script> @@ -440,8 +440,8 @@ HTML-RECTANGLE: <html lang="en-US"> HTML-RECTANGLE: <head> HTML-RECTANGLE: <meta charset="utf-8"/> HTML-RECTANGLE: <title>Rectangle</title> -HTML-RECTANGLE: <link rel="stylesheet" type="text/css" href="./clang-doc-mustache.css"/> -HTML-RECTANGLE: <script src="./mustache-index.js"></script> +HTML-RECTANGLE: <link rel="stylesheet" type="text/css" href="../clang-doc-mustache.css"/> +HTML-RECTANGLE: <script src="../mustache-index.js"></script> HTML-RECTANGLE: <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css"> HTML-RECTANGLE: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js"></script> HTML-RECTANGLE: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/languages/cpp.min.js"></script> @@ -597,8 +597,8 @@ HTML-CIRCLE: <html lang="en-US"> HTML-CIRCLE: <head> HTML-CIRCLE: <meta charset="utf-8"/> HTML-CIRCLE: <title>Circle</title> -HTML-CIRCLE: <link rel="stylesheet" type="text/css" href="./clang-doc-mustache.css"/> -HTML-CIRCLE: <script src="./mustache-index.js"></script> +HTML-CIRCLE: <link rel="stylesheet" type="text/css" href="../clang-doc-mustache.css"/> +HTML-CIRCLE: <script src="../mustache-index.js"></script> HTML-CIRCLE: <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css"> HTML-CIRCLE: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js"></script> HTML-CIRCLE: <script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/languages/cpp.min.js"></script> diff --git a/clang-tools-extra/test/clang-doc/json/class-requires.cpp b/clang-tools-extra/test/clang-doc/json/class-requires.cpp index 5139617..4e5ec3a 100644 --- a/clang-tools-extra/test/clang-doc/json/class-requires.cpp +++ b/clang-tools-extra/test/clang-doc/json/class-requires.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --extra-arg -std=c++20 --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/_ZTV7MyClass.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/_ZTV7MyClass.json template<typename T> concept Addable = requires(T a, T b) { diff --git a/clang-tools-extra/test/clang-doc/json/class-specialization.cpp b/clang-tools-extra/test/clang-doc/json/class-specialization.cpp index d3ad695..60f3b44 100644 --- a/clang-tools-extra/test/clang-doc/json/class-specialization.cpp +++ b/clang-tools-extra/test/clang-doc/json/class-specialization.cpp @@ -1,7 +1,7 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/_ZTV7MyClass.json --check-prefix=BASE -// RUN: FileCheck %s < %t/json/_ZTV7MyClassIiE.json --check-prefix=SPECIALIZATION +// RUN: FileCheck %s < %t/json/GlobalNamespace/_ZTV7MyClass.json --check-prefix=BASE +// RUN: FileCheck %s < %t/json/GlobalNamespace/_ZTV7MyClassIiE.json --check-prefix=SPECIALIZATION template<typename T> struct MyClass {}; diff --git a/clang-tools-extra/test/clang-doc/json/class-template.cpp b/clang-tools-extra/test/clang-doc/json/class-template.cpp index 5ef78f54..de52064 100644 --- a/clang-tools-extra/test/clang-doc/json/class-template.cpp +++ b/clang-tools-extra/test/clang-doc/json/class-template.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/_ZTV7MyClass.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/_ZTV7MyClass.json template<typename T> struct MyClass { T MemberTemplate; diff --git a/clang-tools-extra/test/clang-doc/json/class.cpp b/clang-tools-extra/test/clang-doc/json/class.cpp index adb1ed7..9116058 100644 --- a/clang-tools-extra/test/clang-doc/json/class.cpp +++ b/clang-tools-extra/test/clang-doc/json/class.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/_ZTV7MyClass.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/_ZTV7MyClass.json struct Foo; diff --git a/clang-tools-extra/test/clang-doc/json/compound-constraints.cpp b/clang-tools-extra/test/clang-doc/json/compound-constraints.cpp index 1a73a0d..5b15a88 100644 --- a/clang-tools-extra/test/clang-doc/json/compound-constraints.cpp +++ b/clang-tools-extra/test/clang-doc/json/compound-constraints.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --extra-arg -std=c++20 --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/index.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/index.json template<typename T> concept Incrementable = requires (T a) { a++; diff --git a/clang-tools-extra/test/clang-doc/json/concept.cpp b/clang-tools-extra/test/clang-doc/json/concept.cpp index e96ec14..5d8c47e 100644 --- a/clang-tools-extra/test/clang-doc/json/concept.cpp +++ b/clang-tools-extra/test/clang-doc/json/concept.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --extra-arg -std=c++20 --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/index.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/index.json // Requires that T suports post and pre-incrementing. template<typename T> diff --git a/clang-tools-extra/test/clang-doc/json/function-requires.cpp b/clang-tools-extra/test/clang-doc/json/function-requires.cpp index 9427146..8ba6adc 100644 --- a/clang-tools-extra/test/clang-doc/json/function-requires.cpp +++ b/clang-tools-extra/test/clang-doc/json/function-requires.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --extra-arg -std=c++20 --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/index.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/index.json template<typename T> concept Incrementable = requires(T x) { diff --git a/clang-tools-extra/test/clang-doc/json/function-specifiers.cpp b/clang-tools-extra/test/clang-doc/json/function-specifiers.cpp index faaccb7..6630d9e 100644 --- a/clang-tools-extra/test/clang-doc/json/function-specifiers.cpp +++ b/clang-tools-extra/test/clang-doc/json/function-specifiers.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/index.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/index.json static void myFunction() {} diff --git a/clang-tools-extra/test/clang-doc/json/method-template.cpp b/clang-tools-extra/test/clang-doc/json/method-template.cpp index 87977f8..f4885d9 100644 --- a/clang-tools-extra/test/clang-doc/json/method-template.cpp +++ b/clang-tools-extra/test/clang-doc/json/method-template.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/_ZTV7MyClass.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/_ZTV7MyClass.json struct MyClass { template<class T> T methodTemplate(T param) { diff --git a/clang-tools-extra/test/clang-doc/json/multiple-namespaces.cpp b/clang-tools-extra/test/clang-doc/json/multiple-namespaces.cpp index 04fcfc1..6926998 100644 --- a/clang-tools-extra/test/clang-doc/json/multiple-namespaces.cpp +++ b/clang-tools-extra/test/clang-doc/json/multiple-namespaces.cpp @@ -1,7 +1,7 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/foo_tools.json --check-prefix=CHECK-FOO -// RUN: FileCheck %s < %t/json/bar_tools.json --check-prefix=CHECK-BAR +// RUN: FileCheck %s < %t/json/foo/tools/index.json --check-prefix=CHECK-FOO +// RUN: FileCheck %s < %t/json/bar/tools/index.json --check-prefix=CHECK-BAR namespace foo { namespace tools { diff --git a/clang-tools-extra/test/clang-doc/json/namespace.cpp b/clang-tools-extra/test/clang-doc/json/namespace.cpp index dcf8323..dd7a9af 100644 --- a/clang-tools-extra/test/clang-doc/json/namespace.cpp +++ b/clang-tools-extra/test/clang-doc/json/namespace.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/index.json +// RUN: FileCheck %s < %t/json/GlobalNamespace/index.json class MyClass {}; diff --git a/clang-tools-extra/test/clang-doc/json/nested-namespace.cpp b/clang-tools-extra/test/clang-doc/json/nested-namespace.cpp index cf19e1e..5baca7f 100644 --- a/clang-tools-extra/test/clang-doc/json/nested-namespace.cpp +++ b/clang-tools-extra/test/clang-doc/json/nested-namespace.cpp @@ -1,7 +1,7 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=json --executor=standalone %s -// RUN: FileCheck %s < %t/json/nested.json --check-prefix=NESTED -// RUN: FileCheck %s < %t/json/nested_inner.json --check-prefix=INNER +// RUN: FileCheck %s < %t/json/nested/index.json --check-prefix=NESTED +// RUN: FileCheck %s < %t/json/nested/inner/index.json --check-prefix=INNER namespace nested { int Global; diff --git a/clang-tools-extra/test/clang-doc/long-name.cpp b/clang-tools-extra/test/clang-doc/long-name.cpp index e29c468..77e50b1 100644 --- a/clang-tools-extra/test/clang-doc/long-name.cpp +++ b/clang-tools-extra/test/clang-doc/long-name.cpp @@ -2,8 +2,8 @@ // UNSUPPORTED: system-windows // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --output=%t --format=mustache --executor=standalone %s -// RUN: ls %t/json | FileCheck %s -check-prefix=CHECK-JSON -// RUN: ls %t/html | FileCheck %s -check-prefix=CHECK-HTML +// RUN: ls %t/json/GlobalNamespace | FileCheck %s -check-prefix=CHECK-JSON +// RUN: ls %t/html/GlobalNamespace | FileCheck %s -check-prefix=CHECK-HTML struct ThisStructHasANameThatResultsInAMangledNameThatIsExactly250CharactersLongThatIsSupposedToTestTheFilenameLengthLimitsWithinClangDocInOrdertoSeeifclangdocwillcrashornotdependingonthelengthofthestructIfTheLengthIsTooLongThenClangDocWillCrashAnd12 {}; diff --git a/clang-tools-extra/test/clang-doc/mustache-index.cpp b/clang-tools-extra/test/clang-doc/mustache-index.cpp index f9aad19..7b98c6b 100644 --- a/clang-tools-extra/test/clang-doc/mustache-index.cpp +++ b/clang-tools-extra/test/clang-doc/mustache-index.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --format=mustache --output=%t --executor=standalone %s -// RUN: FileCheck %s < %t/html/index.html +// RUN: FileCheck %s < %t/html/GlobalNamespace/index.html enum Color { RED, @@ -70,9 +70,7 @@ class Foo; // CHECK-NEXT: <ul class="class-container"> // CHECK-NEXT: <li id="{{[0-9A-F]*}}" style="max-height: 40px;"> // CHECK-NEXT: <a href="_ZTV3Foo.html"> -// CHECK-NEXT: <pre> -// CHECK-NEXT: <code class="language-cpp code-clang-doc">class Foo</code> -// CHECK-NEXT: </pre> +// CHECK-NEXT: <pre><code class="language-cpp code-clang-doc">class Foo</code></pre> // CHECK-NEXT: </a> // CHECK-NEXT: </li> // CHECK-NEXT: </ul> diff --git a/clang-tools-extra/test/clang-doc/mustache-separate-namespace.cpp b/clang-tools-extra/test/clang-doc/mustache-separate-namespace.cpp index a73a5ab..ee8b844 100644 --- a/clang-tools-extra/test/clang-doc/mustache-separate-namespace.cpp +++ b/clang-tools-extra/test/clang-doc/mustache-separate-namespace.cpp @@ -1,6 +1,6 @@ // RUN: rm -rf %t && mkdir -p %t // RUN: clang-doc --format=mustache --output=%t --executor=standalone %s -// RUN: FileCheck %s < %t/html/MyNamespace.html +// RUN: FileCheck %s < %t/html/MyNamespace/index.html namespace MyNamespace { class Foo; @@ -9,9 +9,7 @@ namespace MyNamespace { // CHECK: <ul class="class-container"> // CHECK-NEXT: <li id="{{[0-9A-F]*}}" style="max-height: 40px;"> // CHECK-NEXT: <a href="_ZTVN11MyNamespace3FooE.html"> -// CHECK-NEXT: <pre> -// CHECK-NEXT: <code class="language-cpp code-clang-doc">class Foo</code> -// CHECK-NEXT: </pre> +// CHECK-NEXT: <pre><code class="language-cpp code-clang-doc">class Foo</code></pre> // CHECK-NEXT: </a> // CHECK-NEXT: </li> // CHECK-NEXT: </ul> diff --git a/clang-tools-extra/test/clang-tidy/checkers/cert/oop58-cpp.cpp b/clang-tools-extra/test/clang-tidy/checkers/bugprone/copy-constructor-mutates-argument.cpp index 223248c..9fdbb7a 100644 --- a/clang-tools-extra/test/clang-tidy/checkers/cert/oop58-cpp.cpp +++ b/clang-tools-extra/test/clang-tidy/checkers/bugprone/copy-constructor-mutates-argument.cpp @@ -1,4 +1,4 @@ -// RUN: %check_clang_tidy %s cert-oop58-cpp %t +// RUN: %check_clang_tidy %s bugprone-copy-constructor-mutates-argument %t // Example test cases from CERT rule // https://wiki.sei.cmu.edu/confluence/display/cplusplus/OOP58-CPP.+Copy+operations+must+not+mutate+the+source+object diff --git a/clang/Maintainers.rst b/clang/Maintainers.rst index 1d16ea9..847d37d 100644 --- a/clang/Maintainers.rst +++ b/clang/Maintainers.rst @@ -242,7 +242,7 @@ ARM EABI Compiler-Wide Topics -------------------- The following people are responsible for functionality that does not fit into -a single part of the compiler, but instead span multiple components within the +a single part of the compiler, but instead spans multiple components within the compiler. Attributes diff --git a/clang/docs/AddressSanitizer.rst b/clang/docs/AddressSanitizer.rst index 21e1a36..2c2131b 100644 --- a/clang/docs/AddressSanitizer.rst +++ b/clang/docs/AddressSanitizer.rst @@ -159,7 +159,7 @@ eliminating this check (``-fsanitize-address-use-after-return=never``). To summarize: ``-fsanitize-address-use-after-return=<mode>`` * ``never``: Completely disables detection of UAR errors (reduces code size). - * ``runtime``: Adds the code for detection, but it can be disable via the + * ``runtime``: Adds the code for detection, but it can be disabled via the runtime environment (``ASAN_OPTIONS=detect_stack_use_after_return=0``). * ``always``: Enables detection of UAR errors in all cases. (reduces code size, but not as much as ``never``). @@ -239,7 +239,7 @@ from adding redzones around it and detecting out of bounds accesses. AddressSanitizer also supports ``__attribute__((disable_sanitizer_instrumentation))``. This attribute -works similar to ``__attribute__((no_sanitize("address")))``, but it also +works similarly to ``__attribute__((no_sanitize("address")))``, but it also prevents instrumentation performed by other sanitizers. Suppressing Errors in Recompiled Code (Ignorelist) @@ -305,7 +305,7 @@ Limitations =========== * AddressSanitizer uses more real memory than a native run. Exact overhead - depends on the allocations sizes. The smaller the allocations you make the + depends on the allocation sizes. The smaller the allocations you make the bigger the overhead is. * AddressSanitizer uses more stack memory. We have seen up to 3x increase. * On 64-bit platforms AddressSanitizer maps (but not reserves) 16+ Terabytes of diff --git a/clang/docs/BlockLanguageSpec.rst b/clang/docs/BlockLanguageSpec.rst index 3632d56..0c3a000 100644 --- a/clang/docs/BlockLanguageSpec.rst +++ b/clang/docs/BlockLanguageSpec.rst @@ -279,7 +279,7 @@ copy. The net effect is that instance variables can be mutated. The :block-term:`Block_copy` operator retains all objects held in variables of automatic storage referenced within the Block expression -(or form strong references if running under garbage collection). +(or forms strong references if running under garbage collection). Object variables of ``__block`` storage type are assumed to hold normal pointers with no provision for retain and release messages. diff --git a/clang/docs/BoundsSafety.rst b/clang/docs/BoundsSafety.rst index 519c7b6..b0f77c3 100644 --- a/clang/docs/BoundsSafety.rst +++ b/clang/docs/BoundsSafety.rst @@ -58,7 +58,7 @@ adopt, offering these properties that make it widely adoptable in practice: * It has a relatively low adoption cost. This document discusses the key designs of ``-fbounds-safety``. The document is -subject to be actively updated with a more detailed specification. +subject to active updates with a more detailed specification. Programming Model ================= @@ -574,7 +574,7 @@ When ``sizeof()`` takes a type name, the compiler doesn't apply an implicit bounds annotation on the named pointer types. This means if a bounds annotation is not specified, the evaluated pointer type is treated identically to a plain C pointer type. Therefore, ``sizeof(int*)`` remains the same with or without -``-fbounds-safety``. That said, programmers can explicitly add attribute to the +``-fbounds-safety``. That said, programmers can explicitly add attributes to the types, e.g., ``sizeof(int *__bidi_indexable)``, in which case the sizeof evaluates to the size of type ``int *__bidi_indexable`` (the value equivalent to ``3 * sizeof(int*)``). diff --git a/clang/docs/BoundsSafetyImplPlans.rst b/clang/docs/BoundsSafetyImplPlans.rst index 34276c9..b374b0a 100644 --- a/clang/docs/BoundsSafetyImplPlans.rst +++ b/clang/docs/BoundsSafetyImplPlans.rst @@ -154,7 +154,7 @@ verify its bounds safety. The implementation relies on LLVM optimizations to remove redundant run-time checks. Using this optimization strategy, if the original source code already has bounds checks, the fewer additional checks ``-fbounds-safety`` will introduce. The LLVM ``ConstraintElimination`` pass is -design to remove provable redundant checks (please check Florian Hahn’s +designed to remove provable redundant checks (please check Florian Hahn’s presentation in 2021 LLVM Dev Meeting and the implementation to learn more). In the following example, ``-fbounds-safety`` implicitly adds the redundant bounds checks that the optimizer can remove: diff --git a/clang/docs/ClangFormatStyleOptions.rst b/clang/docs/ClangFormatStyleOptions.rst index 0b4a484..94d6f0d 100644 --- a/clang/docs/ClangFormatStyleOptions.rst +++ b/clang/docs/ClangFormatStyleOptions.rst @@ -1673,6 +1673,16 @@ the configuration (without a prefix: ``Auto``). int abcdef; // but this isn't + * ``bool AlignPPAndNotPP`` If comments following preprocessor directive should be aligned with + comments that don't. + + .. code-block:: c++ + + true: false: + #define A // Comment vs. #define A // Comment + #define AB // Aligned #define AB // Aligned + int i; // Aligned int i; // Not aligned + .. _AllowAllArgumentsOnNextLine: diff --git a/clang/docs/ClangLinkerWrapper.rst b/clang/docs/ClangLinkerWrapper.rst index 28f48fc..3637bdb 100644 --- a/clang/docs/ClangLinkerWrapper.rst +++ b/clang/docs/ClangLinkerWrapper.rst @@ -27,7 +27,7 @@ only for the linker wrapper will be forwarded to the wrapped linker job. .. code-block:: console - USAGE: clang-linker-wrapper [options] -- <options to passed to the linker> + USAGE: clang-linker-wrapper [options] -- <options to pass to the linker> OPTIONS: --cuda-path=<dir> Set the system CUDA path diff --git a/clang/docs/ClangNVLinkWrapper.rst b/clang/docs/ClangNVLinkWrapper.rst index 2acdb05..28763b3 100644 --- a/clang/docs/ClangNVLinkWrapper.rst +++ b/clang/docs/ClangNVLinkWrapper.rst @@ -10,7 +10,7 @@ Clang nvlink Wrapper Introduction ============ -This tools works as a wrapper around the NVIDIA ``nvlink`` linker. The purpose +This tool works as a wrapper around the NVIDIA ``nvlink`` linker. The purpose of this wrapper is to provide an interface similar to the ``ld.lld`` linker while still relying on NVIDIA's proprietary linker to produce the final output. @@ -37,7 +37,7 @@ only for the linker wrapper will be forwarded to ``nvlink``. --arch <value> Specify the 'sm_' name of the target architecture. --cuda-path=<dir> Set the system CUDA path --dry-run Print generated commands without running. - --feature <value> Specify the '+ptx' freature to use for LTO. + --feature <value> Specify the '+ptx' feature to use for LTO. -g Specify that this was a debug compile. -help-hidden Display all available options -help Display available options (--help-hidden for more) diff --git a/clang/docs/ClangPlugins.rst b/clang/docs/ClangPlugins.rst index 92e41fb..3bd9e96 100644 --- a/clang/docs/ClangPlugins.rst +++ b/clang/docs/ClangPlugins.rst @@ -150,7 +150,7 @@ passed to the plugin can. -fplugin-arg-call_super_plugin-help \ test.cpp -If your plugin name contains dashes, either rename the plugin or used the +If your plugin name contains dashes, either rename the plugin or use the cc1 command line options listed below. diff --git a/clang/docs/ClangTools.rst b/clang/docs/ClangTools.rst index 3216328..b53c125 100644 --- a/clang/docs/ClangTools.rst +++ b/clang/docs/ClangTools.rst @@ -66,7 +66,7 @@ in a fast, command line interface. It can also accept flags to re-display the diagnostics in different formats with different flags, suitable for use driving an IDE or editor. Furthermore, it can be used in fixit-mode to directly apply fixit-hints offered by clang. See :doc:`HowToSetupToolingForLLVM` for -instructions on how to setup and used `clang-check`. +instructions on how to setup and use `clang-check`. ``clang-format`` ---------------- diff --git a/clang/docs/ConstantInterpreter.rst b/clang/docs/ConstantInterpreter.rst index a71ee4b..3b1bd4b 100644 --- a/clang/docs/ConstantInterpreter.rst +++ b/clang/docs/ConstantInterpreter.rst @@ -140,7 +140,7 @@ pointer goes out of scope, dead blocks are also deallocated. The lifetime of blocks is managed through 3 methods stored in the descriptor of the block: -* **CtorFn**: initializes the metadata which is store in the block, +* **CtorFn**: initializes the metadata which is stored in the block, alongside actual data. Invokes the default constructors of objects which are not trivial (``Pointer``, ``RealFP``, etc.) diff --git a/clang/docs/ControlFlowIntegrity.rst b/clang/docs/ControlFlowIntegrity.rst index baff9ab..cfe5bd8 100644 --- a/clang/docs/ControlFlowIntegrity.rst +++ b/clang/docs/ControlFlowIntegrity.rst @@ -135,7 +135,7 @@ Bad Cast Checking This scheme checks that pointer casts are made to an object of the correct dynamic type; that is, the dynamic type of the object must be a derived class of the pointee type of the cast. The checks are currently only introduced -where the class being casted to is a polymorphic class. +where the class being cast to is a polymorphic class. Bad casts are not in themselves control flow integrity violations, but they can also create security vulnerabilities, and the implementation uses many diff --git a/clang/docs/DataFlowSanitizer.rst b/clang/docs/DataFlowSanitizer.rst index 5ff50b8..154229f 100644 --- a/clang/docs/DataFlowSanitizer.rst +++ b/clang/docs/DataFlowSanitizer.rst @@ -243,7 +243,7 @@ labels of just ``v1`` and ``v2``. This signature is the same when origin tracking is disabled - in this case the dfsan_origin passed in it will always be 0. - The callback will be called when a tained value reach stack/registers + The callback will be called when a tainted value reaches stack/registers in the context of a function. Tainted values can reach a function: * via the arguments of the function * via the return value of a call that occurs in the function diff --git a/clang/docs/HardwareAssistedAddressSanitizerDesign.rst b/clang/docs/HardwareAssistedAddressSanitizerDesign.rst index 014d103..f2e76d6f 100644 --- a/clang/docs/HardwareAssistedAddressSanitizerDesign.rst +++ b/clang/docs/HardwareAssistedAddressSanitizerDesign.rst @@ -15,7 +15,7 @@ Introduction tags every 8 bytes of the application memory with a 1 byte tag (using *shadow memory*), uses *redzones* to find buffer-overflows and *quarantine* to find use-after-free. -The redzones, the quarantine, and, to a less extent, the shadow, are the +The redzones, the quarantine, and, to a lesser extent, the shadow, are the sources of AddressSanitizer's memory overhead. See the `AddressSanitizer paper`_ for details. diff --git a/clang/docs/JSONCompilationDatabase.rst b/clang/docs/JSONCompilationDatabase.rst index f543227..936ba11 100644 --- a/clang/docs/JSONCompilationDatabase.rst +++ b/clang/docs/JSONCompilationDatabase.rst @@ -54,7 +54,7 @@ python bindings also support this (since clang 3.2); see Format ====== -A compilation database is a JSON file, which consist of an array of +A compilation database is a JSON file, which consists of an array of "command objects", where each command object specifies one way a translation unit is compiled in the project. diff --git a/clang/docs/LibASTImporter.rst b/clang/docs/LibASTImporter.rst index f5d4092..e438de6 100644 --- a/clang/docs/LibASTImporter.rst +++ b/clang/docs/LibASTImporter.rst @@ -35,12 +35,12 @@ Importing one AST node copies that node into the destination ``ASTContext``. Why do we have to copy the node? Isn't enough to insert the pointer to that node into the destination context? One reason is that the "from" context may outlive the "to" context. -Also, the Clang AST consider nodes (or certain properties of nodes) equivalent if they have the same address! +Also, the Clang AST considers nodes (or certain properties of nodes) equivalent if they have the same address! The import algorithm has to ensure that the structurally equivalent nodes in the different translation units are not getting duplicated in the merged AST. E.g. if we include the definition of the vector template (``#include <vector>``) in two translation units, then their merged AST should have only one node which represents the template. Also, we have to discover *one definition rule* (ODR) violations. -For instance, if there is a class definition with the same name in both translation units, but one of the definition contains a different number of fields. +For instance, if there is a class definition with the same name in both translation units, but one of the definitions contains a different number of fields. So, we look up existing definitions, and then we check the structural equivalency on those nodes. The following pseudo-code demonstrates the basics of the import mechanism: diff --git a/clang/docs/LibASTMatchers.rst b/clang/docs/LibASTMatchers.rst index 3b9f0a6..0aa7923 100644 --- a/clang/docs/LibASTMatchers.rst +++ b/clang/docs/LibASTMatchers.rst @@ -95,7 +95,7 @@ and flexibility. ``VariadicDynCastAllOfMatcher<Base, Derived>`` ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -Those match all nodes of type *Base* if they can be dynamically casted to +Those match all nodes of type *Base* if they can be dynamically cast to *Derived*. The names of those matchers are nouns, which closely resemble *Derived*. ``VariadicDynCastAllOfMatchers`` are the backbone of the matcher hierarchy. Most often, your match expression will start with one of them, and diff --git a/clang/docs/LibASTMatchersTutorial.rst b/clang/docs/LibASTMatchersTutorial.rst index d288368..e901eb9 100644 --- a/clang/docs/LibASTMatchersTutorial.rst +++ b/clang/docs/LibASTMatchersTutorial.rst @@ -209,7 +209,7 @@ and traversal matchers to get from one kind of AST node to another. For a complete list of AST matchers, take a look at the `AST Matcher References <LibASTMatchersReference.html>`_ -All matcher that are nouns describe entities in the AST and can be +All matchers that are nouns describe entities in the AST and can be bound, so that they can be referred to whenever a match is found. To do so, simply call the method ``bind`` on these matchers, e.g.: diff --git a/clang/docs/LibFormat.rst b/clang/docs/LibFormat.rst index 833f768..9450073 100644 --- a/clang/docs/LibFormat.rst +++ b/clang/docs/LibFormat.rst @@ -3,7 +3,7 @@ LibFormat ========= LibFormat is a library that implements automatic source code formatting based -on Clang. This documents describes the LibFormat interface and design as well +on Clang. This document describes the LibFormat interface and design as well as some basic style discussions. If you just want to use `clang-format` as a tool or integrated into an editor, diff --git a/clang/docs/MatrixTypes.rst b/clang/docs/MatrixTypes.rst index 32949c6..b3a2c8c 100644 --- a/clang/docs/MatrixTypes.rst +++ b/clang/docs/MatrixTypes.rst @@ -53,7 +53,7 @@ type of the *typedef* becomes a matrix type with the given dimensions and an element type of the former underlying type. If a declaration of a *typedef-name* has a ``matrix_type`` attribute, then all -declaration of that *typedef-name* shall have a matrix_type attribute with the +declarations of that *typedef-name* shall have a matrix_type attribute with the same element type, number of rows, and number of columns. Standard Conversions diff --git a/clang/docs/MemorySanitizer.rst b/clang/docs/MemorySanitizer.rst index 9f0d3f1..4f58142 100644 --- a/clang/docs/MemorySanitizer.rst +++ b/clang/docs/MemorySanitizer.rst @@ -176,7 +176,7 @@ for `lifetime <https://eel.is/c++draft/basic.life#1>`_ definition. This feature can be disabled with either: -#. Pass addition Clang option ``-fno-sanitize-memory-use-after-dtor`` during +#. Pass additional Clang option ``-fno-sanitize-memory-use-after-dtor`` during compilation. #. Set environment variable `MSAN_OPTIONS=poison_in_dtor=0` before running the program. diff --git a/clang/docs/Modules.rst b/clang/docs/Modules.rst index e45ee9f..0abb85c 100644 --- a/clang/docs/Modules.rst +++ b/clang/docs/Modules.rst @@ -115,7 +115,7 @@ Objective-C provides syntax for importing a module via an *@import declaration*, @import std; -The ``@import`` declaration above imports the entire contents of the ``std`` module (which would contain, e.g., the entire C or C++ standard library) and make its API available within the current translation unit. To import only part of a module, one may use dot syntax to specific a particular submodule, e.g., +The ``@import`` declaration above imports the entire contents of the ``std`` module (which would contain, e.g., the entire C or C++ standard library) and make its API available within the current translation unit. To import only part of a module, one may use dot syntax to specify a particular submodule, e.g., .. parsed-literal:: diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst index a59f6bd..3a4e1fc 100644 --- a/clang/docs/ReleaseNotes.rst +++ b/clang/docs/ReleaseNotes.rst @@ -572,6 +572,8 @@ X86 Support Arm and AArch64 Support ^^^^^^^^^^^^^^^^^^^^^^^ +- More intrinsics for the following AArch64 instructions: + FCVTZ[US], FCVTN[US], FCVTM[US], FCVTP[US], FCVTA[US] Android Support ^^^^^^^^^^^^^^^ @@ -651,6 +653,7 @@ clang-format - Deprecate ``AlwaysBreak`` and ``BlockIndent`` suboptions from the ``AlignAfterOpenBracket`` option, and make ``AlignAfterOpenBracket`` a ``bool`` type. +- Add ``AlignPPAndNotPP`` suboption to ``AlignTrailingComments``. libclang -------- diff --git a/clang/include/clang/AST/OpenMPClause.h b/clang/include/clang/AST/OpenMPClause.h index 4f50748..3296fbf 100644 --- a/clang/include/clang/AST/OpenMPClause.h +++ b/clang/include/clang/AST/OpenMPClause.h @@ -10068,6 +10068,152 @@ public: Expr *getSize() const { return getStmtAs<Expr>(); } }; +/// This represents 'dyn_groupprivate' clause in '#pragma omp target ...' +/// and '#pragma omp teams ...' directives. +/// +/// \code +/// #pragma omp target [...] dyn_groupprivate(a,b: N) +/// \endcode +class OMPDynGroupprivateClause : public OMPClause, public OMPClauseWithPreInit { + friend class OMPClauseReader; + + /// Location of '('. + SourceLocation LParenLoc; + + /// Modifiers for 'dyn_groupprivate' clause. + enum { SIMPLE, FALLBACK, NUM_MODIFIERS }; + unsigned Modifiers[NUM_MODIFIERS]; + + /// Locations of modifiers. + SourceLocation ModifiersLoc[NUM_MODIFIERS]; + + /// The size of the dyn_groupprivate. + Expr *Size = nullptr; + + /// Set the first dyn_groupprivate modifier. + /// + /// \param M The modifier. + void setDynGroupprivateModifier(OpenMPDynGroupprivateClauseModifier M) { + Modifiers[SIMPLE] = M; + } + + /// Set the second dyn_groupprivate modifier. + /// + /// \param M The modifier. + void setDynGroupprivateFallbackModifier( + OpenMPDynGroupprivateClauseFallbackModifier M) { + Modifiers[FALLBACK] = M; + } + + /// Set location of the first dyn_groupprivate modifier. + void setDynGroupprivateModifierLoc(SourceLocation Loc) { + ModifiersLoc[SIMPLE] = Loc; + } + + /// Set location of the second dyn_groupprivate modifier. + void setDynGroupprivateFallbackModifierLoc(SourceLocation Loc) { + ModifiersLoc[FALLBACK] = Loc; + } + + /// Sets the location of '('. + /// + /// \param Loc Location of '('. + void setLParenLoc(SourceLocation Loc) { LParenLoc = Loc; } + + /// Set size. + /// + /// \param E Size. + void setSize(Expr *E) { Size = E; } + +public: + /// Build 'dyn_groupprivate' clause with a size expression \a Size. + /// + /// \param StartLoc Starting location of the clause. + /// \param LParenLoc Location of '('. + /// \param EndLoc Ending location of the clause. + /// \param Size Size. + /// \param M1 The first modifier applied to 'dyn_groupprivate' clause. + /// \param M1Loc Location of the first modifier. + /// \param M2 The second modifier applied to 'dyn_groupprivate' clause. + /// \param M2Loc Location of the second modifier. + OMPDynGroupprivateClause(SourceLocation StartLoc, SourceLocation LParenLoc, + SourceLocation EndLoc, Expr *Size, Stmt *HelperSize, + OpenMPDirectiveKind CaptureRegion, + OpenMPDynGroupprivateClauseModifier M1, + SourceLocation M1Loc, + OpenMPDynGroupprivateClauseFallbackModifier M2, + SourceLocation M2Loc) + : OMPClause(llvm::omp::OMPC_dyn_groupprivate, StartLoc, EndLoc), + OMPClauseWithPreInit(this), LParenLoc(LParenLoc), Size(Size) { + setPreInitStmt(HelperSize, CaptureRegion); + Modifiers[SIMPLE] = M1; + Modifiers[FALLBACK] = M2; + ModifiersLoc[SIMPLE] = M1Loc; + ModifiersLoc[FALLBACK] = M2Loc; + } + + /// Build an empty clause. + explicit OMPDynGroupprivateClause() + : OMPClause(llvm::omp::OMPC_dyn_groupprivate, SourceLocation(), + SourceLocation()), + OMPClauseWithPreInit(this) { + Modifiers[SIMPLE] = OMPC_DYN_GROUPPRIVATE_unknown; + Modifiers[FALLBACK] = OMPC_DYN_GROUPPRIVATE_FALLBACK_unknown; + } + + /// Get the first modifier of the clause. + OpenMPDynGroupprivateClauseModifier getDynGroupprivateModifier() const { + return static_cast<OpenMPDynGroupprivateClauseModifier>(Modifiers[SIMPLE]); + } + + /// Get the second modifier of the clause. + OpenMPDynGroupprivateClauseFallbackModifier + getDynGroupprivateFallbackModifier() const { + return static_cast<OpenMPDynGroupprivateClauseFallbackModifier>( + Modifiers[FALLBACK]); + } + + /// Get location of '('. + SourceLocation getLParenLoc() { return LParenLoc; } + + /// Get the first modifier location. + SourceLocation getDynGroupprivateModifierLoc() const { + return ModifiersLoc[SIMPLE]; + } + + /// Get the second modifier location. + SourceLocation getDynGroupprivateFallbackModifierLoc() const { + return ModifiersLoc[FALLBACK]; + } + + /// Get size. + Expr *getSize() { return Size; } + + /// Get size. + const Expr *getSize() const { return Size; } + + child_range children() { + return child_range(reinterpret_cast<Stmt **>(&Size), + reinterpret_cast<Stmt **>(&Size) + 1); + } + + const_child_range children() const { + auto Children = const_cast<OMPDynGroupprivateClause *>(this)->children(); + return const_child_range(Children.begin(), Children.end()); + } + + child_range used_children() { + return child_range(child_iterator(), child_iterator()); + } + const_child_range used_children() const { + return const_child_range(const_child_iterator(), const_child_iterator()); + } + + static bool classof(const OMPClause *T) { + return T->getClauseKind() == llvm::omp::OMPC_dyn_groupprivate; + } +}; + /// This represents the 'doacross' clause for the '#pragma omp ordered' /// directive. /// diff --git a/clang/include/clang/AST/RecursiveASTVisitor.h b/clang/include/clang/AST/RecursiveASTVisitor.h index 8cb0a65..8f42742 100644 --- a/clang/include/clang/AST/RecursiveASTVisitor.h +++ b/clang/include/clang/AST/RecursiveASTVisitor.h @@ -4166,6 +4166,14 @@ bool RecursiveASTVisitor<Derived>::VisitOMPXDynCGroupMemClause( } template <typename Derived> +bool RecursiveASTVisitor<Derived>::VisitOMPDynGroupprivateClause( + OMPDynGroupprivateClause *C) { + TRY_TO(VisitOMPClauseWithPreInit(C)); + TRY_TO(TraverseStmt(C->getSize())); + return true; +} + +template <typename Derived> bool RecursiveASTVisitor<Derived>::VisitOMPDoacrossClause( OMPDoacrossClause *C) { TRY_TO(VisitOMPClauseList(C)); diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td index 04f2e8d..3e86447 100644 --- a/clang/include/clang/Basic/DiagnosticSemaKinds.td +++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td @@ -12137,6 +12137,9 @@ def err_omp_unexpected_schedule_modifier : Error< "modifier '%0' cannot be used along with modifier '%1'">; def err_omp_schedule_nonmonotonic_static : Error< "'nonmonotonic' modifier can only be specified with 'dynamic' or 'guided' schedule kind">; +def err_omp_incompatible_dyn_groupprivate_modifier + : Error<"modifier '%0' cannot be used along with modifier '%1' in " + "dyn_groupprivate">; def err_omp_simple_clause_incompatible_with_ordered : Error< "'%0' clause with '%1' modifier cannot be specified if an 'ordered' clause is specified">; def err_omp_ordered_simd : Error< diff --git a/clang/include/clang/Basic/OpenMPKinds.def b/clang/include/clang/Basic/OpenMPKinds.def index 328a0747..166b803 100644 --- a/clang/include/clang/Basic/OpenMPKinds.def +++ b/clang/include/clang/Basic/OpenMPKinds.def @@ -86,6 +86,12 @@ #ifndef OPENMP_GRAINSIZE_MODIFIER #define OPENMP_GRAINSIZE_MODIFIER(Name) #endif +#ifndef OPENMP_DYN_GROUPPRIVATE_MODIFIER +#define OPENMP_DYN_GROUPPRIVATE_MODIFIER(Name) +#endif +#ifndef OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER +#define OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(Name) +#endif #ifndef OPENMP_NUMTASKS_MODIFIER #define OPENMP_NUMTASKS_MODIFIER(Name) #endif @@ -242,6 +248,14 @@ OPENMP_BIND_KIND(thread) // Modifiers for the 'grainsize' clause. OPENMP_GRAINSIZE_MODIFIER(strict) +// Modifiers for the 'dyn_groupprivate' clause. +OPENMP_DYN_GROUPPRIVATE_MODIFIER(cgroup) + +// Fallback modifiers for the 'dyn_groupprivate' clause. +OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(abort) +OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(null) +OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(default_mem) + // Modifiers for the 'num_tasks' clause. OPENMP_NUMTASKS_MODIFIER(strict) @@ -263,6 +277,8 @@ OPENMP_THREADSET_KIND(omp_team) #undef OPENMP_NUMTASKS_MODIFIER #undef OPENMP_NUMTHREADS_MODIFIER +#undef OPENMP_DYN_GROUPPRIVATE_MODIFIER +#undef OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER #undef OPENMP_GRAINSIZE_MODIFIER #undef OPENMP_BIND_KIND #undef OPENMP_ADJUST_ARGS_KIND diff --git a/clang/include/clang/Basic/OpenMPKinds.h b/clang/include/clang/Basic/OpenMPKinds.h index c9ddbcd..41b2c4e 100644 --- a/clang/include/clang/Basic/OpenMPKinds.h +++ b/clang/include/clang/Basic/OpenMPKinds.h @@ -224,6 +224,20 @@ enum OpenMPGrainsizeClauseModifier { OMPC_GRAINSIZE_unknown }; +enum OpenMPDynGroupprivateClauseModifier { +#define OPENMP_DYN_GROUPPRIVATE_MODIFIER(Name) OMPC_DYN_GROUPPRIVATE_##Name, +#include "clang/Basic/OpenMPKinds.def" + OMPC_DYN_GROUPPRIVATE_unknown +}; + +enum OpenMPDynGroupprivateClauseFallbackModifier { + OMPC_DYN_GROUPPRIVATE_FALLBACK_unknown = OMPC_DYN_GROUPPRIVATE_unknown, +#define OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(Name) \ + OMPC_DYN_GROUPPRIVATE_FALLBACK_##Name, +#include "clang/Basic/OpenMPKinds.def" + OMPC_DYN_GROUPPRIVATE_FALLBACK_last +}; + enum OpenMPNumTasksClauseModifier { #define OPENMP_NUMTASKS_MODIFIER(Name) OMPC_NUMTASKS_##Name, #include "clang/Basic/OpenMPKinds.def" diff --git a/clang/include/clang/Basic/arm_neon.td b/clang/include/clang/Basic/arm_neon.td index ef19610..e91d7ce 100644 --- a/clang/include/clang/Basic/arm_neon.td +++ b/clang/include/clang/Basic/arm_neon.td @@ -1466,26 +1466,51 @@ def SCALAR_UCVTFD : SInst<"vcvt_f64", "(1F)(1!)", "SUl">; //////////////////////////////////////////////////////////////////////////////// // Scalar Floating-point Converts def SCALAR_FCVTXN : IInst<"vcvtx_f32", "(1F<)(1!)", "Sd">; -def SCALAR_FCVTNSS : SInst<"vcvtn_s32", "(1S)1", "Sf">; -def SCALAR_FCVTNUS : SInst<"vcvtn_u32", "(1U)1", "Sf">; -def SCALAR_FCVTNSD : SInst<"vcvtn_s64", "(1S)1", "Sd">; -def SCALAR_FCVTNUD : SInst<"vcvtn_u64", "(1U)1", "Sd">; -def SCALAR_FCVTMSS : SInst<"vcvtm_s32", "(1S)1", "Sf">; -def SCALAR_FCVTMUS : SInst<"vcvtm_u32", "(1U)1", "Sf">; -def SCALAR_FCVTMSD : SInst<"vcvtm_s64", "(1S)1", "Sd">; -def SCALAR_FCVTMUD : SInst<"vcvtm_u64", "(1U)1", "Sd">; -def SCALAR_FCVTASS : SInst<"vcvta_s32", "(1S)1", "Sf">; -def SCALAR_FCVTAUS : SInst<"vcvta_u32", "(1U)1", "Sf">; -def SCALAR_FCVTASD : SInst<"vcvta_s64", "(1S)1", "Sd">; -def SCALAR_FCVTAUD : SInst<"vcvta_u64", "(1U)1", "Sd">; -def SCALAR_FCVTPSS : SInst<"vcvtp_s32", "(1S)1", "Sf">; -def SCALAR_FCVTPUS : SInst<"vcvtp_u32", "(1U)1", "Sf">; -def SCALAR_FCVTPSD : SInst<"vcvtp_s64", "(1S)1", "Sd">; -def SCALAR_FCVTPUD : SInst<"vcvtp_u64", "(1U)1", "Sd">; -def SCALAR_FCVTZSS : SInst<"vcvt_s32", "(1S)1", "Sf">; -def SCALAR_FCVTZUS : SInst<"vcvt_u32", "(1U)1", "Sf">; -def SCALAR_FCVTZSD : SInst<"vcvt_s64", "(1S)1", "Sd">; -def SCALAR_FCVTZUD : SInst<"vcvt_u64", "(1U)1", "Sd">; + +def SCALAR_FCVTN_F32toSS : SInst<"vcvtn_s32", "(1S)1", "Sf">; +def SCALAR_FCVTN_F32toUS : SInst<"vcvtn_u32", "(1U)1", "Sf">; +def SCALAR_FCVTN_F64toSS : SInst<"vcvtn_s32", "(1S<)1", "Sd">; +def SCALAR_FCVTN_F64toUS : SInst<"vcvtn_u32", "(1U<)1", "Sd">; +def SCALAR_FCVTN_F32toSD : SInst<"vcvtn_s64", "(1S>)1", "Sf">; +def SCALAR_FCVTN_F32toUD : SInst<"vcvtn_u64", "(1U>)1", "Sf">; +def SCALAR_FCVTN_F64toSD : SInst<"vcvtn_s64", "(1S)1", "Sd">; +def SCALAR_FCVTN_F64toUD : SInst<"vcvtn_u64", "(1U)1", "Sd">; + +def SCALAR_FCVTM_F32toSS : SInst<"vcvtm_s32", "(1S)1", "Sf">; +def SCALAR_FCVTM_F32toUS : SInst<"vcvtm_u32", "(1U)1", "Sf">; +def SCALAR_FCVTM_F64toSS : SInst<"vcvtm_s32", "(1S<)1", "Sd">; +def SCALAR_FCVTM_F64toUS : SInst<"vcvtm_u32", "(1U<)1", "Sd">; +def SCALAR_FCVTM_F32toSD : SInst<"vcvtm_s64", "(1S>)1", "Sf">; +def SCALAR_FCVTM_F32toUD : SInst<"vcvtm_u64", "(1U>)1", "Sf">; +def SCALAR_FCVTM_F64toSD : SInst<"vcvtm_s64", "(1S)1", "Sd">; +def SCALAR_FCVTM_F64toUD : SInst<"vcvtm_u64", "(1U)1", "Sd">; + +def SCALAR_FCVTA_F32toSS : SInst<"vcvta_s32", "(1S)1", "Sf">; +def SCALAR_FCVTA_F32toUS : SInst<"vcvta_u32", "(1U)1", "Sf">; +def SCALAR_FCVTA_F64toSS : SInst<"vcvta_s32", "(1S<)1", "Sd">; +def SCALAR_FCVTA_F64toUS : SInst<"vcvta_u32", "(1U<)1", "Sd">; +def SCALAR_FCVTA_F32toSD : SInst<"vcvta_s64", "(1S>)1", "Sf">; +def SCALAR_FCVTA_F32toUD : SInst<"vcvta_u64", "(1U>)1", "Sf">; +def SCALAR_FCVTA_F64toSD : SInst<"vcvta_s64", "(1S)1", "Sd">; +def SCALAR_FCVTA_F64toUD : SInst<"vcvta_u64", "(1U)1", "Sd">; + +def SCALAR_FCVTP_F32toSS : SInst<"vcvtp_s32", "(1S)1", "Sf">; +def SCALAR_FCVTP_F32toUS : SInst<"vcvtp_u32", "(1U)1", "Sf">; +def SCALAR_FCVTP_F64toSS : SInst<"vcvtp_s32", "(1S<)1", "Sd">; +def SCALAR_FCVTP_F64toUS : SInst<"vcvtp_u32", "(1U<)1", "Sd">; +def SCALAR_FCVTP_F32toSD : SInst<"vcvtp_s64", "(1S>)1", "Sf">; +def SCALAR_FCVTP_F32toUD : SInst<"vcvtp_u64", "(1U>)1", "Sf">; +def SCALAR_FCVTP_F64toSD : SInst<"vcvtp_s64", "(1S)1", "Sd">; +def SCALAR_FCVTP_F64toUD : SInst<"vcvtp_u64", "(1U)1", "Sd">; + +def SCALAR_FCVTZ_F32toSS : SInst<"vcvt_s32", "(1S)1", "Sf">; +def SCALAR_FCVTZ_F32toUS : SInst<"vcvt_u32", "(1U)1", "Sf">; +def SCALAR_FCVTZ_F64toSS : SInst<"vcvt_s32", "(1S<)1", "Sd">; +def SCALAR_FCVTZ_F64toUS : SInst<"vcvt_u32", "(1U<)1", "Sd">; +def SCALAR_FCVTZ_F32toSD : SInst<"vcvt_s64", "(1S>)1", "Sf">; +def SCALAR_FCVTZ_F32toUD : SInst<"vcvt_u64", "(1U>)1", "Sf">; +def SCALAR_FCVTZ_F64toSD : SInst<"vcvt_s64", "(1S)1", "Sd">; +def SCALAR_FCVTZ_F64toUD : SInst<"vcvt_u64", "(1U)1", "Sd">; //////////////////////////////////////////////////////////////////////////////// // Scalar Floating-point Reciprocal Estimate @@ -1896,6 +1921,14 @@ let ArchGuard = "defined(__aarch64__) || defined(__arm64ec__)", TargetGuard = "f def VFMLSL_LANEQ_HIGH : SOpInst<"vfmlsl_laneq_high", "(F>)(F>)F(FQ)I", "hQh", OP_FMLSL_LN_Hi>; } +let ArchGuard = "defined(__aarch64__)", TargetGuard = "f8f16mm,neon" in { + def VMMLA_F16_MF8 : VInst<"vmmla_f16_mf8_fpm", "(>F)(>F)..V", "Qm">; +} + +let ArchGuard = "defined(__aarch64__)", TargetGuard = "f8f32mm,neon" in { + def VMMLA_F32_MF8 : VInst<"vmmla_f32_mf8_fpm", "(>>F)(>>F)..V", "Qm">; +} + let TargetGuard = "i8mm,neon" in { def VMMLA : SInst<"vmmla", "..(<<)(<<)", "QUiQi">; def VUSMMLA : SInst<"vusmmla", "..(<<U)(<<)", "Qi">; diff --git a/clang/include/clang/Format/Format.h b/clang/include/clang/Format/Format.h index f246def..b6f124f 100644 --- a/clang/include/clang/Format/Format.h +++ b/clang/include/clang/Format/Format.h @@ -601,9 +601,19 @@ struct FormatStyle { /// int abcdef; // but this isn't /// \endcode unsigned OverEmptyLines; + /// If comments following preprocessor directive should be aligned with + /// comments that don't. + /// \code + /// true: false: + /// #define A // Comment vs. #define A // Comment + /// #define AB // Aligned #define AB // Aligned + /// int i; // Aligned int i; // Not aligned + /// \endcode + bool AlignPPAndNotPP; bool operator==(const TrailingCommentsAlignmentStyle &R) const { - return Kind == R.Kind && OverEmptyLines == R.OverEmptyLines; + return Kind == R.Kind && OverEmptyLines == R.OverEmptyLines && + AlignPPAndNotPP == R.AlignPPAndNotPP; } bool operator!=(const TrailingCommentsAlignmentStyle &R) const { return !(*this == R); diff --git a/clang/include/clang/Frontend/CompilerInstance.h b/clang/include/clang/Frontend/CompilerInstance.h index 2403cbb..18ad7bf 100644 --- a/clang/include/clang/Frontend/CompilerInstance.h +++ b/clang/include/clang/Frontend/CompilerInstance.h @@ -946,6 +946,12 @@ public: DependencyCollectors.push_back(std::move(Listener)); } + void clearDependencyCollectors() { DependencyCollectors.clear(); } + + std::vector<std::shared_ptr<DependencyCollector>> &getDependencyCollectors() { + return DependencyCollectors; + } + void setExternalSemaSource(IntrusiveRefCntPtr<ExternalSemaSource> ESS); ModuleCache &getModuleCache() const { return *ModCache; } diff --git a/clang/include/clang/Frontend/FrontendActions.h b/clang/include/clang/Frontend/FrontendActions.h index 73308c0..87a9f0d 100644 --- a/clang/include/clang/Frontend/FrontendActions.h +++ b/clang/include/clang/Frontend/FrontendActions.h @@ -320,15 +320,6 @@ protected: bool hasPCHSupport() const override { return true; } }; -class GetDependenciesByModuleNameAction : public PreprocessOnlyAction { - StringRef ModuleName; - void ExecuteAction() override; - -public: - GetDependenciesByModuleNameAction(StringRef ModuleName) - : ModuleName(ModuleName) {} -}; - //===----------------------------------------------------------------------===// // HLSL Specific Actions //===----------------------------------------------------------------------===// diff --git a/clang/include/clang/Lex/Preprocessor.h b/clang/include/clang/Lex/Preprocessor.h index 4120022..b1c648e 100644 --- a/clang/include/clang/Lex/Preprocessor.h +++ b/clang/include/clang/Lex/Preprocessor.h @@ -1327,6 +1327,7 @@ public: std::move(Callbacks)); Callbacks = std::move(C); } + void removePPCallbacks() { Callbacks.reset(); } /// \} /// Get the number of tokens processed so far. diff --git a/clang/include/clang/Sema/SemaOpenMP.h b/clang/include/clang/Sema/SemaOpenMP.h index ba12b40..e5628e8 100644 --- a/clang/include/clang/Sema/SemaOpenMP.h +++ b/clang/include/clang/Sema/SemaOpenMP.h @@ -1411,6 +1411,13 @@ public: SourceLocation LParenLoc, SourceLocation EndLoc); + /// Called on a well-formed 'dyn_groupprivate' clause. + OMPClause *ActOnOpenMPDynGroupprivateClause( + OpenMPDynGroupprivateClauseModifier M1, + OpenMPDynGroupprivateClauseFallbackModifier M2, Expr *Size, + SourceLocation StartLoc, SourceLocation LParenLoc, SourceLocation M1Loc, + SourceLocation M2Loc, SourceLocation EndLoc); + /// Called on well-formed 'doacross' clause. OMPClause * ActOnOpenMPDoacrossClause(OpenMPDoacrossClauseModifier DepType, diff --git a/clang/include/clang/Tooling/DependencyScanning/DependencyScanningTool.h b/clang/include/clang/Tooling/DependencyScanning/DependencyScanningTool.h index f222ded..ed562f4 100644 --- a/clang/include/clang/Tooling/DependencyScanning/DependencyScanningTool.h +++ b/clang/include/clang/Tooling/DependencyScanning/DependencyScanningTool.h @@ -154,11 +154,52 @@ public: /// Given a compilation context specified via the Clang driver command-line, /// gather modular dependencies of module with the given name, and return the /// information needed for explicit build. + /// TODO: this method should be removed as soon as Swift and our C-APIs adopt + /// CompilerInstanceWithContext. We are keeping it here so that it is easier + /// to coordinate with Swift and C-API changes. llvm::Expected<TranslationUnitDeps> getModuleDependencies( StringRef ModuleName, const std::vector<std::string> &CommandLine, StringRef CWD, const llvm::DenseSet<ModuleID> &AlreadySeen, LookupModuleOutputCallback LookupModuleOutput); + /// The following three methods provide a new interface to perform + /// by name dependency scan. The new interface's intention is to improve + /// dependency scanning performance when a sequence of name is looked up + /// with the same current working directory and the command line. + + /// @brief Initializing the context and the compiler instance. + /// This method must be called before calling + /// computeDependenciesByNameWithContext. + /// @param CWD The current working directory used during the scan. + /// @param CommandLine The commandline used for the scan. + /// @return Error if the initializaiton fails. + llvm::Error initializeCompilerInstanceWithContext( + StringRef CWD, const std::vector<std::string> &CommandLine); + + /// @brief Computes the dependeny for the module named ModuleName. + /// @param ModuleName The name of the module for which this method computes + ///. dependencies. + /// @param AlreadySeen This stores modules which have previously been + /// reported. Use the same instance for all calls to this + /// function for a single \c DependencyScanningTool in a + /// single build. Note that this parameter is not part of + /// the context because it can be shared across different + /// worker threads and each worker thread may update it. + /// @param LookupModuleOutput This function is called to fill in + /// "-fmodule-file=", "-o" and other output + /// arguments for dependencies. + /// @return An instance of \c TranslationUnitDeps if the scan is successful. + /// Otherwise it returns an error. + llvm::Expected<TranslationUnitDeps> computeDependenciesByNameWithContext( + StringRef ModuleName, const llvm::DenseSet<ModuleID> &AlreadySeen, + LookupModuleOutputCallback LookupModuleOutput); + + /// @brief This method finializes the compiler instance. It finalizes the + /// diagnostics and deletes the compiler instance. Call this method + /// once all names for a same commandline are scanned. + /// @return Error if an error occured during finalization. + llvm::Error finalizeCompilerInstanceWithContext(); + llvm::vfs::FileSystem &getWorkerVFS() const { return Worker.getVFS(); } private: diff --git a/clang/include/clang/Tooling/DependencyScanning/DependencyScanningWorker.h b/clang/include/clang/Tooling/DependencyScanning/DependencyScanningWorker.h index 6060e4b..e2c353a 100644 --- a/clang/include/clang/Tooling/DependencyScanning/DependencyScanningWorker.h +++ b/clang/include/clang/Tooling/DependencyScanning/DependencyScanningWorker.h @@ -29,6 +29,7 @@ namespace tooling { namespace dependencies { class DependencyScanningWorkerFilesystem; +class CompilerInstanceWithContext; /// A command-line tool invocation that is part of building a TU. /// @@ -89,6 +90,8 @@ public: DependencyScanningWorker(DependencyScanningService &Service, llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS); + ~DependencyScanningWorker(); + /// Run the dependency scanning tool for a given clang driver command-line, /// and report the discovered dependencies to the provided consumer. If /// TUBuffer is not nullopt, it is used as TU input for the dependency @@ -104,18 +107,6 @@ public: std::optional<llvm::MemoryBufferRef> TUBuffer = std::nullopt); /// Run the dependency scanning tool for a given clang driver command-line - /// for a specific module. - /// - /// \returns false if clang errors occurred (with diagnostics reported to - /// \c DiagConsumer), true otherwise. - bool computeDependencies(StringRef WorkingDirectory, - const std::vector<std::string> &CommandLine, - DependencyConsumer &DepConsumer, - DependencyActionController &Controller, - DiagnosticConsumer &DiagConsumer, - StringRef ModuleName); - - /// Run the dependency scanning tool for a given clang driver command-line /// for a specific translation unit via file system or memory buffer. /// /// \returns A \c StringError with the diagnostic output if clang errors @@ -125,16 +116,46 @@ public: DependencyConsumer &Consumer, DependencyActionController &Controller, std::optional<llvm::MemoryBufferRef> TUBuffer = std::nullopt); - /// Run the dependency scanning tool for a given clang driver command-line - /// for a specific module. - /// - /// \returns A \c StringError with the diagnostic output if clang errors - /// occurred, success otherwise. - llvm::Error computeDependencies(StringRef WorkingDirectory, - const std::vector<std::string> &CommandLine, - DependencyConsumer &Consumer, - DependencyActionController &Controller, - StringRef ModuleName); + /// The three method below implements a new interface for by name + /// dependency scanning. They together enable the dependency scanning worker + /// to more effectively perform scanning for a sequence of modules + /// by name when the CWD and CommandLine do not change across the queries. + + /// @brief Initializing the context and the compiler instance. + /// @param CWD The current working directory used during the scan. + /// @param CommandLine The commandline used for the scan. + /// @return Error if the initializaiton fails. + llvm::Error initializeCompilerInstanceWithContextOrError( + StringRef CWD, const std::vector<std::string> &CommandLine); + + /// @brief Performaces dependency scanning for the module whose name is + /// specified. + /// @param ModuleName The name of the module whose dependency will be + /// scanned. + /// @param Consumer The dependency consumer that stores the results. + /// @param Controller The controller for the dependency scanning action. + /// @return Error if the scanner incurs errors. + llvm::Error computeDependenciesByNameWithContextOrError( + StringRef ModuleName, DependencyConsumer &Consumer, + DependencyActionController &Controller); + + /// @brief Finalizes the diagnostics engine and deletes the compiler instance. + /// @return Error if errors occur during finalization. + llvm::Error finalizeCompilerInstanceWithContextOrError(); + + /// The three methods below provides the same functionality as the + /// three methods above. Instead of returning `llvm::Error`s, these + /// three methods return a flag to indicate if the call is successful. + /// The initialization function asks the client for a DiagnosticsConsumer + /// that it direct the diagnostics to. + bool initializeCompilerInstanceWithContext( + StringRef CWD, const std::vector<std::string> &CommandLine, + DiagnosticConsumer *DC = nullptr); + bool + computeDependenciesByNameWithContext(StringRef ModuleName, + DependencyConsumer &Consumer, + DependencyActionController &Controller); + bool finalizeCompilerInstance(); llvm::vfs::FileSystem &getVFS() const { return *BaseFS; } @@ -151,14 +172,16 @@ private: /// (passed in the constructor). llvm::IntrusiveRefCntPtr<DependencyScanningWorkerFilesystem> DepFS; + friend CompilerInstanceWithContext; + std::unique_ptr<CompilerInstanceWithContext> CIWithContext; + /// Private helper functions. bool scanDependencies(StringRef WorkingDirectory, const std::vector<std::string> &CommandLine, DependencyConsumer &Consumer, DependencyActionController &Controller, DiagnosticConsumer &DC, - llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS, - std::optional<StringRef> ModuleName); + llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS); }; } // end namespace dependencies diff --git a/clang/include/clang/Tooling/DependencyScanning/ModuleDepCollector.h b/clang/include/clang/Tooling/DependencyScanning/ModuleDepCollector.h index 4136cb7..b0a91b6 100644 --- a/clang/include/clang/Tooling/DependencyScanning/ModuleDepCollector.h +++ b/clang/include/clang/Tooling/DependencyScanning/ModuleDepCollector.h @@ -288,6 +288,8 @@ public: void attachToPreprocessor(Preprocessor &PP) override; void attachToASTReader(ASTReader &R) override; + PPCallbacks *getPPCallbacks() { return CollectorPPPtr; } + /// Apply any changes implied by the discovered dependencies to the given /// invocation, (e.g. disable implicit modules, add explicit module paths). void applyDiscoveredDependencies(CompilerInvocation &CI); @@ -339,6 +341,11 @@ private: std::optional<P1689ModuleInfo> ProvidedStdCXXModule; std::vector<P1689ModuleInfo> RequiredStdCXXModules; + /// A pointer to the preprocessor callback so we can invoke it directly + /// if needed. The callback is created and added to a Preprocessor instance by + /// attachToPreprocessor and the Preprocessor instance owns it. + ModuleDepCollectorPP *CollectorPPPtr = nullptr; + /// Checks whether the module is known as being prebuilt. bool isPrebuiltModule(const Module *M); diff --git a/clang/lib/AST/OpenMPClause.cpp b/clang/lib/AST/OpenMPClause.cpp index 59d9459..0640fed 100644 --- a/clang/lib/AST/OpenMPClause.cpp +++ b/clang/lib/AST/OpenMPClause.cpp @@ -105,6 +105,8 @@ const OMPClauseWithPreInit *OMPClauseWithPreInit::get(const OMPClause *C) { return static_cast<const OMPFilterClause *>(C); case OMPC_ompx_dyn_cgroup_mem: return static_cast<const OMPXDynCGroupMemClause *>(C); + case OMPC_dyn_groupprivate: + return static_cast<const OMPDynGroupprivateClause *>(C); case OMPC_message: return static_cast<const OMPMessageClause *>(C); case OMPC_default: @@ -2857,6 +2859,24 @@ void OMPClausePrinter::VisitOMPXDynCGroupMemClause( OS << ")"; } +void OMPClausePrinter::VisitOMPDynGroupprivateClause( + OMPDynGroupprivateClause *Node) { + OS << "dyn_groupprivate("; + if (Node->getDynGroupprivateModifier() != OMPC_DYN_GROUPPRIVATE_unknown) { + OS << getOpenMPSimpleClauseTypeName(OMPC_dyn_groupprivate, + Node->getDynGroupprivateModifier()); + if (Node->getDynGroupprivateFallbackModifier() != + OMPC_DYN_GROUPPRIVATE_FALLBACK_unknown) { + OS << ", "; + OS << getOpenMPSimpleClauseTypeName( + OMPC_dyn_groupprivate, Node->getDynGroupprivateFallbackModifier()); + } + OS << ": "; + } + Node->getSize()->printPretty(OS, nullptr, Policy, 0); + OS << ')'; +} + void OMPClausePrinter::VisitOMPDoacrossClause(OMPDoacrossClause *Node) { OS << "doacross("; OpenMPDoacrossClauseModifier DepType = Node->getDependenceType(); diff --git a/clang/lib/AST/StmtProfile.cpp b/clang/lib/AST/StmtProfile.cpp index c909e1b..4a8c638 100644 --- a/clang/lib/AST/StmtProfile.cpp +++ b/clang/lib/AST/StmtProfile.cpp @@ -968,6 +968,12 @@ void OMPClauseProfiler::VisitOMPXDynCGroupMemClause( if (Expr *Size = C->getSize()) Profiler->VisitStmt(Size); } +void OMPClauseProfiler::VisitOMPDynGroupprivateClause( + const OMPDynGroupprivateClause *C) { + VisitOMPClauseWithPreInit(C); + if (auto *Size = C->getSize()) + Profiler->VisitStmt(Size); +} void OMPClauseProfiler::VisitOMPDoacrossClause(const OMPDoacrossClause *C) { VisitOMPClauseList(C); } diff --git a/clang/lib/Basic/OpenMPKinds.cpp b/clang/lib/Basic/OpenMPKinds.cpp index 3d41f2d..8e60fc2 100644 --- a/clang/lib/Basic/OpenMPKinds.cpp +++ b/clang/lib/Basic/OpenMPKinds.cpp @@ -196,6 +196,16 @@ unsigned clang::getOpenMPSimpleClauseType(OpenMPClauseKind Kind, StringRef Str, return OMPC_GRAINSIZE_unknown; return Type; } + case OMPC_dyn_groupprivate: { + return llvm::StringSwitch<unsigned>(Str) +#define OPENMP_DYN_GROUPPRIVATE_MODIFIER(Name) \ + .Case(#Name, OMPC_DYN_GROUPPRIVATE_##Name) +#define OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(Name) \ + .Case(#Name, OMPC_DYN_GROUPPRIVATE_FALLBACK_##Name) \ + .Case("fallback(" #Name ")", OMPC_DYN_GROUPPRIVATE_FALLBACK_##Name) +#include "clang/Basic/OpenMPKinds.def" + .Default(OMPC_DYN_GROUPPRIVATE_unknown); + } case OMPC_num_tasks: { unsigned Type = llvm::StringSwitch<unsigned>(Str) #define OPENMP_NUMTASKS_MODIFIER(Name) .Case(#Name, OMPC_NUMTASKS_##Name) @@ -544,6 +554,20 @@ const char *clang::getOpenMPSimpleClauseTypeName(OpenMPClauseKind Kind, #include "clang/Basic/OpenMPKinds.def" } llvm_unreachable("Invalid OpenMP 'grainsize' clause modifier"); + case OMPC_dyn_groupprivate: + switch (Type) { + case OMPC_DYN_GROUPPRIVATE_unknown: + case OMPC_DYN_GROUPPRIVATE_FALLBACK_last: + return "unknown"; +#define OPENMP_DYN_GROUPPRIVATE_MODIFIER(Name) \ + case OMPC_DYN_GROUPPRIVATE_##Name: \ + return #Name; +#define OPENMP_DYN_GROUPPRIVATE_FALLBACK_MODIFIER(Name) \ + case OMPC_DYN_GROUPPRIVATE_FALLBACK_##Name: \ + return "fallback(" #Name ")"; +#include "clang/Basic/OpenMPKinds.def" + } + llvm_unreachable("Invalid OpenMP 'dyn_groupprivate' clause modifier"); case OMPC_num_tasks: switch (Type) { case OMPC_NUMTASKS_unknown: diff --git a/clang/lib/CIR/CodeGen/CIRGenStmt.cpp b/clang/lib/CIR/CodeGen/CIRGenStmt.cpp index 1eb7199..7bb8c21 100644 --- a/clang/lib/CIR/CodeGen/CIRGenStmt.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenStmt.cpp @@ -66,7 +66,7 @@ static mlir::LogicalResult emitStmtWithResult(CIRGenFunction &cgf, mlir::LogicalResult CIRGenFunction::emitCompoundStmtWithoutScope( const CompoundStmt &s, Address *lastValue, AggValueSlot slot) { mlir::LogicalResult result = mlir::success(); - const Stmt *exprResult = s.getStmtExprResult(); + const Stmt *exprResult = s.body_back(); assert((!lastValue || (lastValue && exprResult)) && "If lastValue is not null then the CompoundStmt must have a " "StmtExprResult"); diff --git a/clang/lib/CodeGen/TargetBuiltins/ARM.cpp b/clang/lib/CodeGen/TargetBuiltins/ARM.cpp index 15fa78d..d4b0b81 100644 --- a/clang/lib/CodeGen/TargetBuiltins/ARM.cpp +++ b/clang/lib/CodeGen/TargetBuiltins/ARM.cpp @@ -590,6 +590,7 @@ struct ARMVectorIntrinsicInfo { Intrinsic::LLVMIntrinsic, Intrinsic::AltLLVMIntrinsic, \ TypeModifier } +// clang-format off static const ARMVectorIntrinsicInfo ARMSIMDIntrinsicMap [] = { NEONMAP1(__a32_vcvt_bf16_f32, arm_neon_vcvtfp2bf, 0), NEONMAP0(splat_lane_v), @@ -1217,35 +1218,55 @@ static const ARMVectorIntrinsicInfo AArch64SISDIntrinsicMap[] = { NEONMAP1(vcales_f32, aarch64_neon_facge, AddRetType | Add1ArgType), NEONMAP1(vcaltd_f64, aarch64_neon_facgt, AddRetType | Add1ArgType), NEONMAP1(vcalts_f32, aarch64_neon_facgt, AddRetType | Add1ArgType), + NEONMAP1(vcvtad_s32_f64, aarch64_neon_fcvtas, AddRetType | Add1ArgType), NEONMAP1(vcvtad_s64_f64, aarch64_neon_fcvtas, AddRetType | Add1ArgType), + NEONMAP1(vcvtad_u32_f64, aarch64_neon_fcvtau, AddRetType | Add1ArgType), NEONMAP1(vcvtad_u64_f64, aarch64_neon_fcvtau, AddRetType | Add1ArgType), NEONMAP1(vcvtas_s32_f32, aarch64_neon_fcvtas, AddRetType | Add1ArgType), + NEONMAP1(vcvtas_s64_f32, aarch64_neon_fcvtas, AddRetType | Add1ArgType), NEONMAP1(vcvtas_u32_f32, aarch64_neon_fcvtau, AddRetType | Add1ArgType), + NEONMAP1(vcvtas_u64_f32, aarch64_neon_fcvtau, AddRetType | Add1ArgType), NEONMAP1(vcvtd_n_f64_s64, aarch64_neon_vcvtfxs2fp, AddRetType | Add1ArgType), NEONMAP1(vcvtd_n_f64_u64, aarch64_neon_vcvtfxu2fp, AddRetType | Add1ArgType), NEONMAP1(vcvtd_n_s64_f64, aarch64_neon_vcvtfp2fxs, AddRetType | Add1ArgType), NEONMAP1(vcvtd_n_u64_f64, aarch64_neon_vcvtfp2fxu, AddRetType | Add1ArgType), + NEONMAP1(vcvtd_s32_f64, aarch64_neon_fcvtzs, AddRetType | Add1ArgType), NEONMAP1(vcvtd_s64_f64, aarch64_neon_fcvtzs, AddRetType | Add1ArgType), + NEONMAP1(vcvtd_u32_f64, aarch64_neon_fcvtzu, AddRetType | Add1ArgType), NEONMAP1(vcvtd_u64_f64, aarch64_neon_fcvtzu, AddRetType | Add1ArgType), NEONMAP0(vcvth_bf16_f32), + NEONMAP1(vcvtmd_s32_f64, aarch64_neon_fcvtms, AddRetType | Add1ArgType), NEONMAP1(vcvtmd_s64_f64, aarch64_neon_fcvtms, AddRetType | Add1ArgType), + NEONMAP1(vcvtmd_u32_f64, aarch64_neon_fcvtmu, AddRetType | Add1ArgType), NEONMAP1(vcvtmd_u64_f64, aarch64_neon_fcvtmu, AddRetType | Add1ArgType), NEONMAP1(vcvtms_s32_f32, aarch64_neon_fcvtms, AddRetType | Add1ArgType), + NEONMAP1(vcvtms_s64_f32, aarch64_neon_fcvtms, AddRetType | Add1ArgType), NEONMAP1(vcvtms_u32_f32, aarch64_neon_fcvtmu, AddRetType | Add1ArgType), + NEONMAP1(vcvtms_u64_f32, aarch64_neon_fcvtmu, AddRetType | Add1ArgType), + NEONMAP1(vcvtnd_s32_f64, aarch64_neon_fcvtns, AddRetType | Add1ArgType), NEONMAP1(vcvtnd_s64_f64, aarch64_neon_fcvtns, AddRetType | Add1ArgType), + NEONMAP1(vcvtnd_u32_f64, aarch64_neon_fcvtnu, AddRetType | Add1ArgType), NEONMAP1(vcvtnd_u64_f64, aarch64_neon_fcvtnu, AddRetType | Add1ArgType), NEONMAP1(vcvtns_s32_f32, aarch64_neon_fcvtns, AddRetType | Add1ArgType), + NEONMAP1(vcvtns_s64_f32, aarch64_neon_fcvtns, AddRetType | Add1ArgType), NEONMAP1(vcvtns_u32_f32, aarch64_neon_fcvtnu, AddRetType | Add1ArgType), + NEONMAP1(vcvtns_u64_f32, aarch64_neon_fcvtnu, AddRetType | Add1ArgType), + NEONMAP1(vcvtpd_s32_f64, aarch64_neon_fcvtps, AddRetType | Add1ArgType), NEONMAP1(vcvtpd_s64_f64, aarch64_neon_fcvtps, AddRetType | Add1ArgType), + NEONMAP1(vcvtpd_u32_f64, aarch64_neon_fcvtpu, AddRetType | Add1ArgType), NEONMAP1(vcvtpd_u64_f64, aarch64_neon_fcvtpu, AddRetType | Add1ArgType), NEONMAP1(vcvtps_s32_f32, aarch64_neon_fcvtps, AddRetType | Add1ArgType), + NEONMAP1(vcvtps_s64_f32, aarch64_neon_fcvtps, AddRetType | Add1ArgType), NEONMAP1(vcvtps_u32_f32, aarch64_neon_fcvtpu, AddRetType | Add1ArgType), + NEONMAP1(vcvtps_u64_f32, aarch64_neon_fcvtpu, AddRetType | Add1ArgType), NEONMAP1(vcvts_n_f32_s32, aarch64_neon_vcvtfxs2fp, AddRetType | Add1ArgType), NEONMAP1(vcvts_n_f32_u32, aarch64_neon_vcvtfxu2fp, AddRetType | Add1ArgType), NEONMAP1(vcvts_n_s32_f32, aarch64_neon_vcvtfp2fxs, AddRetType | Add1ArgType), NEONMAP1(vcvts_n_u32_f32, aarch64_neon_vcvtfp2fxu, AddRetType | Add1ArgType), NEONMAP1(vcvts_s32_f32, aarch64_neon_fcvtzs, AddRetType | Add1ArgType), + NEONMAP1(vcvts_s64_f32, aarch64_neon_fcvtzs, AddRetType | Add1ArgType), NEONMAP1(vcvts_u32_f32, aarch64_neon_fcvtzu, AddRetType | Add1ArgType), + NEONMAP1(vcvts_u64_f32, aarch64_neon_fcvtzu, AddRetType | Add1ArgType), NEONMAP1(vcvtxd_f32_f64, aarch64_sisd_fcvtxn, 0), NEONMAP1(vmaxnmv_f32, aarch64_neon_fmaxnmv, AddRetType | Add1ArgType), NEONMAP1(vmaxnmvq_f32, aarch64_neon_fmaxnmv, AddRetType | Add1ArgType), @@ -1446,6 +1467,7 @@ static const ARMVectorIntrinsicInfo AArch64SISDIntrinsicMap[] = { NEONMAP1(vrsqrteh_f16, aarch64_neon_frsqrte, Add1ArgType), NEONMAP1(vrsqrtsh_f16, aarch64_neon_frsqrts, Add1ArgType), }; +// clang-format on // Some intrinsics are equivalent for codegen. static const std::pair<unsigned, unsigned> NEONEquivalentIntrinsicMap[] = { @@ -7624,6 +7646,16 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned BuiltinID, Int = Intrinsic::aarch64_neon_vluti4q_laneq_x2; return EmitNeonCall(CGM.getIntrinsic(Int, Ty), Ops, "vluti4q_laneq_x2"); } + case NEON::BI__builtin_neon_vmmlaq_f16_mf8_fpm: + return EmitFP8NeonCall(Intrinsic::aarch64_neon_fmmla, + {llvm::FixedVectorType::get(HalfTy, 8), + llvm::FixedVectorType::get(Int8Ty, 16)}, + Ops, E, "fmmla"); + case NEON::BI__builtin_neon_vmmlaq_f32_mf8_fpm: + return EmitFP8NeonCall(Intrinsic::aarch64_neon_fmmla, + {llvm::FixedVectorType::get(FloatTy, 4), + llvm::FixedVectorType::get(Int8Ty, 16)}, + Ops, E, "fmmla"); case NEON::BI__builtin_neon_vcvt1_low_bf16_mf8_fpm: ExtractLow = true; [[fallthrough]]; diff --git a/clang/lib/Format/Format.cpp b/clang/lib/Format/Format.cpp index dd14fcd..9bbb33c 100644 --- a/clang/lib/Format/Format.cpp +++ b/clang/lib/Format/Format.cpp @@ -876,27 +876,28 @@ template <> struct MappingTraits<FormatStyle::TrailingCommentsAlignmentStyle> { FormatStyle::TrailingCommentsAlignmentStyle &Value) { IO.enumCase(Value, "Leave", FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Leave, 0})); + {FormatStyle::TCAS_Leave, 0, true})); IO.enumCase(Value, "Always", FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Always, 0})); + {FormatStyle::TCAS_Always, 0, true})); IO.enumCase(Value, "Never", FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Never, 0})); + {FormatStyle::TCAS_Never, 0, true})); // For backwards compatibility IO.enumCase(Value, "true", FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Always, 0})); + {FormatStyle::TCAS_Always, 0, true})); IO.enumCase(Value, "false", FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Never, 0})); + {FormatStyle::TCAS_Never, 0, true})); } static void mapping(IO &IO, FormatStyle::TrailingCommentsAlignmentStyle &Value) { + IO.mapOptional("AlignPPAndNotPP", Value.AlignPPAndNotPP); IO.mapOptional("Kind", Value.Kind); IO.mapOptional("OverEmptyLines", Value.OverEmptyLines); } @@ -1651,6 +1652,7 @@ FormatStyle getLLVMStyle(FormatStyle::LanguageKind Language) { LLVMStyle.AlignTrailingComments = {}; LLVMStyle.AlignTrailingComments.Kind = FormatStyle::TCAS_Always; LLVMStyle.AlignTrailingComments.OverEmptyLines = 0; + LLVMStyle.AlignTrailingComments.AlignPPAndNotPP = true; LLVMStyle.AllowAllArgumentsOnNextLine = true; LLVMStyle.AllowAllParametersOfDeclarationOnNextLine = true; LLVMStyle.AllowBreakBeforeNoexceptSpecifier = FormatStyle::BBNSS_Never; diff --git a/clang/lib/Format/WhitespaceManager.cpp b/clang/lib/Format/WhitespaceManager.cpp index 406c77c..fece384 100644 --- a/clang/lib/Format/WhitespaceManager.cpp +++ b/clang/lib/Format/WhitespaceManager.cpp @@ -1007,9 +1007,13 @@ void WhitespaceManager::alignTrailingComments() { return; const int Size = Changes.size(); + if (Size == 0) + return; + int MinColumn = 0; int StartOfSequence = 0; bool BreakBeforeNext = false; + bool IsInPP = Changes.front().Tok->Tok.is(tok::hash); int NewLineThreshold = 1; if (Style.AlignTrailingComments.Kind == FormatStyle::TCAS_Always) NewLineThreshold = Style.AlignTrailingComments.OverEmptyLines + 1; @@ -1018,7 +1022,19 @@ void WhitespaceManager::alignTrailingComments() { auto &C = Changes[I]; if (C.StartOfBlockComment) continue; - Newlines += C.NewlinesBefore; + if (C.NewlinesBefore != 0) { + Newlines += C.NewlinesBefore; + const bool WasInPP = std::exchange( + IsInPP, C.Tok->Tok.is(tok::hash) || (IsInPP && C.IsTrailingComment) || + C.ContinuesPPDirective); + if (IsInPP != WasInPP && !Style.AlignTrailingComments.AlignPPAndNotPP) { + alignTrailingComments(StartOfSequence, I, MinColumn); + MinColumn = 0; + MaxColumn = INT_MAX; + StartOfSequence = I; + Newlines = 0; + } + } if (!C.IsTrailingComment) continue; diff --git a/clang/lib/Frontend/FrontendActions.cpp b/clang/lib/Frontend/FrontendActions.cpp index d7d56b8..3595bbc 100644 --- a/clang/lib/Frontend/FrontendActions.cpp +++ b/clang/lib/Frontend/FrontendActions.cpp @@ -1233,20 +1233,6 @@ void PrintDependencyDirectivesSourceMinimizerAction::ExecuteAction() { llvm::outs()); } -void GetDependenciesByModuleNameAction::ExecuteAction() { - CompilerInstance &CI = getCompilerInstance(); - Preprocessor &PP = CI.getPreprocessor(); - SourceManager &SM = PP.getSourceManager(); - FileID MainFileID = SM.getMainFileID(); - SourceLocation FileStart = SM.getLocForStartOfFile(MainFileID); - SmallVector<IdentifierLoc, 2> Path; - IdentifierInfo *ModuleID = PP.getIdentifierInfo(ModuleName); - Path.emplace_back(FileStart, ModuleID); - auto ModResult = CI.loadModule(FileStart, Path, Module::Hidden, false); - PPCallbacks *CB = PP.getPPCallbacks(); - CB->moduleImport(SourceLocation(), Path, ModResult); -} - //===----------------------------------------------------------------------===// // HLSL Specific Actions //===----------------------------------------------------------------------===// diff --git a/clang/lib/Headers/llvm_libc_wrappers/stdlib.h b/clang/lib/Headers/llvm_libc_wrappers/stdlib.h index 1da22abd0..d79e7fa 100644 --- a/clang/lib/Headers/llvm_libc_wrappers/stdlib.h +++ b/clang/lib/Headers/llvm_libc_wrappers/stdlib.h @@ -34,13 +34,13 @@ _Static_assert(__builtin_offsetof(div_t, quot) == 0, "ABI mismatch!"); _Static_assert(__builtin_offsetof(ldiv_t, quot) == 0, "ABI mismatch!"); _Static_assert(__builtin_offsetof(lldiv_t, quot) == 0, "ABI mismatch!"); -#if defined(__GLIBC__) && __cplusplus >= 201703L +#if defined(__GLIBC__) && __cplusplus >= 201103L #define at_quick_exit atexit #endif #include <llvm-libc-decls/stdlib.h> -#if defined(__GLIBC__) && __cplusplus >= 201703L +#if defined(__GLIBC__) && __cplusplus >= 201103L #undef at_quick_exit #endif diff --git a/clang/lib/Parse/ParseOpenMP.cpp b/clang/lib/Parse/ParseOpenMP.cpp index 31bc941..334438e 100644 --- a/clang/lib/Parse/ParseOpenMP.cpp +++ b/clang/lib/Parse/ParseOpenMP.cpp @@ -3178,6 +3178,7 @@ OMPClause *Parser::ParseOpenMPClause(OpenMPDirectiveKind DKind, case OMPC_align: case OMPC_message: case OMPC_ompx_dyn_cgroup_mem: + case OMPC_dyn_groupprivate: // OpenMP [2.5, Restrictions] // At most one num_threads clause can appear on the directive. // OpenMP [2.8.1, simd construct, Restrictions] @@ -3216,7 +3217,7 @@ OMPClause *Parser::ParseOpenMPClause(OpenMPDirectiveKind DKind, PP.LookAhead(/*N=*/0).isNot(tok::l_paren)) Clause = ParseOpenMPClause(CKind, WrongDirective); else if (CKind == OMPC_grainsize || CKind == OMPC_num_tasks || - CKind == OMPC_num_threads) + CKind == OMPC_num_threads || CKind == OMPC_dyn_groupprivate) Clause = ParseOpenMPSingleExprWithArgClause(DKind, CKind, WrongDirective); else Clause = ParseOpenMPSingleExprClause(CKind, WrongDirective); @@ -4009,6 +4010,83 @@ OMPClause *Parser::ParseOpenMPSingleExprWithArgClause(OpenMPDirectiveKind DKind, Arg.push_back(OMPC_GRAINSIZE_unknown); KLoc.emplace_back(); } + } else if (Kind == OMPC_dyn_groupprivate) { + enum { SimpleModifier, ComplexModifier, NumberOfModifiers }; + Arg.resize(NumberOfModifiers); + KLoc.resize(NumberOfModifiers); + Arg[SimpleModifier] = OMPC_DYN_GROUPPRIVATE_unknown; + Arg[ComplexModifier] = OMPC_DYN_GROUPPRIVATE_FALLBACK_unknown; + + auto ConsumeModifier = [&]() { + unsigned Type = NumberOfModifiers; + unsigned Modifier; + SourceLocation Loc; + if (!Tok.isAnnotation() && PP.getSpelling(Tok) == "fallback" && + NextToken().is(tok::l_paren)) { + ConsumeToken(); + BalancedDelimiterTracker ParenT(*this, tok::l_paren, tok::r_paren); + ParenT.consumeOpen(); + + Modifier = getOpenMPSimpleClauseType( + Kind, Tok.isAnnotation() ? "" : PP.getSpelling(Tok), getLangOpts()); + if (Modifier <= OMPC_DYN_GROUPPRIVATE_FALLBACK_unknown || + Modifier >= OMPC_DYN_GROUPPRIVATE_FALLBACK_last) { + Diag(Tok.getLocation(), diag::err_expected) + << "'abort', 'null' or 'default_mem' in fallback modifier"; + SkipUntil(tok::r_paren); + return std::make_tuple(Type, Modifier, Loc); + } + Type = ComplexModifier; + Loc = Tok.getLocation(); + if (Tok.isNot(tok::r_paren) && Tok.isNot(tok::comma) && + Tok.isNot(tok::annot_pragma_openmp_end)) + ConsumeAnyToken(); + ParenT.consumeClose(); + } else { + Modifier = getOpenMPSimpleClauseType( + Kind, Tok.isAnnotation() ? "" : PP.getSpelling(Tok), getLangOpts()); + if (Modifier < OMPC_DYN_GROUPPRIVATE_unknown) { + Type = SimpleModifier; + Loc = Tok.getLocation(); + if (Tok.isNot(tok::r_paren) && Tok.isNot(tok::comma) && + Tok.isNot(tok::annot_pragma_openmp_end)) + ConsumeAnyToken(); + } + } + return std::make_tuple(Type, Modifier, Loc); + }; + + auto SaveModifier = [&](unsigned Type, unsigned Modifier, + SourceLocation Loc) { + assert(Type < NumberOfModifiers && "Unexpected modifier type"); + if (!KLoc[Type].isValid()) { + Arg[Type] = Modifier; + KLoc[Type] = Loc; + } else { + Diag(Loc, diag::err_omp_incompatible_dyn_groupprivate_modifier) + << getOpenMPSimpleClauseTypeName(OMPC_dyn_groupprivate, Modifier) + << getOpenMPSimpleClauseTypeName(OMPC_dyn_groupprivate, Arg[Type]); + } + }; + + // Parse 'modifier' + auto [Type1, Mod1, Loc1] = ConsumeModifier(); + if (Type1 < NumberOfModifiers) { + SaveModifier(Type1, Mod1, Loc1); + if (Tok.is(tok::comma)) { + // Parse ',' 'modifier' + ConsumeAnyToken(); + auto [Type2, Mod2, Loc2] = ConsumeModifier(); + if (Type2 < NumberOfModifiers) + SaveModifier(Type2, Mod2, Loc2); + } + // Parse ':' + if (Tok.is(tok::colon)) + ConsumeAnyToken(); + else + Diag(Tok, diag::warn_pragma_expected_colon) + << "dyn_groupprivate modifier"; + } } else if (Kind == OMPC_num_tasks) { // Parse optional <num_tasks modifier> ':' OpenMPNumTasksClauseModifier Modifier = @@ -4083,11 +4161,11 @@ OMPClause *Parser::ParseOpenMPSingleExprWithArgClause(OpenMPDirectiveKind DKind, } } - bool NeedAnExpression = (Kind == OMPC_schedule && DelimLoc.isValid()) || - (Kind == OMPC_dist_schedule && DelimLoc.isValid()) || - Kind == OMPC_if || Kind == OMPC_device || - Kind == OMPC_grainsize || Kind == OMPC_num_tasks || - Kind == OMPC_num_threads; + bool NeedAnExpression = + (Kind == OMPC_schedule && DelimLoc.isValid()) || + (Kind == OMPC_dist_schedule && DelimLoc.isValid()) || Kind == OMPC_if || + Kind == OMPC_device || Kind == OMPC_grainsize || Kind == OMPC_num_tasks || + Kind == OMPC_num_threads || Kind == OMPC_dyn_groupprivate; if (NeedAnExpression) { SourceLocation ELoc = Tok.getLocation(); ExprResult LHS( diff --git a/clang/lib/Sema/SemaOpenMP.cpp b/clang/lib/Sema/SemaOpenMP.cpp index 256f952..465dab2 100644 --- a/clang/lib/Sema/SemaOpenMP.cpp +++ b/clang/lib/Sema/SemaOpenMP.cpp @@ -16532,6 +16532,7 @@ OMPClause *SemaOpenMP::ActOnOpenMPSingleExprClause(OpenMPClauseKind Kind, case OMPC_holds: Res = ActOnOpenMPHoldsClause(Expr, StartLoc, LParenLoc, EndLoc); break; + case OMPC_dyn_groupprivate: case OMPC_grainsize: case OMPC_num_tasks: case OMPC_num_threads: @@ -16658,6 +16659,8 @@ static OpenMPDirectiveKind getOpenMPCaptureRegionForClause( case OMPC_num_teams: case OMPC_thread_limit: case OMPC_ompx_dyn_cgroup_mem: + case OMPC_dyn_groupprivate: + // TODO: This may need to consider teams too. if (Leafs[0] == OMPD_target) return OMPD_target; break; @@ -17705,7 +17708,7 @@ OMPClause *SemaOpenMP::ActOnOpenMPSingleExprWithArgClause( SourceLocation EndLoc) { OMPClause *Res = nullptr; switch (Kind) { - case OMPC_schedule: + case OMPC_schedule: { enum { Modifier1, Modifier2, ScheduleKind, NumberOfElements }; assert(Argument.size() == NumberOfElements && ArgumentLoc.size() == NumberOfElements); @@ -17716,6 +17719,7 @@ OMPClause *SemaOpenMP::ActOnOpenMPSingleExprWithArgClause( StartLoc, LParenLoc, ArgumentLoc[Modifier1], ArgumentLoc[Modifier2], ArgumentLoc[ScheduleKind], DelimLoc, EndLoc); break; + } case OMPC_if: assert(Argument.size() == 1 && ArgumentLoc.size() == 1); Res = ActOnOpenMPIfClause(static_cast<OpenMPDirectiveKind>(Argument.back()), @@ -17771,6 +17775,20 @@ OMPClause *SemaOpenMP::ActOnOpenMPSingleExprWithArgClause( static_cast<OpenMPNumTasksClauseModifier>(Argument.back()), Expr, StartLoc, LParenLoc, ArgumentLoc.back(), EndLoc); break; + case OMPC_dyn_groupprivate: { + enum { Modifier1, Modifier2, NumberOfElements }; + assert(Argument.size() == NumberOfElements && + ArgumentLoc.size() == NumberOfElements && + "Modifiers for dyn_groupprivate clause and their locations are " + "expected."); + Res = ActOnOpenMPDynGroupprivateClause( + static_cast<OpenMPDynGroupprivateClauseModifier>(Argument[Modifier1]), + static_cast<OpenMPDynGroupprivateClauseFallbackModifier>( + Argument[Modifier2]), + Expr, StartLoc, LParenLoc, ArgumentLoc[Modifier1], + ArgumentLoc[Modifier2], EndLoc); + break; + } case OMPC_num_threads: assert(Argument.size() == 1 && ArgumentLoc.size() == 1 && "Modifier for num_threads clause and its location are expected."); @@ -18127,6 +18145,7 @@ OMPClause *SemaOpenMP::ActOnOpenMPClause(OpenMPClauseKind Kind, case OMPC_affinity: case OMPC_when: case OMPC_ompx_dyn_cgroup_mem: + case OMPC_dyn_groupprivate: default: llvm_unreachable("Clause is not allowed."); } @@ -25246,6 +25265,49 @@ OMPClause *SemaOpenMP::ActOnOpenMPXDynCGroupMemClause(Expr *Size, ValExpr, HelperValStmt, CaptureRegion, StartLoc, LParenLoc, EndLoc); } +OMPClause *SemaOpenMP::ActOnOpenMPDynGroupprivateClause( + OpenMPDynGroupprivateClauseModifier M1, + OpenMPDynGroupprivateClauseFallbackModifier M2, Expr *Size, + SourceLocation StartLoc, SourceLocation LParenLoc, SourceLocation M1Loc, + SourceLocation M2Loc, SourceLocation EndLoc) { + + if ((M1Loc.isValid() && M1 == OMPC_DYN_GROUPPRIVATE_unknown) || + (M2Loc.isValid() && M2 == OMPC_DYN_GROUPPRIVATE_FALLBACK_unknown)) { + std::string Values = getListOfPossibleValues( + OMPC_dyn_groupprivate, /*First=*/0, OMPC_DYN_GROUPPRIVATE_unknown); + Diag((M1Loc.isValid() && M1 == OMPC_DYN_GROUPPRIVATE_unknown) ? M1Loc + : M2Loc, + diag::err_omp_unexpected_clause_value) + << Values << getOpenMPClauseName(OMPC_dyn_groupprivate); + return nullptr; + } + + Expr *ValExpr = Size; + Stmt *HelperValStmt = nullptr; + + // OpenMP [2.5, Restrictions] + // The dyn_groupprivate expression must evaluate to a positive integer + // value. + if (!isNonNegativeIntegerValue(ValExpr, SemaRef, OMPC_dyn_groupprivate, + /*StrictlyPositive=*/false)) + return nullptr; + + OpenMPDirectiveKind DKind = DSAStack->getCurrentDirective(); + OpenMPDirectiveKind CaptureRegion = getOpenMPCaptureRegionForClause( + DKind, OMPC_dyn_groupprivate, getLangOpts().OpenMP); + if (CaptureRegion != OMPD_unknown && + !SemaRef.CurContext->isDependentContext()) { + ValExpr = SemaRef.MakeFullExpr(ValExpr).get(); + llvm::MapVector<const Expr *, DeclRefExpr *> Captures; + ValExpr = tryBuildCapture(SemaRef, ValExpr, Captures).get(); + HelperValStmt = buildPreInits(getASTContext(), Captures); + } + + return new (getASTContext()) OMPDynGroupprivateClause( + StartLoc, LParenLoc, EndLoc, ValExpr, HelperValStmt, CaptureRegion, M1, + M1Loc, M2, M2Loc); +} + OMPClause *SemaOpenMP::ActOnOpenMPDoacrossClause( OpenMPDoacrossClauseModifier DepType, SourceLocation DepLoc, SourceLocation ColonLoc, ArrayRef<Expr *> VarList, SourceLocation StartLoc, diff --git a/clang/lib/Sema/TreeTransform.h b/clang/lib/Sema/TreeTransform.h index 94105f1..c249148 100644 --- a/clang/lib/Sema/TreeTransform.h +++ b/clang/lib/Sema/TreeTransform.h @@ -2463,6 +2463,19 @@ public: LParenLoc, EndLoc); } + /// Build a new OpenMP 'dyn_groupprivate' clause. + /// + /// By default, performs semantic analysis to build the new OpenMP clause. + /// Subclasses may override this routine to provide different behavior. + OMPClause *RebuildOMPDynGroupprivateClause( + OpenMPDynGroupprivateClauseModifier M1, + OpenMPDynGroupprivateClauseFallbackModifier M2, Expr *Size, + SourceLocation StartLoc, SourceLocation LParenLoc, SourceLocation M1Loc, + SourceLocation M2Loc, SourceLocation EndLoc) { + return getSema().OpenMP().ActOnOpenMPDynGroupprivateClause( + M1, M2, Size, StartLoc, LParenLoc, M1Loc, M2Loc, EndLoc); + } + /// Build a new OpenMP 'ompx_attribute' clause. /// /// By default, performs semantic analysis to build the new OpenMP clause. @@ -11726,6 +11739,19 @@ OMPClause *TreeTransform<Derived>::TransformOMPXDynCGroupMemClause( } template <typename Derived> +OMPClause *TreeTransform<Derived>::TransformOMPDynGroupprivateClause( + OMPDynGroupprivateClause *C) { + ExprResult Size = getDerived().TransformExpr(C->getSize()); + if (Size.isInvalid()) + return nullptr; + return getDerived().RebuildOMPDynGroupprivateClause( + C->getDynGroupprivateModifier(), C->getDynGroupprivateFallbackModifier(), + Size.get(), C->getBeginLoc(), C->getLParenLoc(), + C->getDynGroupprivateModifierLoc(), + C->getDynGroupprivateFallbackModifierLoc(), C->getEndLoc()); +} + +template <typename Derived> OMPClause * TreeTransform<Derived>::TransformOMPDoacrossClause(OMPDoacrossClause *C) { llvm::SmallVector<Expr *, 16> Vars; diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp index d552821..a04041c 100644 --- a/clang/lib/Serialization/ASTReader.cpp +++ b/clang/lib/Serialization/ASTReader.cpp @@ -11544,6 +11544,9 @@ OMPClause *OMPClauseReader::readClause() { case llvm::omp::OMPC_ompx_dyn_cgroup_mem: C = new (Context) OMPXDynCGroupMemClause(); break; + case llvm::omp::OMPC_dyn_groupprivate: + C = new (Context) OMPDynGroupprivateClause(); + break; case llvm::omp::OMPC_doacross: { unsigned NumVars = Record.readInt(); unsigned NumLoops = Record.readInt(); @@ -12736,6 +12739,19 @@ void OMPClauseReader::VisitOMPXDynCGroupMemClause(OMPXDynCGroupMemClause *C) { C->setLParenLoc(Record.readSourceLocation()); } +void OMPClauseReader::VisitOMPDynGroupprivateClause( + OMPDynGroupprivateClause *C) { + VisitOMPClauseWithPreInit(C); + C->setDynGroupprivateModifier( + Record.readEnum<OpenMPDynGroupprivateClauseModifier>()); + C->setDynGroupprivateFallbackModifier( + Record.readEnum<OpenMPDynGroupprivateClauseFallbackModifier>()); + C->setSize(Record.readSubExpr()); + C->setLParenLoc(Record.readSourceLocation()); + C->setDynGroupprivateModifierLoc(Record.readSourceLocation()); + C->setDynGroupprivateFallbackModifierLoc(Record.readSourceLocation()); +} + void OMPClauseReader::VisitOMPDoacrossClause(OMPDoacrossClause *C) { C->setLParenLoc(Record.readSourceLocation()); C->setDependenceType( diff --git a/clang/lib/Serialization/ASTWriter.cpp b/clang/lib/Serialization/ASTWriter.cpp index b1fd151..821e7df 100644 --- a/clang/lib/Serialization/ASTWriter.cpp +++ b/clang/lib/Serialization/ASTWriter.cpp @@ -8651,6 +8651,17 @@ void OMPClauseWriter::VisitOMPXDynCGroupMemClause(OMPXDynCGroupMemClause *C) { Record.AddSourceLocation(C->getLParenLoc()); } +void OMPClauseWriter::VisitOMPDynGroupprivateClause( + OMPDynGroupprivateClause *C) { + VisitOMPClauseWithPreInit(C); + Record.push_back(C->getDynGroupprivateModifier()); + Record.push_back(C->getDynGroupprivateFallbackModifier()); + Record.AddStmt(C->getSize()); + Record.AddSourceLocation(C->getLParenLoc()); + Record.AddSourceLocation(C->getDynGroupprivateModifierLoc()); + Record.AddSourceLocation(C->getDynGroupprivateFallbackModifierLoc()); +} + void OMPClauseWriter::VisitOMPDoacrossClause(OMPDoacrossClause *C) { Record.push_back(C->varlist_size()); Record.push_back(C->getNumLoops()); diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.cpp b/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.cpp index eebecdb..33f5640 100644 --- a/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.cpp +++ b/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.cpp @@ -1,4 +1,4 @@ -//===- DependencyScanner.cpp - Performs module dependency scanning --------===// +//===- DependencyScannerImpl.cpp - Implements module dependency scanning --===// // // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. // See https://llvm.org/LICENSE.txt for license information. @@ -12,6 +12,7 @@ #include "clang/Driver/Driver.h" #include "clang/Frontend/FrontendActions.h" #include "clang/Tooling/DependencyScanning/DependencyScanningWorker.h" +#include "llvm/ADT/ScopeExit.h" #include "llvm/TargetParser/Host.h" using namespace clang; @@ -456,7 +457,8 @@ initVFSForTUBuferScanning(IntrusiveRefCntPtr<llvm::vfs::FileSystem> BaseFS, return std::make_pair(ModifiedFS, ModifiedCommandLine); } -std::pair<IntrusiveRefCntPtr<llvm::vfs::FileSystem>, std::vector<std::string>> +std::pair<IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem>, + std::vector<std::string>> initVFSForByNameScanning(IntrusiveRefCntPtr<llvm::vfs::FileSystem> BaseFS, ArrayRef<std::string> CommandLine, StringRef WorkingDirectory, StringRef ModuleName) { @@ -588,7 +590,7 @@ computePrebuiltModulesASTMap(CompilerInstance &ScanInstance, } std::unique_ptr<DependencyOutputOptions> -takeDependencyOutputOptionsFrom(CompilerInstance &ScanInstance) { +takeAndUpdateDependencyOutputOptionsFrom(CompilerInstance &ScanInstance) { // This function moves the existing dependency output options from the // invocation to the collector. The options in the invocation are reset, // which ensures that the compiler won't create new dependency collectors, @@ -675,7 +677,7 @@ bool DependencyScanningAction::runInvocation( if (!MaybePrebuiltModulesASTMap) return false; - auto DepOutputOpts = takeDependencyOutputOptionsFrom(ScanInstance); + auto DepOutputOpts = takeAndUpdateDependencyOutputOptionsFrom(ScanInstance); MDC = initializeScanInstanceDependencyCollector( ScanInstance, std::move(DepOutputOpts), WorkingDirectory, Consumer, @@ -686,8 +688,6 @@ bool DependencyScanningAction::runInvocation( if (Service.getFormat() == ScanningOutputFormat::P1689) Action = std::make_unique<PreprocessOnlyAction>(); - else if (ModuleName) - Action = std::make_unique<GetDependenciesByModuleNameAction>(*ModuleName); else Action = std::make_unique<ReadPCHAndPreprocessAction>(); @@ -704,3 +704,175 @@ bool DependencyScanningAction::runInvocation( return Result; } + +bool CompilerInstanceWithContext::initialize(DiagnosticConsumer *DC) { + if (DC) { + DiagConsumer = DC; + } else { + DiagPrinterWithOS = + std::make_unique<TextDiagnosticsPrinterWithOutput>(CommandLine); + DiagConsumer = &DiagPrinterWithOS->DiagPrinter; + } + + DiagEngineWithCmdAndOpts = std::make_unique<DignosticsEngineWithDiagOpts>( + CommandLine, OverlayFS, *DiagConsumer); + + std::tie(OverlayFS, CommandLine) = initVFSForByNameScanning( + Worker.BaseFS, CommandLine, CWD, "ScanningByName"); + + std::tie(Driver, Compilation) = buildCompilation( + CommandLine, *DiagEngineWithCmdAndOpts->DiagEngine, OverlayFS, Alloc); + + if (!Compilation) + return false; + + assert(Compilation->getJobs().size() && + "Must have a job list of non-zero size"); + const driver::Command &Command = *(Compilation->getJobs().begin()); + const auto &CommandArgs = Command.getArguments(); + assert(!CommandArgs.empty() && "Cannot have a command with 0 args"); + assert(StringRef(CommandArgs[0]) == "-cc1" && "Requires a cc1 job."); + OriginalInvocation = std::make_unique<CompilerInvocation>(); + + if (!CompilerInvocation::CreateFromArgs(*OriginalInvocation, CommandArgs, + *DiagEngineWithCmdAndOpts->DiagEngine, + Command.getExecutable())) { + DiagEngineWithCmdAndOpts->DiagEngine->Report( + diag::err_fe_expected_compiler_job) + << llvm::join(CommandLine, " "); + return false; + } + + if (any(Worker.Service.getOptimizeArgs() & ScanningOptimizations::Macros)) + canonicalizeDefines(OriginalInvocation->getPreprocessorOpts()); + + // Create the CompilerInstance. + IntrusiveRefCntPtr<ModuleCache> ModCache = + makeInProcessModuleCache(Worker.Service.getModuleCacheEntries()); + CIPtr = std::make_unique<CompilerInstance>( + std::make_shared<CompilerInvocation>(*OriginalInvocation), + Worker.PCHContainerOps, ModCache.get()); + auto &CI = *CIPtr; + + if (!initializeScanCompilerInstance( + CI, OverlayFS, DiagEngineWithCmdAndOpts->DiagEngine->getClient(), + Worker.Service, Worker.DepFS)) + return false; + + StableDirs = getInitialStableDirs(CI); + auto MaybePrebuiltModulesASTMap = + computePrebuiltModulesASTMap(CI, StableDirs); + if (!MaybePrebuiltModulesASTMap) + return false; + + PrebuiltModuleASTMap = std::move(*MaybePrebuiltModulesASTMap); + OutputOpts = takeAndUpdateDependencyOutputOptionsFrom(CI); + + // We do not create the target in initializeScanCompilerInstance because + // setting it here is unique for by-name lookups. We create the target only + // once here, and the information is reused for all computeDependencies calls. + // We do not need to call createTarget explicitly if we go through + // CompilerInstance::ExecuteAction to perform scanning. + CI.createTarget(); + + return true; +} + +bool CompilerInstanceWithContext::computeDependencies( + StringRef ModuleName, DependencyConsumer &Consumer, + DependencyActionController &Controller) { + assert(CIPtr && "CIPtr must be initialized before calling this method"); + auto &CI = *CIPtr; + + // We create this cleanup object because computeDependencies may exit + // early with errors. + auto CleanUp = llvm::make_scope_exit([&]() { + CI.clearDependencyCollectors(); + // The preprocessor may not be created at the entry of this method, + // but it must have been created when this method returns, whether + // there are errors during scanning or not. + CI.getPreprocessor().removePPCallbacks(); + }); + + auto MDC = initializeScanInstanceDependencyCollector( + CI, std::make_unique<DependencyOutputOptions>(*OutputOpts), CWD, Consumer, + Worker.Service, + /* The MDC's constructor makes a copy of the OriginalInvocation, so + we can pass it in without worrying that it might be changed across + invocations of computeDependencies. */ + *OriginalInvocation, Controller, PrebuiltModuleASTMap, StableDirs); + + if (!SrcLocOffset) { + // When SrcLocOffset is zero, we are at the beginning of the fake source + // file. In this case, we call BeginSourceFile to initialize. + std::unique_ptr<FrontendAction> Action = + std::make_unique<PreprocessOnlyAction>(); + auto InputFile = CI.getFrontendOpts().Inputs.begin(); + bool ActionBeginSucceeded = Action->BeginSourceFile(CI, *InputFile); + assert(ActionBeginSucceeded && "Action BeginSourceFile must succeed"); + (void)ActionBeginSucceeded; + } + + Preprocessor &PP = CI.getPreprocessor(); + SourceManager &SM = PP.getSourceManager(); + FileID MainFileID = SM.getMainFileID(); + SourceLocation FileStart = SM.getLocForStartOfFile(MainFileID); + SourceLocation IDLocation = FileStart.getLocWithOffset(SrcLocOffset); + PPCallbacks *CB = nullptr; + if (!SrcLocOffset) { + // We need to call EnterSourceFile when SrcLocOffset is zero to initialize + // the preprocessor. + bool PPFailed = PP.EnterSourceFile(MainFileID, nullptr, SourceLocation()); + assert(!PPFailed && "Preprocess must be able to enter the main file."); + (void)PPFailed; + CB = MDC->getPPCallbacks(); + } else { + // When SrcLocOffset is non-zero, the preprocessor has already been + // initialized through a previous call of computeDependencies. We want to + // preserve the PP's state, hence we do not call EnterSourceFile again. + MDC->attachToPreprocessor(PP); + CB = MDC->getPPCallbacks(); + + FileID PrevFID; + SrcMgr::CharacteristicKind FileType = SM.getFileCharacteristic(IDLocation); + CB->LexedFileChanged(MainFileID, + PPChainedCallbacks::LexedFileChangeReason::EnterFile, + FileType, PrevFID, IDLocation); + } + + SrcLocOffset++; + SmallVector<IdentifierLoc, 2> Path; + IdentifierInfo *ModuleID = PP.getIdentifierInfo(ModuleName); + Path.emplace_back(IDLocation, ModuleID); + auto ModResult = CI.loadModule(IDLocation, Path, Module::Hidden, false); + + assert(CB && "Must have PPCallbacks after module loading"); + CB->moduleImport(SourceLocation(), Path, ModResult); + // Note that we are calling the CB's EndOfMainFile function, which + // forwards the results to the dependency consumer. + // It does not indicate the end of processing the fake file. + CB->EndOfMainFile(); + + if (!ModResult) + return false; + + CompilerInvocation ModuleInvocation(*OriginalInvocation); + MDC->applyDiscoveredDependencies(ModuleInvocation); + Consumer.handleBuildCommand( + {CommandLine[0], ModuleInvocation.getCC1CommandLine()}); + + return true; +} + +bool CompilerInstanceWithContext::finalize() { + DiagConsumer->finish(); + return true; +} + +llvm::Error CompilerInstanceWithContext::handleReturnStatus(bool Success) { + assert(DiagPrinterWithOS && "Must use the default DiagnosticConsumer."); + return Success ? llvm::Error::success() + : llvm::make_error<llvm::StringError>( + DiagPrinterWithOS->DiagnosticsOS.str(), + llvm::inconvertibleErrorCode()); +} diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.h b/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.h index 5657317..54166da 100644 --- a/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.h +++ b/clang/lib/Tooling/DependencyScanning/DependencyScannerImpl.h @@ -1,4 +1,4 @@ -//===- DependencyScanner.h - Performs module dependency scanning *- C++ -*-===// +//===- DependencyScannerImpl.h - Implements dependency scanning *- C++ -*--===// // // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. // See https://llvm.org/LICENSE.txt for license information. @@ -23,6 +23,8 @@ class DiagnosticConsumer; namespace tooling { namespace dependencies { class DependencyScanningService; +class DependencyScanningWorker; + class DependencyConsumer; class DependencyActionController; class DependencyScanningWorkerFilesystem; @@ -35,8 +37,7 @@ public: IntrusiveRefCntPtr<DependencyScanningWorkerFilesystem> DepFS, std::optional<StringRef> ModuleName = std::nullopt) : Service(Service), WorkingDirectory(WorkingDirectory), - Consumer(Consumer), Controller(Controller), DepFS(std::move(DepFS)), - ModuleName(ModuleName) {} + Consumer(Consumer), Controller(Controller), DepFS(std::move(DepFS)) {} bool runInvocation(std::unique_ptr<CompilerInvocation> Invocation, IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS, std::shared_ptr<PCHContainerOperations> PCHContainerOps, @@ -66,7 +67,6 @@ private: DependencyConsumer &Consumer; DependencyActionController &Controller; IntrusiveRefCntPtr<DependencyScanningWorkerFilesystem> DepFS; - std::optional<StringRef> ModuleName; std::optional<CompilerInstance> ScanInstanceStorage; std::shared_ptr<ModuleDepCollector> MDC; std::vector<std::string> LastCC1Arguments; @@ -118,7 +118,8 @@ initVFSForTUBuferScanning(IntrusiveRefCntPtr<llvm::vfs::FileSystem> BaseFS, StringRef WorkingDirectory, llvm::MemoryBufferRef TUBuffer); -std::pair<IntrusiveRefCntPtr<llvm::vfs::FileSystem>, std::vector<std::string>> +std::pair<IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem>, + std::vector<std::string>> initVFSForByNameScanning(IntrusiveRefCntPtr<llvm::vfs::FileSystem> BaseFS, ArrayRef<std::string> CommandLine, StringRef WorkingDirectory, StringRef ModuleName); @@ -137,7 +138,7 @@ computePrebuiltModulesASTMap(CompilerInstance &ScanInstance, SmallVector<StringRef> &StableDirs); std::unique_ptr<DependencyOutputOptions> -takeDependencyOutputOptionsFrom(CompilerInstance &ScanInstance); +takeAndUpdateDependencyOutputOptionsFrom(CompilerInstance &ScanInstance); /// Create the dependency collector that will collect the produced /// dependencies. May return the created ModuleDepCollector depending @@ -150,6 +151,60 @@ std::shared_ptr<ModuleDepCollector> initializeScanInstanceDependencyCollector( DependencyActionController &Controller, PrebuiltModulesAttrsMap PrebuiltModulesASTMap, llvm::SmallVector<StringRef> &StableDirs); + +class CompilerInstanceWithContext { + // Context + DependencyScanningWorker &Worker; + llvm::StringRef CWD; + std::vector<std::string> CommandLine; + + // Context - file systems + llvm::IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem> OverlayFS; + + // Context - Diagnostics engine. + std::unique_ptr<TextDiagnosticsPrinterWithOutput> DiagPrinterWithOS; + // DiagConsumer may points to DiagPrinterWithOS->DiagPrinter, or a custom + // DiagnosticConsumer passed in from initialize. + DiagnosticConsumer *DiagConsumer = nullptr; + std::unique_ptr<DignosticsEngineWithDiagOpts> DiagEngineWithCmdAndOpts; + + // Context - compiler invocation + // Compilation's command's arguments may be owned by Alloc when expanded from + // response files, so we need to keep Alloc alive in the context. + llvm::BumpPtrAllocator Alloc; + std::unique_ptr<clang::driver::Driver> Driver; + std::unique_ptr<clang::driver::Compilation> Compilation; + std::unique_ptr<CompilerInvocation> OriginalInvocation; + + // Context - output options + std::unique_ptr<DependencyOutputOptions> OutputOpts; + + // Context - stable directory handling + llvm::SmallVector<StringRef> StableDirs; + PrebuiltModulesAttrsMap PrebuiltModuleASTMap; + + // Compiler Instance + std::unique_ptr<CompilerInstance> CIPtr; + + // Source location offset. + int32_t SrcLocOffset = 0; + +public: + CompilerInstanceWithContext(DependencyScanningWorker &Worker, StringRef CWD, + const std::vector<std::string> &CMD) + : Worker(Worker), CWD(CWD), CommandLine(CMD) {}; + + // The three methods below returns false when they fail, with the detail + // accumulated in DiagConsumer. + bool initialize(DiagnosticConsumer *DC); + bool computeDependencies(StringRef ModuleName, DependencyConsumer &Consumer, + DependencyActionController &Controller); + bool finalize(); + + // The method below turns the return status from the above methods + // into an llvm::Error using a default DiagnosticConsumer. + llvm::Error handleReturnStatus(bool Success); +}; } // namespace dependencies } // namespace tooling } // namespace clang diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScanningTool.cpp b/clang/lib/Tooling/DependencyScanning/DependencyScanningTool.cpp index 27734ff..a1f2db7 100644 --- a/clang/lib/Tooling/DependencyScanning/DependencyScanningTool.cpp +++ b/clang/lib/Tooling/DependencyScanning/DependencyScanningTool.cpp @@ -162,13 +162,45 @@ DependencyScanningTool::getModuleDependencies( LookupModuleOutputCallback LookupModuleOutput) { FullDependencyConsumer Consumer(AlreadySeen); CallbackActionController Controller(LookupModuleOutput); - llvm::Error Result = Worker.computeDependencies(CWD, CommandLine, Consumer, - Controller, ModuleName); + if (auto Error = + Worker.initializeCompilerInstanceWithContextOrError(CWD, CommandLine)) + return std::move(Error); + + auto Result = Worker.computeDependenciesByNameWithContextOrError( + ModuleName, Consumer, Controller); + + if (auto Error = Worker.finalizeCompilerInstanceWithContextOrError()) + return std::move(Error); + if (Result) return std::move(Result); + return Consumer.takeTranslationUnitDeps(); } +llvm::Error DependencyScanningTool::initializeCompilerInstanceWithContext( + StringRef CWD, const std::vector<std::string> &CommandLine) { + return Worker.initializeCompilerInstanceWithContextOrError(CWD, CommandLine); +} + +llvm::Expected<TranslationUnitDeps> +DependencyScanningTool::computeDependenciesByNameWithContext( + StringRef ModuleName, const llvm::DenseSet<ModuleID> &AlreadySeen, + LookupModuleOutputCallback LookupModuleOutput) { + FullDependencyConsumer Consumer(AlreadySeen); + CallbackActionController Controller(LookupModuleOutput); + llvm::Error Result = Worker.computeDependenciesByNameWithContextOrError( + ModuleName, Consumer, Controller); + if (Result) + return std::move(Result); + + return Consumer.takeTranslationUnitDeps(); +} + +llvm::Error DependencyScanningTool::finalizeCompilerInstanceWithContext() { + return Worker.finalizeCompilerInstanceWithContextOrError(); +} + TranslationUnitDeps FullDependencyConsumer::takeTranslationUnitDeps() { TranslationUnitDeps TU; diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp b/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp index 0a1cf6b..dc408b1 100644 --- a/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp +++ b/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp @@ -43,6 +43,9 @@ DependencyScanningWorker::DependencyScanningWorker( } } +DependencyScanningWorker::~DependencyScanningWorker() = default; +DependencyActionController::~DependencyActionController() = default; + llvm::Error DependencyScanningWorker::computeDependencies( StringRef WorkingDirectory, const std::vector<std::string> &CommandLine, DependencyConsumer &Consumer, DependencyActionController &Controller, @@ -58,21 +61,6 @@ llvm::Error DependencyScanningWorker::computeDependencies( DiagPrinterWithOS.DiagnosticsOS.str(), llvm::inconvertibleErrorCode()); } -llvm::Error DependencyScanningWorker::computeDependencies( - StringRef WorkingDirectory, const std::vector<std::string> &CommandLine, - DependencyConsumer &Consumer, DependencyActionController &Controller, - StringRef ModuleName) { - // Capture the emitted diagnostics and report them to the client - // in the case of a failure. - TextDiagnosticsPrinterWithOutput DiagPrinterWithOS(CommandLine); - - if (computeDependencies(WorkingDirectory, CommandLine, Consumer, Controller, - DiagPrinterWithOS.DiagPrinter, ModuleName)) - return llvm::Error::success(); - return llvm::make_error<llvm::StringError>( - DiagPrinterWithOS.DiagnosticsOS.str(), llvm::inconvertibleErrorCode()); -} - static bool forEachDriverJob( ArrayRef<std::string> ArgStrs, DiagnosticsEngine &Diags, IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS, @@ -113,11 +101,11 @@ static bool createAndRunToolInvocation( bool DependencyScanningWorker::scanDependencies( StringRef WorkingDirectory, const std::vector<std::string> &CommandLine, DependencyConsumer &Consumer, DependencyActionController &Controller, - DiagnosticConsumer &DC, llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS, - std::optional<StringRef> ModuleName) { + DiagnosticConsumer &DC, + llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS) { DignosticsEngineWithDiagOpts DiagEngineWithCmdAndOpts(CommandLine, FS, DC); DependencyScanningAction Action(Service, WorkingDirectory, Consumer, - Controller, DepFS, ModuleName); + Controller, DepFS); bool Success = false; if (CommandLine[1] == "-cc1") { @@ -172,24 +160,51 @@ bool DependencyScanningWorker::computeDependencies( auto [FinalFS, FinalCommandLine] = initVFSForTUBuferScanning( BaseFS, CommandLine, WorkingDirectory, *TUBuffer); return scanDependencies(WorkingDirectory, FinalCommandLine, Consumer, - Controller, DC, FinalFS, - /*ModuleName=*/std::nullopt); + Controller, DC, FinalFS); } else { BaseFS->setCurrentWorkingDirectory(WorkingDirectory); return scanDependencies(WorkingDirectory, CommandLine, Consumer, Controller, - DC, BaseFS, /*ModuleName=*/std::nullopt); + DC, BaseFS); } } -bool DependencyScanningWorker::computeDependencies( - StringRef WorkingDirectory, const std::vector<std::string> &CommandLine, - DependencyConsumer &Consumer, DependencyActionController &Controller, - DiagnosticConsumer &DC, StringRef ModuleName) { - auto [OverlayFS, ModifiedCommandLine] = initVFSForByNameScanning( - BaseFS, CommandLine, WorkingDirectory, ModuleName); +llvm::Error +DependencyScanningWorker::initializeCompilerInstanceWithContextOrError( + StringRef CWD, const std::vector<std::string> &CommandLine) { + bool Success = initializeCompilerInstanceWithContext(CWD, CommandLine); + return CIWithContext->handleReturnStatus(Success); +} + +llvm::Error +DependencyScanningWorker::computeDependenciesByNameWithContextOrError( + StringRef ModuleName, DependencyConsumer &Consumer, + DependencyActionController &Controller) { + bool Success = + computeDependenciesByNameWithContext(ModuleName, Consumer, Controller); + return CIWithContext->handleReturnStatus(Success); +} + +llvm::Error +DependencyScanningWorker::finalizeCompilerInstanceWithContextOrError() { + bool Success = finalizeCompilerInstance(); + return CIWithContext->handleReturnStatus(Success); +} - return scanDependencies(WorkingDirectory, ModifiedCommandLine, Consumer, - Controller, DC, OverlayFS, ModuleName); +bool DependencyScanningWorker::initializeCompilerInstanceWithContext( + StringRef CWD, const std::vector<std::string> &CommandLine, + DiagnosticConsumer *DC) { + CIWithContext = + std::make_unique<CompilerInstanceWithContext>(*this, CWD, CommandLine); + return CIWithContext->initialize(DC); } -DependencyActionController::~DependencyActionController() {} +bool DependencyScanningWorker::computeDependenciesByNameWithContext( + StringRef ModuleName, DependencyConsumer &Consumer, + DependencyActionController &Controller) { + assert(CIWithContext && "CompilerInstance with context required!"); + return CIWithContext->computeDependencies(ModuleName, Consumer, Controller); +} + +bool DependencyScanningWorker::finalizeCompilerInstance() { + return CIWithContext->finalize(); +} diff --git a/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp b/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp index a117bec..e07a208 100644 --- a/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp +++ b/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp @@ -965,7 +965,9 @@ ModuleDepCollector::ModuleDepCollector( makeCommonInvocationForModuleBuild(std::move(OriginalCI))) {} void ModuleDepCollector::attachToPreprocessor(Preprocessor &PP) { - PP.addPPCallbacks(std::make_unique<ModuleDepCollectorPP>(*this)); + auto CollectorPP = std::make_unique<ModuleDepCollectorPP>(*this); + CollectorPPPtr = CollectorPP.get(); + PP.addPPCallbacks(std::move(CollectorPP)); } void ModuleDepCollector::attachToASTReader(ASTReader &R) {} diff --git a/clang/test/CIR/CodeGen/statement-exprs.c b/clang/test/CIR/CodeGen/statement-exprs.c index c784ec9..2d05fc1 100644 --- a/clang/test/CIR/CodeGen/statement-exprs.c +++ b/clang/test/CIR/CodeGen/statement-exprs.c @@ -6,7 +6,7 @@ // RUN: FileCheck --input-file=%t.ll %s --check-prefix=OGCG int f19(void) { - return ({ 3;;4;; }); + return ({ 3;;4; }); } // CIR: cir.func dso_local @f19() -> !s32i @@ -42,6 +42,16 @@ int f19(void) { // OGCG: %[[TMP_VAL:.+]] = load i32, ptr %[[TMP]] // OGCG: ret i32 %[[TMP_VAL]] +// PR166036: The trailing NullStmt should result in a void. +void f20(void) { + return ({ 3;;4;; }); +} + +// CIR-LABEL: cir.func dso_local @f20() {{[^-]*}} +// CIR: cir.return {{[^%]*}} + +// LLVM-LABEL: define{{.*}} void @f20 +// LLVM: ret void int nested(void) { ({123;}); diff --git a/clang/test/ClangScanDeps/link-libraries.c b/clang/test/ClangScanDeps/link-libraries.c index cc2e223..3719d71 100644 --- a/clang/test/ClangScanDeps/link-libraries.c +++ b/clang/test/ClangScanDeps/link-libraries.c @@ -32,7 +32,7 @@ module transitive { }] // RUN: sed "s|DIR|%/t|g" %t/cdb.json.template > %t/cdb.json -// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-name=root > %t/result.json +// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-names=root > %t/result.json // RUN: cat %t/result.json | sed 's:\\\\\?:/:g' | FileCheck -DPREFIX=%/t %s // CHECK: { diff --git a/clang/test/ClangScanDeps/modules-full-by-mod-name.c b/clang/test/ClangScanDeps/modules-full-by-mod-name.c index c838614..edb9963 100644 --- a/clang/test/ClangScanDeps/modules-full-by-mod-name.c +++ b/clang/test/ClangScanDeps/modules-full-by-mod-name.c @@ -25,7 +25,7 @@ module transitive { header "transitive.h" } }] // RUN: sed "s|DIR|%/t|g" %t/cdb.json.template > %t/cdb.json -// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-name=root > %t/result.json +// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-names=root > %t/result.json // RUN: cat %t/result.json | sed 's:\\\\\?:/:g' | FileCheck -DPREFIX=%/t %s // CHECK: { diff --git a/clang/test/ClangScanDeps/modules-full-by-mult-mod-names.c b/clang/test/ClangScanDeps/modules-full-by-mult-mod-names.c new file mode 100644 index 0000000..030f7f3 --- /dev/null +++ b/clang/test/ClangScanDeps/modules-full-by-mult-mod-names.c @@ -0,0 +1,108 @@ +// RUN: rm -rf %t +// RUN: split-file %s %t + +//--- module.modulemap +module root { header "root.h" } +module direct { header "direct.h" } +module transitive { header "transitive.h" } +module root1 { header "root1.h"} +//--- root.h +#include "direct.h" +#include "root/textual.h" + +//--- root1.h +#include "direct.h" + +//--- direct.h +#include "transitive.h" +//--- transitive.h +// empty + +//--- root/textual.h +// This is here to verify that the "root" directory doesn't clash with name of +// the "root" module. + +//--- cdb.json.template +[{ + "file": "", + "directory": "DIR", + "command": "clang -fmodules -fmodules-cache-path=DIR/cache -I DIR -x c" +}] + +// RUN: sed "s|DIR|%/t|g" %t/cdb.json.template > %t/cdb.json +// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-names=root,root1,direct > %t/result.json +// RUN: cat %t/result.json | sed 's:\\\\\?:/:g' | FileCheck -DPREFIX=%/t %s + +// CHECK: { +// CHECK-NEXT: "modules": [ +// CHECK-NEXT: { +// CHECK-NEXT: "clang-module-deps": [ +// CHECK-NEXT: { +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "module-name": "transitive" +// CHECK-NEXT: } +// CHECK-NEXT: ], +// CHECK-NEXT: "clang-modulemap-file": "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "command-line": [ +// CHECK: ], +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "file-deps": [ +// CHECK-NEXT: "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "[[PREFIX]]/direct.h" +// CHECK-NEXT: ], +// CHECK-NEXT: "link-libraries": [], +// CHECK-NEXT: "name": "direct" +// CHECK-NEXT: }, +// CHECK-NEXT: { +// CHECK-NEXT: "clang-module-deps": [ +// CHECK-NEXT: { +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "module-name": "direct" +// CHECK-NEXT: } +// CHECK-NEXT: ], +// CHECK-NEXT: "clang-modulemap-file": "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "command-line": [ +// CHECK: ], +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "file-deps": [ +// CHECK-NEXT: "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "[[PREFIX]]/root.h", +// CHECK-NEXT: "[[PREFIX]]/root/textual.h" +// CHECK-NEXT: ], +// CHECK-NEXT: "link-libraries": [], +// CHECK-NEXT: "name": "root" +// CHECK-NEXT: }, +// CHECK-NEXT: { +// CHECK-NEXT: "clang-module-deps": [ +// CHECK-NEXT: { +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "module-name": "direct" +// CHECK-NEXT: } +// CHECK-NEXT: ], +// CHECK-NEXT: "clang-modulemap-file": "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "command-line": [ +// CHECK: ], +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "file-deps": [ +// CHECK-NEXT: "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "[[PREFIX]]/root1.h" +// CHECK-NEXT: ], +// CHECK-NEXT: "link-libraries": [], +// CHECK-NEXT: "name": "root1" +// CHECK-NEXT: }, +// CHECK-NEXT: { +// CHECK-NEXT: "clang-module-deps": [], +// CHECK-NEXT: "clang-modulemap-file": "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "command-line": [ +// CHECK: ], +// CHECK-NEXT: "context-hash": "{{.*}}", +// CHECK-NEXT: "file-deps": [ +// CHECK-NEXT: "[[PREFIX]]/module.modulemap", +// CHECK-NEXT: "[[PREFIX]]/transitive.h" +// CHECK-NEXT: ], +// CHECK-NEXT: "link-libraries": [], +// CHECK-NEXT: "name": "transitive" +// CHECK-NEXT: } +// CHECK-NEXT: ], +// CHECK-NEXT: "translation-units": [] +// CHECK-NEXT: } diff --git a/clang/test/CodeGen/AArch64/neon-fcvt-intrinsics.c b/clang/test/CodeGen/AArch64/neon-fcvt-intrinsics.c index 670b650..929df94 100644 --- a/clang/test/CodeGen/AArch64/neon-fcvt-intrinsics.c +++ b/clang/test/CodeGen/AArch64/neon-fcvt-intrinsics.c @@ -26,16 +26,36 @@ int32_t test_vcvtas_s32_f32(float32_t a) { return (int32_t)vcvtas_s32_f32(a); } -// CHECK-LABEL: define {{[^@]+}}@test_test_vcvtad_s64_f64 +// CHECK-LABEL: define {{[^@]+}}@test_vcvtad_s64_f64 // CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: // CHECK-NEXT: [[VCVTAD_S64_F64_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtas.i64.f64(double [[A]]) // CHECK-NEXT: ret i64 [[VCVTAD_S64_F64_I]] // -int64_t test_test_vcvtad_s64_f64(float64_t a) { +int64_t test_vcvtad_s64_f64(float64_t a) { return (int64_t)vcvtad_s64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtas_s64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTAS_S64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtas.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTAS_S64_F32_I]] +// +int64_t test_vcvtas_s64_f32(float32_t a) { + return (int64_t)vcvtas_s64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtad_s32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTAD_S32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtas.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTAD_S32_F64_I]] +// +int32_t test_vcvtad_s32_f64(float64_t a) { + return (int32_t)vcvtad_s32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtas_u32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -56,6 +76,26 @@ uint64_t test_vcvtad_u64_f64(float64_t a) { return (uint64_t)vcvtad_u64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtas_u64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTAS_U64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtau.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTAS_U64_F32_I]] +// +uint64_t test_vcvtas_u64_f32(float32_t a) { + return (uint64_t)vcvtas_u64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtad_u32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTAD_U32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtau.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTAD_U32_F64_I]] +// +uint32_t test_vcvtad_u32_f64(float64_t a) { + return (uint32_t)vcvtad_u32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtms_s32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -76,6 +116,26 @@ int64_t test_vcvtmd_s64_f64(float64_t a) { return (int64_t)vcvtmd_s64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtms_s64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTMS_S64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtms.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTMS_S64_F32_I]] +// +int64_t test_vcvtms_s64_f32(float32_t a) { + return (int64_t)vcvtms_s64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtmd_s32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTMD_S32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtms.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTMD_S32_F64_I]] +// +int32_t test_vcvtmd_s32_f64(float64_t a) { + return (int32_t)vcvtmd_s32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtms_u32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -96,6 +156,26 @@ uint64_t test_vcvtmd_u64_f64(float64_t a) { return (uint64_t)vcvtmd_u64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtms_u64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTMS_U64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtmu.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTMS_U64_F32_I]] +// +uint64_t test_vcvtms_u64_f32(float32_t a) { + return (uint64_t)vcvtms_u64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtmd_u32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTMD_U32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtmu.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTMD_U32_F64_I]] +// +uint32_t test_vcvtmd_u32_f64(float64_t a) { + return (uint32_t)vcvtmd_u32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtns_s32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -116,6 +196,26 @@ int64_t test_vcvtnd_s64_f64(float64_t a) { return (int64_t)vcvtnd_s64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtns_s64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTNS_S64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtns.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTNS_S64_F32_I]] +// +int64_t test_vcvtns_s64_f32(float32_t a) { + return (int64_t)vcvtns_s64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtnd_s32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTND_S32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtns.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTND_S32_F64_I]] +// +int32_t test_vcvtnd_s32_f64(float64_t a) { + return (int32_t)vcvtnd_s32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtns_u32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -136,6 +236,26 @@ uint64_t test_vcvtnd_u64_f64(float64_t a) { return (uint64_t)vcvtnd_u64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtns_u64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTNS_U64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtnu.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTNS_U64_F32_I]] +// +uint64_t test_vcvtns_u64_f32(float32_t a) { + return (uint64_t)vcvtns_u64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtnd_u32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTND_U32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtnu.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTND_U32_F64_I]] +// +uint32_t test_vcvtnd_u32_f64(float64_t a) { + return (uint32_t)vcvtnd_u32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtps_s32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -156,6 +276,26 @@ int64_t test_vcvtpd_s64_f64(float64_t a) { return (int64_t)vcvtpd_s64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtps_s64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTPS_S64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtps.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTPS_S64_F32_I]] +// +int64_t test_vcvtps_s64_f32(float32_t a) { + return (int64_t)vcvtps_s64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtpd_s32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTPD_S32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtps.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTPD_S32_F64_I]] +// +int32_t test_vcvtpd_s32_f64(float64_t a) { + return (int32_t)vcvtpd_s32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvtps_u32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -176,6 +316,26 @@ uint64_t test_vcvtpd_u64_f64(float64_t a) { return (uint64_t)vcvtpd_u64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvtps_u64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTPS_U64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtpu.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTPS_U64_F32_I]] +// +uint64_t test_vcvtps_u64_f32(float32_t a) { + return (uint64_t)vcvtps_u64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtpd_u32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTPD_U32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtpu.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTPD_U32_F64_I]] +// +uint32_t test_vcvtpd_u32_f64(float64_t a) { + return (uint32_t)vcvtpd_u32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvts_s32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -196,6 +356,26 @@ int64_t test_vcvtd_s64_f64(float64_t a) { return (int64_t)vcvtd_s64_f64(a); } +// CHECK-LABEL: define {{[^@]+}}@test_vcvts_s64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTS_S64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtzs.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTS_S64_F32_I]] +// +int64_t test_vcvts_s64_f32(float32_t a) { + return (int64_t)vcvts_s64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtd_s32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTD_S32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtzs.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTD_S32_F64_I]] +// +int32_t test_vcvtd_s32_f64(float64_t a) { + return (int32_t)vcvtd_s32_f64(a); +} + // CHECK-LABEL: define {{[^@]+}}@test_vcvts_u32_f32 // CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { // CHECK-NEXT: entry: @@ -215,3 +395,24 @@ uint32_t test_vcvts_u32_f32(float32_t a) { uint64_t test_vcvtd_u64_f64(float64_t a) { return (uint64_t)vcvtd_u64_f64(a); } + +// CHECK-LABEL: define {{[^@]+}}@test_vcvts_u64_f32 +// CHECK-SAME: (float noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTS_U64_F32_I:%.*]] = call i64 @llvm.aarch64.neon.fcvtzu.i64.f32(float [[A]]) +// CHECK-NEXT: ret i64 [[VCVTS_U64_F32_I]] +// +uint64_t test_vcvts_u64_f32(float32_t a) { + return (uint64_t)vcvts_u64_f32(a); +} + +// CHECK-LABEL: define {{[^@]+}}@test_vcvtd_u32_f64 +// CHECK-SAME: (double noundef [[A:%.*]]) #[[ATTR0]] { +// CHECK-NEXT: entry: +// CHECK-NEXT: [[VCVTD_U32_F64_I:%.*]] = call i32 @llvm.aarch64.neon.fcvtzu.i32.f64(double [[A]]) +// CHECK-NEXT: ret i32 [[VCVTD_U32_F64_I]] +// +uint32_t test_vcvtd_u32_f64(float64_t a) { + return (uint32_t)vcvtd_u32_f64(a); +} + diff --git a/clang/test/CodeGen/AArch64/v9.6a-neon-f16-intrinsics.c b/clang/test/CodeGen/AArch64/v9.6a-neon-f16-intrinsics.c new file mode 100644 index 0000000..89ee9e3 --- /dev/null +++ b/clang/test/CodeGen/AArch64/v9.6a-neon-f16-intrinsics.c @@ -0,0 +1,23 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5 +// RUN: %clang_cc1 -triple arm64-none-linux-gnu -target-feature +neon -target-feature +v9.6a -target-feature +f8f16mm -target-feature +fp8 \ +// RUN: -disable-O0-optnone -emit-llvm -o - %s \ +// RUN: | opt -S -passes=mem2reg,sroa \ +// RUN: | FileCheck %s + +// REQUIRES: aarch64-registered-target + +#include <arm_neon.h> + +// CHECK-LABEL: define dso_local <8 x half> @test_vmmlaq_f16_mf8( +// CHECK-SAME: <8 x half> noundef [[P0:%.*]], <16 x i8> [[P1:%.*]], <16 x i8> [[P2:%.*]], i64 noundef [[P3:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-NEXT: [[ENTRY:.*:]] +// CHECK-NEXT: [[TMP0:%.*]] = bitcast <8 x half> [[P0]] to <8 x i16> +// CHECK-NEXT: [[TMP1:%.*]] = bitcast <8 x i16> [[TMP0]] to <16 x i8> +// CHECK-NEXT: call void @llvm.aarch64.set.fpmr(i64 [[P3]]) +// CHECK-NEXT: [[FMMLA_I:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x half> +// CHECK-NEXT: [[FMMLA1_I:%.*]] = call <8 x half> @llvm.aarch64.neon.fmmla.v8f16.v16i8(<8 x half> [[FMMLA_I]], <16 x i8> [[P1]], <16 x i8> [[P2]]) +// CHECK-NEXT: ret <8 x half> [[FMMLA1_I]] +// +float16x8_t test_vmmlaq_f16_mf8(float16x8_t p0, mfloat8x16_t p1, mfloat8x16_t p2, fpm_t p3) { + return vmmlaq_f16_mf8_fpm(p0, p1, p2, p3); +} diff --git a/clang/test/CodeGen/AArch64/v9.6a-neon-f32-intrinsics.c b/clang/test/CodeGen/AArch64/v9.6a-neon-f32-intrinsics.c new file mode 100644 index 0000000..13db72c --- /dev/null +++ b/clang/test/CodeGen/AArch64/v9.6a-neon-f32-intrinsics.c @@ -0,0 +1,21 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5 +// RUN: %clang_cc1 -triple arm64-none-linux-gnu -target-feature +neon -target-feature +v9.6a -target-feature +f8f32mm -target-feature +fp8 \ +// RUN: -disable-O0-optnone -emit-llvm -o - %s \ +// RUN: | opt -S -passes=mem2reg,sroa \ +// RUN: | FileCheck %s + +// REQUIRES: aarch64-registered-target + +#include <arm_neon.h> + +// CHECK-LABEL: define dso_local <4 x float> @test_vmmlaq_f32_mf8( +// CHECK-SAME: <4 x float> noundef [[P0:%.*]], <16 x i8> [[P1:%.*]], <16 x i8> [[P2:%.*]], i64 noundef [[P3:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-NEXT: [[ENTRY:.*:]] +// CHECK-NEXT: call void @llvm.aarch64.set.fpmr(i64 [[P3]]) +// CHECK-NEXT: [[FMMLA_I:%.*]] = call <4 x float> @llvm.aarch64.neon.fmmla.v4f32.v16i8(<4 x float> [[P0]], <16 x i8> [[P1]], <16 x i8> [[P2]]) +// CHECK-NEXT: ret <4 x float> [[FMMLA_I]] +// +float32x4_t test_vmmlaq_f32_mf8(float32x4_t p0, mfloat8x16_t p1, mfloat8x16_t p2, fpm_t p3) { + return vmmlaq_f32_mf8_fpm(p0, p1, p2, p3); +} + diff --git a/clang/test/Modules/transitive-system.test b/clang/test/Modules/transitive-system.test index b1f1558..5f6196c 100644 --- a/clang/test/Modules/transitive-system.test +++ b/clang/test/Modules/transitive-system.test @@ -2,9 +2,9 @@ // RUN: split-file %s %t // RUN: sed "s|DIR|%/t|g" %t/cdb.json.template > %t/cdb.json -// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-name=direct > %t/result1.json +// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-names=direct > %t/result1.json // RUN: rm -rf %t/cache -// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-name=transitive > %t/result2.json +// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-names=transitive > %t/result2.json // RUN: %deps-to-rsp %t/result1.json --module-name transitive > %t/1.rsp // RUN: %deps-to-rsp %t/result2.json --module-name transitive > %t/2.rsp // RUN: diff %t/1.rsp %t/2.rsp diff --git a/clang/test/OpenMP/target_dyn_groupprivate_messages.cpp b/clang/test/OpenMP/target_dyn_groupprivate_messages.cpp new file mode 100644 index 0000000..385bd5e --- /dev/null +++ b/clang/test/OpenMP/target_dyn_groupprivate_messages.cpp @@ -0,0 +1,89 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=61 %s -Wuninitialized +// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=61 %s -Wuninitialized + +void foo() { +} + +bool foobool(int argc) { + return argc; +} + +struct S1; // expected-note {{declared here}} + +template <class T, class S> // expected-note {{declared here}} +int tmain(T argc, S **argv) { + T z; + #pragma omp target dyn_groupprivate // expected-error {{expected '(' after 'dyn_groupprivate'}} + foo(); + #pragma omp target dyn_groupprivate ( // expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate () // expected-error {{expected expression}} + foo(); + #pragma omp target dyn_groupprivate (argc // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate (argc)) // expected-warning {{extra tokens at the end of '#pragma omp target' are ignored}} + foo(); + #pragma omp target dyn_groupprivate (argc > 0 ? argv[1] : argv[2]) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} + foo(); + #pragma omp target dyn_groupprivate (foobool(argc)), dyn_groupprivate (true) // expected-error {{directive '#pragma omp target' cannot contain more than one 'dyn_groupprivate' clause}} + foo(); + #pragma omp target dyn_groupprivate (S) // expected-error {{'S' does not refer to a value}} + foo(); + #pragma omp target dyn_groupprivate (argv[1]=2) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate (argc argc) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(argc+z) + foo(); + return 0; +} + +int main(int argc, char **argv) { +constexpr int n = -1; +int z; + #pragma omp target dyn_groupprivate // expected-error {{expected '(' after 'dyn_groupprivate'}} + foo(); + #pragma omp target dyn_groupprivate ( // expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate () // expected-error {{expected expression}} + foo(); + #pragma omp target dyn_groupprivate (argc // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate (argc)) // expected-warning {{extra tokens at the end of '#pragma omp target' are ignored}} + foo(); + #pragma omp target dyn_groupprivate (argc > 0 ? argv[1] : argv[2]) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} + foo(); + #pragma omp target dyn_groupprivate (foobool(argc)), dyn_groupprivate (true) // expected-error {{directive '#pragma omp target' cannot contain more than one 'dyn_groupprivate' clause}} + foo(); + #pragma omp target dyn_groupprivate (S1) // expected-error {{'S1' does not refer to a value}} + foo(); + #pragma omp target dyn_groupprivate (argv[1]=2) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate (argc argc) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate (1 0) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(dyn_groupprivate(tmain(argc, argv) // expected-error2 {{expected ')'}} expected-note2 {{to match this '('}} expected-note {{in instantiation of function template specialization 'tmain<int, char>' requested here}} + foo(); + #pragma omp target dyn_groupprivate(-1) // expected-error {{argument to 'dyn_groupprivate' clause must be a non-negative integer value}} + foo(); + #pragma omp target dyn_groupprivate(cgrou) // expected-error {{use of undeclared identifier 'cgrou'}} + foo(); + #pragma omp target dyn_groupprivate(cgrou: argc) // expected-error {{use of undeclared identifier 'cgrou'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(cgroup,cgroup: argc) // expected-error {{modifier 'cgroup' cannot be used along with modifier 'cgroup' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(default_mem),fallback(abort): argc) // expected-error {{modifier 'fallback(abort)' cannot be used along with modifier 'fallback(default_mem)' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(abort),fallback(null): argc) // expected-error {{modifier 'fallback(null)' cannot be used along with modifier 'fallback(abort)' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(cgroup): argc) // expected-error {{expected 'abort', 'null' or 'default_mem' in fallback modifier}} expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(fallback(): argc) // expected-error {{expected 'abort', 'null' or 'default_mem' in fallback modifier}} expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(: argc) // expected-error {{expected ')'}} expected-error {{expected expression}} expected-note {{to match this '('}} + foo(); + + return tmain(argc, argv); +} + diff --git a/clang/test/OpenMP/target_teams_dyn_groupprivate_messages.cpp b/clang/test/OpenMP/target_teams_dyn_groupprivate_messages.cpp new file mode 100644 index 0000000..ac2cc0d --- /dev/null +++ b/clang/test/OpenMP/target_teams_dyn_groupprivate_messages.cpp @@ -0,0 +1,89 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=61 %s -Wuninitialized +// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=61 %s -Wuninitialized + +void foo() { +} + +bool foobool(int argc) { + return argc; +} + +struct S1; // expected-note {{declared here}} + +template <class T, class S> // expected-note {{declared here}} +int tmain(T argc, S **argv) { + T z; + #pragma omp target teams dyn_groupprivate // expected-error {{expected '(' after 'dyn_groupprivate'}} + foo(); + #pragma omp target teams dyn_groupprivate ( // expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate () // expected-error {{expected expression}} + foo(); + #pragma omp target teams dyn_groupprivate (argc // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate (argc)) // expected-warning {{extra tokens at the end of '#pragma omp target teams' are ignored}} + foo(); + #pragma omp target teams dyn_groupprivate (argc > 0 ? argv[1] : argv[2]) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} + foo(); + #pragma omp target teams dyn_groupprivate (foobool(argc)), dyn_groupprivate (true) // expected-error {{directive '#pragma omp target teams' cannot contain more than one 'dyn_groupprivate' clause}} + foo(); + #pragma omp target teams dyn_groupprivate (S) // expected-error {{'S' does not refer to a value}} + foo(); + #pragma omp target teams dyn_groupprivate (argv[1]=2) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate (argc argc) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate(argc+z) + foo(); + return 0; +} + +int main(int argc, char **argv) { +constexpr int n = -1; +int z; + #pragma omp target teams dyn_groupprivate // expected-error {{expected '(' after 'dyn_groupprivate'}} + foo(); + #pragma omp target teams dyn_groupprivate ( // expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate () // expected-error {{expected expression}} + foo(); + #pragma omp target teams dyn_groupprivate (argc // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate (argc)) // expected-warning {{extra tokens at the end of '#pragma omp target teams' are ignored}} + foo(); + #pragma omp target teams dyn_groupprivate (argc > 0 ? argv[1] : argv[2]) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} + foo(); + #pragma omp target teams dyn_groupprivate (foobool(argc)), dyn_groupprivate (true) // expected-error {{directive '#pragma omp target teams' cannot contain more than one 'dyn_groupprivate' clause}} + foo(); + #pragma omp target teams dyn_groupprivate (S1) // expected-error {{'S1' does not refer to a value}} + foo(); + #pragma omp target teams dyn_groupprivate (argv[1]=2) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate (argc argc) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate (1 0) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate(dyn_groupprivate(tmain(argc, argv) // expected-error2 {{expected ')'}} expected-note2 {{to match this '('}} expected-note {{in instantiation of function template specialization 'tmain<int, char>' requested here}} + foo(); + #pragma omp target teams dyn_groupprivate(-1) // expected-error {{argument to 'dyn_groupprivate' clause must be a non-negative integer value}} + foo(); + #pragma omp target teams dyn_groupprivate(cgrou) // expected-error {{use of undeclared identifier 'cgrou'}} + foo(); + #pragma omp target teams dyn_groupprivate(cgrou: argc) // expected-error {{use of undeclared identifier 'cgrou'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate(cgroup,cgroup: argc) // expected-error {{modifier 'cgroup' cannot be used along with modifier 'cgroup' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(default_mem),fallback(abort): argc) // expected-error {{modifier 'fallback(abort)' cannot be used along with modifier 'fallback(default_mem)' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(abort),fallback(null): argc) // expected-error {{modifier 'fallback(null)' cannot be used along with modifier 'fallback(abort)' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(cgroup): argc) // expected-error {{expected 'abort', 'null' or 'default_mem' in fallback modifier}} expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(fallback(): argc) // expected-error {{expected 'abort', 'null' or 'default_mem' in fallback modifier}} expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target teams dyn_groupprivate(: argc) // expected-error {{expected ')'}} expected-error {{expected expression}} expected-note {{to match this '('}} + foo(); + + return tmain(argc, argv); +} + diff --git a/clang/test/OpenMP/teams_dyn_groupprivate_messages.cpp b/clang/test/OpenMP/teams_dyn_groupprivate_messages.cpp new file mode 100644 index 0000000..701ebfb --- /dev/null +++ b/clang/test/OpenMP/teams_dyn_groupprivate_messages.cpp @@ -0,0 +1,89 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=61 %s -Wuninitialized +// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=61 %s -Wuninitialized + +void foo() { +} + +bool foobool(int argc) { + return argc; +} + +struct S1; // expected-note {{declared here}} + +template <class T, class S> // expected-note {{declared here}} +int tmain(T argc, S **argv) { + T z; + #pragma omp teams dyn_groupprivate // expected-error {{expected '(' after 'dyn_groupprivate'}} + foo(); + #pragma omp teams dyn_groupprivate ( // expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate () // expected-error {{expected expression}} + foo(); + #pragma omp teams dyn_groupprivate (argc // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate (argc)) // expected-warning {{extra tokens at the end of '#pragma omp teams' are ignored}} + foo(); + #pragma omp teams dyn_groupprivate (argc > 0 ? argv[1] : argv[2]) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} + foo(); + #pragma omp teams dyn_groupprivate (foobool(argc)), dyn_groupprivate (true) // expected-error {{directive '#pragma omp teams' cannot contain more than one 'dyn_groupprivate' clause}} + foo(); + #pragma omp teams dyn_groupprivate (S) // expected-error {{'S' does not refer to a value}} + foo(); + #pragma omp teams dyn_groupprivate (argv[1]=2) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate (argc argc) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate(argc+z) + foo(); + return 0; +} + +int main(int argc, char **argv) { +constexpr int n = -1; +int z; + #pragma omp teams dyn_groupprivate // expected-error {{expected '(' after 'dyn_groupprivate'}} + foo(); + #pragma omp teams dyn_groupprivate ( // expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate () // expected-error {{expected expression}} + foo(); + #pragma omp teams dyn_groupprivate (argc // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate (argc)) // expected-warning {{extra tokens at the end of '#pragma omp teams' are ignored}} + foo(); + #pragma omp teams dyn_groupprivate (argc > 0 ? argv[1] : argv[2]) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} + foo(); + #pragma omp teams dyn_groupprivate (foobool(argc)), dyn_groupprivate (true) // expected-error {{directive '#pragma omp teams' cannot contain more than one 'dyn_groupprivate' clause}} + foo(); + #pragma omp teams dyn_groupprivate (S1) // expected-error {{'S1' does not refer to a value}} + foo(); + #pragma omp teams dyn_groupprivate (argv[1]=2) // expected-error {{expression must have integral or unscoped enumeration type, not 'char *'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate (argc argc) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate (1 0) // expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate(dyn_groupprivate(tmain(argc, argv) // expected-error2 {{expected ')'}} expected-note2 {{to match this '('}} expected-note {{in instantiation of function template specialization 'tmain<int, char>' requested here}} + foo(); + #pragma omp teams dyn_groupprivate(-1) // expected-error {{argument to 'dyn_groupprivate' clause must be a non-negative integer value}} + foo(); + #pragma omp teams dyn_groupprivate(cgrou) // expected-error {{use of undeclared identifier 'cgrou'}} + foo(); + #pragma omp teams dyn_groupprivate(cgrou: argc) // expected-error {{use of undeclared identifier 'cgrou'}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate(cgroup,cgroup: argc) // expected-error {{modifier 'cgroup' cannot be used along with modifier 'cgroup' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(default_mem),fallback(abort): argc) // expected-error {{modifier 'fallback(abort)' cannot be used along with modifier 'fallback(default_mem)' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(abort),fallback(null): argc) // expected-error {{modifier 'fallback(null)' cannot be used along with modifier 'fallback(abort)' in dyn_groupprivate}} + foo(); + #pragma omp target dyn_groupprivate(fallback(cgroup): argc) // expected-error {{expected 'abort', 'null' or 'default_mem' in fallback modifier}} expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp target dyn_groupprivate(fallback(): argc) // expected-error {{expected 'abort', 'null' or 'default_mem' in fallback modifier}} expected-error {{expected expression}} expected-error {{expected ')'}} expected-note {{to match this '('}} + foo(); + #pragma omp teams dyn_groupprivate(: argc) // expected-error {{expected ')'}} expected-error {{expected expression}} expected-note {{to match this '('}} + foo(); + + return tmain(argc, argv); +} + diff --git a/clang/tools/clang-scan-deps/ClangScanDeps.cpp b/clang/tools/clang-scan-deps/ClangScanDeps.cpp index c11a348..5f5bf42 100644 --- a/clang/tools/clang-scan-deps/ClangScanDeps.cpp +++ b/clang/tools/clang-scan-deps/ClangScanDeps.cpp @@ -87,7 +87,7 @@ static std::string ModuleFilesDir; static bool EagerLoadModules; static unsigned NumThreads = 0; static std::string CompilationDB; -static std::optional<std::string> ModuleName; +static std::optional<std::string> ModuleNames; static std::vector<std::string> ModuleDepTargets; static std::string TranslationUnitFile; static bool DeprecatedDriverCommand; @@ -205,8 +205,8 @@ static void ParseArgs(int argc, char **argv) { if (const llvm::opt::Arg *A = Args.getLastArg(OPT_compilation_database_EQ)) CompilationDB = A->getValue(); - if (const llvm::opt::Arg *A = Args.getLastArg(OPT_module_name_EQ)) - ModuleName = A->getValue(); + if (const llvm::opt::Arg *A = Args.getLastArg(OPT_module_names_EQ)) + ModuleNames = A->getValue(); for (const llvm::opt::Arg *A : Args.filtered(OPT_dependency_target_EQ)) ModuleDepTargets.emplace_back(A->getValue()); @@ -664,6 +664,16 @@ static bool handleModuleResult(StringRef ModuleName, return false; } +static void handleErrorWithInfoString(StringRef Info, llvm::Error E, + SharedStream &OS, SharedStream &Errs) { + llvm::handleAllErrors(std::move(E), [&Info, &Errs](llvm::StringError &Err) { + Errs.applyLocked([&](raw_ostream &OS) { + OS << "Error: " << Info << ":\n"; + OS << Err.getMessage(); + }); + }); +} + class P1689Deps { public: void printDependencies(raw_ostream &OS) { @@ -1008,7 +1018,7 @@ int clang_scan_deps_main(int argc, char **argv, const llvm::ToolContext &) { }; if (Format == ScanningOutputFormat::Full) - FD.emplace(!ModuleName ? Inputs.size() : 0); + FD.emplace(!ModuleNames ? Inputs.size() : 0); std::atomic<size_t> NumStatusCalls = 0; std::atomic<size_t> NumOpenFileForReadCalls = 0; @@ -1082,13 +1092,48 @@ int clang_scan_deps_main(int argc, char **argv, const llvm::ToolContext &) { MakeformatOS, Errs)) HadErrors = true; } - } else if (ModuleName) { - auto MaybeModuleDepsGraph = WorkerTool.getModuleDependencies( - *ModuleName, Input->CommandLine, CWD, AlreadySeenModules, - LookupOutput); - if (handleModuleResult(*ModuleName, MaybeModuleDepsGraph, *FD, - LocalIndex, DependencyOS, Errs)) - HadErrors = true; + } else if (ModuleNames) { + StringRef ModuleNameRef(*ModuleNames); + SmallVector<StringRef> Names; + ModuleNameRef.split(Names, ','); + + if (Names.size() == 1) { + auto MaybeModuleDepsGraph = WorkerTool.getModuleDependencies( + Names[0], Input->CommandLine, CWD, AlreadySeenModules, + LookupOutput); + if (handleModuleResult(Names[0], MaybeModuleDepsGraph, *FD, + LocalIndex, DependencyOS, Errs)) + HadErrors = true; + } else { + if (llvm::Error Err = + WorkerTool.initializeCompilerInstanceWithContext( + CWD, Input->CommandLine)) { + handleErrorWithInfoString( + "Compiler instance with context setup error", std::move(Err), + DependencyOS, Errs); + HadErrors = true; + continue; + } + + for (auto N : Names) { + auto MaybeModuleDepsGraph = + WorkerTool.computeDependenciesByNameWithContext( + N, AlreadySeenModules, LookupOutput); + if (handleModuleResult(N, MaybeModuleDepsGraph, *FD, LocalIndex, + DependencyOS, Errs)) { + HadErrors = true; + break; + } + } + + if (llvm::Error Err = + WorkerTool.finalizeCompilerInstanceWithContext()) { + handleErrorWithInfoString( + "Compiler instance with context finialization error", + std::move(Err), DependencyOS, Errs); + HadErrors = true; + } + } } else { std::unique_ptr<llvm::MemoryBuffer> TU; std::optional<llvm::MemoryBufferRef> TUBuffer; diff --git a/clang/tools/clang-scan-deps/Opts.td b/clang/tools/clang-scan-deps/Opts.td index 7a63b18..6ea9d82 100644 --- a/clang/tools/clang-scan-deps/Opts.td +++ b/clang/tools/clang-scan-deps/Opts.td @@ -26,7 +26,9 @@ def eager_load_pcm : F<"eager-load-pcm", "Load PCM files eagerly (instead of laz def j : Arg<"j", "Number of worker threads to use (default: use all concurrent threads)">; defm compilation_database : Eq<"compilation-database", "Compilation database">; -defm module_name : Eq<"module-name", "the module of which the dependencies are to be computed">; +defm module_names + : Eq<"module-names", "A comma separated list of names of modules of which " + "the dependencies are to be computed">; defm dependency_target : Eq<"dependency-target", "The names of dependency targets for the dependency file">; defm tu_buffer_path: Eq<"tu-buffer-path", "The path to the translation unit for depscan. Not compatible with -module-name">; diff --git a/clang/tools/libclang/CIndex.cpp b/clang/tools/libclang/CIndex.cpp index 08776d9..f4d6fa7 100644 --- a/clang/tools/libclang/CIndex.cpp +++ b/clang/tools/libclang/CIndex.cpp @@ -2757,6 +2757,11 @@ void OMPClauseEnqueue::VisitOMPXDynCGroupMemClause( VisitOMPClauseWithPreInit(C); Visitor->AddStmt(C->getSize()); } +void OMPClauseEnqueue::VisitOMPDynGroupprivateClause( + const OMPDynGroupprivateClause *C) { + VisitOMPClauseWithPreInit(C); + Visitor->AddStmt(C->getSize()); +} void OMPClauseEnqueue::VisitOMPDoacrossClause(const OMPDoacrossClause *C) { VisitOMPClauseList(C); } diff --git a/clang/unittests/Format/ConfigParseTest.cpp b/clang/unittests/Format/ConfigParseTest.cpp index 43b2117..d578fa7 100644 --- a/clang/unittests/Format/ConfigParseTest.cpp +++ b/clang/unittests/Format/ConfigParseTest.cpp @@ -589,20 +589,20 @@ TEST(ConfigParseTest, ParsesConfiguration) { CHECK_PARSE("AlignTrailingComments: Leave", AlignTrailingComments, FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Leave, 0})); + {FormatStyle::TCAS_Leave, 0, true})); CHECK_PARSE("AlignTrailingComments: Always", AlignTrailingComments, FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Always, 0})); + {FormatStyle::TCAS_Always, 0, true})); CHECK_PARSE("AlignTrailingComments: Never", AlignTrailingComments, FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Never, 0})); + {FormatStyle::TCAS_Never, 0, true})); // For backwards compatibility CHECK_PARSE("AlignTrailingComments: true", AlignTrailingComments, FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Always, 0})); + {FormatStyle::TCAS_Always, 0, true})); CHECK_PARSE("AlignTrailingComments: false", AlignTrailingComments, FormatStyle::TrailingCommentsAlignmentStyle( - {FormatStyle::TCAS_Never, 0})); + {FormatStyle::TCAS_Never, 0, true})); CHECK_PARSE_NESTED_VALUE("Kind: Always", AlignTrailingComments, Kind, FormatStyle::TCAS_Always); CHECK_PARSE_NESTED_VALUE("Kind: Never", AlignTrailingComments, Kind, @@ -611,6 +611,7 @@ TEST(ConfigParseTest, ParsesConfiguration) { FormatStyle::TCAS_Leave); CHECK_PARSE_NESTED_VALUE("OverEmptyLines: 1234", AlignTrailingComments, OverEmptyLines, 1234u); + CHECK_PARSE_NESTED_BOOL(AlignTrailingComments, AlignPPAndNotPP); Style.UseTab = FormatStyle::UT_ForIndentation; CHECK_PARSE("UseTab: Never", UseTab, FormatStyle::UT_Never); diff --git a/clang/unittests/Format/FormatTestComments.cpp b/clang/unittests/Format/FormatTestComments.cpp index d7b2257..684d301 100644 --- a/clang/unittests/Format/FormatTestComments.cpp +++ b/clang/unittests/Format/FormatTestComments.cpp @@ -3378,6 +3378,66 @@ TEST_F(FormatTestComments, DontAlignOverScope) { "int foobar; // group"); } +TEST_F(FormatTestComments, DontAlignOverPPDirective) { + auto Style = getLLVMStyle(); + Style.AlignTrailingComments.AlignPPAndNotPP = false; + + verifyFormat("int i; // Aligned\n" + "int long; // with this\n" + "#define FOO // only aligned\n" + "#define LOOONG // with other pp directives\n" + "int loooong; // new alignment", + "int i;//Aligned\n" + "int long;//with this\n" + "#define FOO //only aligned\n" + "#define LOOONG //with other pp directives\n" + "int loooong; //new alignment", + Style); + + verifyFormat("#define A // Comment\n" + "#define AB // Comment", + Style); + + Style.ColumnLimit = 30; + verifyNoChange("#define A // Comment\n" + " // Continued\n" + "int i = 0; // New Stuff\n" + " // Continued\n" + "#define Func(X) \\\n" + " X(); \\\n" + " X(); // Comment\n" + " // Continued\n" + "long loong = 1; // Dont align", + Style); + + verifyFormat("#define A // Comment that\n" + " // would wrap\n" + "#define FOO // For the\n" + " // alignment\n" + "#define B // Also\n" + " // aligned", + "#define A // Comment that would wrap\n" + "#define FOO // For the alignment\n" + "#define B // Also\n" + " // aligned", + Style); + + Style.AlignTrailingComments.OverEmptyLines = 1; + verifyNoChange("#define A // Comment\n" + "\n" + " // Continued\n" + "int i = 0; // New Stuff\n" + "\n" + " // Continued\n" + "#define Func(X) \\\n" + " X(); \\\n" + " X(); // Comment\n" + "\n" + " // Continued\n" + "long loong = 1; // Dont align", + Style); +} + TEST_F(FormatTestComments, AlignsBlockCommentDecorations) { verifyFormat("/*\n" " */", diff --git a/compiler-rt/test/asan/TestCases/log-path_test.cpp b/compiler-rt/test/asan/TestCases/log-path_test.cpp index 6875d57..22f077f 100644 --- a/compiler-rt/test/asan/TestCases/log-path_test.cpp +++ b/compiler-rt/test/asan/TestCases/log-path_test.cpp @@ -1,6 +1,5 @@ // FIXME: https://code.google.com/p/address-sanitizer/issues/detail?id=316 -// XFAIL: android -// UNSUPPORTED: ios +// UNSUPPORTED: ios, android // // The for loop in the backticks below requires bash. // REQUIRES: shell diff --git a/compiler-rt/test/asan/TestCases/verbose-log-path_test.cpp b/compiler-rt/test/asan/TestCases/verbose-log-path_test.cpp index 53166cc..f4781a7 100644 --- a/compiler-rt/test/asan/TestCases/verbose-log-path_test.cpp +++ b/compiler-rt/test/asan/TestCases/verbose-log-path_test.cpp @@ -9,8 +9,8 @@ // RUN: FileCheck %s --check-prefix=CHECK-ERROR < %t-dir/asan.log.verbose-log-path_test-binary.* // FIXME: only FreeBSD, NetBSD and Linux have verbose log paths now. -// XFAIL: target={{.*windows-msvc.*}},android -// UNSUPPORTED: ios +// XFAIL: target={{.*windows-msvc.*}} +// UNSUPPORTED: ios, android #include <stdlib.h> #include <string.h> diff --git a/compiler-rt/test/hwasan/TestCases/Linux/fixed-shadow.c b/compiler-rt/test/hwasan/TestCases/Linux/fixed-shadow.c index fc83b21..c950e06 100644 --- a/compiler-rt/test/hwasan/TestCases/Linux/fixed-shadow.c +++ b/compiler-rt/test/hwasan/TestCases/Linux/fixed-shadow.c @@ -3,17 +3,17 @@ // Default compiler instrumentation works with any shadow base (dynamic or fixed). // RUN: %clang_hwasan %s -o %t // RUN: %run %t -// RUN: env HWASAN_OPTIONS=fixed_shadow_base=263878495698944 %run %t 2>%t.out || (cat %t.out | FileCheck %s) +// RUN: env HWASAN_OPTIONS=fixed_shadow_base=17592186044416 %run %t // RUN: env HWASAN_OPTIONS=fixed_shadow_base=4398046511104 %run %t // // If -hwasan-mapping-offset is set, then the fixed_shadow_base needs to match. -// RUN: %clang_hwasan %s -mllvm -hwasan-mapping-offset=263878495698944 -o %t -// RUN: env HWASAN_OPTIONS=fixed_shadow_base=263878495698944 %run %t 2>%t.out || (cat %t.out | FileCheck %s) +// RUN: %clang_hwasan %s -mllvm -hwasan-mapping-offset=17592186044416 -o %t +// RUN: env HWASAN_OPTIONS=fixed_shadow_base=17592186044416 %run %t // RUN: env HWASAN_OPTIONS=fixed_shadow_base=4398046511104 not %run %t // RUN: %clang_hwasan %s -mllvm -hwasan-mapping-offset=4398046511104 -o %t // RUN: env HWASAN_OPTIONS=fixed_shadow_base=4398046511104 %run %t -// RUN: env HWASAN_OPTIONS=fixed_shadow_base=263878495698944 not %run %t +// RUN: env HWASAN_OPTIONS=fixed_shadow_base=17592186044416 not %run %t // // Note: if fixed_shadow_base is not set, compiler-rt will dynamically choose a // shadow base, which has a tiny but non-zero probability of matching the @@ -26,8 +26,6 @@ // // UNSUPPORTED: android -// CHECK: FATAL: HWAddressSanitizer: Shadow range {{.*}} is not available - #include <assert.h> #include <sanitizer/allocator_interface.h> #include <sanitizer/hwasan_interface.h> diff --git a/compiler-rt/test/sanitizer_common/TestCases/Posix/sanitizer_set_report_fd_test.cpp b/compiler-rt/test/sanitizer_common/TestCases/Posix/sanitizer_set_report_fd_test.cpp index 6ba7025..e406482 100644 --- a/compiler-rt/test/sanitizer_common/TestCases/Posix/sanitizer_set_report_fd_test.cpp +++ b/compiler-rt/test/sanitizer_common/TestCases/Posix/sanitizer_set_report_fd_test.cpp @@ -7,7 +7,7 @@ // RUN: not %run %t %t-out && FileCheck < %t-out %s // REQUIRES: stable-runtime -// XFAIL: android && asan +// UNSUPPORTED: android && asan #include <sanitizer/common_interface_defs.h> #include <stdio.h> diff --git a/flang-rt/include/flang-rt/runtime/io-stmt.h b/flang-rt/include/flang-rt/runtime/io-stmt.h index f6a81f7..3c6bcfe 100644 --- a/flang-rt/include/flang-rt/runtime/io-stmt.h +++ b/flang-rt/include/flang-rt/runtime/io-stmt.h @@ -730,8 +730,7 @@ public: RT_API_ATTRS bool AdvanceRecord(int = 1); RT_API_ATTRS int EndIoStatement(); RT_API_ATTRS bool CanAdvance() { - return DIR == Direction::Input && - (canAdvance_ || this->mutableModes().inNamelist); + return canAdvance_ || this->mutableModes().inNamelist; } private: diff --git a/flang-rt/lib/runtime/edit-output.cpp b/flang-rt/lib/runtime/edit-output.cpp index f90b6fb..73dba35 100644 --- a/flang-rt/lib/runtime/edit-output.cpp +++ b/flang-rt/lib/runtime/edit-output.cpp @@ -175,9 +175,10 @@ bool RT_API_ATTRS EditIntegerOutput(IoStatementState &io, const DataEdit &edit, } if (edit.IsListDirected()) { int total{std::max(leadingSpaces, 1) + subTotal}; - if (io.GetConnectionState().NeedAdvance(static_cast<std::size_t>(total)) && - !io.AdvanceRecord()) { - return false; + if (io.GetConnectionState().NeedAdvance(static_cast<std::size_t>(total))) { + if (!io.AdvanceRecord()) { + return false; + } } leadingSpaces = 1; } else if (!edit.width) { diff --git a/flang-rt/lib/runtime/io-stmt.cpp b/flang-rt/lib/runtime/io-stmt.cpp index b958f23..a88fbe6 100644 --- a/flang-rt/lib/runtime/io-stmt.cpp +++ b/flang-rt/lib/runtime/io-stmt.cpp @@ -1109,20 +1109,20 @@ ChildListIoStatementState<DIR>::ChildListIoStatementState( ChildIo &child, const char *sourceFile, int sourceLine) : ChildIoStatementState<DIR>{child, sourceFile, sourceLine} { #if !defined(RT_DEVICE_AVOID_RECURSION) - if constexpr (DIR == Direction::Input) { - if (const auto *listInput{child.parent() - .get_if<ListDirectedStatementState<Direction::Input>>()}) { - this->set_eatComma(listInput->eatComma()); - this->namelistGroup_ = listInput->namelistGroup(); - if (auto *childListInput{child.parent() - .get_if<ChildListIoStatementState<Direction::Input>>()}) { - // Child list input whose parent is child list input: can advance - // if the parent can. - this->canAdvance_ = childListInput->CanAdvance(); - } else { - // Child list input of top-level list input: can advance. - this->canAdvance_ = true; - } + if (const auto *listParent{ + child.parent().get_if<ListDirectedStatementState<DIR>>()}) { + if constexpr (DIR == Direction::Input) { + this->set_eatComma(listParent->eatComma()); + this->namelistGroup_ = listParent->namelistGroup(); + } + if (auto *childListParent{ + child.parent().get_if<ChildListIoStatementState<DIR>>()}) { + // Child list I/O whose parent is child list I/O: can advance + // if the parent can. + this->canAdvance_ = childListParent->CanAdvance(); + } else { + // Child list I/O of top-level list I/O: can advance. + this->canAdvance_ = true; } } #else diff --git a/flang/include/flang/Optimizer/Builder/CUDAIntrinsicCall.h b/flang/include/flang/Optimizer/Builder/CUDAIntrinsicCall.h index d735ce9..ae7d566 100644 --- a/flang/include/flang/Optimizer/Builder/CUDAIntrinsicCall.h +++ b/flang/include/flang/Optimizer/Builder/CUDAIntrinsicCall.h @@ -63,9 +63,8 @@ struct CUDAIntrinsicLibrary : IntrinsicLibrary { mlir::Value genThisGrid(mlir::Type, llvm::ArrayRef<mlir::Value>); mlir::Value genThisThreadBlock(mlir::Type, llvm::ArrayRef<mlir::Value>); mlir::Value genThisWarp(mlir::Type, llvm::ArrayRef<mlir::Value>); + template <mlir::NVVM::MemScopeKind scope> void genThreadFence(llvm::ArrayRef<fir::ExtendedValue>); - void genThreadFenceBlock(llvm::ArrayRef<fir::ExtendedValue>); - void genThreadFenceSystem(llvm::ArrayRef<fir::ExtendedValue>); void genTMABulkCommitGroup(llvm::ArrayRef<fir::ExtendedValue>); void genTMABulkG2S(llvm::ArrayRef<fir::ExtendedValue>); void genTMABulkLoadC4(llvm::ArrayRef<fir::ExtendedValue>); diff --git a/flang/lib/Evaluate/check-expression.cpp b/flang/lib/Evaluate/check-expression.cpp index 839717d..656fc50 100644 --- a/flang/lib/Evaluate/check-expression.cpp +++ b/flang/lib/Evaluate/check-expression.cpp @@ -379,8 +379,11 @@ bool IsInitialProcedureTarget(const semantics::Symbol &symbol) { common::visitors{ [&](const semantics::SubprogramDetails &subp) { return !subp.isDummy() && !subp.stmtFunction() && - symbol.owner().kind() != semantics::Scope::Kind::MainProgram && - symbol.owner().kind() != semantics::Scope::Kind::Subprogram; + ((symbol.owner().kind() != + semantics::Scope::Kind::MainProgram && + symbol.owner().kind() != + semantics::Scope::Kind::Subprogram) || + ultimate.attrs().test(semantics::Attr::EXTERNAL)); }, [](const semantics::SubprogramNameDetails &x) { return x.kind() != semantics::SubprogramKind::Internal; diff --git a/flang/lib/Evaluate/common.cpp b/flang/lib/Evaluate/common.cpp index ed6a0ef..119ea3c 100644 --- a/flang/lib/Evaluate/common.cpp +++ b/flang/lib/Evaluate/common.cpp @@ -16,25 +16,26 @@ namespace Fortran::evaluate { void FoldingContext::RealFlagWarnings( const RealFlags &flags, const char *operation) { static constexpr auto warning{common::UsageWarning::FoldingException}; + if (!realFlagWarningContext_.empty()) { + // Override 'operation' with a string like + // "compilation-time evaluation of a call to '...'" + operation = realFlagWarningContext_.c_str(); + } if (flags.test(RealFlag::Overflow)) { - Warn(warning, "overflow on %s%s"_warn_en_US, operation, - realFlagWarningContext_); + Warn(warning, "overflow on %s"_warn_en_US, operation); } if (flags.test(RealFlag::DivideByZero)) { if (std::strcmp(operation, "division") == 0) { - Warn(warning, "division by zero%s"_warn_en_US, realFlagWarningContext_); + Warn(warning, "division by zero"_warn_en_US); } else { - Warn(warning, "division by zero on %s%s"_warn_en_US, operation, - realFlagWarningContext_); + Warn(warning, "division by zero on %s"_warn_en_US, operation); } } if (flags.test(RealFlag::InvalidArgument)) { - Warn(warning, "invalid argument on %s%s"_warn_en_US, operation, - realFlagWarningContext_); + Warn(warning, "invalid argument on %s"_warn_en_US, operation); } if (flags.test(RealFlag::Underflow)) { - Warn(warning, "underflow on %s%s"_warn_en_US, operation, - realFlagWarningContext_); + Warn(warning, "underflow on %s"_warn_en_US, operation); } } diff --git a/flang/lib/Evaluate/intrinsics-library.cpp b/flang/lib/Evaluate/intrinsics-library.cpp index d8af524..54726ac5 100644 --- a/flang/lib/Evaluate/intrinsics-library.cpp +++ b/flang/lib/Evaluate/intrinsics-library.cpp @@ -1052,7 +1052,7 @@ std::optional<HostRuntimeWrapper> GetHostRuntimeWrapper(const std::string &name, .value()); } auto restorer{context.SetRealFlagWarningContext( - " after folding a call to '"s + name + "'"s)}; + "compilation-time evaluation of a call to '"s + name + "'"s)}; return Fold(context, ConvertToType( resultType, hostFolderWithChecks(context, std::move(args))) diff --git a/flang/lib/Optimizer/Builder/CUDAIntrinsicCall.cpp b/flang/lib/Optimizer/Builder/CUDAIntrinsicCall.cpp index 4c0d266..323d1ef 100644 --- a/flang/lib/Optimizer/Builder/CUDAIntrinsicCall.cpp +++ b/flang/lib/Optimizer/Builder/CUDAIntrinsicCall.cpp @@ -472,17 +472,17 @@ static constexpr IntrinsicHandler cudaHandlers[]{ /*isElemental=*/false}, {"threadfence", static_cast<CUDAIntrinsicLibrary::SubroutineGenerator>( - &CI::genThreadFence), + &CI::genThreadFence<mlir::NVVM::MemScopeKind::GPU>), {}, /*isElemental=*/false}, {"threadfence_block", static_cast<CUDAIntrinsicLibrary::SubroutineGenerator>( - &CI::genThreadFenceBlock), + &CI::genThreadFence<mlir::NVVM::MemScopeKind::CTA>), {}, /*isElemental=*/false}, {"threadfence_system", static_cast<CUDAIntrinsicLibrary::SubroutineGenerator>( - &CI::genThreadFenceSystem), + &CI::genThreadFence<mlir::NVVM::MemScopeKind::SYS>), {}, /*isElemental=*/false}, {"tma_bulk_commit_group", @@ -1306,37 +1306,12 @@ CUDAIntrinsicLibrary::genThisWarp(mlir::Type resultType, return res; } -// THREADFENCE +// THREADFENCE, THREADFENCE_BLOCK, THREADFENCE_SYSTEM +template <mlir::NVVM::MemScopeKind scope> void CUDAIntrinsicLibrary::genThreadFence( llvm::ArrayRef<fir::ExtendedValue> args) { - constexpr llvm::StringLiteral funcName = "llvm.nvvm.membar.gl"; - mlir::FunctionType funcType = - mlir::FunctionType::get(builder.getContext(), {}, {}); - auto funcOp = builder.createFunction(loc, funcName, funcType); - llvm::SmallVector<mlir::Value> noArgs; - fir::CallOp::create(builder, loc, funcOp, noArgs); -} - -// THREADFENCE_BLOCK -void CUDAIntrinsicLibrary::genThreadFenceBlock( - llvm::ArrayRef<fir::ExtendedValue> args) { - constexpr llvm::StringLiteral funcName = "llvm.nvvm.membar.cta"; - mlir::FunctionType funcType = - mlir::FunctionType::get(builder.getContext(), {}, {}); - auto funcOp = builder.createFunction(loc, funcName, funcType); - llvm::SmallVector<mlir::Value> noArgs; - fir::CallOp::create(builder, loc, funcOp, noArgs); -} - -// THREADFENCE_SYSTEM -void CUDAIntrinsicLibrary::genThreadFenceSystem( - llvm::ArrayRef<fir::ExtendedValue> args) { - constexpr llvm::StringLiteral funcName = "llvm.nvvm.membar.sys"; - mlir::FunctionType funcType = - mlir::FunctionType::get(builder.getContext(), {}, {}); - auto funcOp = builder.createFunction(loc, funcName, funcType); - llvm::SmallVector<mlir::Value> noArgs; - fir::CallOp::create(builder, loc, funcOp, noArgs); + assert(args.size() == 0); + mlir::NVVM::MembarOp::create(builder, loc, scope); } // TMA_BULK_COMMIT_GROUP diff --git a/flang/lib/Optimizer/OpenMP/MapInfoFinalization.cpp b/flang/lib/Optimizer/OpenMP/MapInfoFinalization.cpp index bd07d7f..d60da89 100644 --- a/flang/lib/Optimizer/OpenMP/MapInfoFinalization.cpp +++ b/flang/lib/Optimizer/OpenMP/MapInfoFinalization.cpp @@ -477,58 +477,6 @@ class MapInfoFinalizationPass return false; } - mlir::omp::MapInfoOp genBoxcharMemberMap(mlir::omp::MapInfoOp op, - fir::FirOpBuilder &builder) { - if (!op.getMembers().empty()) - return op; - mlir::Location loc = op.getVarPtr().getLoc(); - mlir::Value boxChar = op.getVarPtr(); - - if (mlir::isa<fir::ReferenceType>(op.getVarPtr().getType())) - boxChar = fir::LoadOp::create(builder, loc, op.getVarPtr()); - - fir::BoxCharType boxCharType = - mlir::dyn_cast<fir::BoxCharType>(boxChar.getType()); - mlir::Value boxAddr = fir::BoxOffsetOp::create( - builder, loc, op.getVarPtr(), fir::BoxFieldAttr::base_addr); - - mlir::ArrayAttr newMembersAttr; - llvm::SmallVector<llvm::SmallVector<int64_t>> memberIdx = {{0}}; - newMembersAttr = builder.create2DI64ArrayAttr(memberIdx); - - mlir::Value varPtr = op.getVarPtr(); - mlir::omp::MapInfoOp memberMapInfoOp = mlir::omp::MapInfoOp::create( - builder, op.getLoc(), varPtr.getType(), varPtr, - mlir::TypeAttr::get(boxCharType.getEleTy()), - builder.getAttr<mlir::omp::ClauseMapFlagsAttr>( - mlir::omp::ClauseMapFlags::to | - mlir::omp::ClauseMapFlags::implicit), - builder.getAttr<mlir::omp::VariableCaptureKindAttr>( - mlir::omp::VariableCaptureKind::ByRef), - /*varPtrPtr=*/boxAddr, - /*members=*/llvm::SmallVector<mlir::Value>{}, - /*member_index=*/mlir::ArrayAttr{}, - /*bounds=*/op.getBounds(), - /*mapperId=*/mlir::FlatSymbolRefAttr(), /*name=*/op.getNameAttr(), - builder.getBoolAttr(false)); - - mlir::omp::MapInfoOp newMapInfoOp = mlir::omp::MapInfoOp::create( - builder, op.getLoc(), op.getResult().getType(), varPtr, - mlir::TypeAttr::get( - llvm::cast<mlir::omp::PointerLikeType>(varPtr.getType()) - .getElementType()), - op.getMapTypeAttr(), op.getMapCaptureTypeAttr(), - /*varPtrPtr=*/mlir::Value{}, - /*members=*/llvm::SmallVector<mlir::Value>{memberMapInfoOp}, - /*member_index=*/newMembersAttr, - /*bounds=*/llvm::SmallVector<mlir::Value>{}, - /*mapperId=*/mlir::FlatSymbolRefAttr(), op.getNameAttr(), - /*partial_map=*/builder.getBoolAttr(false)); - op.replaceAllUsesWith(newMapInfoOp.getResult()); - op->erase(); - return newMapInfoOp; - } - // Expand mappings of type(C_PTR) to map their `__address` field explicitly // as a single pointer-sized member (USM-gated at callsite). This helps in // USM scenarios to ensure the pointer-sized mapping is used. @@ -956,6 +904,14 @@ class MapInfoFinalizationPass baseAddr.erase(); } + static bool hasADescriptor(mlir::Operation *varOp, mlir::Type varType) { + if (fir::isTypeWithDescriptor(varType) || + mlir::isa<fir::BoxCharType>(varType) || + mlir::isa_and_present<fir::BoxAddrOp>(varOp)) + return true; + return false; + } + // This pass executes on omp::MapInfoOp's containing descriptor based types // (allocatables, pointers, assumed shape etc.) and expanding them into // multiple omp::MapInfoOp's for each pointer member contained within the @@ -1209,36 +1165,6 @@ class MapInfoFinalizationPass return mlir::WalkResult::advance(); }); - func->walk([&](mlir::omp::MapInfoOp op) { - if (!op.getMembers().empty()) - return; - - if (!mlir::isa<fir::BoxCharType>(fir::unwrapRefType(op.getVarType()))) - return; - - // POSSIBLE_HACK_ALERT: If the boxchar has been implicitly mapped then - // it is likely that the underlying pointer to the data - // (!fir.ref<fir.char<k,?>>) has already been mapped. So, skip such - // boxchars. We are primarily interested in boxchars that were mapped - // by passes such as MapsForPrivatizedSymbols that map boxchars that - // are privatized. At present, such boxchar maps are not marked - // implicit. Should they be? I don't know. If they should be then - // we need to change this check for early return OR live with - // over-mapping. - bool hasImplicitMap = - (op.getMapType() & mlir::omp::ClauseMapFlags::implicit) == - mlir::omp::ClauseMapFlags::implicit; - if (hasImplicitMap) - return; - - assert(llvm::hasSingleElement(op->getUsers()) && - "OMPMapInfoFinalization currently only supports single users " - "of a MapInfoOp"); - - builder.setInsertionPoint(op); - genBoxcharMemberMap(op, builder); - }); - // Expand type(C_PTR) only when unified_shared_memory is required, // to ensure device-visible pointer size/behavior in USM scenarios // without changing default expectations elsewhere. @@ -1266,9 +1192,8 @@ class MapInfoFinalizationPass "OMPMapInfoFinalization currently only supports single users " "of a MapInfoOp"); - if (fir::isTypeWithDescriptor(op.getVarType()) || - mlir::isa_and_present<fir::BoxAddrOp>( - op.getVarPtr().getDefiningOp())) { + if (hasADescriptor(op.getVarPtr().getDefiningOp(), + fir::unwrapRefType(op.getVarType()))) { builder.setInsertionPoint(op); mlir::Operation *targetUser = getFirstTargetUser(op); assert(targetUser && "expected user of map operation was not found"); diff --git a/flang/test/Driver/convert.f90 b/flang/test/Driver/convert.f90 index 0ba31d2..0b4da02 100755 --- a/flang/test/Driver/convert.f90 +++ b/flang/test/Driver/convert.f90 @@ -1,5 +1,5 @@ ! Ensure argument -fconvert=<value> accepts all relevant options and produces an -! error if an invalid value is specified. +! error if an invalid value is specified. !-------------------------- ! FLANG DRIVER (flang) diff --git a/flang/test/Driver/do_concurrent_to_omp_cli.f90 b/flang/test/Driver/do_concurrent_to_omp_cli.f90 index bdb603f..e44db04 100644 --- a/flang/test/Driver/do_concurrent_to_omp_cli.f90 +++ b/flang/test/Driver/do_concurrent_to_omp_cli.f90 @@ -3,12 +3,12 @@ ! RUN: %flang --help | FileCheck %s --check-prefix=FLANG ! FLANG: -fdo-concurrent-to-openmp=<value> -! FLANG-NEXT: Try to map `do concurrent` loops to OpenMP [none|host|device] +! FLANG-NEXT: Try to map `do concurrent` loops to OpenMP [none|host|device] ! RUN: bbc --help | FileCheck %s --check-prefix=BBC ! BBC: -fdo-concurrent-to-openmp=<string> -! BBC-SAME: Try to map `do concurrent` loops to OpenMP [none|host|device] +! BBC-SAME: Try to map `do concurrent` loops to OpenMP [none|host|device] ! RUN: %flang -c -fdo-concurrent-to-openmp=host %s 2>&1 \ ! RUN: | FileCheck %s --check-prefix=OPT diff --git a/flang/test/Driver/emit-mlir.f90 b/flang/test/Driver/emit-mlir.f90 index de5a62d..f2a4b6c 100644 --- a/flang/test/Driver/emit-mlir.f90 +++ b/flang/test/Driver/emit-mlir.f90 @@ -21,7 +21,7 @@ ! CHECK-NEXT: func.func @main(%arg0: i32, %arg1: !llvm.ptr, %arg2: !llvm.ptr) -> i32 { ! CHECK-NEXT: %c0_i32 = arith.constant 0 : i32 ! CHECK-NEXT: %0 = fir.zero_bits !fir.ref<tuple<i32, !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>> -! CHECK-NEXT: fir.call @_FortranAProgramStart(%arg0, %arg1, %arg2, %0) {{.*}} : (i32, !llvm.ptr, !llvm.ptr, !fir.ref<tuple<i32, !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>>) +! CHECK-NEXT: fir.call @_FortranAProgramStart(%arg0, %arg1, %arg2, %0) {{.*}} : (i32, !llvm.ptr, !llvm.ptr, !fir.ref<tuple<i32, !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>>) ! CHECK-NEXT: fir.call @_QQmain() fastmath<contract> : () -> () ! CHECK-NEXT: fir.call @_FortranAProgramEndStatement() {{.*}} : () -> () ! CHECK-NEXT: return %c0_i32 : i32 diff --git a/flang/test/Driver/fatal-errors-parsing.f90 b/flang/test/Driver/fatal-errors-parsing.f90 index 185a6e0..fd8e167 100644 --- a/flang/test/Driver/fatal-errors-parsing.f90 +++ b/flang/test/Driver/fatal-errors-parsing.f90 @@ -7,7 +7,7 @@ program p ! CHECK2: fatal-errors-parsing.f90:{{.*}} error: continue end - + subroutine s contains ! CHECK1-NOT: error: diff --git a/flang/test/Driver/fatal-errors-semantics.f90 b/flang/test/Driver/fatal-errors-semantics.f90 index 54740dd..3d3f642 100644 --- a/flang/test/Driver/fatal-errors-semantics.f90 +++ b/flang/test/Driver/fatal-errors-semantics.f90 @@ -37,4 +37,3 @@ module m call soa(null()) end end -
\ No newline at end of file diff --git a/flang/test/Driver/flang-ld-aarch64.f90 b/flang/test/Driver/flang-ld-aarch64.f90 index 61cd46c..4039859 100644 --- a/flang/test/Driver/flang-ld-aarch64.f90 +++ b/flang/test/Driver/flang-ld-aarch64.f90 @@ -1,4 +1,4 @@ -! Check linker flags for AArch64 linux, since it needs both libgcc and +! Check linker flags for AArch64 linux, since it needs both libgcc and ! compiler-rt, with compiler-rt second when -rtlib=libgcc. ! RUN: %flang -### -rtlib=libgcc --target=aarch64-linux-gnu %S/Inputs/hello.f90 2>&1 | FileCheck %s diff --git a/flang/test/Driver/flang-ld-powerpc.f90 b/flang/test/Driver/flang-ld-powerpc.f90 index 5328077ac..9058679 100644 --- a/flang/test/Driver/flang-ld-powerpc.f90 +++ b/flang/test/Driver/flang-ld-powerpc.f90 @@ -4,7 +4,7 @@ !! -static-libflang_rt in the future. Need to add that option here. !! Because flang-rt currently only supports -!! LLVM_ENABLE_PER_TARGET_RUNTIME_DIR=ON, use +!! LLVM_ENABLE_PER_TARGET_RUNTIME_DIR=ON, use !! resource_dir_with_per_target_subdir as inputs. ! Check powerpc64-ibm-aix 64-bit linking to static flang-rt by default @@ -26,7 +26,7 @@ ! AIX64-LD-PER-TARGET-DEFAULT-NOT: "-L/[[RESOURCE_DIR]]{{/|\\\\}}lib{{/|\\\\}}powerpc64-ibm-aix" -! Check powerpc64-ibm-aix 64-bit linking to static flang-rt by option +! Check powerpc64-ibm-aix 64-bit linking to static flang-rt by option ! RUN: %flang -static-libflangrt -Werror %s -### 2>&1 \ ! RUN: --target=powerpc64-ibm-aix \ ! RUN: -resource-dir=%S/../../../clang/test/Driver/Inputs/resource_dir_with_per_target_subdir \ @@ -44,7 +44,7 @@ ! AIX64-LD-PER-TARGET-STATIC-SAME: "-lpthread" -! Check powerpc64-ibm-aix 64-bit linking to shared flang-rt by option +! Check powerpc64-ibm-aix 64-bit linking to shared flang-rt by option ! RUN: %flang -shared-libflangrt -Werror %s -### 2>&1 \ ! RUN: --target=powerpc64-ibm-aix \ ! RUN: -resource-dir=%S/../../../clang/test/Driver/Inputs/resource_dir_with_per_target_subdir \ diff --git a/flang/test/Driver/gcc-toolchain-install-dir.f90 b/flang/test/Driver/gcc-toolchain-install-dir.f90 index e195bdd..05b73bc 100644 --- a/flang/test/Driver/gcc-toolchain-install-dir.f90 +++ b/flang/test/Driver/gcc-toolchain-install-dir.f90 @@ -5,10 +5,10 @@ ! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=i386-unknown-linux-gnu --gcc-install-dir=%S/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0 | FileCheck %s --check-prefix=CHECK-I386 ! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=i386-unknown-linux-gnu --gcc-toolchain=%S/Inputs/basic_cross_linux_tree/usr | FileCheck %s --check-prefix=CHECK-I386 ! CHECK-I386: Selected GCC installation: [[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0 -! CHECK-I386: "-fc1" "-triple" "i386-unknown-linux-gnu" +! CHECK-I386: "-fc1" "-triple" "i386-unknown-linux-gnu" ! CHECK-I386: "[[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/../../../../i386-unknown-linux-gnu/bin{{/|\\\\}}as" ! CHECK-I386: "[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/../../../../i386-unknown-linux-gnu/bin{{/|\\\\}}ld" {{.*}} "-m" "elf_i386" -! CHECK-I386-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0" +! CHECK-I386-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0" ! CHECK-I386-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/../../../../i386-unknown-linux-gnu/lib" ! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=x86_64-unknown-linux-gnu --gcc-install-dir=%S/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0 | FileCheck %s --check-prefix=CHECK-X86-64 @@ -17,5 +17,5 @@ ! CHECK-X86-64: "-fc1" "-triple" "x86_64-unknown-linux-gnu" ! CHECK-X86-64: "[[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/../../../../x86_64-unknown-linux-gnu/bin{{/|\\\\}}as" "--64" ! CHECK-X86-64: "[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/../../../../x86_64-unknown-linux-gnu/bin{{/|\\\\}}ld" {{.*}} "-m" "elf_x86_64" -! CHECK-X86-64-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0" +! CHECK-X86-64-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0" ! CHECK-X86-64-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/../../../../x86_64-unknown-linux-gnu/lib" diff --git a/flang/test/Driver/large-data-threshold.f90 b/flang/test/Driver/large-data-threshold.f90 index 6a7eef7..fa2d4ae 100644 --- a/flang/test/Driver/large-data-threshold.f90 +++ b/flang/test/Driver/large-data-threshold.f90 @@ -5,8 +5,8 @@ ! RUN: %flang -### -c --target=x86_64 -mlarge-data-threshold=32768 %s 2>&1 | FileCheck %s --check-prefix=NO-MCMODEL ! RUN: %flang -### -c --target=x86_64 -mcmodel=small -mlarge-data-threshold=32768 %s 2>&1 | FileCheck %s --check-prefix=NO-MCMODEL ! RUN: not %flang -### -c --target=aarch64 -mcmodel=small -mlarge-data-threshold=32768 %s 2>&1 | FileCheck %s --check-prefix=NOT-SUPPORTED - - + + ! CHECK: "{{.*}}flang" "-fc1" ! CHECK-SAME: "-mlarge-data-threshold=32768" ! CHECK-59000: "{{.*}}flang" "-fc1" diff --git a/flang/test/Driver/lto-fatlto.f90 b/flang/test/Driver/lto-fatlto.f90 index c52d6e3..2ea251e 100644 --- a/flang/test/Driver/lto-fatlto.f90 +++ b/flang/test/Driver/lto-fatlto.f90 @@ -1,5 +1,5 @@ ! REQUIRES: x86-registered-target -! checks fatlto objects: that valid bitcode is included in the object file generated. +! checks fatlto objects: that valid bitcode is included in the object file generated. ! RUN: %flang -fc1 -triple x86_64-unknown-linux-gnu -flto -ffat-lto-objects -emit-obj %s -o %t.o ! RUN: llvm-readelf -S %t.o | FileCheck %s --check-prefixes=ELF diff --git a/flang/test/Driver/mlir-debug-pass-pipeline.f90 b/flang/test/Driver/mlir-debug-pass-pipeline.f90 index eb5165e..0138d9b 100644 --- a/flang/test/Driver/mlir-debug-pass-pipeline.f90 +++ b/flang/test/Driver/mlir-debug-pass-pipeline.f90 @@ -100,7 +100,7 @@ end program ! ALL-NEXT: CSE ! ALL-NEXT: (S) 0 num-cse'd - Number of operations CSE'd ! ALL-NEXT: (S) 0 num-dce'd - Number of operations DCE'd -! ALL-NEXT: MIFOpConversion +! ALL-NEXT: MIFOpConversion ! ALL-NEXT: BoxedProcedurePass ! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'gpu.module', 'omp.declare_reduction', 'omp.private'] @@ -109,10 +109,10 @@ end program ! ALL-NEXT: 'func.func' Pipeline ! ALL-NEXT: AbstractResultOpt ! ALL-NEXT: 'gpu.module' Pipeline -! ALL-NEXT: Pipeline Collection : ['func.func', 'gpu.func'] -! ALL-NEXT: 'func.func' Pipeline +! ALL-NEXT: Pipeline Collection : ['func.func', 'gpu.func'] +! ALL-NEXT: 'func.func' Pipeline ! ALL-NEXT: AbstractResultOpt -! ALL-NEXT: 'gpu.func' Pipeline +! ALL-NEXT: 'gpu.func' Pipeline ! ALL-NEXT: AbstractResultOpt ! ALL-NEXT: 'omp.declare_reduction' Pipeline ! ALL-NEXT: AbstractResultOpt diff --git a/flang/test/Driver/mlir-pass-pipeline.f90 b/flang/test/Driver/mlir-pass-pipeline.f90 index 3b6a9d7..0d68191 100644 --- a/flang/test/Driver/mlir-pass-pipeline.f90 +++ b/flang/test/Driver/mlir-pass-pipeline.f90 @@ -142,7 +142,7 @@ end program ! ALL-NEXT: (S) 0 num-dce'd - Number of operations DCE'd ! O2-NEXT: 'func.func' Pipeline ! O2-NEXT: SetRuntimeCallAttributes -! ALL-NEXT: MIFOpConversion +! ALL-NEXT: MIFOpConversion ! ALL-NEXT: BoxedProcedurePass ! O2-NEXT: AddAliasTags @@ -152,10 +152,10 @@ end program ! ALL-NEXT: 'func.func' Pipeline ! ALL-NEXT: AbstractResultOpt ! ALL-NEXT: 'gpu.module' Pipeline -! ALL-NEXT: Pipeline Collection : ['func.func', 'gpu.func'] -! ALL-NEXT: 'func.func' Pipeline +! ALL-NEXT: Pipeline Collection : ['func.func', 'gpu.func'] +! ALL-NEXT: 'func.func' Pipeline ! ALL-NEXT: AbstractResultOpt -! ALL-NEXT: 'gpu.func' Pipeline +! ALL-NEXT: 'gpu.func' Pipeline ! ALL-NEXT: AbstractResultOpt ! ALL-NEXT: 'omp.declare_reduction' Pipeline ! ALL-NEXT: AbstractResultOpt diff --git a/flang/test/Driver/multiple-fc1-input.f90 b/flang/test/Driver/multiple-fc1-input.f90 index 57f7c5e..e142f35 100644 --- a/flang/test/Driver/multiple-fc1-input.f90 +++ b/flang/test/Driver/multiple-fc1-input.f90 @@ -5,5 +5,5 @@ ! RUN: %flang_fc1 -emit-fir %s %s -o - | FileCheck %s subroutine foo() end subroutine -! CHECK: func @_QPfoo() -! CHECK: func @_QPfoo() +! CHECK: func @_QPfoo() +! CHECK: func @_QPfoo() diff --git a/flang/test/Driver/omp-driver-offload.f90 b/flang/test/Driver/omp-driver-offload.f90 index 0924857..8660bec 100644 --- a/flang/test/Driver/omp-driver-offload.f90 +++ b/flang/test/Driver/omp-driver-offload.f90 @@ -1,9 +1,9 @@ -! Test that flang OpenMP and OpenMP offload related -! commands forward or expand to the appropriate commands +! Test that flang OpenMP and OpenMP offload related +! commands forward or expand to the appropriate commands ! for flang -fc1 as expected. Assumes a gfx90a, aarch64, -! and sm_70 architecture, but doesn't require one to be -! installed or compiled for, just testing the appropriate -! generation of jobs are created with the correct +! and sm_70 architecture, but doesn't require one to be +! installed or compiled for, just testing the appropriate +! generation of jobs are created with the correct ! corresponding arguments. ! Test regular -fopenmp with no offload @@ -47,7 +47,7 @@ ! OFFLOAD-DEVICE-NEXT: "{{[^"]*}}flang" "-fc1" "-triple" "nvptx64-nvidia-cuda" ! OFFLOAD-DEVICE-NOT: "{{[^"]*}}flang" "-fc1" "-triple" "aarch64-unknown-linux-gnu" -! Test regular -fopenmp with offload for basic fopenmp-is-target-device flag addition and correct fopenmp +! Test regular -fopenmp with offload for basic fopenmp-is-target-device flag addition and correct fopenmp ! RUN: %flang -### -fopenmp --offload-arch=gfx90a -fopenmp-targets=amdgcn-amd-amdhsa -nogpulib %s 2>&1 | FileCheck --check-prefixes=CHECK-OPENMP-IS-TARGET-DEVICE %s ! CHECK-OPENMP-IS-TARGET-DEVICE: "{{[^"]*}}flang" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" {{.*}}.f90" @@ -169,7 +169,7 @@ ! RUN: | FileCheck %s --check-prefixes=CHECK-OPENMP-VERSION ! CHECK-OPENMP-VERSION: "{{[^"]*}}flang" "-fc1" {{.*}} "-fopenmp" "-fopenmp-version=45" {{.*}}.f90" -! Test diagnostic error when host IR file is non-existent +! Test diagnostic error when host IR file is non-existent ! RUN: not %flang_fc1 %s -o %t 2>&1 -fopenmp -fopenmp-is-target-device \ ! RUN: -fopenmp-host-ir-file-path non-existant-file.bc \ ! RUN: | FileCheck %s --check-prefix=HOST-IR-MISSING diff --git a/flang/test/Driver/tune-cpu-fir.f90 b/flang/test/Driver/tune-cpu-fir.f90 index 43c13b4..843feeb 100644 --- a/flang/test/Driver/tune-cpu-fir.f90 +++ b/flang/test/Driver/tune-cpu-fir.f90 @@ -14,7 +14,7 @@ ! ARMTUNE-SAME: fir.tune_cpu = "neoverse-n1" ! ARMBOTH-SAME: fir.target_cpu = "aarch64" -! ARMBOTH-SAME: fir.tune_cpu = "neoverse-n1" +! ARMBOTH-SAME: fir.tune_cpu = "neoverse-n1" ! X86CPU-SAME: fir.target_cpu = "x86-64" ! X86CPU-NOT: fir.tune_cpu = "pentium4" diff --git a/flang/test/Driver/version-loops.f90 b/flang/test/Driver/version-loops.f90 index d206393..c4caf46 100644 --- a/flang/test/Driver/version-loops.f90 +++ b/flang/test/Driver/version-loops.f90 @@ -1,22 +1,22 @@ -! Test that flang forwards the -f{no-,}version-loops-for-stride +! Test that flang forwards the -f{no-,}version-loops-for-stride ! options correctly to flang -fc1 for different variants of optimisation ! and explicit flags. ! RUN: %flang -### %s -o %t 2>&1 -O3 \ ! RUN: | FileCheck %s - + ! RUN: %flang -### %s -o %t 2>&1 -O2 \ ! RUN: | FileCheck %s --check-prefix=CHECK-O2 ! RUN: %flang -### %s -o %t 2>&1 -O2 -fversion-loops-for-stride \ ! RUN: | FileCheck %s --check-prefix=CHECK-O2-with - + ! RUN: %flang -### %s -o %t 2>&1 -O4 \ ! RUN: | FileCheck %s --check-prefix=CHECK-O4 - + ! RUN: %flang -### %s -o %t 2>&1 -Ofast \ ! RUN: | FileCheck %s --check-prefix=CHECK-Ofast - + ! RUN: %flang -### %s -o %t 2>&1 -Ofast -fno-version-loops-for-stride \ ! RUN: | FileCheck %s --check-prefix=CHECK-Ofast-no @@ -29,12 +29,12 @@ ! CHECK-O2: "{{.*}}flang" "-fc1" ! CHECK-O2-NOT: "-fversion-loops-for-stride" -! CHECK-O2-SAME: "-O2" +! CHECK-O2-SAME: "-O2" ! CHECK-O2-with: "{{.*}}flang" "-fc1" ! CHECK-O2-with-SAME: "-fversion-loops-for-stride" -! CHECK-O2-with-SAME: "-O2" - +! CHECK-O2-with-SAME: "-O2" + ! CHECK-O4: "{{.*}}flang" "-fc1" ! CHECK-O4-SAME: "-fversion-loops-for-stride" ! CHECK-O4-SAME: "-O3" diff --git a/flang/test/Evaluate/folding33.f90 b/flang/test/Evaluate/folding33.f90 index fb5a23cf..299cb7e 100644 --- a/flang/test/Evaluate/folding33.f90 +++ b/flang/test/Evaluate/folding33.f90 @@ -1,4 +1,4 @@ !RUN: %flang_fc1 -fsyntax-only %s 2>&1 | FileCheck %s -!CHECK: warning: overflow on REAL(4) to REAL(2) conversion after folding a call to 'exp' [-Wfolding-exception] +!CHECK: warning: overflow on compilation-time evaluation of a call to 'exp' [-Wfolding-exception] print *, exp((11.265625_2,1._2)) end diff --git a/flang/test/Lower/CUDA/cuda-device-proc.cuf b/flang/test/Lower/CUDA/cuda-device-proc.cuf index 9f8f74a..3a255af 100644 --- a/flang/test/Lower/CUDA/cuda-device-proc.cuf +++ b/flang/test/Lower/CUDA/cuda-device-proc.cuf @@ -22,9 +22,6 @@ attributes(global) subroutine devsub() call syncthreads() call syncwarp(1) - call threadfence() - call threadfence_block() - call threadfence_system() ret = syncthreads_and(1) res = syncthreads_and(tid > offset) ret = syncthreads_count(1) @@ -106,9 +103,6 @@ end ! CHECK-LABEL: func.func @_QPdevsub() attributes {cuf.proc_attr = #cuf.cuda_proc<global>} ! CHECK: nvvm.barrier0 ! CHECK: nvvm.bar.warp.sync %c1{{.*}} : i32 -! CHECK: fir.call @llvm.nvvm.membar.gl() fastmath<contract> : () -> () -! CHECK: fir.call @llvm.nvvm.membar.cta() fastmath<contract> : () -> () -! CHECK: fir.call @llvm.nvvm.membar.sys() fastmath<contract> : () -> () ! CHECK: %{{.*}} = fir.call @llvm.nvvm.barrier0.and(%c1{{.*}}) fastmath<contract> : (i32) -> i32 ! CHECK: %[[A:.*]] = fir.load %{{.*}} : !fir.ref<i32> ! CHECK: %[[B:.*]] = fir.load %{{.*}} : !fir.ref<i32> diff --git a/flang/test/Lower/CUDA/cuda-synchronization.cuf b/flang/test/Lower/CUDA/cuda-synchronization.cuf new file mode 100644 index 0000000..6e2e234 --- /dev/null +++ b/flang/test/Lower/CUDA/cuda-synchronization.cuf @@ -0,0 +1,14 @@ +! RUN: bbc -emit-hlfir -fcuda %s -o - | FileCheck %s + +! Test CUDA Fortran instrinsics lowerings for synchronization. + +attributes(global) subroutine sync() + call threadfence() + call threadfence_block() + call threadfence_system() +end subroutine + +! CHECK-LABEL: func.func @_QPsync() attributes {cuf.proc_attr = #cuf.cuda_proc<global>} +! CHECK: nvvm.memory.barrier <gpu> +! CHECK: nvvm.memory.barrier <cta> +! CHECK: nvvm.memory.barrier <sys> diff --git a/flang/test/Lower/OpenMP/DelayedPrivatization/target-private-multiple-variables.f90 b/flang/test/Lower/OpenMP/DelayedPrivatization/target-private-multiple-variables.f90 index f3b9397..a6394ea 100644 --- a/flang/test/Lower/OpenMP/DelayedPrivatization/target-private-multiple-variables.f90 +++ b/flang/test/Lower/OpenMP/DelayedPrivatization/target-private-multiple-variables.f90 @@ -156,7 +156,7 @@ end subroutine target_allocatable ! CHECK-SAME: %[[REAL_ARR_DESC_MAP]] -> %[[MAPPED_ARG2:[^,]+]] ! CHECK-SAME: %[[CHAR_VAR_DESC_MAP]] -> %[[MAPPED_ARG3:.[^,]+]] ! CHECK-SAME: %[[MAPPED_MI0]] -> %[[MAPPED_ARG0:[^,]+]] -! CHECK-SAME: !fir.ref<!fir.box<!fir.heap<i32>>>, !fir.ref<!fir.box<!fir.array<?xf32>>>, !fir.ref<!fir.boxchar<1>>, !fir.ref<i32>, !fir.llvm_ptr<!fir.ref<i32>>, !fir.llvm_ptr<!fir.ref<!fir.array<?xf32>>>, !fir.ref<!fir.boxchar<1>> +! CHECK-SAME: !fir.ref<!fir.box<!fir.heap<i32>>>, !fir.ref<!fir.box<!fir.array<?xf32>>>, !fir.ref<!fir.boxchar<1>>, !fir.ref<i32>, !fir.llvm_ptr<!fir.ref<i32>>, !fir.llvm_ptr<!fir.ref<!fir.array<?xf32>>>, !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> ! CHECK-SAME: private( ! CHECK-SAME: @[[ALLOC_PRIVATIZER_SYM]] %{{[^[:space:]]+}}#0 -> %[[ALLOC_ARG:[^,]+]] [map_idx=0], ! CHECK-SAME: @[[REAL_PRIVATIZER_SYM]] %{{[^[:space:]]+}}#0 -> %[[REAL_ARG:[^,]+]], diff --git a/flang/test/Lower/OpenMP/map-character.f90 b/flang/test/Lower/OpenMP/map-character.f90 index cefd3ac..9a11423 100644 --- a/flang/test/Lower/OpenMP/map-character.f90 +++ b/flang/test/Lower/OpenMP/map-character.f90 @@ -39,8 +39,11 @@ end subroutine TestOfCharacter !CHECK: %[[A1_UB:.*]] = arith.subi %[[UNBOXED_ARG1]]#1, %[[CONST_ONE]] : index !CHECK: %[[BOUNDS_A1_BOXCHAR:.*]] = omp.map.bounds lower_bound(%[[CONST_ZERO]] : index) upper_bound(%[[A1_UB]] : index) extent(%[[UNBOXED_ARG1]]#1 : index) !CHECK-SAME: stride(%[[CONST_ONE]] : index) start_idx(%[[CONST_ZERO]] : index) {stride_in_bytes = true} -!CHECK: %[[A1_BOXCHAR_MAP:.*]] = omp.map.info var_ptr(%[[A1_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(implicit, to) -!CHECK-SAME: capture(ByRef) bounds(%[[BOUNDS_A1_BOXCHAR]]) -> !fir.ref<!fir.boxchar<1>> {name = ""} +!CHECK: %[[A1_BOX_ADDR:.*]] = fir.box_offset %[[A1_BOXCHAR_ALLOCA]] base_addr : (!fir.ref<!fir.boxchar<1>>) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> +!CHECK: %[[A1_BOXCHAR_MAP:.*]] = omp.map.info var_ptr(%[[A1_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(implicit, to) +!CHECK-SAME: capture(ByRef) var_ptr_ptr(%[[A1_BOX_ADDR]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%[[BOUNDS_A1_BOXCHAR]]) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> {name = ""} +!CHECK: %[[A1_BOXCHAR_MAP_2:.*]] = omp.map.info var_ptr(%[[A1_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) +!CHECK-SAME: map_clauses(implicit, to) capture(ByRef) members(%[[A1_BOXCHAR_MAP]] : [0] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) -> !fir.ref<!fir.boxchar<1>> {name = ""} !CHECK: fir.store %[[ARG0]] to %[[A0_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>> !CHECK: %[[CONST_ZERO:.*]] = arith.constant 0 : index !CHECK: %[[CONST_ONE:.*]] = arith.constant 1 : index @@ -48,9 +51,12 @@ end subroutine TestOfCharacter !CHECK: %[[A0_UB:.*]] = arith.subi %[[UNBOXED_ARG0]]#1, %[[CONST_ONE]] : index !CHECK: %[[BOUNDS_A0_BOXCHAR:.*]] = omp.map.bounds lower_bound(%[[CONST_ZERO]] : index) upper_bound(%[[A0_UB]] : index) extent(%[[UNBOXED_ARG0]]#1 : index) !CHECK-SAME: stride(%[[CONST_ONE]] : index) start_idx(%[[CONST_ZERO]] : index) {stride_in_bytes = true} -!CHECK: %[[A0_BOXCHAR_MAP:.*]] = omp.map.info var_ptr(%[[A0_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(implicit, to) -!CHECK-SAME: capture(ByRef) bounds(%[[BOUNDS_A0_BOXCHAR]]) -> !fir.ref<!fir.boxchar<1>> {name = ""} -!CHECK: omp.target map_entries(%[[A0_MAP]] -> %[[TGT_A0:.*]], %[[A1_MAP]] -> %[[TGT_A1:.*]], %[[A1_BOXCHAR_MAP]] -> %[[TGT_A1_BOXCHAR:.*]], %[[A0_BOXCHAR_MAP]] -> %[[TGT_A0_BOXCHAR:.*]] : !fir.ref<!fir.char<1,?>>, !fir.ref<!fir.char<1,?>>, !fir.ref<!fir.boxchar<1>>, !fir.ref<!fir.boxchar<1>>) { +!CHECK: %[[A0_BOX_ADDR:.*]] = fir.box_offset %[[A0_BOXCHAR_ALLOCA]] base_addr : (!fir.ref<!fir.boxchar<1>>) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> +!CHECK: %[[A0_BOXCHAR_MAP:.*]] = omp.map.info var_ptr(%[[A0_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(implicit, to) +!CHECK-SAME: capture(ByRef) var_ptr_ptr(%[[A0_BOX_ADDR]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%24) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> {name = ""} +!CHECK: %[[A0_BOXCHAR_MAP_2:.*]] = omp.map.info var_ptr(%[[A0_BOXCHAR_ALLOCA]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(implicit, to) +!CHECK-SAME: capture(ByRef) members(%[[A0_BOXCHAR_MAP]] : [0] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) -> !fir.ref<!fir.boxchar<1>> {name = ""} +!CHECK: omp.target map_entries(%[[A0_MAP]] -> %[[TGT_A0:.*]], %[[A1_MAP]] -> %[[TGT_A1:.*]], %[[A1_BOXCHAR_MAP_2]] -> %[[TGT_A1_BOXCHAR:.*]], %[[A0_BOXCHAR_MAP_2]] -> %[[TGT_A0_BOXCHAR:.*]], %[[A1_BOXCHAR_MAP]] -> %[[TGT_A1_BOXCHAR2:.*]], %[[A0_BOXCHAR_MAP]] -> %[[TGT_A0_BOXCHAR2:.*]] : !fir.ref<!fir.char<1,?>>, !fir.ref<!fir.char<1,?>>, !fir.ref<!fir.boxchar<1>>, !fir.ref<!fir.boxchar<1>>, !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>, !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) { !CHECK: %[[TGT_A0_BC_LD:.*]] = fir.load %[[TGT_A0_BOXCHAR]] : !fir.ref<!fir.boxchar<1>> !CHECK: %[[TGT_A1_BC_LD:.*]] = fir.load %[[TGT_A1_BOXCHAR]] : !fir.ref<!fir.boxchar<1>> !CHECK: %[[UNBOXED_TGT_A1:.*]]:2 = fir.unboxchar %[[TGT_A1_BC_LD]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) diff --git a/flang/test/Lower/OpenMP/optional-argument-map-2.f90 b/flang/test/Lower/OpenMP/optional-argument-map-2.f90 index 791d509..a787ad7 100644 --- a/flang/test/Lower/OpenMP/optional-argument-map-2.f90 +++ b/flang/test/Lower/OpenMP/optional-argument-map-2.f90 @@ -71,11 +71,10 @@ end module mod ! CHECK-FPRIV: %[[VAL_12:.*]]:2 = fir.unboxchar %[[VAL_8]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) ! CHECK-FPRIV: %[[VAL_13:.*]] = arith.subi %[[VAL_12]]#1, %[[VAL_11]] : index ! CHECK-FPRIV: %[[VAL_14:.*]] = omp.map.bounds lower_bound(%[[VAL_10]] : index) upper_bound(%[[VAL_13]] : index) extent(%[[VAL_12]]#1 : index) stride(%[[VAL_11]] : index) start_idx(%[[VAL_10]] : index) {stride_in_bytes = true} -! CHECK-FPRIV: %[[VAL_15:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.boxchar<1>> ! CHECK-FPRIV: %[[VAL_16:.*]] = fir.box_offset %[[VAL_0]] base_addr : (!fir.ref<!fir.boxchar<1>>) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> -! CHECK-FPRIV: %[[VAL_17:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(implicit, to) capture(ByRef) var_ptr_ptr(%[[VAL_16]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%[[VAL_14]]) -> !fir.ref<!fir.boxchar<1>> -! CHECK-FPRIV: %[[VAL_18:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(to) capture(ByRef) members(%[[VAL_17]] : [0] : !fir.ref<!fir.boxchar<1>>) -> !fir.ref<!fir.boxchar<1>> -! CHECK-FPRIV: omp.target map_entries(%[[VAL_7]] -> %[[VAL_19:.*]], %[[VAL_18]] -> %[[VAL_20:.*]], %[[VAL_17]] -> %[[VAL_21:.*]] : !fir.ref<!fir.char<1,4>>, !fir.ref<!fir.boxchar<1>>, !fir.ref<!fir.boxchar<1>>) private(@_QMmodFroutine_boxcharEa_firstprivate_boxchar_c8xU %[[VAL_3]]#0 -> %[[VAL_22:.*]] [map_idx=1] : !fir.boxchar<1>) { +! CHECK-FPRIV: %[[VAL_17:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(to) capture(ByRef) var_ptr_ptr(%[[VAL_16]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%[[VAL_14]]) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> {name = ""} +! CHECK-FPRIV: %[[VAL_18:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(to) capture(ByRef) members(%[[VAL_17]] : [0] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) -> !fir.ref<!fir.boxchar<1>> +! CHECK-FPRIV: omp.target map_entries(%[[VAL_7]] -> %[[VAL_19:.*]], %[[VAL_18]] -> %[[VAL_20:.*]], %[[VAL_17]] -> %[[VAL_21:.*]] : !fir.ref<!fir.char<1,4>>, !fir.ref<!fir.boxchar<1>>, !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) private(@_QMmodFroutine_boxcharEa_firstprivate_boxchar_c8xU %[[VAL_3]]#0 -> %[[VAL_22:.*]] [map_idx=1] : !fir.boxchar<1>) { ! CHECK-FPRIV: %[[VAL_23:.*]] = arith.constant 4 : index ! CHECK-FPRIV: %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] typeparams %[[VAL_23]] {uniq_name = "_QMmodFroutine_boxcharEb"} : (!fir.ref<!fir.char<1,4>>, index) -> (!fir.ref<!fir.char<1,4>>, !fir.ref<!fir.char<1,4>>) ! CHECK-FPRIV: %[[VAL_25:.*]]:2 = fir.unboxchar %[[VAL_22]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) @@ -103,14 +102,16 @@ end module mod ! CHECK-NO-FPRIV: %[[VAL_19:.*]]:2 = fir.unboxchar %[[ARG0]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) ! CHECK-NO-FPRIV: %[[VAL_20:.*]] = arith.subi %[[VAL_19]]#1, %[[VAL_18]] : index ! CHECK-NO-FPRIV: %[[VAL_21:.*]] = omp.map.bounds lower_bound(%[[VAL_17]] : index) upper_bound(%[[VAL_20]] : index) extent(%[[VAL_19]]#1 : index) stride(%[[VAL_18]] : index) start_idx(%[[VAL_17]] : index) {stride_in_bytes = true} -! CHECK-NO-FPRIV: %[[VAL_22:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(implicit, to) capture(ByRef) bounds(%[[VAL_21]]) -> !fir.ref<!fir.boxchar<1>> {name = ""} -! CHECK-NO-FPRIV: omp.target map_entries(%[[VAL_7]] -> %[[VAL_23:.*]], %[[VAL_16]] -> %[[VAL_24:.*]], %[[VAL_22]] -> %[[VAL_25:.*]] : !fir.ref<!fir.char<1,4>>, !fir.ref<!fir.char<1,?>>, !fir.ref<!fir.boxchar<1>>) { -! CHECK-NO-FPRIV: %[[VAL_26:.*]] = fir.load %[[VAL_25]] : !fir.ref<!fir.boxchar<1>> -! CHECK-NO-FPRIV: %[[VAL_27:.*]]:2 = fir.unboxchar %[[VAL_26]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) -! CHECK-NO-FPRIV: %[[VAL_28:.*]] = arith.constant 4 : index -! CHECK-NO-FPRIV: %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_23]] typeparams %[[VAL_28]] {uniq_name = "_QMmodFroutine_boxcharEb"} : (!fir.ref<!fir.char<1,4>>, index) -> (!fir.ref<!fir.char<1,4>>, !fir.ref<!fir.char<1,4>>) -! CHECK-NO-FPRIV: %[[VAL_30:.*]]:2 = hlfir.declare %[[VAL_24]] typeparams %[[VAL_27]]#1 {fortran_attrs = #fir.var_attrs<intent_in, optional>, uniq_name = "_QMmodFroutine_boxcharEa"} : (!fir.ref<!fir.char<1,?>>, index) -> (!fir.boxchar<1>, !fir.ref<!fir.char<1,?>>) -! CHECK-NO-FPRIV: hlfir.assign %[[VAL_30]]#0 to %[[VAL_29]]#0 : !fir.boxchar<1>, !fir.ref<!fir.char<1,4>> +! CHECK-NO-FPRIV: %[[VAL_22:.*]] = fir.box_offset %[[VAL_0]] base_addr : (!fir.ref<!fir.boxchar<1>>) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> +! CHECK-NO-FPRIV: %[[VAL_23:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(implicit, to) capture(ByRef) var_ptr_ptr(%[[VAL_22]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%14) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> {name = ""} +! CHECK-NO-FPRIV: %[[VAL_24:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(implicit, to) capture(ByRef) members(%[[VAL_23]] : [0] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) -> !fir.ref<!fir.boxchar<1>> {name = ""} +! CHECK-NO-FPRIV: omp.target map_entries(%[[VAL_7]] -> %[[VAL_25:.*]], %[[VAL_16]] -> %[[VAL_26:.*]], %[[VAL_24]] -> %[[VAL_27:.*]], %[[VAL_23]] -> %[[VAL_28:.*]] : !fir.ref<!fir.char<1,4>>, !fir.ref<!fir.char<1,?>>, !fir.ref<!fir.boxchar<1>>, !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) { +! CHECK-NO-FPRIV: %[[VAL_29:.*]] = fir.load %[[VAL_27]] : !fir.ref<!fir.boxchar<1>> +! CHECK-NO-FPRIV: %[[VAL_30:.*]]:2 = fir.unboxchar %[[VAL_29]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) +! CHECK-NO-FPRIV: %[[VAL_31:.*]] = arith.constant 4 : index +! CHECK-NO-FPRIV: %[[VAL_32:.*]]:2 = hlfir.declare %[[VAL_25]] typeparams %[[VAL_31]] {uniq_name = "_QMmodFroutine_boxcharEb"} : (!fir.ref<!fir.char<1,4>>, index) -> (!fir.ref<!fir.char<1,4>>, !fir.ref<!fir.char<1,4>>) +! CHECK-NO-FPRIV: %[[VAL_33:.*]]:2 = hlfir.declare %[[VAL_26]] typeparams %[[VAL_30]]#1 {fortran_attrs = #fir.var_attrs<intent_in, optional>, uniq_name = "_QMmodFroutine_boxcharEa"} : (!fir.ref<!fir.char<1,?>>, index) -> (!fir.boxchar<1>, !fir.ref<!fir.char<1,?>>) +! CHECK-NO-FPRIV: hlfir.assign %[[VAL_33]]#0 to %[[VAL_32]]#0 : !fir.boxchar<1>, !fir.ref<!fir.char<1,4>> ! CHECK-NO-FPRIV: omp.terminator ! CHECK-NO-FPRIV: } ! CHECK-NO-FPRIV: return diff --git a/flang/test/Semantics/structconst12.f90 b/flang/test/Semantics/structconst12.f90 new file mode 100644 index 0000000..345016b --- /dev/null +++ b/flang/test/Semantics/structconst12.f90 @@ -0,0 +1,12 @@ +!RUN: %flang_fc1 -fdebug-unparse %s 2>&1 | FileCheck %s +!CHECK: TYPE(t) :: x = t(pp=f) +!CHECK-NOT: error: +interface + function f() + end +end interface +type t + procedure(f), nopass, pointer :: pp +end type +type(t) :: x = t(pp=f) +end diff --git a/flang/test/Transforms/omp-map-info-finalization.fir b/flang/test/Transforms/omp-map-info-finalization.fir index b30a2fc..5b0fd9f 100644 --- a/flang/test/Transforms/omp-map-info-finalization.fir +++ b/flang/test/Transforms/omp-map-info-finalization.fir @@ -381,11 +381,10 @@ func.func @_QPrealtest(%arg0: !fir.boxchar<1>) { // CHECK: %[[VAL_8:.*]]:2 = fir.unboxchar %[[VAL_4]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) // CHECK: %[[VAL_9:.*]] = arith.subi %[[VAL_8]]#1, %[[VAL_7]] : index // CHECK: %[[VAL_10:.*]] = omp.map.bounds lower_bound(%[[VAL_6]] : index) upper_bound(%[[VAL_9]] : index) extent(%[[VAL_8]]#1 : index) stride(%[[VAL_7]] : index) start_idx(%[[VAL_6]] : index) {stride_in_bytes = true} -// CHECK: %[[VAL_11:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.boxchar<1>> // CHECK: %[[VAL_12:.*]] = fir.box_offset %[[VAL_0]] base_addr : (!fir.ref<!fir.boxchar<1>>) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> -// CHECK: %[[VAL_13:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(implicit, to) capture(ByRef) var_ptr_ptr(%[[VAL_12]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%[[VAL_10]]) -> !fir.ref<!fir.boxchar<1>> -// CHECK: %[[VAL_14:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(to) capture(ByRef) members(%[[VAL_13]] : [0] : !fir.ref<!fir.boxchar<1>>) -> !fir.ref<!fir.boxchar<1>> -// CHECK: omp.target map_entries(%[[VAL_14]] -> %[[VAL_15:.*]], %[[VAL_13]] -> %[[VAL_16:.*]] : !fir.ref<!fir.boxchar<1>>, !fir.ref<!fir.boxchar<1>>) private(@boxchar.privatizer %[[VAL_3]]#0 -> %[[VAL_17:.*]] [map_idx=0] : !fir.boxchar<1>) { +// CHECK: %[[VAL_13:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.char<1,?>) map_clauses(to) capture(ByRef) var_ptr_ptr(%[[VAL_12]] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) bounds(%[[VAL_10]]) -> !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>> +// CHECK: %[[VAL_14:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.boxchar<1>>, !fir.boxchar<1>) map_clauses(to) capture(ByRef) members(%[[VAL_13]] : [0] : !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) -> !fir.ref<!fir.boxchar<1>> +// CHECK: omp.target map_entries(%[[VAL_14]] -> %[[VAL_15:.*]], %[[VAL_13]] -> %[[VAL_16:.*]] : !fir.ref<!fir.boxchar<1>>, !fir.llvm_ptr<!fir.ref<!fir.char<1,?>>>) private(@boxchar.privatizer %[[VAL_3]]#0 -> %[[VAL_17:.*]] [map_idx=0] : !fir.boxchar<1>) { // CHECK: %[[VAL_18:.*]]:2 = fir.unboxchar %[[VAL_17]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index) // CHECK: %[[VAL_19:.*]]:2 = hlfir.declare %[[VAL_18]]#0 typeparams %[[VAL_18]]#1 {uniq_name = "tgt_a0"} : (!fir.ref<!fir.char<1,?>>, index) -> (!fir.boxchar<1>, !fir.ref<!fir.char<1,?>>) // CHECK: omp.terminator diff --git a/libc/cmake/modules/LLVMLibCCheckCpuFeatures.cmake b/libc/cmake/modules/LLVMLibCCheckCpuFeatures.cmake index c09d475..d76f3b1 100644 --- a/libc/cmake/modules/LLVMLibCCheckCpuFeatures.cmake +++ b/libc/cmake/modules/LLVMLibCCheckCpuFeatures.cmake @@ -9,7 +9,7 @@ if(LIBC_TARGET_ARCHITECTURE_IS_X86_64) set(ALL_CPU_FEATURES SSE2 SSE4_2 AVX AVX2 AVX512F AVX512BW FMA) set(LIBC_COMPILE_OPTIONS_NATIVE -march=native) elseif(LIBC_TARGET_ARCHITECTURE_IS_AARCH64) - set(ALL_CPU_FEATURES "FullFP16") + set(ALL_CPU_FEATURES FullFP16 MOPS SVE SVE2) set(LIBC_COMPILE_OPTIONS_NATIVE -mcpu=native) endif() diff --git a/libc/cmake/modules/cpu_features/check_MOPS.cpp b/libc/cmake/modules/cpu_features/check_MOPS.cpp new file mode 100644 index 0000000..314fe9b --- /dev/null +++ b/libc/cmake/modules/cpu_features/check_MOPS.cpp @@ -0,0 +1,5 @@ +#include "src/__support/macros/properties/cpu_features.h" + +#ifndef LIBC_TARGET_CPU_HAS_MOPS +#error unsupported +#endif diff --git a/libc/cmake/modules/cpu_features/check_SVE.cpp b/libc/cmake/modules/cpu_features/check_SVE.cpp new file mode 100644 index 0000000..725f42f --- /dev/null +++ b/libc/cmake/modules/cpu_features/check_SVE.cpp @@ -0,0 +1,5 @@ +#include "src/__support/macros/properties/cpu_features.h" + +#ifndef LIBC_TARGET_CPU_HAS_SVE +#error unsupported +#endif diff --git a/libc/cmake/modules/cpu_features/check_SVE2.cpp b/libc/cmake/modules/cpu_features/check_SVE2.cpp new file mode 100644 index 0000000..37f4b4f --- /dev/null +++ b/libc/cmake/modules/cpu_features/check_SVE2.cpp @@ -0,0 +1,5 @@ +#include "src/__support/macros/properties/cpu_features.h" + +#ifndef LIBC_TARGET_CPU_HAS_SVE2 +#error unsupported +#endif diff --git a/libc/src/__support/macros/properties/cpu_features.h b/libc/src/__support/macros/properties/cpu_features.h index fc6099c..1fe20d9 100644 --- a/libc/src/__support/macros/properties/cpu_features.h +++ b/libc/src/__support/macros/properties/cpu_features.h @@ -18,6 +18,18 @@ #define LIBC_TARGET_CPU_HAS_FULLFP16 #endif +#if defined(__ARM_FEATURE_SVE) +#define LIBC_TARGET_CPU_HAS_SVE +#endif + +#if defined(__ARM_FEATURE_SVE2) +#define LIBC_TARGET_CPU_HAS_SVE2 +#endif + +#if defined(__ARM_FEATURE_MOPS) +#define LIBC_TARGET_CPU_HAS_MOPS +#endif + #if defined(__SSE2__) #define LIBC_TARGET_CPU_HAS_SSE2 #define LIBC_TARGET_CPU_HAS_FPU_FLOAT diff --git a/libcxx/test/libcxx/time/time.traits.is.clock/trait.is.clock.compile.verify.cpp b/libcxx/test/libcxx/time/time.traits/is.clock.verify.cpp index e9ad59a..f4f438d 100644 --- a/libcxx/test/libcxx/time/time.traits.is.clock/trait.is.clock.compile.verify.cpp +++ b/libcxx/test/libcxx/time/time.traits/is.clock.verify.cpp @@ -8,6 +8,18 @@ // REQUIRES: std-at-least-c++20 +// <chrono> +// +// template<class T> struct is_clock; +// template<class T> constexpr bool is_clock_v = is_clock<T>::value; + +// [time.traits.is.clock]/3: +// The behavior of a program that adds specializations for is_clock is undefined. + +// [namespace.std]/3: +// The behavior of a C++ program is undefined if it declares an explicit or partial specialization of any standard +// library variable template, except where explicitly permitted by the specification of that variable template. + #include <chrono> #include <ratio> diff --git a/libcxx/test/std/time/time.traits.is.clock/trait.is.clock.compile.pass.cpp b/libcxx/test/std/time/time.traits/is.clock.compile.pass.cpp index 4168fa7..4af29d2 100644 --- a/libcxx/test/std/time/time.traits.is.clock/trait.is.clock.compile.pass.cpp +++ b/libcxx/test/std/time/time.traits/is.clock.compile.pass.cpp @@ -8,6 +8,11 @@ // REQUIRES: std-at-least-c++20 +// <chrono> +// +// template<class T> struct is_clock; +// template<class T> constexpr bool is_clock_v = is_clock<T>::value; + #include <chrono> #include <ratio> diff --git a/lld/MachO/ICF.cpp b/lld/MachO/ICF.cpp index 7b31378..e0fc897 100644 --- a/lld/MachO/ICF.cpp +++ b/lld/MachO/ICF.cpp @@ -173,14 +173,37 @@ bool ICF::equalsConstant(const ConcatInputSection *ia, // a valid offset in the literal section. return isecA->getOffset(valueA) == isecB->getOffset(valueB) && ra.addend == rb.addend; - else { - assert(valueA == 0 && valueB == 0); - // For section relocs, we compare the content at the section offset. - return isecA->getOffset(ra.addend) == isecB->getOffset(rb.addend); - } + assert(valueA == 0 && valueB == 0); + // For section relocs, we compare the content at the section offset. + return isecA->getOffset(ra.addend) == isecB->getOffset(rb.addend); }; - return std::equal(ia->relocs.begin(), ia->relocs.end(), ib->relocs.begin(), - f); + if (!llvm::equal(ia->relocs, ib->relocs, f)) + return false; + + // Check unwind info structural compatibility: if there are symbols with + // associated unwind info, check that both sections have compatible symbol + // layouts. For simplicity, we only attempt folding when all symbols are at + // offset zero within the section (which is typically the case with + // .subsections_via_symbols.) + auto hasUnwind = [](Defined *d) { return d->unwindEntry() != nullptr; }; + const auto *itA = llvm::find_if(ia->symbols, hasUnwind); + const auto *itB = llvm::find_if(ib->symbols, hasUnwind); + if (itA == ia->symbols.end()) + return itB == ib->symbols.end(); + if (itB == ib->symbols.end()) + return false; + const Defined *da = *itA; + const Defined *db = *itB; + if (da->value != 0 || db->value != 0) + return false; + auto isZero = [](Defined *d) { return d->value == 0; }; + // Since symbols are stored in order of value, and since we have already + // checked that da/db have value zero, we just need to do the isZero check on + // the subsequent symbols. + return std::find_if_not(std::next(itA), ia->symbols.end(), isZero) == + ia->symbols.end() && + std::find_if_not(std::next(itB), ib->symbols.end(), isZero) == + ib->symbols.end(); } // Compare the "moving" parts of two ConcatInputSections -- i.e. everything not @@ -217,31 +240,19 @@ bool ICF::equalsVariable(const ConcatInputSection *ia, } return isecA->icfEqClass[icfPass % 2] == isecB->icfEqClass[icfPass % 2]; }; - if (!std::equal(ia->relocs.begin(), ia->relocs.end(), ib->relocs.begin(), f)) + if (!llvm::equal(ia->relocs, ib->relocs, f)) return false; - // If there are symbols with associated unwind info, check that the unwind - // info matches. For simplicity, we only handle the case where there are only - // symbols at offset zero within the section (which is typically the case with - // .subsections_via_symbols.) + // Compare unwind info equivalence classes. auto hasUnwind = [](Defined *d) { return d->unwindEntry() != nullptr; }; const auto *itA = llvm::find_if(ia->symbols, hasUnwind); - const auto *itB = llvm::find_if(ib->symbols, hasUnwind); if (itA == ia->symbols.end()) - return itB == ib->symbols.end(); - if (itB == ib->symbols.end()) - return false; + return true; const Defined *da = *itA; - const Defined *db = *itB; - if (da->unwindEntry()->icfEqClass[icfPass % 2] != - db->unwindEntry()->icfEqClass[icfPass % 2] || - da->value != 0 || db->value != 0) - return false; - auto isZero = [](Defined *d) { return d->value == 0; }; - return std::find_if_not(std::next(itA), ia->symbols.end(), isZero) == - ia->symbols.end() && - std::find_if_not(std::next(itB), ib->symbols.end(), isZero) == - ib->symbols.end(); + // equalsConstant() guarantees that both sections have unwind info. + const Defined *db = *llvm::find_if(ib->symbols, hasUnwind); + return da->unwindEntry()->icfEqClass[icfPass % 2] == + db->unwindEntry()->icfEqClass[icfPass % 2]; } // Find the first InputSection after BEGIN whose equivalence class differs diff --git a/lld/docs/ReleaseNotes.rst b/lld/docs/ReleaseNotes.rst index 29db1cd..60db612 100644 --- a/lld/docs/ReleaseNotes.rst +++ b/lld/docs/ReleaseNotes.rst @@ -52,5 +52,8 @@ MachO Improvements WebAssembly Improvements ------------------------ +* The ``--stack-first`` flag is now enabled by default. The old + behavior can be enabled using ``--no-stack-first``. + Fixes ##### diff --git a/lld/test/wasm/alias.s b/lld/test/wasm/alias.s index 0bb035b..83f40a8 100644 --- a/lld/test/wasm/alias.s +++ b/lld/test/wasm/alias.s @@ -24,7 +24,7 @@ _start: # CHECK-NEXT: FunctionTypes: [ 0 ] # CHECK-NEXT: - Type: MEMORY # CHECK-NEXT: Memories: -# CHECK-NEXT: - Minimum: 0x2 +# CHECK-NEXT: - Minimum: 0x1 # CHECK-NEXT: - Type: GLOBAL # CHECK-NEXT: Globals: # CHECK-NEXT: - Index: 0 @@ -32,7 +32,7 @@ _start: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory diff --git a/lld/test/wasm/bss-only.s b/lld/test/wasm/bss-only.s index 1c0500f..bec7592 100644 --- a/lld/test/wasm/bss-only.s +++ b/lld/test/wasm/bss-only.s @@ -26,13 +26,13 @@ b: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 67568 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: I32 # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 2028 +# CHECK-NEXT: Value: 66540 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory diff --git a/lld/test/wasm/build-id.test b/lld/test/wasm/build-id.test index dd6e223710..5fafd21 100644 --- a/lld/test/wasm/build-id.test +++ b/lld/test/wasm/build-id.test @@ -43,12 +43,12 @@ foo: # DEFAULT: Contents of section build_id: -# DEFAULT-NEXT: 0079 10299168 1e3c845a 3c8f80ae 2f16cc22 .).h.<.Z<.../.." -# DEFAULT-NEXT: 0089 2d +# DEFAULT-NEXT: 0079 103f86e6 3bb81959 2e99ffa9 acfed331 .?..;..Y.......1 +# DEFAULT-NEXT: 0089 3a # SHA1: Contents of section build_id: -# SHA1-NEXT: 0079 145abdda 387a9bc4 e3aed3c3 3319cd37 .Z..8z......3..7 -# SHA1-NEXT: 0089 0212237c e4 ..#|. +# SHA1-NEXT: 0079 1410ade4 e75d1c9d 71023465 03b7572f .....]..q.4e..W/ +# SHA1-NEXT: 0089 c06c5ae0 74 .lZ.t # UUID: Contents of section build_id: # UUID-NEXT: 0079 10 diff --git a/lld/test/wasm/call-indirect.s b/lld/test/wasm/call-indirect.s index 7bf39a9..64eaa59 100644 --- a/lld/test/wasm/call-indirect.s +++ b/lld/test/wasm/call-indirect.s @@ -82,13 +82,13 @@ indirect_func: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66576 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: I32 # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1032 +# CHECK-NEXT: Value: 65544 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory @@ -125,23 +125,23 @@ indirect_func: # CHECK-NEXT: Body: 42010B # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Locals: -# CHECK-NEXT: Body: 410028028088808000118080808000001A410028028488808000118180808000001A0B +# CHECK-NEXT: Body: 410028028080848000118080808000001A410028028480848000118180808000001A0B # CHECK-NEXT: - Index: 2 # CHECK-NEXT: Locals: # CHECK-NEXT: Body: 41020B # CHECK-NEXT: - Index: 3 # CHECK-NEXT: Locals: -# CHECK-NEXT: Body: 410028028888808000118180808000001A0B +# CHECK-NEXT: Body: 410028028880848000118180808000001A0B # CHECK-NEXT: - Index: 4 # CHECK-NEXT: Locals: # CHECK-NEXT: Body: 42012000118280808000001A0B # CHECK-NEXT: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: '010000000200000002000000' # CHECK-NEXT: - Type: CUSTOM # CHECK-NEXT: Name: name diff --git a/lld/test/wasm/comdats.ll b/lld/test/wasm/comdats.ll index 2dd687f..1662a98 100644 --- a/lld/test/wasm/comdats.ll +++ b/lld/test/wasm/comdats.ll @@ -23,13 +23,13 @@ entry: ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66576 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Index: 1 ; CHECK-NEXT: Type: I32 ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Type: EXPORT ; CHECK-NEXT: Exports: ; CHECK-NEXT: - Name: memory @@ -69,7 +69,7 @@ entry: ; CHECK-NEXT: Body: 1080808080001082808080001A0B ; CHECK-NEXT: - Index: 2 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4180888080000B +; CHECK-NEXT: Body: 4180808480000B ; CHECK-NEXT: - Index: 3 ; CHECK-NEXT: Locals: ; CHECK-NEXT: Body: 0B @@ -81,9 +81,9 @@ entry: ; CHECK-NEXT: Body: 4181808080000B ; CHECK-NEXT: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: '616263' diff --git a/lld/test/wasm/compress-relocs.s b/lld/test/wasm/compress-relocs.s index 41d4ff5..37f1b3b 100644 --- a/lld/test/wasm/compress-relocs.s +++ b/lld/test/wasm/compress-relocs.s @@ -47,16 +47,16 @@ test_memory_and_indirect_call_relocs: end_function # CHECK: test_memory_and_indirect_call_relocs -# CHECK: 41 90 88 80 80 00 i32.const 1040 +# CHECK: 41 90 80 84 80 00 i32.const 65552 # CHECK: 11 80 80 80 80 00 80 80 80 80 00 call_indirect 0 -# CHECK: 28 02 94 88 80 80 00 i32.load 1044 +# CHECK: 28 02 94 80 84 80 00 i32.load 65556 # CHECK: 11 81 80 80 80 00 80 80 80 80 00 call_indirect 1 # CHECK: 41 81 80 80 80 00 i32.const 1 # CHECK: 11 80 80 80 80 00 80 80 80 80 00 call_indirect 0 # COMPRESS: test_memory_and_indirect_call_relocs -# COMPRESS: 41 90 08 i32.const 1040 +# COMPRESS: 41 90 80 04 i32.const 65552 # COMPRESS: 11 00 00 call_indirect 0 -# COMPRESS: 28 02 94 08 i32.load 1044 +# COMPRESS: 28 02 94 80 04 i32.load 65556 # COMPRESS: 11 01 00 call_indirect 1 # COMPRESS: 41 01 i32.const 1 # COMPRESS: 11 00 00 call_indirect 0 @@ -91,11 +91,11 @@ test_relative_relocs: end_function # CHECK: test_relative_relocs -# CHECK: 41 90 88 80 80 00 i32.const 1040 +# CHECK: 41 90 80 84 80 00 i32.const 65552 # CHECK: 41 81 80 80 80 00 i32.const 1 # CHECK: 41 83 80 80 80 00 i32.const 3 # COMPRESS: test_relative_relocs -# COMPRESS: 41 90 08 i32.const 1040 +# COMPRESS: 41 90 80 04 i32.const 65552 # COMPRESS: 41 01 i32.const 1 # COMPRESS: 41 03 i32.const 3 diff --git a/lld/test/wasm/compress-relocs64.s b/lld/test/wasm/compress-relocs64.s index 44e7a08..f3ff646 100644 --- a/lld/test/wasm/compress-relocs64.s +++ b/lld/test/wasm/compress-relocs64.s @@ -36,12 +36,12 @@ test_memory_and_indirect_call_relocs: end_function # CHECK: test_memory_and_indirect_call_relocs -# CHECK: 42 90 88 80 80 80 80 80 80 80 00 i64.const 1040 -# CHECK: 29 03 98 88 80 80 80 80 80 80 80 00 i64.load 1048 +# CHECK: 42 90 80 84 80 80 80 80 80 80 00 i64.const 65552 +# CHECK: 29 03 98 80 84 80 80 80 80 80 80 00 i64.load 65560 # CHECK: 42 81 80 80 80 80 80 80 80 80 00 i64.const 1 # COMPRESS: test_memory_and_indirect_call_relocs -# COMPRESS: 42 90 08 i64.const 1040 -# COMPRESS: 29 03 98 08 i64.load 1048 +# COMPRESS: 42 90 80 04 i64.const 65552 +# COMPRESS: 29 03 98 80 04 i64.load 65560 # COMPRESS: 42 01 i64.const 1 .globl test_relative_relocs @@ -56,11 +56,11 @@ test_relative_relocs: end_function # CHECK: test_relative_relocs -# CHECK: 42 90 88 80 80 80 80 80 80 80 00 i64.const 1040 +# CHECK: 42 90 80 84 80 80 80 80 80 80 00 i64.const 65552 # CHECK: 42 81 80 80 80 80 80 80 80 80 00 i64.const 1 # CHECK: 42 83 80 80 80 80 80 80 80 80 00 i64.const 3 # COMPRESS: test_relative_relocs -# COMPRESS: 42 90 08 i64.const 1040 +# COMPRESS: 42 90 80 04 i64.const 65552 # COMPRESS: 42 01 i64.const 1 # COMPRESS: 42 03 i64.const 3 diff --git a/lld/test/wasm/custom-section-name.ll b/lld/test/wasm/custom-section-name.ll index 8799fbf..89cb72f 100644 --- a/lld/test/wasm/custom-section-name.ll +++ b/lld/test/wasm/custom-section-name.ll @@ -16,29 +16,29 @@ target triple = "wasm32-unknown-unknown" ; CHECK-LABEL: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: '00000000' -; CHECK-NEXT: - SectionOffset: 17 +; CHECK-NEXT: - SectionOffset: 19 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1028 +; CHECK-NEXT: Value: 65540 ; CHECK-NEXT: Content: 2A000000 -; CHECK-NEXT: - SectionOffset: 27 +; CHECK-NEXT: - SectionOffset: 30 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1032 +; CHECK-NEXT: Value: 65544 ; CHECK-NEXT: Content: '07000000' -; BSS-NEXT: - SectionOffset: 37 +; BSS-NEXT: - SectionOffset: 41 ; BSS-NEXT: InitFlags: 0 ; BSS-NEXT: Offset: ; BSS-NEXT: Opcode: I32_CONST -; BSS-NEXT: Value: 1036 +; BSS-NEXT: Value: 65548 ; BSS-NEXT: Content: '00000000' ; NO-BSS-NOT: - SectionOffset: diff --git a/lld/test/wasm/data-layout.s b/lld/test/wasm/data-layout.s index a68bc03..8df834d 100644 --- a/lld/test/wasm/data-layout.s +++ b/lld/test/wasm/data-layout.s @@ -63,33 +63,33 @@ local_struct_internal_ptr: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: [[PTR]]_CONST -# CHECK-NEXT: Value: 66624 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: [[PTR]] # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: [[PTR]]_CONST -# CHECK-NEXT: Value: 1080 +# CHECK-NEXT: Value: 65592 # CHECK-NEXT: - Index: 2 # CHECK-NEXT: Type: [[PTR]] # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: [[PTR]]_CONST -# CHECK-NEXT: Value: 66624 +# CHECK-NEXT: Value: 65600 # CHECK: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: [[PTR]]_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: 68656C6C6F0A00 -# CHECK-NEXT: - SectionOffset: 20 +# CHECK-NEXT: - SectionOffset: 22 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: [[PTR]]_CONST -# CHECK-NEXT: Value: 1040 +# CHECK-NEXT: Value: 65552 # RUN: wasm-ld -no-gc-sections --allow-undefined --no-entry \ diff --git a/lld/test/wasm/data-segment-merging.ll b/lld/test/wasm/data-segment-merging.ll index e6f3c5e..34c49e8 100644 --- a/lld/test/wasm/data-segment-merging.ll +++ b/lld/test/wasm/data-segment-merging.ll @@ -15,11 +15,11 @@ ; MERGE-LABEL: - Type: DATA ; MERGE-NEXT: Segments: -; MERGE-NEXT: - SectionOffset: 7 +; MERGE-NEXT: - SectionOffset: 8 ; MERGE-NEXT: InitFlags: 0 ; MERGE-NEXT: Offset: ; MERGE: Content: 636F6E7374616E74000000002B -; MERGE-NEXT: - SectionOffset: 26 +; MERGE-NEXT: - SectionOffset: 28 ; MERGE-NEXT: InitFlags: 0 ; MERGE-NEXT: Offset: ; MERGE: Content: 68656C6C6F00676F6F6462796500776861746576657200002A000000 @@ -41,27 +41,27 @@ ; SEPARATE-NOT: DATACOUNT ; SEPARATE-LABEL: - Type: DATA ; SEPARATE-NEXT: Segments: -; SEPARATE-NEXT: - SectionOffset: 7 +; SEPARATE-NEXT: - SectionOffset: 8 ; SEPARATE-NEXT: InitFlags: 0 ; SEPARATE-NEXT: Offset: ; SEPARATE: Content: 636F6E7374616E7400 -; SEPARATE-NEXT: - SectionOffset: 22 +; SEPARATE-NEXT: - SectionOffset: 24 ; SEPARATE-NEXT: InitFlags: 0 ; SEPARATE-NEXT: Offset: ; SEPARATE: Content: 2B -; SEPARATE-NEXT: - SectionOffset: 29 +; SEPARATE-NEXT: - SectionOffset: 32 ; SEPARATE-NEXT: InitFlags: 0 ; SEPARATE-NEXT: Offset: ; SEPARATE: Content: 68656C6C6F00 -; SEPARATE-NEXT: - SectionOffset: 41 +; SEPARATE-NEXT: - SectionOffset: 45 ; SEPARATE-NEXT: InitFlags: 0 ; SEPARATE-NEXT: Offset: ; SEPARATE: Content: 676F6F6462796500 -; SEPARATE-NEXT: - SectionOffset: 55 +; SEPARATE-NEXT: - SectionOffset: 60 ; SEPARATE-NEXT: InitFlags: 0 ; SEPARATE-NEXT: Offset: ; SEPARATE: Content: '776861746576657200' -; SEPARATE-NEXT: - SectionOffset: 70 +; SEPARATE-NEXT: - SectionOffset: 76 ; SEPARATE-NEXT: InitFlags: 0 ; SEPARATE-NEXT: Offset: ; SEPARATE: Content: 2A000000 diff --git a/lld/test/wasm/data-segments.ll b/lld/test/wasm/data-segments.ll index 6c401c4..237f428 100644 --- a/lld/test/wasm/data-segments.ll +++ b/lld/test/wasm/data-segments.ll @@ -61,20 +61,20 @@ ; ACTIVE-NEXT: Body: 0B ; ACTIVE-NEXT: - Type: DATA ; ACTIVE-NEXT: Segments: -; ACTIVE-NEXT: - SectionOffset: 7 +; ACTIVE-NEXT: - SectionOffset: 8 ; ACTIVE-NEXT: InitFlags: 0 ; ACTIVE-NEXT: Offset: ; ACTIVE32-NEXT: Opcode: I32_CONST ; ACTIVE64-NEXT: Opcode: I64_CONST -; ACTIVE-NEXT: Value: 1024 +; ACTIVE-NEXT: Value: 65536 ; ACTIVE-NEXT: Content: 636F6E7374616E74000000002B -; ACTIVE-NEXT: - SectionOffset: 26 +; ACTIVE-NEXT: - SectionOffset: 28 ; ACTIVE-NEXT: InitFlags: 0 ; ACTIVE-NEXT: Offset: ; ACTIVE32-NEXT: Opcode: I32_CONST ; ACTIVE64-NEXT: Opcode: I64_CONST -; ACTIVE-NEXT: Value: 1040 -; ACTIVE-NEXT: Content: 68656C6C6F00676F6F646279650000002A000000 +; ACTIVE-NEXT: Value: 65552 +; ACTIVE-NEXT: Content: 68656C6C6F00676F6F646279650000002A00000063000000 ; ACTIVE-NEXT: - Type: CUSTOM ; ACTIVE-NEXT: Name: name ; ACTIVE-NEXT: FunctionNames: @@ -201,7 +201,7 @@ ; DIS-NEXT: block ; DIS-NEXT: block -; NOPIC-DIS-NEXT: [[PTR]].const 11064 +; NOPIC-DIS-NEXT: [[PTR]].const 75576 ; PIC-DIS-NEXT: local.get 0 ; DIS-NEXT: i32.const 0 @@ -211,8 +211,8 @@ ; DIS-NEXT: # 2: down to label0 ; DIS-NEXT: end -; NOPIC-DIS-NEXT: [[PTR]].const 1024 -; NOPIC-DIS-NEXT: [[PTR]].const 1024 +; NOPIC-DIS-NEXT: [[PTR]].const 65536 +; NOPIC-DIS-NEXT: [[PTR]].const 65536 ; NOPIC-DIS-NEXT: global.set 1 ; PIC-DIS-NEXT: [[PTR]].const 0 ; PIC-DIS-NEXT: global.get 1 @@ -224,7 +224,7 @@ ; DIS-NEXT: i32.const 4 ; DIS-NEXT: memory.init 0, 0 -; NOPIC-DIS-NEXT: [[PTR]].const 1028 +; NOPIC-DIS-NEXT: [[PTR]].const 65540 ; PIC-DIS-NEXT: [[PTR]].const 4 ; PIC-DIS-NEXT: global.get 1 ; PIC-DIS-NEXT: [[PTR]].add @@ -233,7 +233,7 @@ ; DIS-NEXT: i32.const 13 ; DIS-NEXT: memory.init 1, 0 -; NOPIC-DIS-NEXT: [[PTR]].const 1044 +; NOPIC-DIS-NEXT: [[PTR]].const 65556 ; PIC-DIS-NEXT: [[PTR]].const 20 ; PIC-DIS-NEXT: global.get 1 ; PIC-DIS-NEXT: [[PTR]].add @@ -241,7 +241,7 @@ ; DIS-NEXT: i32.const 0 ; DIS-NEXT: i32.const 20 ; DIS-NEXT: memory.init 2, 0 -; NOPIC-DIS-NEXT: [[PTR]].const 1064 +; NOPIC-DIS-NEXT: [[PTR]].const 65576 ; PIC-DIS-NEXT: [[PTR]].const 40 ; PIC-DIS-NEXT: global.get 1 ; PIC-DIS-NEXT: [[PTR]].add @@ -249,13 +249,13 @@ ; DIS-NEXT: [[PTR]].const 10000 ; DIS-NEXT: memory.fill 0 -; NOPIC-DIS-NEXT: [[PTR]].const 11064 +; NOPIC-DIS-NEXT: [[PTR]].const 75576 ; PIC-DIS-NEXT: local.get 0 ; DIS-NEXT: i32.const 2 ; DIS-NEXT: i32.atomic.store 0 -; NOPIC-DIS-NEXT: [[PTR]].const 11064 +; NOPIC-DIS-NEXT: [[PTR]].const 75576 ; PIC-DIS-NEXT: local.get 0 ; DIS-NEXT: i32.const -1 @@ -264,7 +264,7 @@ ; DIS-NEXT: br 1 # 1: down to label1 ; DIS-NEXT: end -; NOPIC-DIS-NEXT: [[PTR]].const 11064 +; NOPIC-DIS-NEXT: [[PTR]].const 75576 ; PIC-DIS-NEXT: local.get 0 ; DIS-NEXT: i32.const 1 diff --git a/lld/test/wasm/debuginfo.test b/lld/test/wasm/debuginfo.test index 9cb1cc3..7e6bd51 100644 --- a/lld/test/wasm/debuginfo.test +++ b/lld/test/wasm/debuginfo.test @@ -50,7 +50,7 @@ CHECK-NEXT: DW_AT_type (0x000000ac "int[2]") CHECK-NEXT: DW_AT_external (true) CHECK-NEXT: DW_AT_decl_file ("{{.*}}hi_foo.c") CHECK-NEXT: DW_AT_decl_line (1) -CHECK: DW_AT_location (DW_OP_addr 0x400) +CHECK: DW_AT_location (DW_OP_addr 0x10000) CHECK: DW_TAG_array_type diff --git a/lld/test/wasm/dylink-non-pie.s b/lld/test/wasm/dylink-non-pie.s index 3157b8c..fddfddb 100755 --- a/lld/test/wasm/dylink-non-pie.s +++ b/lld/test/wasm/dylink-non-pie.s @@ -32,7 +32,7 @@ f_p: # DIS: <__wasm_apply_data_relocs>: # DIS-EMPTY: -# DIS-NEXT: i32.const 1024 +# DIS-NEXT: i32.const 65536 # DIS-NEXT: global.get 0 # DIS-NEXT: i32.store 0 # DIS-NEXT: end diff --git a/lld/test/wasm/emit-relocs.s b/lld/test/wasm/emit-relocs.s index 385344c..3df345c 100644 --- a/lld/test/wasm/emit-relocs.s +++ b/lld/test/wasm/emit-relocs.s @@ -41,11 +41,11 @@ foo: # CHECK: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: '00000000' # There should be a single relocation in this section (just the live symbol) @@ -75,5 +75,5 @@ foo: # CHECK-NEXT: Kind: DATA # CHECK-NEXT: Name: __stack_low # CHECK-NEXT: Flags: [ VISIBILITY_HIDDEN, ABSOLUTE ] -# CHECK-NEXT: Offset: 1040 # CHECK-NEXT: Size: 0 +# CHECK-NEXT: - Index: 3 diff --git a/lld/test/wasm/externref.s b/lld/test/wasm/externref.s index ffc63a6..1443e5f 100644 --- a/lld/test/wasm/externref.s +++ b/lld/test/wasm/externref.s @@ -35,7 +35,7 @@ _start: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: EXTERNREF # CHECK-NEXT: Mutable: true diff --git a/lld/test/wasm/gc-sections.ll b/lld/test/wasm/gc-sections.ll index e709ab7..69d7ed2 100644 --- a/lld/test/wasm/gc-sections.ll +++ b/lld/test/wasm/gc-sections.ll @@ -57,7 +57,7 @@ entry: ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66576 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Index: 1 ; CHECK-NEXT: Type: I64 ; CHECK-NEXT: Mutable: true @@ -67,11 +67,11 @@ entry: ; CHECK: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: '02000000' ; CHECK-NEXT: - Type: CUSTOM ; CHECK-NEXT: Name: name @@ -123,7 +123,7 @@ entry: ; NO-GC-NEXT: Mutable: true ; NO-GC-NEXT: InitExpr: ; NO-GC-NEXT: Opcode: I32_CONST -; NO-GC-NEXT: Value: 66576 +; NO-GC-NEXT: Value: 65536 ; NO-GC-NEXT: - Index: 1 ; NO-GC-NEXT: Type: I64 ; NO-GC-NEXT: Mutable: true @@ -139,11 +139,11 @@ entry: ; NO-GC: - Type: DATA ; NO-GC-NEXT: Segments: -; NO-GC-NEXT: - SectionOffset: 7 +; NO-GC-NEXT: - SectionOffset: 8 ; NO-GC-NEXT: InitFlags: 0 ; NO-GC-NEXT: Offset: ; NO-GC-NEXT: Opcode: I32_CONST -; NO-GC-NEXT: Value: 1024 +; NO-GC-NEXT: Value: 65536 ; NO-GC-NEXT: Content: '010000000000000002000000' ; NO-GC-NEXT: - Type: CUSTOM ; NO-GC-NEXT: Name: name diff --git a/lld/test/wasm/global-base.test b/lld/test/wasm/global-base.test index 0e65f0c..e84b8ec 100644 --- a/lld/test/wasm/global-base.test +++ b/lld/test/wasm/global-base.test @@ -19,19 +19,19 @@ CHECK-1024-NEXT: Type: I32 CHECK-1024-NEXT: Mutable: true CHECK-1024-NEXT: InitExpr: CHECK-1024-NEXT: Opcode: I32_CONST -CHECK-1024-NEXT: Value: 66560 +CHECK-1024-NEXT: Value: 65536 CHECK-1024-NEXT: - Index: 1 CHECK-1024-NEXT: Type: I32 CHECK-1024-NEXT: Mutable: false CHECK-1024-NEXT: InitExpr: CHECK-1024-NEXT: Opcode: I32_CONST -CHECK-1024-NEXT: Value: 1024 +CHECK-1024-NEXT: Value: 65536 CHECK-1024-NEXT: - Index: 2 CHECK-1024-NEXT: Type: I32 CHECK-1024-NEXT: Mutable: false CHECK-1024-NEXT: InitExpr: CHECK-1024-NEXT: Opcode: I32_CONST -CHECK-1024-NEXT: Value: 1024 +CHECK-1024-NEXT: Value: 65536 CHECK-1024: - Type: EXPORT CHECK-1024: - Name: __data_end @@ -50,7 +50,7 @@ CHECK-16777216-NEXT: Type: I32 CHECK-16777216-NEXT: Mutable: true CHECK-16777216-NEXT: InitExpr: CHECK-16777216-NEXT: Opcode: I32_CONST -CHECK-16777216-NEXT: Value: 16842752 +CHECK-16777216-NEXT: Value: 65536 CHECK-16777216-NEXT: - Index: 1 CHECK-16777216-NEXT: Type: I32 CHECK-16777216-NEXT: Mutable: false diff --git a/lld/test/wasm/globals.s b/lld/test/wasm/globals.s index 6e049e1..47d9ba8 100644 --- a/lld/test/wasm/globals.s +++ b/lld/test/wasm/globals.s @@ -42,7 +42,7 @@ immutable_global: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: I32 # CHECK-NEXT: Mutable: false diff --git a/lld/test/wasm/import-memory.test b/lld/test/wasm/import-memory.test index dd7066d..9b8148e 100644 --- a/lld/test/wasm/import-memory.test +++ b/lld/test/wasm/import-memory.test @@ -10,7 +10,7 @@ # CHECK-NEXT: Field: memory # CHECK-NEXT: Kind: MEMORY # CHECK-NEXT: Memory: -# CHECK-NEXT: Minimum: 0x2 +# CHECK-NEXT: Minimum: 0x1 # CHECK-NEXT: - Type: diff --git a/lld/test/wasm/init-fini.ll b/lld/test/wasm/init-fini.ll index ef2f41f..7471ebb 100644 --- a/lld/test/wasm/init-fini.ll +++ b/lld/test/wasm/init-fini.ll @@ -78,7 +78,7 @@ entry: ; CHECK-NEXT: Body: 10041005100A100F1012100F10141004100C100F10161002100E0B ; CHECK: - Index: 22 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 02404186808080004100418088808000108080808000450D00000B0B +; CHECK-NEXT: Body: 02404186808080004100418080848000108080808000450D00000B0B ; CHECK-NEXT: - Type: CUSTOM ; CHECK-NEXT: Name: name ; CHECK-NEXT: FunctionNames: diff --git a/lld/test/wasm/large-memory.test b/lld/test/wasm/large-memory.test index 5b737e4..a2888c6 100644 --- a/lld/test/wasm/large-memory.test +++ b/lld/test/wasm/large-memory.test @@ -12,7 +12,7 @@ RUN: obj2yaml %t2.wasm | FileCheck %s --check-prefixes=CHECK,CHECK-4G CHECK: - Type: MEMORY CHECK-NEXT: Memories: CHECK-NEXT: - Flags: [ HAS_MAX ] -CHECK-NEXT: Minimum: 0x2 +CHECK-NEXT: Minimum: 0x1 CHECK-2G-NEXT: Maximum: 0x8000 CHECK-4G-NEXT: Maximum: 0x10000 diff --git a/lld/test/wasm/local-symbols.ll b/lld/test/wasm/local-symbols.ll index 8faee64..6c639a8 100644 --- a/lld/test/wasm/local-symbols.ll +++ b/lld/test/wasm/local-symbols.ll @@ -45,13 +45,13 @@ entry: ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66576 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Index: 1 ; CHECK-NEXT: Type: I32 ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Type: EXPORT ; CHECK-NEXT: Exports: ; CHECK-NEXT: - Name: memory @@ -67,17 +67,17 @@ entry: ; CHECK-NEXT: Functions: ; CHECK-NEXT: - Index: 0 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4100280284888080000B +; CHECK-NEXT: Body: 4100280284808480000B ; CHECK-NEXT: - Index: 1 ; CHECK-NEXT: Locals: ; CHECK-NEXT: Body: 1080808080001A0B ; CHECK-NEXT: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: '0100000003000000' ; CHECK-NEXT: - Type: CUSTOM ; CHECK-NEXT: Name: name diff --git a/lld/test/wasm/locals-duplicate.test b/lld/test/wasm/locals-duplicate.test index 5c3135a..88819b2 100644 --- a/lld/test/wasm/locals-duplicate.test +++ b/lld/test/wasm/locals-duplicate.test @@ -26,7 +26,7 @@ ; CHECK-NEXT: Maximum: 0x7 ; CHECK-NEXT: - Type: MEMORY ; CHECK-NEXT: Memories: -; CHECK-NEXT: - Minimum: 0x2 +; CHECK-NEXT: - Minimum: 0x2 ; CHECK-NEXT: - Type: GLOBAL ; CHECK-NEXT: Globals: ; CHECK-NEXT: - Index: 0 @@ -34,19 +34,19 @@ ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66592 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Index: 1 ; CHECK-NEXT: Type: I32 ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1028 +; CHECK-NEXT: Value: 65540 ; CHECK-NEXT: - Index: 2 ; CHECK-NEXT: Type: I32 ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1036 +; CHECK-NEXT: Value: 65548 ; CHECK-NEXT: - Type: EXPORT ; CHECK-NEXT: Exports: ; CHECK-NEXT: - Name: memory @@ -119,13 +119,13 @@ ; CHECK-NEXT: Body: 41020B ; CHECK-NEXT: - Index: 3 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4180888080000B +; CHECK-NEXT: Body: 4180808480000B ; CHECK-NEXT: - Index: 4 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4184888080000B +; CHECK-NEXT: Body: 4184808480000B ; CHECK-NEXT: - Index: 5 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4188888080000B +; CHECK-NEXT: Body: 4188808480000B ; CHECK-NEXT: - Index: 6 ; CHECK-NEXT: Locals: ; CHECK-NEXT: Body: 4181808080000B @@ -146,13 +146,13 @@ ; CHECK-NEXT: Body: 41020B ; CHECK-NEXT: - Index: 12 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 418C888080000B +; CHECK-NEXT: Body: 418C808480000B ; CHECK-NEXT: - Index: 13 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4190888080000B +; CHECK-NEXT: Body: 4190808480000B ; CHECK-NEXT: - Index: 14 ; CHECK-NEXT: Locals: -; CHECK-NEXT: Body: 4194888080000B +; CHECK-NEXT: Body: 4194808480000B ; CHECK-NEXT: - Index: 15 ; CHECK-NEXT: Locals: ; CHECK-NEXT: Body: 4184808080000B @@ -164,11 +164,11 @@ ; CHECK-NEXT: Body: 4186808080000B ; CHECK-NEXT: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: '010000000100000001000000010000000100000001000000' ; CHECK-NEXT: - Type: CUSTOM ; CHECK-NEXT: Name: name diff --git a/lld/test/wasm/lto/tls.ll b/lld/test/wasm/lto/tls.ll index b61edfb..9c1642e 100644 --- a/lld/test/wasm/lto/tls.ll +++ b/lld/test/wasm/lto/tls.ll @@ -30,13 +30,13 @@ attributes #0 = { noinline nounwind optnone "target-features"="+atomics,+bulk-me ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66576 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Index: 1 ; CHECK-NEXT: Type: I32 ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK: GlobalNames: ; CHECK-NEXT: - Index: 0 diff --git a/lld/test/wasm/lto/used.ll b/lld/test/wasm/lto/used.ll index a185103..dd36259 100644 --- a/lld/test/wasm/lto/used.ll +++ b/lld/test/wasm/lto/used.ll @@ -26,11 +26,11 @@ return: ; CHECK: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: '01000000' ; CHECK: - Type: CUSTOM diff --git a/lld/test/wasm/map-file.s b/lld/test/wasm/map-file.s index 2757f50..380ab57 100644 --- a/lld/test/wasm/map-file.s +++ b/lld/test/wasm/map-file.s @@ -59,15 +59,15 @@ somezeroes: # CHECK-NEXT: - 66 b write_global # CHECK-NEXT: - 71 f {{.*}}{{/|\\}}map-file.s.tmp1.o:(_start) # CHECK-NEXT: - 71 f _start -# CHECK-NEXT: - 82 11 DATA -# CHECK-NEXT: 400 83 8 .data -# CHECK-NEXT: 400 89 8 {{.*}}{{/|\\}}map-file.s.tmp1.o:(.data.somedata) -# CHECK-NEXT: 400 89 8 somedata -# CHECK-NEXT: 408 82 4 .bss -# CHECK-NEXT: 408 0 4 {{.*}}{{/|\\}}map-file.s.tmp1.o:(.bss.somezeroes) -# CHECK-NEXT: 408 0 4 somezeroes -# CHECK-NEXT: - 93 12 CUSTOM(.debug_info) -# CHECK-NEXT: - a5 61 CUSTOM(name) +# CHECK-NEXT: - 82 12 DATA +# CHECK-NEXT: 10000 83 8 .data +# CHECK-NEXT: 10000 8a 8 {{.*}}{{/|\\}}map-file.s.tmp1.o:(.data.somedata) +# CHECK-NEXT: 10000 8a 8 somedata +# CHECK-NEXT: 10008 82 4 .bss +# CHECK-NEXT: 10008 0 4 {{.*}}{{/|\\}}map-file.s.tmp1.o:(.bss.somezeroes) +# CHECK-NEXT: 10008 0 4 somezeroes +# CHECK-NEXT: - 94 12 CUSTOM(.debug_info) +# CHECK-NEXT: - a6 61 CUSTOM(name) # RUN: not wasm-ld %t1.o -o /dev/null -Map=/ 2>&1 \ # RUN: | FileCheck -check-prefix=FAIL %s diff --git a/lld/test/wasm/memory-naming.test b/lld/test/wasm/memory-naming.test index 766d9cd5..66143c3 100644 --- a/lld/test/wasm/memory-naming.test +++ b/lld/test/wasm/memory-naming.test @@ -57,7 +57,7 @@ # CHECK-IMPORT-NEXT: Field: bar # CHECK-IMPORT-NEXT: Kind: MEMORY # CHECK-IMPORT-NEXT: Memory: -# CHECK-IMPORT-NEXT: Minimum: 0x2 +# CHECK-IMPORT-NEXT: Minimum: 0x1 # CHECK-IMPORT: - Type: EXPORT # CHECK-IMPORT-NEXT: Exports: # CHECK-IMPORT-NEXT: - Name: _start @@ -77,7 +77,7 @@ # CHECK-IMPORT-DEFAULT-NEXT: Field: foo # CHECK-IMPORT-DEFAULT-NEXT: Kind: MEMORY # CHECK-IMPORT-DEFAULT-NEXT: Memory: -# CHECK-IMPORT-DEFAULT-NEXT: Minimum: 0x2 +# CHECK-IMPORT-DEFAULT-NEXT: Minimum: 0x1 # CHECK-IMPORT-DEFAULT-NEXT: - Type: # RUN:wasm-ld --import-memory=foo,bar --export-memory=qux -o %t.both.wasm %t.start.o @@ -91,7 +91,7 @@ # CHECK-BOTH-NEXT: Field: bar # CHECK-BOTH-NEXT: Kind: MEMORY # CHECK-BOTH-NEXT: Memory: -# CHECK-BOTH-NEXT: Minimum: 0x2 +# CHECK-BOTH-NEXT: Minimum: 0x1 # CHECK-BOTH: - Type: EXPORT # CHECK-BOTH-NEXT: Exports: # CHECK-BOTH-NEXT: - Name: qux diff --git a/lld/test/wasm/merge-string.s b/lld/test/wasm/merge-string.s index a4b89abf..229f393 100644 --- a/lld/test/wasm/merge-string.s +++ b/lld/test/wasm/merge-string.s @@ -41,21 +41,21 @@ negative_addend: // COMMON-NEXT: Mutable: true // COMMON-NEXT: InitExpr: // COMMON-NEXT: Opcode: I32_CONST -// COMMON-NEXT: Value: 66576 +// COMMON-NEXT: Value: 65536 // COMMON-NEXT: - Index: 1 // COMMON-NEXT: Type: I32 // COMMON-NEXT: Mutable: false // COMMON-NEXT: InitExpr: // COMMON-NEXT: Opcode: I32_CONST -// MERGE-NEXT: Value: 1024 -// NOMERGE-NEXT: Value: 1028 +// MERGE-NEXT: Value: 65536 +// NOMERGE-NEXT: Value: 65540 // COMMON-NEXT: - Index: 2 // COMMON-NEXT: Type: I32 // COMMON-NEXT: Mutable: false // COMMON-NEXT: InitExpr: // COMMON-NEXT: Opcode: I32_CONST -// MERGE-NEXT: Value: 1025 -// NOMERGE-NEXT: Value: 1029 +// MERGE-NEXT: Value: 65537 +// NOMERGE-NEXT: Value: 65541 // COMMON-NEXT: - Type: EXPORT // COMMON-NEXT: Exports: // COMMON-NEXT: - Name: memory @@ -71,11 +71,11 @@ negative_addend: // // COMMON: - Type: DATA // COMMON-NEXT: Segments: -// COMMON-NEXT: - SectionOffset: 7 +// COMMON-NEXT: - SectionOffset: 8 // COMMON-NEXT: InitFlags: 0 // COMMON-NEXT: Offset: // COMMON-NEXT: Opcode: I32_CONST -// COMMON-NEXT: Value: 1024 +// COMMON-NEXT: Value: 65536 // MERGE-NEXT: Content: '61626300' // NOMERGE-NEXT: Content: '6162630061626300626300' diff --git a/lld/test/wasm/multi-table.s b/lld/test/wasm/multi-table.s index afe8dda..31cba9f 100644 --- a/lld/test/wasm/multi-table.s +++ b/lld/test/wasm/multi-table.s @@ -87,7 +87,7 @@ call_indirect_explicit_tables: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66576 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory @@ -127,14 +127,14 @@ call_indirect_explicit_tables: # CHECK-NEXT: Body: 42010B # CHECK-NEXT: - Index: 3 # CHECK-NEXT: Locals: [] -# CHECK-NEXT: Body: 41002802808880800011818080800083808080001A41002802848880800011828080800083808080001A0B +# CHECK-NEXT: Body: 41002802808084800011818080800083808080001A41002802848084800011828080800083808080001A0B # CHECK-NEXT: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: '0100000002000000' # CHECK-NEXT: - Type: CUSTOM # CHECK-NEXT: Name: name diff --git a/lld/test/wasm/no-strip-segment.s b/lld/test/wasm/no-strip-segment.s index e70acae..2b79ed6 100644 --- a/lld/test/wasm/no-strip-segment.s +++ b/lld/test/wasm/no-strip-segment.s @@ -47,16 +47,16 @@ grab_liba: # "greetings" section # CHECK: - Type: DATA # CHECK: Segments: -# CHECK: - SectionOffset: 7 +# CHECK: - SectionOffset: 8 # CHECK: InitFlags: 0 # CHECK: Offset: # CHECK: Opcode: I32_CONST -# CHECK: Value: 1024 +# CHECK: Value: 65536 # CHECK: Content: 68656C6C6F00776F726C6400 # "weahters" section. -# CHECK: - SectionOffset: 25 +# CHECK: - SectionOffset: 27 # CHECK: InitFlags: 0 # CHECK: Offset: # CHECK: Opcode: I32_CONST -# CHECK: Value: 1036 +# CHECK: Value: 65548 # CHECK: Content: 636C6F75647900 diff --git a/lld/test/wasm/no-tls.s b/lld/test/wasm/no-tls.s index c0786c8..c082c1e 100644 --- a/lld/test/wasm/no-tls.s +++ b/lld/test/wasm/no-tls.s @@ -28,7 +28,7 @@ _start: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # __tls_base # CHECK-NEXT: - Index: 1 diff --git a/lld/test/wasm/page-size.s b/lld/test/wasm/page-size.s index a2bf694..17850b5 100644 --- a/lld/test/wasm/page-size.s +++ b/lld/test/wasm/page-size.s @@ -19,7 +19,7 @@ foo: # CHECK-CUSTOM: - Type: MEMORY # CHECK-CUSTOM-NEXT: Memories: # CHECK-CUSTOM-NEXT: - Flags: [ HAS_PAGE_SIZE ] -# CHECK-CUSTOM-NEXT: Minimum: 0x10410 +# CHECK-CUSTOM-NEXT: Minimum: 0x10004 # CHECK-CUSTOM-NEXT: PageSize: 0x1 # RUN: llvm-objdump --disassemble-symbols=_start %t.custom.wasm | FileCheck %s --check-prefix=CHECK-CUSTOM-DIS @@ -51,7 +51,7 @@ foo: # CHECK-CUSTOM-IMPORT-NEXT: Kind: MEMORY # CHECK-CUSTOM-IMPORT-NEXT: Memory: # CHECK-CUSTOM-IMPORT-NEXT: Flags: [ HAS_PAGE_SIZE ] -# CHECK-CUSTOM-IMPORT-NEXT: Minimum: 0x10410 +# CHECK-CUSTOM-IMPORT-NEXT: Minimum: 0x10004 # CHECK-CUSTOM-IMPORT-NEXT: PageSize: 0x1 # RUN: llvm-objdump --disassemble-symbols=_start %t.custom-import.wasm | FileCheck %s --check-prefix=CHECK-CUSTOM-IMPORT-DIS diff --git a/lld/test/wasm/pic-static.ll b/lld/test/wasm/pic-static.ll index 794b721..12ac4c3 100644 --- a/lld/test/wasm/pic-static.ll +++ b/lld/test/wasm/pic-static.ll @@ -62,7 +62,7 @@ entry: ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66576 +; CHECK-NEXT: Value: 65536 ; GOT.func.ret32 ; CHECK-NEXT: - Index: 1 @@ -70,7 +70,7 @@ entry: ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1 +; CHECK-NEXT: Value: 1 ; GOT.func.missing_function ; CHECK-NEXT: - Index: 2 @@ -102,7 +102,7 @@ entry: ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; GOT.mem.ret32_ptr ; CHECK-NEXT: - Index: 6 @@ -110,7 +110,7 @@ entry: ; CHECK-NEXT: Mutable: false ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1032 +; CHECK-NEXT: Value: 65544 ; __memory_base ; CHECK-NEXT: - Index: 7 diff --git a/lld/test/wasm/reloc-relative.s b/lld/test/wasm/reloc-relative.s index fde1d1d..5aab061 100644 --- a/lld/test/wasm/reloc-relative.s +++ b/lld/test/wasm/reloc-relative.s @@ -50,40 +50,40 @@ far: # CHECK: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: 68656C6C6F0A00 -# CHECK-NEXT: - SectionOffset: 20 +# CHECK-NEXT: - SectionOffset: 22 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1031 +# CHECK-NEXT: Value: 65543 # CHECK-NEXT: Content: 000000002A000000 -# CHECK-NEXT: - SectionOffset: 34 +# CHECK-NEXT: - SectionOffset: 37 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1039 +# CHECK-NEXT: Value: 65551 # CHECK-NEXT: Content: FCFFFFFFFCFFFFFF -# CHECK-NEXT: - SectionOffset: 48 +# CHECK-NEXT: - SectionOffset: 52 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1047 +# CHECK-NEXT: Value: 65559 # CHECK-NEXT: Content: E9FFFFFFE9FFFFFF -# CHECK-NEXT: - SectionOffset: 62 +# CHECK-NEXT: - SectionOffset: 67 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1055 +# CHECK-NEXT: Value: 65567 # CHECK-NEXT: Content: '0800000008000000' -# CHECK-NEXT: - SectionOffset: 76 +# CHECK-NEXT: - SectionOffset: 82 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1063 +# CHECK-NEXT: Value: 65575 # CHECK-NEXT: Content: '15000000' diff --git a/lld/test/wasm/shared-memory-no-atomics.yaml b/lld/test/wasm/shared-memory-no-atomics.yaml index 942c690..62f4ac9 100644 --- a/lld/test/wasm/shared-memory-no-atomics.yaml +++ b/lld/test/wasm/shared-memory-no-atomics.yaml @@ -55,7 +55,7 @@ Sections: # NO-SHARED: - Type: MEMORY # NO-SHARED-NEXT: Memories: -# NO-SHARED-NEXT: - Minimum: 0x2 +# NO-SHARED-NEXT: - Minimum: 0x1 # NO-SHARED-NOT: Maximum: # SHARED: --shared-memory is disallowed by {{.*}}shared-memory-no-atomics.yaml.tmp1.o because it was not compiled with 'atomics' or 'bulk-memory' features. diff --git a/lld/test/wasm/shared-memory.yaml b/lld/test/wasm/shared-memory.yaml index 4cdbb95..b3490c8 100644 --- a/lld/test/wasm/shared-memory.yaml +++ b/lld/test/wasm/shared-memory.yaml @@ -1,16 +1,16 @@ # RUN: yaml2obj %s -o %t1.o -# RUN: wasm-ld --no-entry --shared-memory --features=atomics,bulk-memory %t1.o -o - | obj2yaml | FileCheck %s --check-prefix SHARED +# RUN: wasm-ld --no-entry --no-gc-sections --shared-memory --features=atomics,bulk-memory %t1.o -o - | obj2yaml | FileCheck %s --check-prefix SHARED -# RUN: not wasm-ld --no-entry --shared-memory --max-memory=100000 %t1.o -o - 2>&1 | FileCheck %s --check-prefix SHARED-UNALIGNED +# RUN: not wasm-ld --no-entry --no-gc-sections --shared-memory --max-memory=100000 %t1.o -o - 2>&1 | FileCheck %s --check-prefix SHARED-UNALIGNED -# RUN: not wasm-ld --no-entry --shared-memory --max-memory=131072 --features=bulk-memory %t1.o -o - 2>&1 | FileCheck %s --check-prefix SHARED-NO-ATOMICS +# RUN: not wasm-ld --no-entry --no-gc-sections --shared-memory --max-memory=131072 --features=bulk-memory %t1.o -o - 2>&1 | FileCheck %s --check-prefix SHARED-NO-ATOMICS -# RUN: not wasm-ld --no-entry --shared-memory --max-memory=131072 --features=atomics %t1.o -o - 2>&1 | FileCheck %s --check-prefix SHARED-NO-BULK-MEM +# RUN: not wasm-ld --no-entry --no-gc-sections --shared-memory --max-memory=131072 --features=atomics %t1.o -o - 2>&1 | FileCheck %s --check-prefix SHARED-NO-BULK-MEM # RUN: wasm-ld --relocatable --features=atomics %t1.o -o - | obj2yaml | FileCheck %s --check-prefix ATOMICS-RELOCATABLE -# RUN: wasm-ld --no-entry --shared-memory --max-memory=131072 --features=atomics,bulk-memory %t1.o -o - | obj2yaml | FileCheck %s --check-prefix SHARED +# XUN: wasm-ld --no-entry --no-gc-sections --shared-memory --max-memory=131072 --features=atomics,bulk-memory %t1.o -o - | obj2yaml | FileCheck %s --check-prefix SHARED --- !WASM FileHeader: @@ -22,7 +22,7 @@ Sections: Field: __linear_memory Kind: MEMORY Memory: - Minimum: 0x00000001 + Minimum: 0x00000009 - Module: env Field: __indirect_function_table Kind: TABLE diff --git a/lld/test/wasm/stack-first.test b/lld/test/wasm/stack-first.test index 46e8c6eb..91f06a4 100644 --- a/lld/test/wasm/stack-first.test +++ b/lld/test/wasm/stack-first.test @@ -5,16 +5,18 @@ ; Also test that __heap_base is still aligned with the --stack-first option. RUN: llvm-mc -filetype=obj -triple=wasm32-unknown-unknown %p/Inputs/stack-first.s -o %t.o -RUN: wasm-ld -z stack-size=512 --stack-first --export=__data_end --export=__heap_base --export=someByte -o %t.wasm %t.o + +; Check that the default is `--stack-first` +RUN: wasm-ld -z stack-size=512 --export=__data_end --export=__heap_base --export=someByte -o %t.wasm %t.o RUN: obj2yaml %t.wasm | FileCheck %s ; Check `--no-stack-first` -RUN: wasm-ld -z stack-size=512 --stack-first --no-stack-first --export=__data_end --export=__heap_base --export=someByte -o %t.wasm %t.o +RUN: wasm-ld -z stack-size=512 --no-stack-first --export=__data_end --export=__heap_base --export=someByte -o %t.wasm %t.o RUN: obj2yaml %t.wasm | FileCheck %s --check-prefix=NOT-FIRST -; Check that the default is no-stack-first -RUN: wasm-ld -z stack-size=512 --export=__data_end --export=__heap_base --export=someByte -o %t.wasm %t.o -RUN: obj2yaml %t.wasm | FileCheck %s --check-prefix=NOT-FIRST +; Check `--stack-first` +RUN: wasm-ld -z stack-size=512 --no-stack-first --stack-first --export=__data_end --export=__heap_base --export=someByte -o %t.wasm %t.o +RUN: obj2yaml %t.wasm | FileCheck %s CHECK: - Type: GLOBAL diff --git a/lld/test/wasm/startstop.ll b/lld/test/wasm/startstop.ll index e7a5c80..c22956f 100644 --- a/lld/test/wasm/startstop.ll +++ b/lld/test/wasm/startstop.ll @@ -27,19 +27,19 @@ entry: ; CHECK: - Type: DATA ; CHECK-NEXT: Segments: -; CHECK-NEXT: - SectionOffset: 7 +; CHECK-NEXT: - SectionOffset: 8 ; CHECK-NEXT: InitFlags: 0 ; CHECK-NEXT: Offset: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 1024 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: Content: 03000000040000002A0000002B000000 -; ASM: 0000006e <get_start>: +; ASM: 00000070 <get_start>: ; ASM-EMPTY: -; ASM-NEXT: 70: i32.const 1024 -; ASM-NEXT: 76: end +; ASM-NEXT: 72: i32.const 65536 +; ASM-NEXT: 78: end -; ASM: 00000077 <get_end>: +; ASM: 00000079 <get_end>: ; ASM-EMPTY: -; ASM-NEXT: 79: i32.const 1040 -; ASM-NEXT: 7f: end +; ASM-NEXT: 7b: i32.const 65552 +; ASM-NEXT: 81: end diff --git a/lld/test/wasm/table-base.s b/lld/test/wasm/table-base.s index 56fff41..92156c4 100644 --- a/lld/test/wasm/table-base.s +++ b/lld/test/wasm/table-base.s @@ -29,7 +29,7 @@ _start: # CHECK-DEFAULT-NEXT: Mutable: true # CHECK-DEFAULT-NEXT: InitExpr: # CHECK-DEFAULT-NEXT: Opcode: I32_CONST -# CHECK-DEFAULT-NEXT: Value: 66560 +# CHECK-DEFAULT-NEXT: Value: 65536 # CHECK-DEFAULT-NEXT: - Index: 1 # CHECK-DEFAULT-NEXT: Type: I32 # CHECK-DEFAULT-NEXT: Mutable: false @@ -58,7 +58,7 @@ _start: # CHECK-100-NEXT: Mutable: true # CHECK-100-NEXT: InitExpr: # CHECK-100-NEXT: Opcode: I32_CONST -# CHECK-100-NEXT: Value: 66560 +# CHECK-100-NEXT: Value: 65536 # CHECK-100-NEXT: - Index: 1 # CHECK-100-NEXT: Type: I32 # CHECK-100-NEXT: Mutable: false diff --git a/lld/test/wasm/tls-align.s b/lld/test/wasm/tls-align.s index 4fd296e..3b51165 100644 --- a/lld/test/wasm/tls-align.s +++ b/lld/test/wasm/tls-align.s @@ -65,7 +65,7 @@ tls2: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66592 +# CHECK-NEXT: Value: 65536 # __tls_base # CHECK-NEXT: - Index: 1 diff --git a/lld/test/wasm/tls-non-shared-memory-basic.s b/lld/test/wasm/tls-non-shared-memory-basic.s index 8ef0173..66ccf8f 100644 --- a/lld/test/wasm/tls-non-shared-memory-basic.s +++ b/lld/test/wasm/tls-non-shared-memory-basic.s @@ -27,11 +27,11 @@ tls1: # CHECK: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: 2B000000 # CHECK-NEXT: - Type: CUSTOM # CHECK-NOT: - Type: IMPORT diff --git a/lld/test/wasm/tls-non-shared-memory.s b/lld/test/wasm/tls-non-shared-memory.s index 04fbb62..0d73acb 100644 --- a/lld/test/wasm/tls-non-shared-memory.s +++ b/lld/test/wasm/tls-non-shared-memory.s @@ -63,38 +63,38 @@ tls1: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66576 +# CHECK-NEXT: Value: 65536 # __tls_base # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: I32 # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # GOT.data.internal.tls1 # CHECK-NEXT: - Index: 2 # CHECK-NEXT: Type: I32 # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Type: EXPORT # CHECK: - Type: DATA # .data # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: 2B000000 # .tdata -# CHECK-NEXT: - SectionOffset: 17 +# CHECK-NEXT: - SectionOffset: 19 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1028 +# CHECK-NEXT: Value: 65540 # CHECK-NEXT: Content: 2A000000 # CHECK-NEXT: - Type: CUSTOM diff --git a/lld/test/wasm/tls.s b/lld/test/wasm/tls.s index b1f47f6..21f25f5 100644 --- a/lld/test/wasm/tls.s +++ b/lld/test/wasm/tls.s @@ -98,7 +98,7 @@ tls3: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66592 +# CHECK-NEXT: Value: 65536 # __tls_base # CHECK-NEXT: - Index: 1 diff --git a/lld/test/wasm/undefined-weak-call.s b/lld/test/wasm/undefined-weak-call.s index 7490104..47775c8 100644 --- a/lld/test/wasm/undefined-weak-call.s +++ b/lld/test/wasm/undefined-weak-call.s @@ -61,7 +61,7 @@ callWeakFuncs: # CHECK-NEXT: Maximum: 0x1 # CHECK-NEXT: - Type: MEMORY # CHECK-NEXT: Memories: -# CHECK-NEXT: - Minimum: 0x2 +# CHECK-NEXT: - Minimum: 0x1 # CHECK-NEXT: - Type: GLOBAL # CHECK-NEXT: Globals: # CHECK-NEXT: - Index: 0 @@ -69,7 +69,7 @@ callWeakFuncs: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory diff --git a/lld/test/wasm/weak-alias-overide.ll b/lld/test/wasm/weak-alias-overide.ll index ca6f4bf..30bf7cf 100644 --- a/lld/test/wasm/weak-alias-overide.ll +++ b/lld/test/wasm/weak-alias-overide.ll @@ -44,7 +44,7 @@ entry: ; CHECK-NEXT: Maximum: 0x3 ; CHECK-NEXT: - Type: MEMORY ; CHECK-NEXT: Memories: -; CHECK-NEXT: - Minimum: 0x2 +; CHECK-NEXT: - Minimum: 0x1 ; CHECK-NEXT: - Type: GLOBAL ; CHECK-NEXT: Globals: ; CHECK-NEXT: - Index: 0 @@ -52,7 +52,7 @@ entry: ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66560 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Type: EXPORT ; CHECK-NEXT: Exports: ; CHECK-NEXT: - Name: memory diff --git a/lld/test/wasm/weak-alias.ll b/lld/test/wasm/weak-alias.ll index 1768b8f..86e42a7 100644 --- a/lld/test/wasm/weak-alias.ll +++ b/lld/test/wasm/weak-alias.ll @@ -41,7 +41,7 @@ entry: ; CHECK-NEXT: Maximum: 0x2 ; CHECK-NEXT: - Type: MEMORY ; CHECK-NEXT: Memories: -; CHECK-NEXT: - Minimum: 0x2 +; CHECK-NEXT: - Minimum: 0x1 ; CHECK-NEXT: - Type: GLOBAL ; CHECK-NEXT: Globals: ; CHECK-NEXT: - Index: 0 @@ -49,7 +49,7 @@ entry: ; CHECK-NEXT: Mutable: true ; CHECK-NEXT: InitExpr: ; CHECK-NEXT: Opcode: I32_CONST -; CHECK-NEXT: Value: 66560 +; CHECK-NEXT: Value: 65536 ; CHECK-NEXT: - Type: EXPORT ; CHECK-NEXT: Exports: ; CHECK-NEXT: - Name: memory diff --git a/lld/test/wasm/weak-symbols.s b/lld/test/wasm/weak-symbols.s index 165ec17..ed85851 100644 --- a/lld/test/wasm/weak-symbols.s +++ b/lld/test/wasm/weak-symbols.s @@ -48,13 +48,13 @@ _start: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66576 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Index: 1 # CHECK-NEXT: Type: I32 # CHECK-NEXT: Mutable: false # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory @@ -97,11 +97,11 @@ _start: # CHECK-NEXT: Body: 4181808080000B # CHECK-NEXT: - Type: DATA # CHECK-NEXT: Segments: -# CHECK-NEXT: - SectionOffset: 7 +# CHECK-NEXT: - SectionOffset: 8 # CHECK-NEXT: InitFlags: 0 # CHECK-NEXT: Offset: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 1024 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: Content: '01000000' # CHECK-NEXT: - Type: CUSTOM # CHECK-NEXT: Name: name diff --git a/lld/test/wasm/weak-undefined-pic.s b/lld/test/wasm/weak-undefined-pic.s index 5937380..1a3a171 100644 --- a/lld/test/wasm/weak-undefined-pic.s +++ b/lld/test/wasm/weak-undefined-pic.s @@ -45,7 +45,7 @@ _start: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # Global 'undefined_weak:foo' representing the GOT entry for foo # Unlike other internal GOT entries that need to be mutable this one # is immutable and not updated by `__wasm_apply_global_relocs` diff --git a/lld/test/wasm/weak-undefined.s b/lld/test/wasm/weak-undefined.s index e1f551d..558cac5 100644 --- a/lld/test/wasm/weak-undefined.s +++ b/lld/test/wasm/weak-undefined.s @@ -67,7 +67,7 @@ _start: # CHECK-NEXT: Maximum: 0x1 # CHECK-NEXT: - Type: MEMORY # CHECK-NEXT: Memories: -# CHECK-NEXT: - Minimum: 0x2 +# CHECK-NEXT: - Minimum: 0x1 # CHECK-NEXT: - Type: GLOBAL # CHECK-NEXT: Globals: # CHECK-NEXT: - Index: 0 @@ -75,7 +75,7 @@ _start: # CHECK-NEXT: Mutable: true # CHECK-NEXT: InitExpr: # CHECK-NEXT: Opcode: I32_CONST -# CHECK-NEXT: Value: 66560 +# CHECK-NEXT: Value: 65536 # CHECK-NEXT: - Type: EXPORT # CHECK-NEXT: Exports: # CHECK-NEXT: - Name: memory diff --git a/lld/wasm/Driver.cpp b/lld/wasm/Driver.cpp index aad8095..fac16658 100644 --- a/lld/wasm/Driver.cpp +++ b/lld/wasm/Driver.cpp @@ -595,7 +595,7 @@ static void readConfigs(opt::InputArgList &args) { ctx.arg.shlibSigCheck = !args.hasArg(OPT_no_shlib_sigcheck); ctx.arg.stripAll = args.hasArg(OPT_strip_all); ctx.arg.stripDebug = args.hasArg(OPT_strip_debug); - ctx.arg.stackFirst = args.hasFlag(OPT_stack_first, OPT_no_stack_first, false); + ctx.arg.stackFirst = args.hasFlag(OPT_stack_first, OPT_no_stack_first, true); ctx.arg.trace = args.hasArg(OPT_trace); ctx.arg.thinLTOCacheDir = args.getLastArgValue(OPT_thinlto_cache_dir); ctx.arg.thinLTOCachePolicy = CHECK( diff --git a/lld/wasm/Options.td b/lld/wasm/Options.td index 0ed6925..33ecf03 100644 --- a/lld/wasm/Options.td +++ b/lld/wasm/Options.td @@ -251,8 +251,8 @@ def no_shlib_sigcheck: FF<"no-shlib-sigcheck">, HelpText<"Do not check signatures of functions defined in shared libraries.">; defm stack_first: B<"stack-first", - "Place stack at start of linear memory", - "Place the stack after static data refion (default)">; + "Place stack at start of linear memory (default)", + "Place the stack after static data region">; def table_base: JJ<"table-base=">, HelpText<"Table offset at which to place address taken functions (Defaults to 1)">; diff --git a/lldb/bindings/interface/SBSectionDocstrings.i b/lldb/bindings/interface/SBSectionDocstrings.i index 231e9e8..9c9cb813 100644 --- a/lldb/bindings/interface/SBSectionDocstrings.i +++ b/lldb/bindings/interface/SBSectionDocstrings.i @@ -4,7 +4,7 @@ SBSection supports iteration through its subsection, represented as SBSection as well. For example, :: - for sec in exe_module: + for sec in exe_module.section_iter(): if sec.GetName() == '__TEXT': print sec break diff --git a/lldb/source/Commands/CommandObjectTarget.cpp b/lldb/source/Commands/CommandObjectTarget.cpp index 8de6521..30bca63 100644 --- a/lldb/source/Commands/CommandObjectTarget.cpp +++ b/lldb/source/Commands/CommandObjectTarget.cpp @@ -5121,6 +5121,15 @@ public: : CommandObjectParsed(interpreter, "target stop-hook delete", "Delete a stop-hook.", "target stop-hook delete [<idx>]") { + SetHelpLong( + R"( +Deletes the stop hook by index. + +At any given stop, all enabled stop hooks that pass the stop filter will +get a chance to run. That means if one stop-hook deletes another stop hook +while executing, the deleted stop hook will still fire for the stop at which +it was deleted. + )"); AddSimpleArgumentList(eArgTypeStopHookID, eArgRepeatStar); } diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionParser.cpp b/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionParser.cpp index 6bab880..75ed87b 100644 --- a/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionParser.cpp +++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionParser.cpp @@ -115,6 +115,7 @@ class ClangExpressionParser::LLDBPreprocessorCallbacks : public PPCallbacks { ClangModulesDeclVendor &m_decl_vendor; ClangPersistentVariables &m_persistent_vars; clang::SourceManager &m_source_mgr; + /// Accumulates error messages across all moduleImport calls. StreamString m_error_stream; bool m_has_errors = false; @@ -140,11 +141,12 @@ public: module.path.push_back( ConstString(component.getIdentifierInfo()->getName())); - StreamString error_stream; - ClangModulesDeclVendor::ModuleVector exported_modules; - if (!m_decl_vendor.AddModule(module, &exported_modules, m_error_stream)) + if (auto err = m_decl_vendor.AddModule(module, &exported_modules)) { m_has_errors = true; + m_error_stream.PutCString(llvm::toString(std::move(err))); + m_error_stream.PutChar('\n'); + } for (ClangModulesDeclVendor::ModuleID module : exported_modules) m_persistent_vars.AddHandLoadedClangModule(module); diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp b/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp index ff9ed9c..ad48d29 100644 --- a/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp +++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp @@ -383,10 +383,11 @@ bool ClangExpressionSourceCode::GetText( block->CalculateSymbolContext(&sc); if (sc.comp_unit) { - StreamString error_stream; - - decl_vendor->AddModulesForCompileUnit( - *sc.comp_unit, modules_for_macros, error_stream); + if (auto err = decl_vendor->AddModulesForCompileUnit( + *sc.comp_unit, modules_for_macros)) + LLDB_LOG_ERROR( + GetLog(LLDBLog::Expressions), std::move(err), + "Error while loading hand-imported modules:\n{0}"); } } } diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp b/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp index b77e269..7635e49 100644 --- a/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp +++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp @@ -92,11 +92,11 @@ public: ~ClangModulesDeclVendorImpl() override = default; - bool AddModule(const SourceModule &module, ModuleVector *exported_modules, - Stream &error_stream) override; + llvm::Error AddModule(const SourceModule &module, + ModuleVector *exported_modules) override; - bool AddModulesForCompileUnit(CompileUnit &cu, ModuleVector &exported_modules, - Stream &error_stream) override; + llvm::Error AddModulesForCompileUnit(CompileUnit &cu, + ModuleVector &exported_modules) override; uint32_t FindDecls(ConstString name, bool append, uint32_t max_matches, std::vector<CompilerDecl> &decls) override; @@ -273,16 +273,14 @@ void ClangModulesDeclVendorImpl::ReportModuleExports( exports.push_back(module); } -bool ClangModulesDeclVendorImpl::AddModule(const SourceModule &module, - ModuleVector *exported_modules, - Stream &error_stream) { +llvm::Error +ClangModulesDeclVendorImpl::AddModule(const SourceModule &module, + ModuleVector *exported_modules) { // Fail early. - if (m_compiler_instance->hadModuleLoaderFatalFailure()) { - error_stream.PutCString("error: Couldn't load a module because the module " - "loader is in a fatal state.\n"); - return false; - } + if (m_compiler_instance->hadModuleLoaderFatalFailure()) + return llvm::createStringError( + "couldn't load a module because the module loader is in a fatal state"); // Check if we've already imported this module. @@ -297,7 +295,7 @@ bool ClangModulesDeclVendorImpl::AddModule(const SourceModule &module, if (mi != m_imported_modules.end()) { if (exported_modules) ReportModuleExports(*exported_modules, mi->second); - return true; + return llvm::Error::success(); } } @@ -315,30 +313,30 @@ bool ClangModulesDeclVendorImpl::AddModule(const SourceModule &module, std::equal(sysroot_begin, sysroot_end, path_begin); // No need to inject search paths to modules in the sysroot. if (!is_system_module) { - auto error = [&]() { - error_stream.Printf("error: No module map file in %s\n", - module.search_path.AsCString()); - return false; - }; - bool is_system = true; bool is_framework = false; auto dir = HS.getFileMgr().getOptionalDirectoryRef( module.search_path.GetStringRef()); if (!dir) - return error(); + return llvm::createStringError( + "couldn't find module search path directory %s", + module.search_path.GetCString()); + auto file = HS.lookupModuleMapFile(*dir, is_framework); if (!file) - return error(); + return llvm::createStringError("couldn't find modulemap file in %s", + module.search_path.GetCString()); + if (HS.parseAndLoadModuleMapFile(*file, is_system)) - return error(); + return llvm::createStringError( + "failed to parse and load modulemap file in %s", + module.search_path.GetCString()); } } - if (!HS.lookupModule(module.path.front().GetStringRef())) { - error_stream.Printf("error: Header search couldn't locate module '%s'\n", - module.path.front().AsCString()); - return false; - } + + if (!HS.lookupModule(module.path.front().GetStringRef())) + return llvm::createStringError("header search couldn't locate module '%s'", + module.path.front().AsCString()); llvm::SmallVector<clang::IdentifierLoc, 4> clang_path; @@ -364,22 +362,29 @@ bool ClangModulesDeclVendorImpl::AddModule(const SourceModule &module, clang::Module *top_level_module = DoGetModule(clang_path.front(), false); if (!top_level_module) { + lldb_private::StreamString error_stream; diagnostic_consumer->DumpDiagnostics(error_stream); - error_stream.Printf("error: Couldn't load top-level module %s\n", - module.path.front().AsCString()); - return false; + + return llvm::createStringError(llvm::formatv( + "couldn't load top-level module {0}:\n{1}", + module.path.front().GetStringRef(), error_stream.GetString())); } clang::Module *submodule = top_level_module; for (auto &component : llvm::ArrayRef<ConstString>(module.path).drop_front()) { - submodule = submodule->findSubmodule(component.GetStringRef()); - if (!submodule) { + clang::Module *found = submodule->findSubmodule(component.GetStringRef()); + if (!found) { + lldb_private::StreamString error_stream; diagnostic_consumer->DumpDiagnostics(error_stream); - error_stream.Printf("error: Couldn't load submodule %s\n", - component.GetCString()); - return false; + + return llvm::createStringError(llvm::formatv( + "couldn't load submodule '{0}' of module '{1}':\n{2}", + component.GetStringRef(), submodule->getFullModuleName(), + error_stream.GetString())); } + + submodule = found; } // If we didn't make the submodule visible here, Clang wouldn't allow LLDB to @@ -399,10 +404,12 @@ bool ClangModulesDeclVendorImpl::AddModule(const SourceModule &module, m_enabled = true; - return true; + return llvm::Error::success(); } - return false; + return llvm::createStringError( + llvm::formatv("unknown error while loading module {0}\n", + module.path.front().GetStringRef())); } bool ClangModulesDeclVendor::LanguageSupportsClangModules( @@ -424,15 +431,18 @@ bool ClangModulesDeclVendor::LanguageSupportsClangModules( } } -bool ClangModulesDeclVendorImpl::AddModulesForCompileUnit( - CompileUnit &cu, ClangModulesDeclVendor::ModuleVector &exported_modules, - Stream &error_stream) { - if (LanguageSupportsClangModules(cu.GetLanguage())) { - for (auto &imported_module : cu.GetImportedModules()) - if (!AddModule(imported_module, &exported_modules, error_stream)) - return false; - } - return true; +llvm::Error ClangModulesDeclVendorImpl::AddModulesForCompileUnit( + CompileUnit &cu, ClangModulesDeclVendor::ModuleVector &exported_modules) { + if (!LanguageSupportsClangModules(cu.GetLanguage())) + return llvm::Error::success(); + + llvm::Error errors = llvm::Error::success(); + + for (auto &imported_module : cu.GetImportedModules()) + if (auto err = AddModule(imported_module, &exported_modules)) + errors = llvm::joinErrors(std::move(errors), std::move(err)); + + return errors; } // ClangImporter::lookupValue diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.h b/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.h index debf476..0436320 100644 --- a/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.h +++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.h @@ -45,17 +45,12 @@ public: /// If non-NULL, a pointer to a vector to populate with the ID of every /// module that is re-exported by the specified module. /// - /// \param[out] error_stream - /// A stream to populate with the output of the Clang parser when - /// it tries to load the module. - /// /// \return /// True if the module could be loaded; false if not. If the /// compiler encountered a fatal error during a previous module /// load, then this will always return false for this ModuleImporter. - virtual bool AddModule(const SourceModule &module, - ModuleVector *exported_modules, - Stream &error_stream) = 0; + virtual llvm::Error AddModule(const SourceModule &module, + ModuleVector *exported_modules) = 0; /// Add all modules referred to in a given compilation unit to the list /// of modules to search. @@ -67,18 +62,13 @@ public: /// A vector to populate with the ID of each module loaded (directly /// and via re-exports) in this way. /// - /// \param[out] error_stream - /// A stream to populate with the output of the Clang parser when - /// it tries to load the modules. - /// /// \return /// True if all modules referred to by the compilation unit could be /// loaded; false if one could not be loaded. If the compiler /// encountered a fatal error during a previous module /// load, then this will always return false for this ModuleImporter. - virtual bool AddModulesForCompileUnit(CompileUnit &cu, - ModuleVector &exported_modules, - Stream &error_stream) = 0; + virtual llvm::Error + AddModulesForCompileUnit(CompileUnit &cu, ModuleVector &exported_modules) = 0; /// Enumerate all the macros that are defined by a given set of modules /// that are already imported. diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangUserExpression.cpp b/lldb/source/Plugins/ExpressionParser/Clang/ClangUserExpression.cpp index e8d5ec3..d1feda1 100644 --- a/lldb/source/Plugins/ExpressionParser/Clang/ClangUserExpression.cpp +++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangUserExpression.cpp @@ -371,26 +371,20 @@ static void SetupDeclVendor(ExecutionContext &exe_ctx, Target *target, if (!sc.comp_unit) return; - StreamString error_stream; - ClangModulesDeclVendor::ModuleVector modules_for_macros = persistent_state->GetHandLoadedClangModules(); - if (decl_vendor->AddModulesForCompileUnit(*sc.comp_unit, modules_for_macros, - error_stream)) - return; - // Failed to load some modules, so emit the error stream as a diagnostic. - if (!error_stream.Empty()) { - // The error stream already contains several Clang diagnostics that might - // be either errors or warnings, so just print them all as one remark - // diagnostic to prevent that the message starts with "error: error:". - diagnostic_manager.PutString(lldb::eSeverityInfo, error_stream.GetString()); + auto err = + decl_vendor->AddModulesForCompileUnit(*sc.comp_unit, modules_for_macros); + if (!err) return; - } - diagnostic_manager.PutString(lldb::eSeverityError, - "Unknown error while loading modules needed for " - "current compilation unit."); + // Module load errors aren't fatal to the expression evaluator. Printing + // them as diagnostics to the console would be too noisy and misleading + // Hence just print them to the expression log. + llvm::handleAllErrors(std::move(err), [](const llvm::StringError &e) { + LLDB_LOG(GetLog(LLDBLog::Expressions), "{0}", e.getMessage()); + }); } ClangExpressionSourceCode::WrapKind ClangUserExpression::GetWrapKind() const { diff --git a/lldb/source/Target/Process.cpp b/lldb/source/Target/Process.cpp index 42ce198..69edea5 100644 --- a/lldb/source/Target/Process.cpp +++ b/lldb/source/Target/Process.cpp @@ -3256,6 +3256,7 @@ Status Process::ConnectRemote(llvm::StringRef remote_url) { if (state == eStateStopped || state == eStateCrashed) { // If we attached and actually have a process on the other end, then // this ended up being the equivalent of an attach. + SetShouldDetach(true); CompleteAttach(); // This delays passing the stopped event to listeners till diff --git a/lldb/source/Target/Target.cpp b/lldb/source/Target/Target.cpp index ae6c4f7..3b51e17 100644 --- a/lldb/source/Target/Target.cpp +++ b/lldb/source/Target/Target.cpp @@ -3207,6 +3207,11 @@ bool Target::RunStopHooks(bool at_initial_stop) { bool should_stop = false; bool requested_continue = false; + // A stop hook might get deleted while running stop hooks. + // We have to decide what that means. We will follow the rule that deleting + // a stop hook while processing these stop hooks will delete it for FUTURE + // stops but not this stop. Fortunately, copying the m_stop_hooks to the + // active_hooks list before iterating over the hooks has this effect. for (auto cur_hook_sp : active_hooks) { bool any_thread_matched = false; for (auto exc_ctx : exc_ctx_with_reasons) { diff --git a/lldb/test/API/commands/target/stop-hooks/TestStopHookScripted.py b/lldb/test/API/commands/target/stop-hooks/TestStopHookScripted.py index 954cac1..8e91781 100644 --- a/lldb/test/API/commands/target/stop-hooks/TestStopHookScripted.py +++ b/lldb/test/API/commands/target/stop-hooks/TestStopHookScripted.py @@ -48,6 +48,39 @@ class TestStopHooks(TestBase): "Got the right error", ) + def test_self_deleting(self): + """Test that we can handle a stop hook that deletes itself""" + self.script_setup() + # Run to the first breakpoint before setting the stop hook + # so we don't have to figure out where it showed up in the new + # target. + (target, process, thread, bkpt) = lldbutil.run_to_source_breakpoint( + self, "Stop here first", self.main_source_file + ) + + # Now add our stop hook and register it: + result = lldb.SBCommandReturnObject() + command = "target stop-hook add -P stop_hook.self_deleting_stop" + self.interp.HandleCommand(command, result) + self.assertCommandReturn(result, f"Added my stop hook: {result.GetError()}") + + result_str = result.GetOutput() + p = re.compile("Stop hook #([0-9]+) added.") + m = p.match(result_str) + current_stop_hook_id = m.group(1) + command = "command script add -o -f stop_hook.handle_stop_hook_id handle_id" + self.interp.HandleCommand(command, result) + self.assertCommandReturn(result, "Added my command") + + command = f"handle_id {current_stop_hook_id}" + self.interp.HandleCommand(command, result) + self.assertCommandReturn(result, "Registered my stop ID") + + # Now step the process and make sure the stop hook was deleted. + thread.StepOver() + self.interp.HandleCommand("target stop-hook list", result) + self.assertEqual(result.GetOutput().rstrip(), "No stop hooks.", "Deleted hook") + def test_stop_hooks_scripted(self): """Test that a scripted stop hook works with no specifiers""" self.stop_hooks_scripted(5, "-I false") diff --git a/lldb/test/API/commands/target/stop-hooks/stop_hook.py b/lldb/test/API/commands/target/stop-hooks/stop_hook.py index cb7a433..a41190ba 100644 --- a/lldb/test/API/commands/target/stop-hooks/stop_hook.py +++ b/lldb/test/API/commands/target/stop-hooks/stop_hook.py @@ -48,3 +48,28 @@ class bad_handle_stop: class no_handle_stop: def __init__(self, target, extra_args, dict): print("I am okay") + + +class self_deleting_stop: + def __init__(self, target, extra_args, dict): + self.target = target + + def handle_stop(self, exe_ctx, stream): + interp = exe_ctx.target.debugger.GetCommandInterpreter() + result = lldb.SBCommandReturnObject() + interp.HandleCommand("handle_id", result) + id_str = result.GetOutput().rstrip() + + command = f"target stop-hook delete {id_str}" + interp.HandleCommand(command, result) + + +stop_hook_id = 0 + + +def handle_stop_hook_id(debugger, command, exe_ctx, result, extra_args): + global stop_hook_id + if command == "": + result.AppendMessage(str(stop_hook_id)) + else: + stop_hook_id = int(command) diff --git a/lldb/test/API/functionalities/gdb_remote_client/TestConnectRemoteDetach.py b/lldb/test/API/functionalities/gdb_remote_client/TestConnectRemoteDetach.py new file mode 100644 index 0000000..4380455 --- /dev/null +++ b/lldb/test/API/functionalities/gdb_remote_client/TestConnectRemoteDetach.py @@ -0,0 +1,67 @@ +""" +Test that ConnectRemote sets ShouldDetach flag correctly. + +When connecting to a remote process that stops after connection, +the process should be marked for detach (not kill) on destruction. +""" + +import lldb +from lldbsuite.test.lldbtest import * +from lldbsuite.test.decorators import * +from lldbsuite.test.gdbclientutils import * +from lldbsuite.test.lldbgdbclient import GDBRemoteTestBase +from lldbsuite.test import lldbutil + + +class TestConnectRemoteDetach(GDBRemoteTestBase): + """Test that ConnectRemote properly sets ShouldDetach flag.""" + + class StoppedResponder(MockGDBServerResponder): + """A responder that returns a stopped process.""" + + def qfThreadInfo(self): + return "m1" + + def qsThreadInfo(self): + return "l" + + def qC(self): + return "QC1" + + def haltReason(self): + # Return that we're stopped + return "T05thread:1;" + + def cont(self): + # Stay stopped + return "T05thread:1;" + + def D(self): + # Detach packet: this is what we want to verify gets called. + return "OK" + + def k(self): + # Kill packet: this is what we want to verify doesn't get called. + raise RuntimeError("should not receive k(ill) packet") + + def test_connect_remote_sets_detach(self): + """Test that ConnectRemote to a stopped process sets ShouldDetach.""" + self.server.responder = self.StoppedResponder() + + target = self.createTarget("a.yaml") + process = self.connect(target) + + # Wait for the process to be in stopped state after connecting. + # When ConnectRemote connects to a remote process that is stopped, + # it should call SetShouldDetach(true) before CompleteAttach(). + lldbutil.expect_state_changes( + self, self.dbg.GetListener(), process, [lldb.eStateStopped] + ) + + # Now destroy the process. Because ShouldDetach was set to true + # during ConnectRemote, this should send a 'D' (detach) packet + # rather than a 'k' (kill) packet when the process is destroyed. + process.Destroy() + + # Verify that the (D)etach packet was sent. + self.assertPacketLogReceived(["D"]) diff --git a/lldb/test/API/lang/objc/modules-compile-error/Makefile b/lldb/test/API/lang/objc/modules-compile-error/Makefile deleted file mode 100644 index e031aa0..0000000 --- a/lldb/test/API/lang/objc/modules-compile-error/Makefile +++ /dev/null @@ -1,5 +0,0 @@ -OBJC_SOURCES := main.m - -CFLAGS_EXTRAS = $(MANDATORY_MODULE_BUILD_CFLAGS) -I$(BUILDDIR) -DONLY_CLANG=1 - -include Makefile.rules diff --git a/lldb/test/API/lang/objc/modules-compile-error/TestModulesCompileError.py b/lldb/test/API/lang/objc/modules-compile-error/TestModulesCompileError.py deleted file mode 100644 index 36e302be..0000000 --- a/lldb/test/API/lang/objc/modules-compile-error/TestModulesCompileError.py +++ /dev/null @@ -1,28 +0,0 @@ -import lldb -from lldbsuite.test.decorators import * -from lldbsuite.test.lldbtest import * -from lldbsuite.test import lldbutil - - -class TestCase(TestBase): - @skipIf(compiler="clang", compiler_version=["<", "11.0"]) - def test(self): - self.build() - lldbutil.run_to_source_breakpoint( - self, "// break here", lldb.SBFileSpec("main.m") - ) - - # Try importing our custom module. This will fail as LLDB won't define - # the CLANG_ONLY define when it compiles the module for the expression - # evaluator. - # Check that the error message shows file/line/column, prints the relevant - # line from the source code and mentions the module that failed to build. - self.expect( - "expr @import LLDBTestModule", - error=True, - substrs=[ - "module.h:4:1: error: use of undeclared identifier 'syntax_error_for_lldb_to_find'", - "syntax_error_for_lldb_to_find // comment that tests source printing", - "could not build module 'LLDBTestModule'", - ], - ) diff --git a/lldb/test/API/lang/objc/modules-compile-error/main.m b/lldb/test/API/lang/objc/modules-compile-error/main.m deleted file mode 100644 index 35259dd..0000000 --- a/lldb/test/API/lang/objc/modules-compile-error/main.m +++ /dev/null @@ -1,5 +0,0 @@ -@import LLDBTestModule; - -int main() { - return foo(); // break here -} diff --git a/lldb/test/API/lang/objc/modules-compile-error/module.h b/lldb/test/API/lang/objc/modules-compile-error/module.h deleted file mode 100644 index 2edd13b..0000000 --- a/lldb/test/API/lang/objc/modules-compile-error/module.h +++ /dev/null @@ -1,5 +0,0 @@ -int foo() { return 123; } - -#ifndef ONLY_CLANG -syntax_error_for_lldb_to_find // comment that tests source printing -#endif diff --git a/lldb/test/API/lang/objc/modules-compile-error/module.modulemap b/lldb/test/API/lang/objc/modules-compile-error/module.modulemap deleted file mode 100644 index 3d44faf..0000000 --- a/lldb/test/API/lang/objc/modules-compile-error/module.modulemap +++ /dev/null @@ -1 +0,0 @@ -module LLDBTestModule { header "module.h" export * } diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_CompileFailure.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_CompileFailure.test new file mode 100644 index 0000000..49ee277 --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_CompileFailure.test @@ -0,0 +1,46 @@ +## Tests the case where module compilation fails. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -DSHOULD_COMPILE=1 \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s + +#--- main.m +@import foo; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#ifndef SHOULD_COMPILE +#error "Compilation failure." +#endif + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +#--- commands.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# CHECK: Finished building Clang module foo +# CHECK: couldn't load top-level module foo: +# CHECK: While building module 'foo' imported from LLDBModulesMemoryBuffer +# CHEKC: {{.*}}sources/foo.h{{.*}}: error: "Compilation failure." +# CHECK: LLDBModulesMemoryBuffer:1:1: fatal error: could not build module 'foo' + +# CHECK: Error while loading hand-imported modules: +# CHECK: couldn't load top-level module foo: +# CHECK-NOT: Compilation failure diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_FromExpression.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_FromExpression.test new file mode 100644 index 0000000..b964e9b --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_FromExpression.test @@ -0,0 +1,54 @@ +## Tests the case where we fail to import modules from @import +## statements that are part of the expression being run. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# +# RUN: sed -i '' -e 's/foo\.h/baz\.h/' %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s + +#--- main.m +@import foo; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr @import Foo; @import Bar +expr @import foo + +# CHECK: error: while importing modules: +# CHECK-NEXT: header search couldn't locate module 'Foo' +# CHECK-NEXT: header search couldn't locate module 'Bar' +# +# CHECK: expr @import foo +# CHECK: error: while importing modules: +# CHECK-NEXT: couldn't load top-level module foo +## No mention of the previous import errors. +# CHECK-NOT: Foo +# CHECK-NOT: Bar diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidNestedSubmodule.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidNestedSubmodule.test new file mode 100644 index 0000000..1e8075d --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidNestedSubmodule.test @@ -0,0 +1,70 @@ +## Tests the case where we fail to load a submodule of a submodule. We force this +## by removing the submodule 'module qux' of 'module baz' from the modulemap. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# RUN: sed -i '' -e 's/module qux/module quz/' %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo.baz.qux; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- baz.h +struct baz {}; + +#--- qux.h +struct qux {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * + + module baz { + header "baz.h" + export * + + module qux { + header "qux.h" + export * + } + } +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: couldn't load submodule 'qux' of module 'foo.baz' + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: couldn't load submodule 'qux' of module 'foo.baz' diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidSearchPath.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidSearchPath.test new file mode 100644 index 0000000..35ba580 --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidSearchPath.test @@ -0,0 +1,59 @@ +## Tests the case where the DW_AT_LLVM_include_path of the module is invalid. +## We forces this by just removing that directory (which in our case is 'sources'). +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# +# RUN: cp %t/sources/commands.input %t/commands.input +# RUN: cp %t/sources/commands-with-log.input %t/commands-with-log.input +# RUN: rm -r %t/sources +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: couldn't find module search path directory {{.*}}sources +# NO_LOG-NOT: couldn't find module search path directory {{.*}}sources + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: couldn't find module search path directory {{.*}}sources +# LOG: couldn't find module search path directory {{.*}}sources diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidSubmodule.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidSubmodule.test new file mode 100644 index 0000000..1bfbbcf --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidSubmodule.test @@ -0,0 +1,62 @@ +## Tests the case where we fail to load a submodule. We force this by removing +## the submodule 'module baz' from the modulemap. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# RUN: sed -i '' -e 's/module baz/module qux/' %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo.baz; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- baz.h +struct baz {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * + + module baz { + header "baz.h" + export * + } +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: couldn't load submodule 'baz' of module 'foo' + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: couldn't load submodule 'baz' of module 'foo' diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidTopLevelModule.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidTopLevelModule.test new file mode 100644 index 0000000..ad181ee --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_InvalidTopLevelModule.test @@ -0,0 +1,59 @@ +## Tests the case where a module fails to load. We force this by +## replacing the contents of the 'module foo' declaration with garbage. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# RUN: sed -i '' -e 's/foo\.h/baz\.h/' %t/sources/module.modulemap +# RUN: sed -i '' -e 's/bar\.h/qux\.h/' %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: couldn't load top-level module foo +# NO_LOG-NOT: error: header + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: couldn't load top-level module foo +# LOG: error: header 'baz.h' +# LOG: couldn't load top-level module bar +# LOG: error: header 'qux.h' diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_ModulemapParsing.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_ModulemapParsing.test new file mode 100644 index 0000000..6d8e665 --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_ModulemapParsing.test @@ -0,0 +1,57 @@ +## Tests the case where the modulemap is semantically invalid and thus +## Clang fails to load it on behalf of LLDB. We force this error by +## creating a redefinition of 'module bar'. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# RUN: sed -i '' -e 's/module foo/module bar/' %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: failed to parse and load +# NO_LOG-NOT: failed to parse and load + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: failed to parse and load modulemap file in {{.*}}sources +# LOG: failed to parse and load modulemap file in {{.*}}sources diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_NoModule.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_NoModule.test new file mode 100644 index 0000000..bcb8a7d --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_NoModule.test @@ -0,0 +1,58 @@ +## Tests the case where the module LLDB is trying to load isn't +## present in the modulemap. We force this by replacing 'module foo' +## in the modulemap. +# +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# RUN: sed -i '' -e 's/module foo/module baz/' %t/sources/module.modulemap +# RUN: sed -i '' -e 's/module bar/module qux/' %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: header search couldn't locate module 'foo' +# NO_LOG-NOT: header search couldn't locate module 'bar' + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: header search couldn't locate module 'foo' +# LOG: header search couldn't locate module 'bar' diff --git a/lldb/test/Shell/Expr/TestClangModuleLoadError_NoModuleMap.test b/lldb/test/Shell/Expr/TestClangModuleLoadError_NoModuleMap.test new file mode 100644 index 0000000..57f7f16 --- /dev/null +++ b/lldb/test/Shell/Expr/TestClangModuleLoadError_NoModuleMap.test @@ -0,0 +1,53 @@ +# REQUIRES: system-darwin +# +# RUN: split-file %s %t/sources +# RUN: %clang_host -g %t/sources/main.m -fmodules -fcxx-modules \ +# RUN: -fmodule-map-file=%t/sources/module.modulemap \ +# RUN: -fmodules-cache-path=%t/ModuleCache -o %t.out +# RUN: rm %t/sources/module.modulemap +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=NO_LOG +# +# RUN: %lldb -x -o "settings set interpreter.stop-command-source-on-error false" \ +# RUN: -s %t/sources/commands-with-log.input %t.out -o exit 2>&1 | FileCheck %s --check-prefix=LOG + +#--- main.m +@import foo; +@import bar; + +int main() { __builtin_debugtrap(); } + +#--- foo.h +struct foo {}; + +#--- bar.h +struct bar {}; + +#--- module.modulemap +module foo { + header "foo.h" + export * +} + +module bar { + header "bar.h" + export * +} + +#--- commands.input +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# NO_LOG-NOT: couldn't find modulemap +# NO_LOG-NOT: couldn't find modulemap + +#--- commands-with-log.input +log enable lldb expr +run +## Make sure expression fails so the 'note' diagnostics get printed. +expr blah + +# LOG: couldn't find modulemap file in {{.*}}sources +# LOG: couldn't find modulemap file in {{.*}}sources diff --git a/lldb/test/Shell/Unwind/Inputs/call-asm.c b/lldb/test/Shell/Unwind/Inputs/call-asm.c index 30d0069..b154c1a 100644 --- a/lldb/test/Shell/Unwind/Inputs/call-asm.c +++ b/lldb/test/Shell/Unwind/Inputs/call-asm.c @@ -1,2 +1,3 @@ -int asm_main(); +int asm_main() asm("asm_main"); + int main() { return asm_main(); } diff --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst index 30b22a4..30b24b2 100644 --- a/llvm/docs/AMDGPUUsage.rst +++ b/llvm/docs/AMDGPUUsage.rst @@ -1180,6 +1180,51 @@ is conservatively correct for OpenCL. other operations within the same address space. ======================= =================================================== +Target Types +------------ + +The AMDGPU backend implements some target extension types. + +.. _amdgpu-types-named-barriers: + +Named Barriers +~~~~~~~~~~~~~~ + +Named barriers are fixed function hardware barrier objects that are available +in gfx12.5+ in addition to the traditional default barriers. + +In LLVM IR, named barriers are represented by global variables of type +``target("amdgcn.named.barrier", 0)`` in the LDS address space. Named barrier +global variables do not occupy actual LDS memory, but their lifetime and +allocation scope matches that of global variables in LDS. Programs in LLVM IR +refer to named barriers using pointers. + +The following named barrier types are supported in global variables, defined +recursively: + +* a single, standalone ``target("amdgcn.named.barrier", 0)`` +* an array of supported types +* a struct containing a single element of supported type + +.. code-block:: llvm + + @bar = addrspace(3) global target("amdgcn.named.barrier", 0) undef + @foo = addrspace(3) global [2 x target("amdgcn.named.barrier", 0)] undef + @baz = addrspace(3) global { target("amdgcn.named.barrier", 0) } undef + + ... + + %foo.i = getelementptr [2 x target("amdgcn.named.barrier", 0)], ptr addrspace(3) @foo, i32 0, i32 %i + call void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(3) %foo.i, i32 0) + +Named barrier types may not be used in ``alloca``. + +Named barriers do not have an underlying byte representation. +It is undefined behavior to use a pointer to any part of a named barrier object +as the pointer operand of a regular memory access instruction or intrinsic. +Pointers to named barrier objects are intended to be used with dedicated +intrinsics. Reading from or writing to such pointers is undefined behavior. + LLVM IR Intrinsics ------------------ diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst index bd0337f..ab085ca 100644 --- a/llvm/docs/LangRef.rst +++ b/llvm/docs/LangRef.rst @@ -20368,6 +20368,77 @@ Arguments: """""""""" The argument to this intrinsic must be a vector of floating-point values. +Vector Partial Reduction Intrinsics +----------------------------------- + +Partial reductions of vectors can be expressed using the intrinsics described in +this section. Each one reduces the concatenation of the two vector arguments +down to the number of elements of the result vector type. + +Other than the reduction operator (e.g. add, fadd), the way in which the +concatenated arguments is reduced is entirely unspecified. By their nature these +intrinsics are not expected to be useful in isolation but can instead be used to +implement the first phase of an overall reduction operation. + +The typical use case is loop vectorization where reductions are split into an +in-loop phase, where maintaining an unordered vector result is important for +performance, and an out-of-loop phase is required to calculate the final scalar +result. + +By avoiding the introduction of new ordering constraints, these intrinsics +enhance the ability to leverage a target's accumulation instructions. + +'``llvm.vector.partial.reduce.add.*``' Intrinsic +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Syntax: +""""""" +This is an overloaded intrinsic. + +:: + + declare <4 x i32> @llvm.vector.partial.reduce.add.v4i32.v4i32.v8i32(<4 x i32> %a, <8 x i32> %b) + declare <4 x i32> @llvm.vector.partial.reduce.add.v4i32.v4i32.v16i32(<4 x i32> %a, <16 x i32> %b) + declare <vscale x 4 x i32> @llvm.vector.partial.reduce.add.nxv4i32.nxv4i32.nxv8i32(<vscale x 4 x i32> %a, <vscale x 8 x i32> %b) + declare <vscale x 4 x i32> @llvm.vector.partial.reduce.add.nxv4i32.nxv4i32.nxv16i32(<vscale x 4 x i32> %a, <vscale x 16 x i32> %b) + +Arguments: +"""""""""" + +The first argument is an integer vector with the same type as the result. + +The second argument is a vector with a length that is a known integer multiple +of the result's type, while maintaining the same element type. + +'``llvm.vector.partial.reduce.fadd.*``' Intrinsic +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Syntax: +""""""" +This is an overloaded intrinsic. + +:: + + declare <4 x f32> @llvm.vector.partial.reduce.fadd.v4f32.v8f32(<4 x f32> %a, <8 x f32> %b) + declare <vscale x 4 x f32> @llvm.vector.partial.reduce.fadd.nxv4f32.nxv8f32(<vscale x 4 x f32> %a, <vscale x 8 x f32> %b) + +Arguments: +"""""""""" + +The first argument is a floating-point vector with the same type as the result. + +The second argument is a vector with a length that is a known integer multiple +of the result's type, while maintaining the same element type. + +Semantics: +"""""""""" + +As the way in which the arguments to this floating-point intrinsic are reduced +is unspecified, this intrinsic will assume floating-point reassociation and +contraction can be leveraged to implement the reduction, which may result in +variations to the results due to reordering or by lowering to different +instructions (including combining multiple instructions into a single one). + '``llvm.vector.insert``' Intrinsic ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ @@ -20741,50 +20812,6 @@ Note that it has the following implications: - If ``%cnt`` is non-zero, the return value is non-zero as well. - If ``%cnt`` is less than or equal to ``%max_lanes``, the return value is equal to ``%cnt``. -'``llvm.vector.partial.reduce.add.*``' Intrinsic -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - -Syntax: -""""""" -This is an overloaded intrinsic. - -:: - - declare <4 x i32> @llvm.vector.partial.reduce.add.v4i32.v4i32.v8i32(<4 x i32> %a, <8 x i32> %b) - declare <4 x i32> @llvm.vector.partial.reduce.add.v4i32.v4i32.v16i32(<4 x i32> %a, <16 x i32> %b) - declare <vscale x 4 x i32> @llvm.vector.partial.reduce.add.nxv4i32.nxv4i32.nxv8i32(<vscale x 4 x i32> %a, <vscale x 8 x i32> %b) - declare <vscale x 4 x i32> @llvm.vector.partial.reduce.add.nxv4i32.nxv4i32.nxv16i32(<vscale x 4 x i32> %a, <vscale x 16 x i32> %b) - -Overview: -""""""""" - -The '``llvm.vector.partial.reduce.add.*``' intrinsics reduce the -concatenation of the two vector arguments down to the number of elements of the -result vector type. - -Arguments: -"""""""""" - -The first argument is an integer vector with the same type as the result. - -The second argument is a vector with a length that is a known integer multiple -of the result's type, while maintaining the same element type. - -Semantics: -"""""""""" - -Other than the reduction operator (e.g., add) the way in which the concatenated -arguments is reduced is entirely unspecified. By their nature these intrinsics -are not expected to be useful in isolation but instead implement the first phase -of an overall reduction operation. - -The typical use case is loop vectorization where reductions are split into an -in-loop phase, where maintaining an unordered vector result is important for -performance, and an out-of-loop phase to calculate the final scalar result. - -By avoiding the introduction of new ordering constraints, these intrinsics -enhance the ability to leverage a target's accumulation instructions. - '``llvm.experimental.vector.histogram.*``' Intrinsic ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ diff --git a/llvm/include/llvm/CodeGen/ISDOpcodes.h b/llvm/include/llvm/CodeGen/ISDOpcodes.h index af60c04..b3a2ced 100644 --- a/llvm/include/llvm/CodeGen/ISDOpcodes.h +++ b/llvm/include/llvm/CodeGen/ISDOpcodes.h @@ -1516,6 +1516,7 @@ enum NodeType { PARTIAL_REDUCE_SMLA, // sext, sext PARTIAL_REDUCE_UMLA, // zext, zext PARTIAL_REDUCE_SUMLA, // sext, zext + PARTIAL_REDUCE_FMLA, // fpext, fpext // The `llvm.experimental.stackmap` intrinsic. // Operands: input chain, glue, <id>, <numShadowBytes>, [live0[, live1...]] diff --git a/llvm/include/llvm/CodeGen/RDFRegisters.h b/llvm/include/llvm/CodeGen/RDFRegisters.h index 3b7454e..4c15bf5 100644 --- a/llvm/include/llvm/CodeGen/RDFRegisters.h +++ b/llvm/include/llvm/CodeGen/RDFRegisters.h @@ -361,13 +361,6 @@ template <> struct hash<llvm::rdf::RegisterAggr> { } }; -template <> struct equal_to<llvm::rdf::RegisterAggr> { - bool operator()(const llvm::rdf::RegisterAggr &A, - const llvm::rdf::RegisterAggr &B) const { - return A == B; - } -}; - } // namespace std namespace llvm::rdf { diff --git a/llvm/include/llvm/CodeGen/SelectionDAGNodes.h b/llvm/include/llvm/CodeGen/SelectionDAGNodes.h index 1759463..cd466dc 100644 --- a/llvm/include/llvm/CodeGen/SelectionDAGNodes.h +++ b/llvm/include/llvm/CodeGen/SelectionDAGNodes.h @@ -1949,6 +1949,10 @@ LLVM_ABI bool isNullOrNullSplat(SDValue V, bool AllowUndefs = false); /// be zero. LLVM_ABI bool isOneOrOneSplat(SDValue V, bool AllowUndefs = false); +/// Return true if the value is a constant floating-point value, or a splatted +/// vector of a constant floating-point value, of 1.0 (with no undefs). +LLVM_ABI bool isOneOrOneSplatFP(SDValue V, bool AllowUndefs = false); + /// Return true if the value is a constant -1 integer or a splatted vector of a /// constant -1 integer (with no undefs). /// Does not permit build vector implicit truncation. diff --git a/llvm/include/llvm/CodeGen/TargetLowering.h b/llvm/include/llvm/CodeGen/TargetLowering.h index 2550c2b..98565f4 100644 --- a/llvm/include/llvm/CodeGen/TargetLowering.h +++ b/llvm/include/llvm/CodeGen/TargetLowering.h @@ -1679,7 +1679,7 @@ public: LegalizeAction getPartialReduceMLAAction(unsigned Opc, EVT AccVT, EVT InputVT) const { assert(Opc == ISD::PARTIAL_REDUCE_SMLA || Opc == ISD::PARTIAL_REDUCE_UMLA || - Opc == ISD::PARTIAL_REDUCE_SUMLA); + Opc == ISD::PARTIAL_REDUCE_SUMLA || Opc == ISD::PARTIAL_REDUCE_FMLA); PartialReduceActionTypes Key = {Opc, AccVT.getSimpleVT().SimpleTy, InputVT.getSimpleVT().SimpleTy}; auto It = PartialReduceMLAActions.find(Key); @@ -2793,7 +2793,7 @@ protected: void setPartialReduceMLAAction(unsigned Opc, MVT AccVT, MVT InputVT, LegalizeAction Action) { assert(Opc == ISD::PARTIAL_REDUCE_SMLA || Opc == ISD::PARTIAL_REDUCE_UMLA || - Opc == ISD::PARTIAL_REDUCE_SUMLA); + Opc == ISD::PARTIAL_REDUCE_SUMLA || Opc == ISD::PARTIAL_REDUCE_FMLA); assert(AccVT.isValid() && InputVT.isValid() && "setPartialReduceMLAAction types aren't valid"); PartialReduceActionTypes Key = {Opc, AccVT.SimpleTy, InputVT.SimpleTy}; diff --git a/llvm/include/llvm/Frontend/OpenMP/OMP.td b/llvm/include/llvm/Frontend/OpenMP/OMP.td index bebab9b..a01858f 100644 --- a/llvm/include/llvm/Frontend/OpenMP/OMP.td +++ b/llvm/include/llvm/Frontend/OpenMP/OMP.td @@ -187,6 +187,7 @@ def OMPC_DynamicAllocators : Clause<[Spelling<"dynamic_allocators">]> { let isValueOptional = true; } def OMPC_DynGroupprivate : Clause<[Spelling<"dyn_groupprivate">]> { + let clangClass = "OMPDynGroupprivateClause"; let flangClass = "OmpDynGroupprivateClause"; } def OMPC_Enter : Clause<[Spelling<"enter">]> { diff --git a/llvm/include/llvm/IR/Intrinsics.td b/llvm/include/llvm/IR/Intrinsics.td index 520130c..07aa2fa 100644 --- a/llvm/include/llvm/IR/Intrinsics.td +++ b/llvm/include/llvm/IR/Intrinsics.td @@ -2813,6 +2813,10 @@ def int_vector_partial_reduce_add : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [IntrNoMem, IntrSpeculatable]>; +def int_vector_partial_reduce_fadd : DefaultAttrsIntrinsic<[LLVMMatchType<0>], + [llvm_anyfloat_ty, llvm_anyfloat_ty], + [IntrNoMem]>; + //===----------------- Pointer Authentication Intrinsics ------------------===// // diff --git a/llvm/include/llvm/IR/IntrinsicsAArch64.td b/llvm/include/llvm/IR/IntrinsicsAArch64.td index b81edc3..4cab6e0 100644 --- a/llvm/include/llvm/IR/IntrinsicsAArch64.td +++ b/llvm/include/llvm/IR/IntrinsicsAArch64.td @@ -499,6 +499,7 @@ let TargetPrefix = "aarch64" in { def int_aarch64_neon_ummla : AdvSIMD_MatMul_Intrinsic; def int_aarch64_neon_smmla : AdvSIMD_MatMul_Intrinsic; def int_aarch64_neon_usmmla : AdvSIMD_MatMul_Intrinsic; + def int_aarch64_neon_fmmla : AdvSIMD_MatMul_Intrinsic; def int_aarch64_neon_usdot : AdvSIMD_Dot_Intrinsic; def int_aarch64_neon_bfdot : AdvSIMD_Dot_Intrinsic; def int_aarch64_neon_bfmmla diff --git a/llvm/include/llvm/Target/TargetSelectionDAG.td b/llvm/include/llvm/Target/TargetSelectionDAG.td index 07a858f..a9750a5 100644 --- a/llvm/include/llvm/Target/TargetSelectionDAG.td +++ b/llvm/include/llvm/Target/TargetSelectionDAG.td @@ -527,6 +527,8 @@ def partial_reduce_smla : SDNode<"ISD::PARTIAL_REDUCE_SMLA", SDTPartialReduceMLA>; def partial_reduce_sumla : SDNode<"ISD::PARTIAL_REDUCE_SUMLA", SDTPartialReduceMLA>; +def partial_reduce_fmla : SDNode<"ISD::PARTIAL_REDUCE_FMLA", + SDTPartialReduceMLA>; def fadd : SDNode<"ISD::FADD" , SDTFPBinOp, [SDNPCommutative]>; def fsub : SDNode<"ISD::FSUB" , SDTFPBinOp>; diff --git a/llvm/lib/Analysis/DependenceAnalysis.cpp b/llvm/lib/Analysis/DependenceAnalysis.cpp index e45d1f7..b3b62cf 100644 --- a/llvm/lib/Analysis/DependenceAnalysis.cpp +++ b/llvm/lib/Analysis/DependenceAnalysis.cpp @@ -407,9 +407,10 @@ static void dumpExampleDependence(raw_ostream &OS, DependenceInfo *DA, continue; Value *Ptr = getLoadStorePointerOperand(&Inst); const Loop *L = LI.getLoopFor(Inst.getParent()); + const Loop *OutermostLoop = L ? L->getOutermostLoop() : nullptr; const SCEV *PtrSCEV = SE.getSCEVAtScope(Ptr, L); const SCEV *AccessFn = SE.removePointerBase(PtrSCEV); - SCEVMonotonicity Mon = Checker.checkMonotonicity(AccessFn, L); + SCEVMonotonicity Mon = Checker.checkMonotonicity(AccessFn, OutermostLoop); OS.indent(2) << "Inst: " << Inst << "\n"; OS.indent(4) << "Expr: " << *AccessFn << "\n"; Mon.print(OS, 4); @@ -945,6 +946,8 @@ SCEVMonotonicity SCEVMonotonicityChecker::invariantOrUnknown(const SCEV *Expr) { SCEVMonotonicity SCEVMonotonicityChecker::checkMonotonicity(const SCEV *Expr, const Loop *OutermostLoop) { + assert((!OutermostLoop || OutermostLoop->isOutermost()) && + "OutermostLoop must be outermost"); assert(Expr->getType()->isIntegerTy() && "Expr must be integer type"); this->OutermostLoop = OutermostLoop; return visit(Expr); diff --git a/llvm/lib/CAS/ObjectStore.cpp b/llvm/lib/CAS/ObjectStore.cpp index 3110577..c3f7a0c 100644 --- a/llvm/lib/CAS/ObjectStore.cpp +++ b/llvm/lib/CAS/ObjectStore.cpp @@ -213,10 +213,13 @@ Expected<ObjectRef> ObjectStore::importObject(ObjectStore &Upstream, // Remove the current node and its IDs from the stack. PrimaryRefStack.truncate(PrimaryRefStack.size() - Cur.RefsCount); - CursorStack.pop_back(); + // Push new node into created objects. PrimaryRefStack.push_back(*NewNode); CreatedObjects.try_emplace(Cur.Ref, *NewNode); + + // Pop the cursor in the end after all uses. + CursorStack.pop_back(); continue; } diff --git a/llvm/lib/CAS/UnifiedOnDiskCache.cpp b/llvm/lib/CAS/UnifiedOnDiskCache.cpp index ae9d818..7b790bb 100644 --- a/llvm/lib/CAS/UnifiedOnDiskCache.cpp +++ b/llvm/lib/CAS/UnifiedOnDiskCache.cpp @@ -174,7 +174,7 @@ getAllDBDirs(StringRef Path, bool IncludeCorrupt = false) { return createFileError(Path, EC); llvm::sort(FoundDBDirs, [](const DBDir &LHS, const DBDir &RHS) -> bool { - return LHS.Order <= RHS.Order; + return LHS.Order < RHS.Order; }); SmallVector<std::string, 4> DBDirs; diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp index 10daca5..f144f17 100644 --- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp @@ -2042,6 +2042,7 @@ SDValue DAGCombiner::visit(SDNode *N) { case ISD::PARTIAL_REDUCE_SMLA: case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: return visitPARTIAL_REDUCE_MLA(N); case ISD::VECTOR_COMPRESS: return visitVECTOR_COMPRESS(N); case ISD::LIFETIME_END: return visitLIFETIME_END(N); @@ -13006,6 +13007,9 @@ SDValue DAGCombiner::visitPARTIAL_REDUCE_MLA(SDNode *N) { // // partial_reduce_*mla(acc, mul(ext(x), splat(C)), splat(1)) // -> partial_reduce_*mla(acc, x, C) +// +// partial_reduce_fmla(acc, fmul(fpext(a), fpext(b)), splat(1.0)) +// -> partial_reduce_fmla(acc, a, b) SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { SDLoc DL(N); auto *Context = DAG.getContext(); @@ -13014,7 +13018,7 @@ SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { SDValue Op2 = N->getOperand(2); unsigned Opc = Op1->getOpcode(); - if (Opc != ISD::MUL && Opc != ISD::SHL) + if (Opc != ISD::MUL && Opc != ISD::FMUL && Opc != ISD::SHL) return SDValue(); SDValue LHS = Op1->getOperand(0); @@ -13033,13 +13037,16 @@ SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { Opc = ISD::MUL; } - APInt C; - if (Opc != ISD::MUL || !ISD::isConstantSplatVector(Op2.getNode(), C) || - !C.isOne()) + if (!(Opc == ISD::MUL && llvm::isOneOrOneSplat(Op2)) && + !(Opc == ISD::FMUL && llvm::isOneOrOneSplatFP(Op2))) return SDValue(); + auto IsIntOrFPExtOpcode = [](unsigned int Opcode) { + return (ISD::isExtOpcode(Opcode) || Opcode == ISD::FP_EXTEND); + }; + unsigned LHSOpcode = LHS->getOpcode(); - if (!ISD::isExtOpcode(LHSOpcode)) + if (!IsIntOrFPExtOpcode(LHSOpcode)) return SDValue(); SDValue LHSExtOp = LHS->getOperand(0); @@ -13047,6 +13054,7 @@ SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { // partial_reduce_*mla(acc, mul(ext(x), splat(C)), splat(1)) // -> partial_reduce_*mla(acc, x, C) + APInt C; if (ISD::isConstantSplatVector(RHS.getNode(), C)) { // TODO: Make use of partial_reduce_sumla here APInt CTrunc = C.trunc(LHSExtOpVT.getScalarSizeInBits()); @@ -13071,7 +13079,7 @@ SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { } unsigned RHSOpcode = RHS->getOpcode(); - if (!ISD::isExtOpcode(RHSOpcode)) + if (!IsIntOrFPExtOpcode(RHSOpcode)) return SDValue(); SDValue RHSExtOp = RHS->getOperand(0); @@ -13088,6 +13096,8 @@ SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { else if (LHSOpcode == ISD::ZERO_EXTEND && RHSOpcode == ISD::SIGN_EXTEND) { NewOpc = ISD::PARTIAL_REDUCE_SUMLA; std::swap(LHSExtOp, RHSExtOp); + } else if (LHSOpcode == ISD::FP_EXTEND && RHSOpcode == ISD::FP_EXTEND) { + NewOpc = ISD::PARTIAL_REDUCE_FMLA; } else return SDValue(); // For a 2-stage extend the signedness of both of the extends must match @@ -13115,30 +13125,33 @@ SDValue DAGCombiner::foldPartialReduceMLAMulOp(SDNode *N) { // -> partial.reduce.smla(acc, op, splat(trunc(1))) // partial.reduce.sumla(acc, sext(op), splat(1)) // -> partial.reduce.smla(acc, op, splat(trunc(1))) +// partial.reduce.fmla(acc, fpext(op), splat(1.0)) +// -> partial.reduce.fmla(acc, op, splat(1.0)) SDValue DAGCombiner::foldPartialReduceAdd(SDNode *N) { SDLoc DL(N); SDValue Acc = N->getOperand(0); SDValue Op1 = N->getOperand(1); SDValue Op2 = N->getOperand(2); - APInt ConstantOne; - if (!ISD::isConstantSplatVector(Op2.getNode(), ConstantOne) || - !ConstantOne.isOne()) + if (!llvm::isOneOrOneSplat(Op2) && !llvm::isOneOrOneSplatFP(Op2)) return SDValue(); unsigned Op1Opcode = Op1.getOpcode(); - if (!ISD::isExtOpcode(Op1Opcode)) + if (!ISD::isExtOpcode(Op1Opcode) && Op1Opcode != ISD::FP_EXTEND) return SDValue(); - bool Op1IsSigned = Op1Opcode == ISD::SIGN_EXTEND; + bool Op1IsSigned = + Op1Opcode == ISD::SIGN_EXTEND || Op1Opcode == ISD::FP_EXTEND; bool NodeIsSigned = N->getOpcode() != ISD::PARTIAL_REDUCE_UMLA; EVT AccElemVT = Acc.getValueType().getVectorElementType(); if (Op1IsSigned != NodeIsSigned && Op1.getValueType().getVectorElementType() != AccElemVT) return SDValue(); - unsigned NewOpcode = - Op1IsSigned ? ISD::PARTIAL_REDUCE_SMLA : ISD::PARTIAL_REDUCE_UMLA; + unsigned NewOpcode = N->getOpcode() == ISD::PARTIAL_REDUCE_FMLA + ? ISD::PARTIAL_REDUCE_FMLA + : Op1IsSigned ? ISD::PARTIAL_REDUCE_SMLA + : ISD::PARTIAL_REDUCE_UMLA; SDValue UnextOp1 = Op1.getOperand(0); EVT UnextOp1VT = UnextOp1.getValueType(); @@ -13148,8 +13161,12 @@ SDValue DAGCombiner::foldPartialReduceAdd(SDNode *N) { TLI.getTypeToTransformTo(*Context, UnextOp1VT))) return SDValue(); + SDValue Constant = N->getOpcode() == ISD::PARTIAL_REDUCE_FMLA + ? DAG.getConstantFP(1, DL, UnextOp1VT) + : DAG.getConstant(1, DL, UnextOp1VT); + return DAG.getNode(NewOpcode, DL, N->getValueType(0), Acc, UnextOp1, - DAG.getConstant(1, DL, UnextOp1VT)); + Constant); } SDValue DAGCombiner::visitVP_STRIDED_LOAD(SDNode *N) { diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp index 8e423c4..94751be5 100644 --- a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp @@ -534,6 +534,7 @@ SDValue VectorLegalizer::LegalizeOp(SDValue Op) { case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SMLA: case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: Action = TLI.getPartialReduceMLAAction(Op.getOpcode(), Node->getValueType(0), Node->getOperand(1).getValueType()); @@ -1243,6 +1244,7 @@ void VectorLegalizer::Expand(SDNode *Node, SmallVectorImpl<SDValue> &Results) { case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SMLA: case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: Results.push_back(TLI.expandPartialReduceMLA(Node, DAG)); return; case ISD::VECREDUCE_SEQ_FADD: diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp index bb4a8d9..dd5c011 100644 --- a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp @@ -1474,6 +1474,7 @@ void DAGTypeLegalizer::SplitVectorResult(SDNode *N, unsigned ResNo) { case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SMLA: case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: SplitVecRes_PARTIAL_REDUCE_MLA(N, Lo, Hi); break; case ISD::GET_ACTIVE_LANE_MASK: @@ -3689,6 +3690,7 @@ bool DAGTypeLegalizer::SplitVectorOperand(SDNode *N, unsigned OpNo) { case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SMLA: case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: Res = SplitVecOp_PARTIAL_REDUCE_MLA(N); break; } diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp index 0a06752..bbc1d73 100644 --- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp @@ -8404,7 +8404,8 @@ SDValue SelectionDAG::getNode(unsigned Opcode, const SDLoc &DL, EVT VT, } case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SMLA: - case ISD::PARTIAL_REDUCE_SUMLA: { + case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: { [[maybe_unused]] EVT AccVT = N1.getValueType(); [[maybe_unused]] EVT Input1VT = N2.getValueType(); [[maybe_unused]] EVT Input2VT = N3.getValueType(); @@ -13064,6 +13065,11 @@ bool llvm::isOneOrOneSplat(SDValue N, bool AllowUndefs) { return C && C->isOne(); } +bool llvm::isOneOrOneSplatFP(SDValue N, bool AllowUndefs) { + ConstantFPSDNode *C = isConstOrConstSplatFP(N, AllowUndefs); + return C && C->isExactlyValue(1.0); +} + bool llvm::isAllOnesOrAllOnesSplat(SDValue N, bool AllowUndefs) { N = peekThroughBitcasts(N); unsigned BitWidth = N.getScalarValueSizeInBits(); diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp index 2f598b2..88b0809 100644 --- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp @@ -8187,6 +8187,14 @@ void SelectionDAGBuilder::visitIntrinsicCall(const CallInst &I, Input, DAG.getConstant(1, sdl, Input.getValueType()))); return; } + case Intrinsic::vector_partial_reduce_fadd: { + SDValue Acc = getValue(I.getOperand(0)); + SDValue Input = getValue(I.getOperand(1)); + setValue(&I, DAG.getNode( + ISD::PARTIAL_REDUCE_FMLA, sdl, Acc.getValueType(), Acc, + Input, DAG.getConstantFP(1.0, sdl, Input.getValueType()))); + return; + } case Intrinsic::experimental_cttz_elts: { auto DL = getCurSDLoc(); SDValue Op = getValue(I.getOperand(0)); diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp index d3e1628..ec5edd5 100644 --- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp @@ -590,6 +590,8 @@ std::string SDNode::getOperationName(const SelectionDAG *G) const { return "partial_reduce_smla"; case ISD::PARTIAL_REDUCE_SUMLA: return "partial_reduce_sumla"; + case ISD::PARTIAL_REDUCE_FMLA: + return "partial_reduce_fmla"; case ISD::LOOP_DEPENDENCE_WAR_MASK: return "loop_dep_war"; case ISD::LOOP_DEPENDENCE_RAW_MASK: diff --git a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp index 9bdf822..b51d664 100644 --- a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp @@ -12074,22 +12074,32 @@ SDValue TargetLowering::expandPartialReduceMLA(SDNode *N, EVT::getVectorVT(*DAG.getContext(), AccVT.getVectorElementType(), MulOpVT.getVectorElementCount()); - unsigned ExtOpcLHS = N->getOpcode() == ISD::PARTIAL_REDUCE_UMLA - ? ISD::ZERO_EXTEND - : ISD::SIGN_EXTEND; - unsigned ExtOpcRHS = N->getOpcode() == ISD::PARTIAL_REDUCE_SMLA - ? ISD::SIGN_EXTEND - : ISD::ZERO_EXTEND; + unsigned ExtOpcLHS, ExtOpcRHS; + switch (N->getOpcode()) { + default: + llvm_unreachable("Unexpected opcode"); + case ISD::PARTIAL_REDUCE_UMLA: + ExtOpcLHS = ExtOpcRHS = ISD::ZERO_EXTEND; + break; + case ISD::PARTIAL_REDUCE_SMLA: + ExtOpcLHS = ExtOpcRHS = ISD::SIGN_EXTEND; + break; + case ISD::PARTIAL_REDUCE_FMLA: + ExtOpcLHS = ExtOpcRHS = ISD::FP_EXTEND; + break; + } if (ExtMulOpVT != MulOpVT) { MulLHS = DAG.getNode(ExtOpcLHS, DL, ExtMulOpVT, MulLHS); MulRHS = DAG.getNode(ExtOpcRHS, DL, ExtMulOpVT, MulRHS); } SDValue Input = MulLHS; - APInt ConstantOne; - if (!ISD::isConstantSplatVector(MulRHS.getNode(), ConstantOne) || - !ConstantOne.isOne()) + if (N->getOpcode() == ISD::PARTIAL_REDUCE_FMLA) { + if (!llvm::isOneOrOneSplatFP(MulRHS)) + Input = DAG.getNode(ISD::FMUL, DL, ExtMulOpVT, MulLHS, MulRHS); + } else if (!llvm::isOneOrOneSplat(MulRHS)) { Input = DAG.getNode(ISD::MUL, DL, ExtMulOpVT, MulLHS, MulRHS); + } unsigned Stride = AccVT.getVectorMinNumElements(); unsigned ScaleFactor = MulOpVT.getVectorMinNumElements() / Stride; @@ -12099,10 +12109,13 @@ SDValue TargetLowering::expandPartialReduceMLA(SDNode *N, for (unsigned I = 0; I < ScaleFactor; I++) Subvectors.push_back(DAG.getExtractSubvector(DL, AccVT, Input, I * Stride)); + unsigned FlatNode = + N->getOpcode() == ISD::PARTIAL_REDUCE_FMLA ? ISD::FADD : ISD::ADD; + // Flatten the subvector tree while (Subvectors.size() > 1) { Subvectors.push_back( - DAG.getNode(ISD::ADD, DL, AccVT, {Subvectors[0], Subvectors[1]})); + DAG.getNode(FlatNode, DL, AccVT, {Subvectors[0], Subvectors[1]})); Subvectors.pop_front(); Subvectors.pop_front(); } diff --git a/llvm/lib/DWARFLinker/Parallel/SyntheticTypeNameBuilder.cpp b/llvm/lib/DWARFLinker/Parallel/SyntheticTypeNameBuilder.cpp index 34174f9..ca918f6 100644 --- a/llvm/lib/DWARFLinker/Parallel/SyntheticTypeNameBuilder.cpp +++ b/llvm/lib/DWARFLinker/Parallel/SyntheticTypeNameBuilder.cpp @@ -377,8 +377,10 @@ Error SyntheticTypeNameBuilder::addTypeName(UnitEntryPairTy InputUnitEntryPair, } break; } - // If name for the DIE is not determined yet add referenced types to the name. - if (!HasLinkageName && !HasShortName && !HasDeclFileName) { + // If name for the DIE is not determined yet or if the DIE is a typedef, add + // referenced types to the name. + if ((!HasLinkageName && !HasShortName && !HasDeclFileName) || + InputUnitEntryPair.DieEntry->getTag() == dwarf::DW_TAG_typedef) { if (InputUnitEntryPair.CU->find(InputUnitEntryPair.DieEntry, getODRAttributes())) if (Error Err = addReferencedODRDies(InputUnitEntryPair, AddParentNames, diff --git a/llvm/lib/Demangle/ItaniumDemangle.cpp b/llvm/lib/Demangle/ItaniumDemangle.cpp index 1009cc9..8e476cd 100644 --- a/llvm/lib/Demangle/ItaniumDemangle.cpp +++ b/llvm/lib/Demangle/ItaniumDemangle.cpp @@ -25,10 +25,6 @@ using namespace llvm; using namespace llvm::itanium_demangle; -constexpr const char *itanium_demangle::FloatData<float>::spec; -constexpr const char *itanium_demangle::FloatData<double>::spec; -constexpr const char *itanium_demangle::FloatData<long double>::spec; - // <discriminator> := _ <non-negative number> # when number < 10 // := __ <non-negative number> _ # when number >= 10 // extension := decimal-digit+ # at the end of string diff --git a/llvm/lib/IR/Verifier.cpp b/llvm/lib/IR/Verifier.cpp index f1e473a..59eb870 100644 --- a/llvm/lib/IR/Verifier.cpp +++ b/llvm/lib/IR/Verifier.cpp @@ -6583,6 +6583,7 @@ void Verifier::visitIntrinsicCall(Intrinsic::ID ID, CallBase &Call) { } break; } + case Intrinsic::vector_partial_reduce_fadd: case Intrinsic::vector_partial_reduce_add: { VectorType *AccTy = cast<VectorType>(Call.getArgOperand(0)->getType()); VectorType *VecTy = cast<VectorType>(Call.getArgOperand(1)->getType()); diff --git a/llvm/lib/Object/ELFObjectFile.cpp b/llvm/lib/Object/ELFObjectFile.cpp index f9fda23..3f0ecbe 100644 --- a/llvm/lib/Object/ELFObjectFile.cpp +++ b/llvm/lib/Object/ELFObjectFile.cpp @@ -311,6 +311,10 @@ static std::optional<std::string> hexagonAttrToFeatureString(unsigned Attr) { return "v73"; case 75: return "v75"; + case 79: + return "v79"; + case 81: + return "v81"; default: return {}; } diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp index 40e6400..c8a038f 100644 --- a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp +++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp @@ -1916,6 +1916,12 @@ AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM, setPartialReduceMLAAction(MLAOps, MVT::nxv4i32, MVT::nxv8i16, Legal); setPartialReduceMLAAction(MLAOps, MVT::nxv8i16, MVT::nxv16i8, Legal); } + + // Handle floating-point partial reduction + if (Subtarget->hasSVE2p1() || Subtarget->hasSME2()) { + setPartialReduceMLAAction(ISD::PARTIAL_REDUCE_FMLA, MVT::nxv4f32, + MVT::nxv8f16, Legal); + } } // Handle non-aliasing elements mask @@ -2283,6 +2289,11 @@ void AArch64TargetLowering::addTypeForFixedLengthSVE(MVT VT) { MVT::getVectorVT(MVT::i8, NumElts * 8), Custom); } + if (Subtarget->hasSVE2p1() && VT.getVectorElementType() == MVT::f32) { + setPartialReduceMLAAction(ISD::PARTIAL_REDUCE_FMLA, VT, + MVT::getVectorVT(MVT::f16, NumElts * 2), Custom); + } + // Lower fixed length vector operations to scalable equivalents. setOperationAction(ISD::ABDS, VT, Default); setOperationAction(ISD::ABDU, VT, Default); @@ -7875,6 +7886,7 @@ SDValue AArch64TargetLowering::LowerOperation(SDValue Op, case ISD::PARTIAL_REDUCE_SMLA: case ISD::PARTIAL_REDUCE_UMLA: case ISD::PARTIAL_REDUCE_SUMLA: + case ISD::PARTIAL_REDUCE_FMLA: return LowerPARTIAL_REDUCE_MLA(Op, DAG); } } diff --git a/llvm/lib/Target/AArch64/AArch64InstrFormats.td b/llvm/lib/Target/AArch64/AArch64InstrFormats.td index 58a53af..bb2f083 100644 --- a/llvm/lib/Target/AArch64/AArch64InstrFormats.td +++ b/llvm/lib/Target/AArch64/AArch64InstrFormats.td @@ -13292,18 +13292,24 @@ multiclass AtomicFPStore<bit R, bits<3> op0, string asm> { def H : BaseAtomicFPStore<FPR16, 0b01, R, op0, asm>; } -class BaseSIMDThreeSameVectorFP8MatrixMul<string asm, bits<2> size, string kind> +class BaseSIMDThreeSameVectorFP8MatrixMul<string asm, bits<2> size, string kind, list<dag> pattern> : BaseSIMDThreeSameVectorTied<1, 1, {size, 0}, 0b11101, - V128, asm, ".16b", []> { + V128, asm, ".16b", pattern> { let AsmString = !strconcat(asm, "{\t$Rd", kind, ", $Rn.16b, $Rm.16b", "|", kind, "\t$Rd, $Rn, $Rm}"); } -multiclass SIMDThreeSameVectorFP8MatrixMul<string asm>{ - def v8f16: BaseSIMDThreeSameVectorFP8MatrixMul<asm, 0b00, ".8h">{ +multiclass SIMDThreeSameVectorFP8MatrixMul<string asm, SDPatternOperator OpNode>{ + def v8f16: BaseSIMDThreeSameVectorFP8MatrixMul<asm, 0b00, ".8h", + [(set (v8f16 V128:$dst), (OpNode (v8f16 V128:$Rd), + (v16i8 V128:$Rn), + (v16i8 V128:$Rm)))]> { let Predicates = [HasNEON, HasF8F16MM]; } - def v4f32: BaseSIMDThreeSameVectorFP8MatrixMul<asm, 0b10, ".4s">{ + def v4f32: BaseSIMDThreeSameVectorFP8MatrixMul<asm, 0b10, ".4s", + [(set (v4f32 V128:$dst), (OpNode (v4f32 V128:$Rd), + (v16i8 V128:$Rn), + (v16i8 V128:$Rm)))]> { let Predicates = [HasNEON, HasF8F32MM]; } } diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.td b/llvm/lib/Target/AArch64/AArch64InstrInfo.td index e6954f7..76f076a 100644 --- a/llvm/lib/Target/AArch64/AArch64InstrInfo.td +++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.td @@ -11417,7 +11417,7 @@ let Predicates = [HasF16F32MM] in defm FMMLA : SIMDThreeSameVectorFMLAWiden<"fmmla">; let Uses = [FPMR, FPCR] in - defm FMMLA : SIMDThreeSameVectorFP8MatrixMul<"fmmla">; + defm FMMLA : SIMDThreeSameVectorFP8MatrixMul<"fmmla", int_aarch64_neon_fmmla>; //===----------------------------------------------------------------------===// // Contention Management Hints (FEAT_CMH) diff --git a/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td b/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td index 3b268dc..e1f4386 100644 --- a/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td +++ b/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td @@ -375,6 +375,11 @@ def AArch64fclamp : PatFrags<(ops node:$Zd, node:$Zn, node:$Zm), node:$Zm) ]>; +def AArch64fdot : PatFrags<(ops node:$Zd, node:$Zn, node:$Zm), + [(int_aarch64_sve_fdot_x2 node:$Zd, node:$Zn, node:$Zm), + (partial_reduce_fmla node:$Zd, node:$Zn, node:$Zm) + ]>; + def SDT_AArch64FCVT : SDTypeProfile<1, 3, [ SDTCisVec<0>, SDTCisVec<1>, SDTCisVec<2>, SDTCisVec<3>, SDTCVecEltisVT<1,i1>, SDTCisSameNumEltsAs<0,1>, SDTCisSameAs<0,3> @@ -4251,7 +4256,7 @@ defm PSEL_PPPRI : sve2_int_perm_sel_p<"psel", int_aarch64_sve_psel>; let Predicates = [HasSVE2p1_or_SME2] in { defm FCLAMP_ZZZ : sve_fp_clamp<"fclamp", AArch64fclamp>; -defm FDOT_ZZZ_S : sve_float_dot<0b0, 0b0, ZPR32, ZPR16, "fdot", nxv8f16, int_aarch64_sve_fdot_x2>; +defm FDOT_ZZZ_S : sve_float_dot<0b0, 0b0, ZPR32, ZPR16, "fdot", nxv8f16, AArch64fdot>; defm FDOT_ZZZI_S : sve_float_dot_indexed<0b0, 0b00, ZPR16, ZPR3b16, "fdot", nxv8f16, int_aarch64_sve_fdot_lane_x2>; defm BFMLSLB_ZZZ_S : sve2_fp_mla_long<0b110, "bfmlslb", nxv4f32, nxv8bf16, int_aarch64_sve_bfmlslb>; diff --git a/llvm/lib/Target/AMDGPU/AMDGPUUniformIntrinsicCombine.cpp b/llvm/lib/Target/AMDGPU/AMDGPUUniformIntrinsicCombine.cpp index 65e6ed9..c52eb4e 100644 --- a/llvm/lib/Target/AMDGPU/AMDGPUUniformIntrinsicCombine.cpp +++ b/llvm/lib/Target/AMDGPU/AMDGPUUniformIntrinsicCombine.cpp @@ -57,10 +57,11 @@ static bool optimizeUniformIntrinsic(IntrinsicInst &II, const UniformityInfo &UI, ValueMap<const Value *, bool> &Tracker) { llvm::Intrinsic::ID IID = II.getIntrinsicID(); - + /// We deliberately do not simplify readfirstlane with a uniform argument, so + /// that frontends can use it to force a copy to SGPR and thereby prevent the + /// backend from generating unwanted waterfall loops. switch (IID) { case Intrinsic::amdgcn_permlane64: - case Intrinsic::amdgcn_readfirstlane: case Intrinsic::amdgcn_readlane: { Value *Src = II.getArgOperand(0); if (isDivergentUseWithNew(II.getOperandUse(0), UI, Tracker)) @@ -107,7 +108,7 @@ static bool optimizeUniformIntrinsic(IntrinsicInst &II, return Changed; } default: - llvm_unreachable("Unexpected intrinsic ID in optimizeUniformIntrinsic"); + return false; } return false; } @@ -121,16 +122,6 @@ static bool runUniformIntrinsicCombine(Function &F, const UniformityInfo &UI) { auto *II = dyn_cast<IntrinsicInst>(&I); if (!II) continue; - - switch (II->getIntrinsicID()) { - case Intrinsic::amdgcn_permlane64: - case Intrinsic::amdgcn_readfirstlane: - case Intrinsic::amdgcn_readlane: - case Intrinsic::amdgcn_ballot: - break; - default: - continue; - } IsChanged |= optimizeUniformIntrinsic(*II, UI, Tracker); } return IsChanged; diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp index 6ce18ea..9c74c65 100644 --- a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp +++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp @@ -10163,7 +10163,7 @@ static bool followSubRegDef(MachineInstr &MI, } MachineInstr *llvm::getVRegSubRegDef(const TargetInstrInfo::RegSubRegPair &P, - MachineRegisterInfo &MRI) { + const MachineRegisterInfo &MRI) { assert(MRI.isSSA()); if (!P.Reg.isVirtual()) return nullptr; @@ -10628,6 +10628,8 @@ bool SIInstrInfo::analyzeCompare(const MachineInstr &MI, Register &SrcReg, static bool optimizeSCC(MachineInstr *SCCValid, MachineInstr *SCCRedefine, const SIRegisterInfo &RI) { MachineInstr *KillsSCC = nullptr; + if (SCCValid->getParent() != SCCRedefine->getParent()) + return false; for (MachineInstr &MI : make_range(std::next(SCCValid->getIterator()), SCCRedefine->getIterator())) { if (MI.modifiesRegister(AMDGPU::SCC, &RI)) @@ -10672,8 +10674,8 @@ bool SIInstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg, if (CmpValue != 0) return false; - MachineInstr *Def = MRI->getUniqueVRegDef(SrcReg); - if (!Def || Def->getParent() != CmpInstr.getParent()) + MachineInstr *Def = MRI->getVRegDef(SrcReg); + if (!Def) return false; // For S_OP that set SCC = DST!=0, do the transformation @@ -10692,6 +10694,32 @@ bool SIInstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg, if (!optimizeSCC(Def, &CmpInstr, RI)) return false; + // If s_or_b32 result, sY, is unused (i.e. it is effectively a 64-bit + // s_cmp_lg of a register pair) and the inputs are the hi and lo-halves of a + // 64-bit foldableSelect then delete s_or_b32 in the sequence: + // sX = s_cselect_b64 (non-zero imm), 0 + // sLo = copy sX.sub0 + // sHi = copy sX.sub1 + // sY = s_or_b32 sLo, sHi + if (Def->getOpcode() == AMDGPU::S_OR_B32 && + MRI->use_nodbg_empty(Def->getOperand(0).getReg())) { + const MachineOperand &OrOpnd1 = Def->getOperand(1); + const MachineOperand &OrOpnd2 = Def->getOperand(2); + if (OrOpnd1.isReg() && OrOpnd2.isReg()) { + MachineInstr *Def1 = MRI->getVRegDef(OrOpnd1.getReg()); + MachineInstr *Def2 = MRI->getVRegDef(OrOpnd2.getReg()); + if (Def1 && Def1->getOpcode() == AMDGPU::COPY && Def2 && + Def2->getOpcode() == AMDGPU::COPY && Def1->getOperand(1).isReg() && + Def2->getOperand(1).isReg() && + Def1->getOperand(1).getSubReg() == AMDGPU::sub0 && + Def2->getOperand(1).getSubReg() == AMDGPU::sub1 && + Def1->getOperand(1).getReg() == Def2->getOperand(1).getReg()) { + MachineInstr *Select = MRI->getVRegDef(Def1->getOperand(1).getReg()); + if (Select && foldableSelect(*Select)) + optimizeSCC(Select, Def, RI); + } + } + } return true; }; @@ -10721,8 +10749,8 @@ bool SIInstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg, // s_cmp_lg_i32 (s_and_b32 $src, 1 << n), 1 << n => s_bitcmp0_b32 $src, n // s_cmp_lg_u64 (s_and_b64 $src, 1 << n), 1 << n => s_bitcmp0_b64 $src, n - MachineInstr *Def = MRI->getUniqueVRegDef(SrcReg); - if (!Def || Def->getParent() != CmpInstr.getParent()) + MachineInstr *Def = MRI->getVRegDef(SrcReg); + if (!Def) return false; if (Def->getOpcode() != AMDGPU::S_AND_B32 && diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.h b/llvm/lib/Target/AMDGPU/SIInstrInfo.h index 0643b53..8d693b1 100644 --- a/llvm/lib/Target/AMDGPU/SIInstrInfo.h +++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.h @@ -1687,7 +1687,7 @@ TargetInstrInfo::RegSubRegPair getRegSequenceSubReg(MachineInstr &MI, /// skipping copy like instructions and subreg-manipulation pseudos. /// Following another subreg of a reg:subreg isn't supported. MachineInstr *getVRegSubRegDef(const TargetInstrInfo::RegSubRegPair &P, - MachineRegisterInfo &MRI); + const MachineRegisterInfo &MRI); /// \brief Return false if EXEC is not changed between the def of \p VReg at \p /// DefMI and the use at \p UseMI. Should be run on SSA. Currently does not diff --git a/llvm/lib/Target/PowerPC/PPCISelLowering.cpp b/llvm/lib/Target/PowerPC/PPCISelLowering.cpp index 20fc849..dd233e2 100644 --- a/llvm/lib/Target/PowerPC/PPCISelLowering.cpp +++ b/llvm/lib/Target/PowerPC/PPCISelLowering.cpp @@ -657,6 +657,17 @@ PPCTargetLowering::PPCTargetLowering(const PPCTargetMachine &TM, setOperationAction(ISD::EH_DWARF_CFA, MVT::i32, Custom); setOperationAction(ISD::EH_DWARF_CFA, MVT::i64, Custom); + if (Subtarget.isISA3_0() && isPPC64) { + setOperationAction(ISD::VP_STORE, MVT::v16i1, Custom); + setOperationAction(ISD::VP_STORE, MVT::v8i1, Custom); + setOperationAction(ISD::VP_STORE, MVT::v4i1, Custom); + setOperationAction(ISD::VP_STORE, MVT::v2i1, Custom); + setOperationAction(ISD::VP_LOAD, MVT::v16i1, Custom); + setOperationAction(ISD::VP_LOAD, MVT::v8i1, Custom); + setOperationAction(ISD::VP_LOAD, MVT::v4i1, Custom); + setOperationAction(ISD::VP_LOAD, MVT::v2i1, Custom); + } + // We want to custom lower some of our intrinsics. setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::Other, Custom); setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::f64, Custom); @@ -11917,6 +11928,62 @@ SDValue PPCTargetLowering::LowerIS_FPCLASS(SDValue Op, return getDataClassTest(LHS, Category, Dl, DAG, Subtarget); } +// Adjust the length value for a load/store with length to account for the +// instructions requiring a left justified length, and for non-byte element +// types requiring scaling by element size. +static SDValue AdjustLength(SDValue Val, unsigned Bits, bool Left, + SelectionDAG &DAG) { + SDLoc dl(Val); + EVT VT = Val->getValueType(0); + unsigned LeftAdj = Left ? VT.getSizeInBits() - 8 : 0; + unsigned TypeAdj = llvm::countr_zero<uint32_t>(Bits / 8); + SDValue SHLAmt = DAG.getConstant(LeftAdj + TypeAdj, dl, VT); + return DAG.getNode(ISD::SHL, dl, VT, Val, SHLAmt); +} + +SDValue PPCTargetLowering::LowerVP_LOAD(SDValue Op, SelectionDAG &DAG) const { + auto VPLD = cast<VPLoadSDNode>(Op); + bool Future = Subtarget.isISAFuture(); + SDLoc dl(Op); + assert(ISD::isConstantSplatVectorAllOnes(Op->getOperand(3).getNode(), true) && + "Mask predication not supported"); + EVT PtrVT = getPointerTy(DAG.getDataLayout()); + SDValue Len = DAG.getNode(ISD::ANY_EXTEND, dl, PtrVT, VPLD->getOperand(4)); + unsigned IID = Future ? Intrinsic::ppc_vsx_lxvrl : Intrinsic::ppc_vsx_lxvl; + unsigned EltBits = Op->getValueType(0).getScalarType().getSizeInBits(); + Len = AdjustLength(Len, EltBits, !Future, DAG); + SDValue Ops[] = {VPLD->getChain(), DAG.getConstant(IID, dl, MVT::i32), + VPLD->getOperand(1), Len}; + SDVTList Tys = DAG.getVTList(Op->getValueType(0), MVT::Other); + SDValue VPL = + DAG.getMemIntrinsicNode(ISD::INTRINSIC_W_CHAIN, dl, Tys, Ops, + VPLD->getMemoryVT(), VPLD->getMemOperand()); + return VPL; +} + +SDValue PPCTargetLowering::LowerVP_STORE(SDValue Op, SelectionDAG &DAG) const { + auto VPST = cast<VPStoreSDNode>(Op); + assert(ISD::isConstantSplatVectorAllOnes(Op->getOperand(4).getNode(), true) && + "Mask predication not supported"); + EVT PtrVT = getPointerTy(DAG.getDataLayout()); + SDLoc dl(Op); + SDValue Len = DAG.getNode(ISD::ANY_EXTEND, dl, PtrVT, VPST->getOperand(5)); + unsigned EltBits = + Op->getOperand(1).getValueType().getScalarType().getSizeInBits(); + bool Future = Subtarget.isISAFuture(); + unsigned IID = Future ? Intrinsic::ppc_vsx_stxvrl : Intrinsic::ppc_vsx_stxvl; + Len = AdjustLength(Len, EltBits, !Future, DAG); + SDValue Ops[] = { + VPST->getChain(), DAG.getConstant(IID, dl, MVT::i32), + DAG.getNode(ISD::BITCAST, dl, MVT::v4i32, VPST->getOperand(1)), + VPST->getOperand(2), Len}; + SDVTList Tys = DAG.getVTList(MVT::Other); + SDValue VPS = + DAG.getMemIntrinsicNode(ISD::INTRINSIC_VOID, dl, Tys, Ops, + VPST->getMemoryVT(), VPST->getMemOperand()); + return VPS; +} + SDValue PPCTargetLowering::LowerSCALAR_TO_VECTOR(SDValue Op, SelectionDAG &DAG) const { SDLoc dl(Op); @@ -12771,6 +12838,10 @@ SDValue PPCTargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const { if (Op->getFlags().hasNoFPExcept()) return Op; return SDValue(); + case ISD::VP_LOAD: + return LowerVP_LOAD(Op, DAG); + case ISD::VP_STORE: + return LowerVP_STORE(Op, DAG); } } diff --git a/llvm/lib/Target/PowerPC/PPCISelLowering.h b/llvm/lib/Target/PowerPC/PPCISelLowering.h index 880aca7..d967018 100644 --- a/llvm/lib/Target/PowerPC/PPCISelLowering.h +++ b/llvm/lib/Target/PowerPC/PPCISelLowering.h @@ -1345,6 +1345,9 @@ namespace llvm { SDValue LowerFP_ROUND(SDValue Op, SelectionDAG &DAG) const; SDValue LowerROTL(SDValue Op, SelectionDAG &DAG) const; + SDValue LowerVP_LOAD(SDValue Op, SelectionDAG &DAG) const; + SDValue LowerVP_STORE(SDValue Op, SelectionDAG &DAG) const; + SDValue LowerVectorLoad(SDValue Op, SelectionDAG &DAG) const; SDValue LowerVectorStore(SDValue Op, SelectionDAG &DAG) const; SDValue LowerDMFVectorLoad(SDValue Op, SelectionDAG &DAG) const; diff --git a/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.cpp b/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.cpp index b04e887..e74f1bd 100644 --- a/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.cpp +++ b/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.cpp @@ -24,6 +24,10 @@ using namespace llvm; #define DEBUG_TYPE "ppctti" +static cl::opt<bool> Pwr9EVL("ppc-pwr9-evl", + cl::desc("Allow vp.load and vp.store for pwr9"), + cl::init(false), cl::Hidden); + static cl::opt<bool> VecMaskCost("ppc-vec-mask-cost", cl::desc("add masking cost for i1 vectors"), cl::init(true), cl::Hidden); @@ -1031,3 +1035,42 @@ bool PPCTTIImpl::getTgtMemIntrinsic(IntrinsicInst *Inst, bool PPCTTIImpl::supportsTailCallFor(const CallBase *CB) const { return TLI->supportsTailCallFor(CB); } + +// Target hook used by CodeGen to decide whether to expand vector predication +// intrinsics into scalar operations or to use special ISD nodes to represent +// them. The Target will not see the intrinsics. +TargetTransformInfo::VPLegalization +PPCTTIImpl::getVPLegalizationStrategy(const VPIntrinsic &PI) const { + using VPLegalization = TargetTransformInfo::VPLegalization; + unsigned Directive = ST->getCPUDirective(); + VPLegalization DefaultLegalization = BaseT::getVPLegalizationStrategy(PI); + if (Directive != PPC::DIR_PWR10 && Directive != PPC::DIR_PWR_FUTURE && + (!Pwr9EVL || Directive != PPC::DIR_PWR9)) + return DefaultLegalization; + + if (!ST->isPPC64()) + return DefaultLegalization; + + unsigned IID = PI.getIntrinsicID(); + if (IID != Intrinsic::vp_load && IID != Intrinsic::vp_store) + return DefaultLegalization; + + bool IsLoad = IID == Intrinsic::vp_load; + Type *VecTy = IsLoad ? PI.getType() : PI.getOperand(0)->getType(); + EVT VT = TLI->getValueType(DL, VecTy, true); + if (VT != MVT::v2i64 && VT != MVT::v4i32 && VT != MVT::v8i16 && + VT != MVT::v16i8) + return DefaultLegalization; + + auto IsAllTrueMask = [](Value *MaskVal) { + if (Value *SplattedVal = getSplatValue(MaskVal)) + if (auto *ConstValue = dyn_cast<Constant>(SplattedVal)) + return ConstValue->isAllOnesValue(); + return false; + }; + unsigned MaskIx = IsLoad ? 1 : 2; + if (!IsAllTrueMask(PI.getOperand(MaskIx))) + return DefaultLegalization; + + return VPLegalization(VPLegalization::Legal, VPLegalization::Legal); +} diff --git a/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h b/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h index 8d7f255..f80ebdb 100644 --- a/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h +++ b/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h @@ -150,6 +150,9 @@ public: ArrayRef<Type *> Types) const override; bool supportsTailCallFor(const CallBase *CB) const override; + TargetTransformInfo::VPLegalization + getVPLegalizationStrategy(const VPIntrinsic &PI) const override; + private: // The following constant is used for estimating costs on power9. static const InstructionCost::CostType P9PipelineFlushEstimate = 80; diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp index 7bc0b5b..332433b 100644 --- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp +++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp @@ -2140,7 +2140,8 @@ InstructionCost RISCVTTIImpl::getMemoryOpCost(unsigned Opcode, Type *Src, // Assume memory ops cost scale with the number of vector registers // possible accessed by the instruction. Note that BasicTTI already // handles the LT.first term for us. - if (LT.second.isVector() && CostKind != TTI::TCK_CodeSize) + if (ST->hasVInstructions() && LT.second.isVector() && + CostKind != TTI::TCK_CodeSize) BaseCost *= TLI->getLMULCost(LT.second); return Cost + BaseCost; } diff --git a/llvm/lib/Target/SystemZ/SystemZLongBranch.cpp b/llvm/lib/Target/SystemZ/SystemZLongBranch.cpp index 21a233b2..b7a93e7 100644 --- a/llvm/lib/Target/SystemZ/SystemZLongBranch.cpp +++ b/llvm/lib/Target/SystemZ/SystemZLongBranch.cpp @@ -216,6 +216,7 @@ static unsigned getInstSizeInBytes(const MachineInstr &MI, MI.isDebugOrPseudoInstr() || MI.isPosition() || MI.isKill() || MI.isImplicitDef() || MI.getOpcode() == TargetOpcode::MEMBARRIER || MI.getOpcode() == TargetOpcode::INIT_UNDEF || MI.isFakeUse() || + MI.getOpcode() == TargetOpcode::RELOC_NONE || // These have a size that may be zero: MI.isInlineAsm() || MI.getOpcode() == SystemZ::STACKMAP || MI.getOpcode() == SystemZ::PATCHPOINT || diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp index 090f649..05a854a 100644 --- a/llvm/lib/Target/X86/X86ISelLowering.cpp +++ b/llvm/lib/Target/X86/X86ISelLowering.cpp @@ -45022,11 +45022,16 @@ bool X86TargetLowering::isGuaranteedNotToBeUndefOrPoisonForTargetNode( case X86ISD::INSERTPS: case X86ISD::BLENDI: case X86ISD::PSHUFB: + case X86ISD::VZEXT_MOVL: case X86ISD::PSHUFD: + case X86ISD::PSHUFHW: + case X86ISD::PSHUFLW: + case X86ISD::SHUFP: case X86ISD::UNPCKL: case X86ISD::UNPCKH: case X86ISD::VPERMILPV: case X86ISD::VPERMILPI: + case X86ISD::VPERMI: case X86ISD::VPERMV: case X86ISD::VPERMV3: { SmallVector<int, 8> Mask; @@ -45052,6 +45057,16 @@ bool X86TargetLowering::isGuaranteedNotToBeUndefOrPoisonForTargetNode( } break; } + case X86ISD::VBROADCAST: { + SDValue Src = Op.getOperand(0); + MVT SrcVT = Src.getSimpleValueType(); + if (SrcVT.isVector()) { + APInt DemandedSrc = APInt::getOneBitSet(SrcVT.getVectorNumElements(), 0); + return DAG.isGuaranteedNotToBeUndefOrPoison(Src, DemandedSrc, PoisonOnly, + Depth + 1); + } + return DAG.isGuaranteedNotToBeUndefOrPoison(Src, PoisonOnly, Depth + 1); + } } return TargetLowering::isGuaranteedNotToBeUndefOrPoisonForTargetNode( Op, DemandedElts, DAG, PoisonOnly, Depth); @@ -45096,13 +45111,19 @@ bool X86TargetLowering::canCreateUndefOrPoisonForTargetNode( // SSE target shuffles. case X86ISD::INSERTPS: case X86ISD::PSHUFB: + case X86ISD::VZEXT_MOVL: case X86ISD::PSHUFD: + case X86ISD::PSHUFHW: + case X86ISD::PSHUFLW: + case X86ISD::SHUFP: case X86ISD::UNPCKL: case X86ISD::UNPCKH: case X86ISD::VPERMILPV: case X86ISD::VPERMILPI: + case X86ISD::VPERMI: case X86ISD::VPERMV: case X86ISD::VPERMV3: + case X86ISD::VBROADCAST: return false; // SSE comparisons handle all icmp/fcmp cases. // TODO: Add CMPM/MM with test coverage. diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp index b7224a3..666033b 100644 --- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp +++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp @@ -7488,11 +7488,12 @@ void EpilogueVectorizerEpilogueLoop::printDebugTracesAtEnd() { }); } -VPWidenMemoryRecipe * -VPRecipeBuilder::tryToWidenMemory(Instruction *I, ArrayRef<VPValue *> Operands, - VFRange &Range) { - assert((isa<LoadInst>(I) || isa<StoreInst>(I)) && +VPWidenMemoryRecipe *VPRecipeBuilder::tryToWidenMemory(VPInstruction *VPI, + VFRange &Range) { + assert((VPI->getOpcode() == Instruction::Load || + VPI->getOpcode() == Instruction::Store) && "Must be called with either a load or store"); + Instruction *I = VPI->getUnderlyingInstr(); auto WillWiden = [&](ElementCount VF) -> bool { LoopVectorizationCostModel::InstWidening Decision = @@ -7522,7 +7523,8 @@ VPRecipeBuilder::tryToWidenMemory(Instruction *I, ArrayRef<VPValue *> Operands, bool Consecutive = Reverse || Decision == LoopVectorizationCostModel::CM_Widen; - VPValue *Ptr = isa<LoadInst>(I) ? Operands[0] : Operands[1]; + VPValue *Ptr = VPI->getOpcode() == Instruction::Load ? VPI->getOperand(0) + : VPI->getOperand(1); if (Consecutive) { auto *GEP = dyn_cast<GetElementPtrInst>( Ptr->getUnderlyingValue()->stripPointerCasts()); @@ -7536,77 +7538,78 @@ VPRecipeBuilder::tryToWidenMemory(Instruction *I, ArrayRef<VPValue *> Operands, CM.foldTailByMasking() || !GEP ? GEPNoWrapFlags::none() : GEP->getNoWrapFlags().withoutNoUnsignedWrap(); - VectorPtr = - new VPVectorEndPointerRecipe(Ptr, &Plan.getVF(), getLoadStoreType(I), - /*Stride*/ -1, Flags, I->getDebugLoc()); + VectorPtr = new VPVectorEndPointerRecipe( + Ptr, &Plan.getVF(), getLoadStoreType(I), + /*Stride*/ -1, Flags, VPI->getDebugLoc()); } else { VectorPtr = new VPVectorPointerRecipe(Ptr, getLoadStoreType(I), GEP ? GEP->getNoWrapFlags() : GEPNoWrapFlags::none(), - I->getDebugLoc()); + VPI->getDebugLoc()); } Builder.insert(VectorPtr); Ptr = VectorPtr; } - if (LoadInst *Load = dyn_cast<LoadInst>(I)) + if (VPI->getOpcode() == Instruction::Load) { + auto *Load = cast<LoadInst>(I); return new VPWidenLoadRecipe(*Load, Ptr, Mask, Consecutive, Reverse, VPIRMetadata(*Load, LVer), I->getDebugLoc()); + } StoreInst *Store = cast<StoreInst>(I); - return new VPWidenStoreRecipe(*Store, Ptr, Operands[0], Mask, Consecutive, - Reverse, VPIRMetadata(*Store, LVer), - I->getDebugLoc()); + return new VPWidenStoreRecipe(*Store, Ptr, VPI->getOperand(0), Mask, + Consecutive, Reverse, + VPIRMetadata(*Store, LVer), VPI->getDebugLoc()); } -/// Creates a VPWidenIntOrFpInductionRecpipe for \p Phi. If needed, it will also -/// insert a recipe to expand the step for the induction recipe. +/// Creates a VPWidenIntOrFpInductionRecipe for \p PhiR. If needed, it will +/// also insert a recipe to expand the step for the induction recipe. static VPWidenIntOrFpInductionRecipe * -createWidenInductionRecipes(PHINode *Phi, Instruction *PhiOrTrunc, - VPValue *Start, const InductionDescriptor &IndDesc, - VPlan &Plan, ScalarEvolution &SE, Loop &OrigLoop) { - assert(IndDesc.getStartValue() == - Phi->getIncomingValueForBlock(OrigLoop.getLoopPreheader())); +createWidenInductionRecipes(VPInstruction *PhiR, + const InductionDescriptor &IndDesc, VPlan &Plan, + ScalarEvolution &SE, Loop &OrigLoop) { assert(SE.isLoopInvariant(IndDesc.getStep(), &OrigLoop) && "step must be loop invariant"); + VPValue *Start = PhiR->getOperand(0); + assert(Plan.getLiveIn(IndDesc.getStartValue()) == Start && + "Start VPValue must match IndDesc's start value"); + VPValue *Step = vputils::getOrCreateVPValueForSCEVExpr(Plan, IndDesc.getStep()); - if (auto *TruncI = dyn_cast<TruncInst>(PhiOrTrunc)) { - return new VPWidenIntOrFpInductionRecipe(Phi, Start, Step, &Plan.getVF(), - IndDesc, TruncI, - TruncI->getDebugLoc()); - } - assert(isa<PHINode>(PhiOrTrunc) && "must be a phi node here"); + PHINode *Phi = cast<PHINode>(PhiR->getUnderlyingInstr()); return new VPWidenIntOrFpInductionRecipe(Phi, Start, Step, &Plan.getVF(), - IndDesc, Phi->getDebugLoc()); + IndDesc, PhiR->getDebugLoc()); } -VPHeaderPHIRecipe *VPRecipeBuilder::tryToOptimizeInductionPHI( - PHINode *Phi, ArrayRef<VPValue *> Operands, VFRange &Range) { +VPHeaderPHIRecipe * +VPRecipeBuilder::tryToOptimizeInductionPHI(VPInstruction *VPI, VFRange &Range) { + auto *Phi = cast<PHINode>(VPI->getUnderlyingInstr()); // Check if this is an integer or fp induction. If so, build the recipe that // produces its scalar and vector values. if (auto *II = Legal->getIntOrFpInductionDescriptor(Phi)) - return createWidenInductionRecipes(Phi, Phi, Operands[0], *II, Plan, - *PSE.getSE(), *OrigLoop); + return createWidenInductionRecipes(VPI, *II, Plan, *PSE.getSE(), *OrigLoop); // Check if this is pointer induction. If so, build the recipe for it. if (auto *II = Legal->getPointerInductionDescriptor(Phi)) { VPValue *Step = vputils::getOrCreateVPValueForSCEVExpr(Plan, II->getStep()); return new VPWidenPointerInductionRecipe( - Phi, Operands[0], Step, &Plan.getVFxUF(), *II, + Phi, VPI->getOperand(0), Step, &Plan.getVFxUF(), *II, LoopVectorizationPlanner::getDecisionAndClampRange( [&](ElementCount VF) { return CM.isScalarAfterVectorization(Phi, VF); }, Range), - Phi->getDebugLoc()); + VPI->getDebugLoc()); } return nullptr; } -VPWidenIntOrFpInductionRecipe *VPRecipeBuilder::tryToOptimizeInductionTruncate( - TruncInst *I, ArrayRef<VPValue *> Operands, VFRange &Range) { +VPWidenIntOrFpInductionRecipe * +VPRecipeBuilder::tryToOptimizeInductionTruncate(VPInstruction *VPI, + VFRange &Range) { + auto *I = cast<TruncInst>(VPI->getUnderlyingInstr()); // Optimize the special case where the source is a constant integer // induction variable. Notice that we can only optimize the 'trunc' case // because (a) FP conversions lose precision, (b) sext/zext may wrap, and @@ -7621,21 +7624,24 @@ VPWidenIntOrFpInductionRecipe *VPRecipeBuilder::tryToOptimizeInductionTruncate( }; }; - if (LoopVectorizationPlanner::getDecisionAndClampRange( - IsOptimizableIVTruncate(I), Range)) { + if (!LoopVectorizationPlanner::getDecisionAndClampRange( + IsOptimizableIVTruncate(I), Range)) + return nullptr; - auto *Phi = cast<PHINode>(I->getOperand(0)); - const InductionDescriptor &II = *Legal->getIntOrFpInductionDescriptor(Phi); - VPValue *Start = Plan.getOrAddLiveIn(II.getStartValue()); - return createWidenInductionRecipes(Phi, I, Start, II, Plan, *PSE.getSE(), - *OrigLoop); - } - return nullptr; + auto *WidenIV = cast<VPWidenIntOrFpInductionRecipe>( + VPI->getOperand(0)->getDefiningRecipe()); + PHINode *Phi = WidenIV->getPHINode(); + VPValue *Start = WidenIV->getStartValue(); + const InductionDescriptor &IndDesc = WidenIV->getInductionDescriptor(); + VPValue *Step = + vputils::getOrCreateVPValueForSCEVExpr(Plan, IndDesc.getStep()); + return new VPWidenIntOrFpInductionRecipe(Phi, Start, Step, &Plan.getVF(), + IndDesc, I, VPI->getDebugLoc()); } -VPSingleDefRecipe *VPRecipeBuilder::tryToWidenCall(CallInst *CI, - ArrayRef<VPValue *> Operands, +VPSingleDefRecipe *VPRecipeBuilder::tryToWidenCall(VPInstruction *VPI, VFRange &Range) { + CallInst *CI = cast<CallInst>(VPI->getUnderlyingInstr()); bool IsPredicated = LoopVectorizationPlanner::getDecisionAndClampRange( [this, CI](ElementCount VF) { return CM.isScalarWithPredication(CI, VF); @@ -7652,7 +7658,8 @@ VPSingleDefRecipe *VPRecipeBuilder::tryToWidenCall(CallInst *CI, ID == Intrinsic::experimental_noalias_scope_decl)) return nullptr; - SmallVector<VPValue *, 4> Ops(Operands.take_front(CI->arg_size())); + SmallVector<VPValue *, 4> Ops(VPI->op_begin(), + VPI->op_begin() + CI->arg_size()); // Is it beneficial to perform intrinsic call compared to lib call? bool ShouldUseVectorIntrinsic = @@ -7664,7 +7671,7 @@ VPSingleDefRecipe *VPRecipeBuilder::tryToWidenCall(CallInst *CI, Range); if (ShouldUseVectorIntrinsic) return new VPWidenIntrinsicRecipe(*CI, ID, Ops, CI->getType(), - CI->getDebugLoc()); + VPI->getDebugLoc()); Function *Variant = nullptr; std::optional<unsigned> MaskPos; @@ -7711,13 +7718,13 @@ VPSingleDefRecipe *VPRecipeBuilder::tryToWidenCall(CallInst *CI, Mask = getBlockInMask(Builder.getInsertBlock()); else Mask = Plan.getOrAddLiveIn( - ConstantInt::getTrue(IntegerType::getInt1Ty(CI->getContext()))); + ConstantInt::getTrue(IntegerType::getInt1Ty(Plan.getContext()))); Ops.insert(Ops.begin() + *MaskPos, Mask); } - Ops.push_back(Operands.back()); - return new VPWidenCallRecipe(CI, Variant, Ops, CI->getDebugLoc()); + Ops.push_back(VPI->getOperand(VPI->getNumOperands() - 1)); + return new VPWidenCallRecipe(CI, Variant, Ops, VPI->getDebugLoc()); } return nullptr; @@ -7737,9 +7744,9 @@ bool VPRecipeBuilder::shouldWiden(Instruction *I, VFRange &Range) const { Range); } -VPWidenRecipe *VPRecipeBuilder::tryToWiden(Instruction *I, - ArrayRef<VPValue *> Operands) { - switch (I->getOpcode()) { +VPWidenRecipe *VPRecipeBuilder::tryToWiden(VPInstruction *VPI) { + auto *I = VPI->getUnderlyingInstr(); + switch (VPI->getOpcode()) { default: return nullptr; case Instruction::SDiv: @@ -7749,10 +7756,11 @@ VPWidenRecipe *VPRecipeBuilder::tryToWiden(Instruction *I, // If not provably safe, use a select to form a safe divisor before widening the // div/rem operation itself. Otherwise fall through to general handling below. if (CM.isPredicatedInst(I)) { - SmallVector<VPValue *> Ops(Operands); + SmallVector<VPValue *> Ops(VPI->operands()); VPValue *Mask = getBlockInMask(Builder.getInsertBlock()); VPValue *One = Plan.getConstantInt(I->getType(), 1u); - auto *SafeRHS = Builder.createSelect(Mask, Ops[1], One, I->getDebugLoc()); + auto *SafeRHS = + Builder.createSelect(Mask, Ops[1], One, VPI->getDebugLoc()); Ops[1] = SafeRHS; return new VPWidenRecipe(*I, Ops); } @@ -7777,8 +7785,8 @@ VPWidenRecipe *VPRecipeBuilder::tryToWiden(Instruction *I, case Instruction::Sub: case Instruction::Xor: case Instruction::Freeze: { - SmallVector<VPValue *> NewOps(Operands); - if (Instruction::isBinaryOp(I->getOpcode())) { + SmallVector<VPValue *> NewOps(VPI->operands()); + if (Instruction::isBinaryOp(VPI->getOpcode())) { // The legacy cost model uses SCEV to check if some of the operands are // constants. To match the legacy cost model's behavior, use SCEV to try // to replace operands with constants. @@ -7795,7 +7803,7 @@ VPWidenRecipe *VPRecipeBuilder::tryToWiden(Instruction *I, return Plan.getOrAddLiveIn(C->getValue()); }; // For Mul, the legacy cost model checks both operands. - if (I->getOpcode() == Instruction::Mul) + if (VPI->getOpcode() == Instruction::Mul) NewOps[0] = GetConstantViaSCEV(NewOps[0]); // For other binops, the legacy cost model only checks the second operand. NewOps[1] = GetConstantViaSCEV(NewOps[1]); @@ -7803,7 +7811,7 @@ VPWidenRecipe *VPRecipeBuilder::tryToWiden(Instruction *I, return new VPWidenRecipe(*I, NewOps); } case Instruction::ExtractValue: { - SmallVector<VPValue *> NewOps(Operands); + SmallVector<VPValue *> NewOps(VPI->operands()); auto *EVI = cast<ExtractValueInst>(I); assert(EVI->getNumIndices() == 1 && "Expected one extractvalue index"); unsigned Idx = EVI->getIndices()[0]; @@ -7813,9 +7821,8 @@ VPWidenRecipe *VPRecipeBuilder::tryToWiden(Instruction *I, }; } -VPHistogramRecipe * -VPRecipeBuilder::tryToWidenHistogram(const HistogramInfo *HI, - ArrayRef<VPValue *> Operands) { +VPHistogramRecipe *VPRecipeBuilder::tryToWidenHistogram(const HistogramInfo *HI, + VPInstruction *VPI) { // FIXME: Support other operations. unsigned Opcode = HI->Update->getOpcode(); assert((Opcode == Instruction::Add || Opcode == Instruction::Sub) && @@ -7823,7 +7830,7 @@ VPRecipeBuilder::tryToWidenHistogram(const HistogramInfo *HI, SmallVector<VPValue *, 3> HGramOps; // Bucket address. - HGramOps.push_back(Operands[1]); + HGramOps.push_back(VPI->getOperand(1)); // Increment value. HGramOps.push_back(getVPValueOrAddLiveIn(HI->Update->getOperand(1))); @@ -7832,12 +7839,12 @@ VPRecipeBuilder::tryToWidenHistogram(const HistogramInfo *HI, if (Legal->isMaskRequired(HI->Store)) HGramOps.push_back(getBlockInMask(Builder.getInsertBlock())); - return new VPHistogramRecipe(Opcode, HGramOps, HI->Store->getDebugLoc()); + return new VPHistogramRecipe(Opcode, HGramOps, VPI->getDebugLoc()); } -VPReplicateRecipe * -VPRecipeBuilder::handleReplication(Instruction *I, ArrayRef<VPValue *> Operands, - VFRange &Range) { +VPReplicateRecipe *VPRecipeBuilder::handleReplication(VPInstruction *VPI, + VFRange &Range) { + auto *I = VPI->getUnderlyingInstr(); bool IsUniform = LoopVectorizationPlanner::getDecisionAndClampRange( [&](ElementCount VF) { return CM.isUniformAfterVectorization(I, VF); }, Range); @@ -7893,8 +7900,8 @@ VPRecipeBuilder::handleReplication(Instruction *I, ArrayRef<VPValue *> Operands, assert((Range.Start.isScalar() || !IsUniform || !IsPredicated || (Range.Start.isScalable() && isa<IntrinsicInst>(I))) && "Should not predicate a uniform recipe"); - auto *Recipe = new VPReplicateRecipe(I, Operands, IsUniform, BlockInMask, - VPIRMetadata(*I, LVer)); + auto *Recipe = new VPReplicateRecipe(I, VPI->operands(), IsUniform, + BlockInMask, VPIRMetadata(*I, LVer)); return Recipe; } @@ -8075,8 +8082,6 @@ VPRecipeBase *VPRecipeBuilder::tryToCreateWidenRecipe(VPSingleDefRecipe *R, // First, check for specific widening recipes that deal with inductions, Phi // nodes, calls and memory operations. VPRecipeBase *Recipe; - Instruction *Instr = R->getUnderlyingInstr(); - SmallVector<VPValue *, 4> Operands(R->operands()); if (auto *PhiR = dyn_cast<VPPhi>(R)) { VPBasicBlock *Parent = PhiR->getParent(); [[maybe_unused]] VPRegionBlock *LoopRegionOf = @@ -8084,15 +8089,15 @@ VPRecipeBase *VPRecipeBuilder::tryToCreateWidenRecipe(VPSingleDefRecipe *R, assert(LoopRegionOf && LoopRegionOf->getEntry() == Parent && "Non-header phis should have been handled during predication"); auto *Phi = cast<PHINode>(R->getUnderlyingInstr()); - assert(Operands.size() == 2 && "Must have 2 operands for header phis"); - if ((Recipe = tryToOptimizeInductionPHI(Phi, Operands, Range))) + assert(R->getNumOperands() == 2 && "Must have 2 operands for header phis"); + if ((Recipe = tryToOptimizeInductionPHI(PhiR, Range))) return Recipe; VPHeaderPHIRecipe *PhiRecipe = nullptr; assert((Legal->isReductionVariable(Phi) || Legal->isFixedOrderRecurrence(Phi)) && "can only widen reductions and fixed-order recurrences here"); - VPValue *StartV = Operands[0]; + VPValue *StartV = R->getOperand(0); if (Legal->isReductionVariable(Phi)) { const RecurrenceDescriptor &RdxDesc = Legal->getRecurrenceDescriptor(Phi); assert(RdxDesc.getRecurrenceStartValue() == @@ -8112,13 +8117,15 @@ VPRecipeBase *VPRecipeBuilder::tryToCreateWidenRecipe(VPSingleDefRecipe *R, PhiRecipe = new VPFirstOrderRecurrencePHIRecipe(Phi, *StartV); } // Add backedge value. - PhiRecipe->addOperand(Operands[1]); + PhiRecipe->addOperand(R->getOperand(1)); return PhiRecipe; } assert(!R->isPhi() && "only VPPhi nodes expected at this point"); - if (isa<TruncInst>(Instr) && (Recipe = tryToOptimizeInductionTruncate( - cast<TruncInst>(Instr), Operands, Range))) + auto *VPI = cast<VPInstruction>(R); + Instruction *Instr = R->getUnderlyingInstr(); + if (VPI->getOpcode() == Instruction::Trunc && + (Recipe = tryToOptimizeInductionTruncate(VPI, Range))) return Recipe; // All widen recipes below deal only with VF > 1. @@ -8126,46 +8133,46 @@ VPRecipeBase *VPRecipeBuilder::tryToCreateWidenRecipe(VPSingleDefRecipe *R, [&](ElementCount VF) { return VF.isScalar(); }, Range)) return nullptr; - if (auto *CI = dyn_cast<CallInst>(Instr)) - return tryToWidenCall(CI, Operands, Range); + if (VPI->getOpcode() == Instruction::Call) + return tryToWidenCall(VPI, Range); - if (StoreInst *SI = dyn_cast<StoreInst>(Instr)) - if (auto HistInfo = Legal->getHistogramInfo(SI)) - return tryToWidenHistogram(*HistInfo, Operands); + if (VPI->getOpcode() == Instruction::Store) + if (auto HistInfo = Legal->getHistogramInfo(cast<StoreInst>(Instr))) + return tryToWidenHistogram(*HistInfo, VPI); - if (isa<LoadInst>(Instr) || isa<StoreInst>(Instr)) - return tryToWidenMemory(Instr, Operands, Range); + if (VPI->getOpcode() == Instruction::Load || + VPI->getOpcode() == Instruction::Store) + return tryToWidenMemory(VPI, Range); if (std::optional<unsigned> ScaleFactor = getScalingForReduction(Instr)) - return tryToCreatePartialReduction(Instr, Operands, ScaleFactor.value()); + return tryToCreatePartialReduction(VPI, ScaleFactor.value()); if (!shouldWiden(Instr, Range)) return nullptr; - if (auto *GEP = dyn_cast<GetElementPtrInst>(Instr)) - return new VPWidenGEPRecipe(GEP, Operands); + if (VPI->getOpcode() == Instruction::GetElementPtr) + return new VPWidenGEPRecipe(cast<GetElementPtrInst>(Instr), R->operands()); - if (auto *SI = dyn_cast<SelectInst>(Instr)) { - return new VPWidenSelectRecipe(*SI, Operands); - } + if (VPI->getOpcode() == Instruction::Select) + return new VPWidenSelectRecipe(*cast<SelectInst>(Instr), R->operands()); - if (auto *CI = dyn_cast<CastInst>(Instr)) { - return new VPWidenCastRecipe(CI->getOpcode(), Operands[0], CI->getType(), - *CI); + if (Instruction::isCast(VPI->getOpcode())) { + auto *CI = cast<CastInst>(Instr); + return new VPWidenCastRecipe(CI->getOpcode(), VPI->getOperand(0), + CI->getType(), *CI); } - return tryToWiden(Instr, Operands); + return tryToWiden(VPI); } VPRecipeBase * -VPRecipeBuilder::tryToCreatePartialReduction(Instruction *Reduction, - ArrayRef<VPValue *> Operands, +VPRecipeBuilder::tryToCreatePartialReduction(VPInstruction *Reduction, unsigned ScaleFactor) { - assert(Operands.size() == 2 && + assert(Reduction->getNumOperands() == 2 && "Unexpected number of operands for partial reduction"); - VPValue *BinOp = Operands[0]; - VPValue *Accumulator = Operands[1]; + VPValue *BinOp = Reduction->getOperand(0); + VPValue *Accumulator = Reduction->getOperand(1); VPRecipeBase *BinOpRecipe = BinOp->getDefiningRecipe(); if (isa<VPReductionPHIRecipe>(BinOpRecipe) || isa<VPPartialReductionRecipe>(BinOpRecipe)) @@ -8176,28 +8183,29 @@ VPRecipeBuilder::tryToCreatePartialReduction(Instruction *Reduction, "all accumulators in chain must have same scale factor"); unsigned ReductionOpcode = Reduction->getOpcode(); + auto *ReductionI = Reduction->getUnderlyingInstr(); if (ReductionOpcode == Instruction::Sub) { - auto *const Zero = ConstantInt::get(Reduction->getType(), 0); + auto *const Zero = ConstantInt::get(ReductionI->getType(), 0); SmallVector<VPValue *, 2> Ops; Ops.push_back(Plan.getOrAddLiveIn(Zero)); Ops.push_back(BinOp); - BinOp = new VPWidenRecipe(*Reduction, Ops); + BinOp = new VPWidenRecipe(*ReductionI, Ops); Builder.insert(BinOp->getDefiningRecipe()); ReductionOpcode = Instruction::Add; } VPValue *Cond = nullptr; - if (CM.blockNeedsPredicationForAnyReason(Reduction->getParent())) { + if (CM.blockNeedsPredicationForAnyReason(ReductionI->getParent())) { assert((ReductionOpcode == Instruction::Add || ReductionOpcode == Instruction::Sub) && "Expected an ADD or SUB operation for predicated partial " "reductions (because the neutral element in the mask is zero)!"); Cond = getBlockInMask(Builder.getInsertBlock()); - VPValue *Zero = Plan.getConstantInt(Reduction->getType(), 0); + VPValue *Zero = Plan.getConstantInt(ReductionI->getType(), 0); BinOp = Builder.createSelect(Cond, BinOp, Zero, Reduction->getDebugLoc()); } return new VPPartialReductionRecipe(ReductionOpcode, Accumulator, BinOp, Cond, - ScaleFactor, Reduction); + ScaleFactor, ReductionI); } void LoopVectorizationPlanner::buildVPlansWithVPRecipes(ElementCount MinVF, @@ -8382,7 +8390,8 @@ VPlanPtr LoopVectorizationPlanner::tryToBuildVPlanWithVPRecipes( VPRecipeBase *Recipe = RecipeBuilder.tryToCreateWidenRecipe(SingleDef, Range); if (!Recipe) - Recipe = RecipeBuilder.handleReplication(Instr, R.operands(), Range); + Recipe = RecipeBuilder.handleReplication(cast<VPInstruction>(SingleDef), + Range); RecipeBuilder.setRecipe(Instr, Recipe); if (isa<VPWidenIntOrFpInductionRecipe>(Recipe) && isa<TruncInst>(Instr)) { diff --git a/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h b/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h index 41878e3..a7000af 100644 --- a/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h +++ b/llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h @@ -93,42 +93,37 @@ class VPRecipeBuilder { /// Range. The function should not be called for memory instructions or calls. bool shouldWiden(Instruction *I, VFRange &Range) const; - /// Check if the load or store instruction \p I should widened for \p + /// Check if the load or store instruction \p VPI should widened for \p /// Range.Start and potentially masked. Such instructions are handled by a /// recipe that takes an additional VPInstruction for the mask. - VPWidenMemoryRecipe *tryToWidenMemory(Instruction *I, - ArrayRef<VPValue *> Operands, - VFRange &Range); + VPWidenMemoryRecipe *tryToWidenMemory(VPInstruction *VPI, VFRange &Range); - /// Check if an induction recipe should be constructed for \p Phi. If so build + /// Check if an induction recipe should be constructed for \p VPI. If so build /// and return it. If not, return null. - VPHeaderPHIRecipe *tryToOptimizeInductionPHI(PHINode *Phi, - ArrayRef<VPValue *> Operands, + VPHeaderPHIRecipe *tryToOptimizeInductionPHI(VPInstruction *VPI, VFRange &Range); - /// Optimize the special case where the operand of \p I is a constant integer - /// induction variable. + /// Optimize the special case where the operand of \p VPI is a constant + /// integer induction variable. VPWidenIntOrFpInductionRecipe * - tryToOptimizeInductionTruncate(TruncInst *I, ArrayRef<VPValue *> Operands, - VFRange &Range); + tryToOptimizeInductionTruncate(VPInstruction *VPI, VFRange &Range); - /// Handle call instructions. If \p CI can be widened for \p Range.Start, + /// Handle call instructions. If \p VPI can be widened for \p Range.Start, /// return a new VPWidenCallRecipe or VPWidenIntrinsicRecipe. Range.End may be /// decreased to ensure same decision from \p Range.Start to \p Range.End. - VPSingleDefRecipe *tryToWidenCall(CallInst *CI, ArrayRef<VPValue *> Operands, - VFRange &Range); + VPSingleDefRecipe *tryToWidenCall(VPInstruction *VPI, VFRange &Range); - /// Check if \p I has an opcode that can be widened and return a VPWidenRecipe - /// if it can. The function should only be called if the cost-model indicates - /// that widening should be performed. - VPWidenRecipe *tryToWiden(Instruction *I, ArrayRef<VPValue *> Operands); + /// Check if \p VPI has an opcode that can be widened and return a + /// VPWidenRecipe if it can. The function should only be called if the + /// cost-model indicates that widening should be performed. + VPWidenRecipe *tryToWiden(VPInstruction *VPI); /// Makes Histogram count operations safe for vectorization, by emitting a /// llvm.experimental.vector.histogram.add intrinsic in place of the /// Load + Add|Sub + Store operations that perform the histogram in the /// original scalar loop. VPHistogramRecipe *tryToWidenHistogram(const HistogramInfo *HI, - ArrayRef<VPValue *> Operands); + VPInstruction *VPI); /// Examines reduction operations to see if the target can use a cheaper /// operation with a wider per-iteration input VF and narrower PHI VF. @@ -171,8 +166,7 @@ public: /// Create and return a partial reduction recipe for a reduction instruction /// along with binary operation and reduction phi operands. - VPRecipeBase *tryToCreatePartialReduction(Instruction *Reduction, - ArrayRef<VPValue *> Operands, + VPRecipeBase *tryToCreatePartialReduction(VPInstruction *Reduction, unsigned ScaleFactor); /// Set the recipe created for given ingredient. @@ -197,12 +191,10 @@ public: return Ingredient2Recipe[I]; } - /// Build a VPReplicationRecipe for \p I using \p Operands. If it is - /// predicated, add the mask as last operand. Range.End may be decreased to - /// ensure same recipe behavior from \p Range.Start to \p Range.End. - VPReplicateRecipe *handleReplication(Instruction *I, - ArrayRef<VPValue *> Operands, - VFRange &Range); + /// Build a VPReplicationRecipe for \p VPI. If it is predicated, add the mask + /// as last operand. Range.End may be decreased to ensure same recipe behavior + /// from \p Range.Start to \p Range.End. + VPReplicateRecipe *handleReplication(VPInstruction *VPI, VFRange &Range); VPValue *getVPValueOrAddLiveIn(Value *V) { if (auto *I = dyn_cast<Instruction>(V)) { diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp index 48bd697..634df51 100644 --- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp +++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp @@ -1288,8 +1288,9 @@ static void simplifyRecipe(VPSingleDefRecipe *Def, VPTypeAnalysis &TypeInfo) { // Look through broadcast of single-scalar when used as select conditions; in // that case the scalar condition can be used directly. if (match(Def, - m_Select(m_Broadcast(m_VPValue(C)), m_VPValue(), m_VPValue())) && - vputils::isSingleScalar(C)) { + m_Select(m_Broadcast(m_VPValue(C)), m_VPValue(), m_VPValue()))) { + assert(vputils::isSingleScalar(C) && + "broadcast operand must be single-scalar"); Def->setOperand(0, C); return; } diff --git a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp index 27a8bbd..ed3a0a0 100644 --- a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp +++ b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp @@ -696,11 +696,11 @@ bool VectorCombine::foldExtractExtract(Instruction &I) { /// shuffle. bool VectorCombine::foldInsExtFNeg(Instruction &I) { // Match an insert (op (extract)) pattern. - Value *DestVec; - uint64_t Index; + Value *DstVec; + uint64_t ExtIdx, InsIdx; Instruction *FNeg; - if (!match(&I, m_InsertElt(m_Value(DestVec), m_OneUse(m_Instruction(FNeg)), - m_ConstantInt(Index)))) + if (!match(&I, m_InsertElt(m_Value(DstVec), m_OneUse(m_Instruction(FNeg)), + m_ConstantInt(InsIdx)))) return false; // Note: This handles the canonical fneg instruction and "fsub -0.0, X". @@ -708,67 +708,74 @@ bool VectorCombine::foldInsExtFNeg(Instruction &I) { Instruction *Extract; if (!match(FNeg, m_FNeg(m_CombineAnd( m_Instruction(Extract), - m_ExtractElt(m_Value(SrcVec), m_SpecificInt(Index)))))) + m_ExtractElt(m_Value(SrcVec), m_ConstantInt(ExtIdx)))))) return false; - auto *VecTy = cast<FixedVectorType>(I.getType()); - auto *ScalarTy = VecTy->getScalarType(); + auto *DstVecTy = cast<FixedVectorType>(DstVec->getType()); + auto *DstVecScalarTy = DstVecTy->getScalarType(); auto *SrcVecTy = dyn_cast<FixedVectorType>(SrcVec->getType()); - if (!SrcVecTy || ScalarTy != SrcVecTy->getScalarType()) + if (!SrcVecTy || DstVecScalarTy != SrcVecTy->getScalarType()) return false; - // Ignore bogus insert/extract index. - unsigned NumElts = VecTy->getNumElements(); - if (Index >= NumElts) + // Ignore if insert/extract index is out of bounds or destination vector has + // one element + unsigned NumDstElts = DstVecTy->getNumElements(); + unsigned NumSrcElts = SrcVecTy->getNumElements(); + if (ExtIdx > NumSrcElts || InsIdx >= NumDstElts || NumDstElts == 1) return false; // We are inserting the negated element into the same lane that we extracted // from. This is equivalent to a select-shuffle that chooses all but the // negated element from the destination vector. - SmallVector<int> Mask(NumElts); + SmallVector<int> Mask(NumDstElts); std::iota(Mask.begin(), Mask.end(), 0); - Mask[Index] = Index + NumElts; + Mask[InsIdx] = (ExtIdx % NumDstElts) + NumDstElts; InstructionCost OldCost = - TTI.getArithmeticInstrCost(Instruction::FNeg, ScalarTy, CostKind) + - TTI.getVectorInstrCost(I, VecTy, CostKind, Index); + TTI.getArithmeticInstrCost(Instruction::FNeg, DstVecScalarTy, CostKind) + + TTI.getVectorInstrCost(I, DstVecTy, CostKind, InsIdx); // If the extract has one use, it will be eliminated, so count it in the // original cost. If it has more than one use, ignore the cost because it will // be the same before/after. if (Extract->hasOneUse()) - OldCost += TTI.getVectorInstrCost(*Extract, VecTy, CostKind, Index); + OldCost += TTI.getVectorInstrCost(*Extract, SrcVecTy, CostKind, ExtIdx); InstructionCost NewCost = - TTI.getArithmeticInstrCost(Instruction::FNeg, VecTy, CostKind) + - TTI.getShuffleCost(TargetTransformInfo::SK_PermuteTwoSrc, VecTy, VecTy, - Mask, CostKind); + TTI.getArithmeticInstrCost(Instruction::FNeg, SrcVecTy, CostKind) + + TTI.getShuffleCost(TargetTransformInfo::SK_PermuteTwoSrc, DstVecTy, + DstVecTy, Mask, CostKind); - bool NeedLenChg = SrcVecTy->getNumElements() != NumElts; + bool NeedLenChg = SrcVecTy->getNumElements() != NumDstElts; // If the lengths of the two vectors are not equal, // we need to add a length-change vector. Add this cost. SmallVector<int> SrcMask; if (NeedLenChg) { - SrcMask.assign(NumElts, PoisonMaskElem); - SrcMask[Index] = Index; + SrcMask.assign(NumDstElts, PoisonMaskElem); + SrcMask[ExtIdx % NumDstElts] = ExtIdx; NewCost += TTI.getShuffleCost(TargetTransformInfo::SK_PermuteSingleSrc, - VecTy, SrcVecTy, SrcMask, CostKind); + DstVecTy, SrcVecTy, SrcMask, CostKind); } + LLVM_DEBUG(dbgs() << "Found an insertion of (extract)fneg : " << I + << "\n OldCost: " << OldCost << " vs NewCost: " << NewCost + << "\n"); if (NewCost > OldCost) return false; - Value *NewShuf; - // insertelt DestVec, (fneg (extractelt SrcVec, Index)), Index + Value *NewShuf, *LenChgShuf = nullptr; + // insertelt DstVec, (fneg (extractelt SrcVec, Index)), Index Value *VecFNeg = Builder.CreateFNegFMF(SrcVec, FNeg); if (NeedLenChg) { - // shuffle DestVec, (shuffle (fneg SrcVec), poison, SrcMask), Mask - Value *LenChgShuf = Builder.CreateShuffleVector(VecFNeg, SrcMask); - NewShuf = Builder.CreateShuffleVector(DestVec, LenChgShuf, Mask); + // shuffle DstVec, (shuffle (fneg SrcVec), poison, SrcMask), Mask + LenChgShuf = Builder.CreateShuffleVector(VecFNeg, SrcMask); + NewShuf = Builder.CreateShuffleVector(DstVec, LenChgShuf, Mask); + Worklist.pushValue(LenChgShuf); } else { - // shuffle DestVec, (fneg SrcVec), Mask - NewShuf = Builder.CreateShuffleVector(DestVec, VecFNeg, Mask); + // shuffle DstVec, (fneg SrcVec), Mask + NewShuf = Builder.CreateShuffleVector(DstVec, VecFNeg, Mask); } + Worklist.pushValue(VecFNeg); replaceValue(I, *NewShuf); return true; } diff --git a/llvm/test/Analysis/CostModel/AArch64/sve-arith-fp.ll b/llvm/test/Analysis/CostModel/AArch64/sve-arith-fp.ll index 1c40354..ec848c2 100644 --- a/llvm/test/Analysis/CostModel/AArch64/sve-arith-fp.ll +++ b/llvm/test/Analysis/CostModel/AArch64/sve-arith-fp.ll @@ -5,214 +5,214 @@ target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128" define void @fadd() { ; CHECK-LABEL: 'fadd' -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fadd <vscale x 4 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fadd <vscale x 8 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fadd <vscale x 16 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of Invalid for: %V1F32 = fadd <vscale x 1 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fadd <vscale x 2 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fadd <vscale x 4 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fadd <vscale x 8 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fadd <vscale x 2 x double> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fadd <vscale x 4 x double> undef, undef +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fadd <vscale x 4 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fadd <vscale x 8 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fadd <vscale x 16 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of Invalid for: %V1F32 = fadd <vscale x 1 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fadd <vscale x 2 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fadd <vscale x 4 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fadd <vscale x 8 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fadd <vscale x 2 x double> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fadd <vscale x 4 x double> poison, poison ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = fadd <vscale x 4 x half> undef, undef - %V8F16 = fadd <vscale x 8 x half> undef, undef - %V16F16 = fadd <vscale x 16 x half> undef, undef + %V4F16 = fadd <vscale x 4 x half> poison, poison + %V8F16 = fadd <vscale x 8 x half> poison, poison + %V16F16 = fadd <vscale x 16 x half> poison, poison - %V1F32 = fadd <vscale x 1 x float> undef, undef - %V2F32 = fadd <vscale x 2 x float> undef, undef - %V4F32 = fadd <vscale x 4 x float> undef, undef - %V8F32 = fadd <vscale x 8 x float> undef, undef + %V1F32 = fadd <vscale x 1 x float> poison, poison + %V2F32 = fadd <vscale x 2 x float> poison, poison + %V4F32 = fadd <vscale x 4 x float> poison, poison + %V8F32 = fadd <vscale x 8 x float> poison, poison - %V2F64 = fadd <vscale x 2 x double> undef, undef - %V4F64 = fadd <vscale x 4 x double> undef, undef + %V2F64 = fadd <vscale x 2 x double> poison, poison + %V4F64 = fadd <vscale x 4 x double> poison, poison ret void } define void @fsub() { ; CHECK-LABEL: 'fsub' -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fsub <vscale x 4 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fsub <vscale x 8 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fsub <vscale x 16 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of Invalid for: %V1F32 = fsub <vscale x 1 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fsub <vscale x 2 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fsub <vscale x 4 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fsub <vscale x 8 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fsub <vscale x 2 x double> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fsub <vscale x 4 x double> undef, undef +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fsub <vscale x 4 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fsub <vscale x 8 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fsub <vscale x 16 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of Invalid for: %V1F32 = fsub <vscale x 1 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fsub <vscale x 2 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fsub <vscale x 4 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fsub <vscale x 8 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fsub <vscale x 2 x double> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fsub <vscale x 4 x double> poison, poison ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = fsub <vscale x 4 x half> undef, undef - %V8F16 = fsub <vscale x 8 x half> undef, undef - %V16F16 = fsub <vscale x 16 x half> undef, undef + %V4F16 = fsub <vscale x 4 x half> poison, poison + %V8F16 = fsub <vscale x 8 x half> poison, poison + %V16F16 = fsub <vscale x 16 x half> poison, poison - %V1F32 = fsub <vscale x 1 x float> undef, undef - %V2F32 = fsub <vscale x 2 x float> undef, undef - %V4F32 = fsub <vscale x 4 x float> undef, undef - %V8F32 = fsub <vscale x 8 x float> undef, undef + %V1F32 = fsub <vscale x 1 x float> poison, poison + %V2F32 = fsub <vscale x 2 x float> poison, poison + %V4F32 = fsub <vscale x 4 x float> poison, poison + %V8F32 = fsub <vscale x 8 x float> poison, poison - %V2F64 = fsub <vscale x 2 x double> undef, undef - %V4F64 = fsub <vscale x 4 x double> undef, undef + %V2F64 = fsub <vscale x 2 x double> poison, poison + %V4F64 = fsub <vscale x 4 x double> poison, poison ret void } define void @fneg() { ; CHECK-LABEL: 'fneg' -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F16 = fneg <vscale x 2 x half> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fneg <vscale x 4 x half> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fneg <vscale x 8 x half> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fneg <vscale x 16 x half> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fneg <vscale x 2 x float> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fneg <vscale x 4 x float> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fneg <vscale x 8 x float> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fneg <vscale x 2 x double> undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fneg <vscale x 4 x double> undef +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F16 = fneg <vscale x 2 x half> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fneg <vscale x 4 x half> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fneg <vscale x 8 x half> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fneg <vscale x 16 x half> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fneg <vscale x 2 x float> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fneg <vscale x 4 x float> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fneg <vscale x 8 x float> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:1 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fneg <vscale x 2 x double> poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fneg <vscale x 4 x double> poison ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V2F16 = fneg <vscale x 2 x half> undef - %V4F16 = fneg <vscale x 4 x half> undef - %V8F16 = fneg <vscale x 8 x half> undef - %V16F16 = fneg <vscale x 16 x half> undef + %V2F16 = fneg <vscale x 2 x half> poison + %V4F16 = fneg <vscale x 4 x half> poison + %V8F16 = fneg <vscale x 8 x half> poison + %V16F16 = fneg <vscale x 16 x half> poison - %V2F32 = fneg <vscale x 2 x float> undef - %V4F32 = fneg <vscale x 4 x float> undef - %V8F32 = fneg <vscale x 8 x float> undef + %V2F32 = fneg <vscale x 2 x float> poison + %V4F32 = fneg <vscale x 4 x float> poison + %V8F32 = fneg <vscale x 8 x float> poison - %V2F64 = fneg <vscale x 2 x double> undef - %V4F64 = fneg <vscale x 4 x double> undef + %V2F64 = fneg <vscale x 2 x double> poison + %V4F64 = fneg <vscale x 4 x double> poison ret void } define void @fmul() { ; CHECK-LABEL: 'fmul' -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fmul <vscale x 4 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fmul <vscale x 8 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fmul <vscale x 16 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fmul <vscale x 2 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fmul <vscale x 4 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fmul <vscale x 8 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fmul <vscale x 2 x double> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fmul <vscale x 4 x double> undef, undef +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = fmul <vscale x 4 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = fmul <vscale x 8 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = fmul <vscale x 16 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = fmul <vscale x 2 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = fmul <vscale x 4 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = fmul <vscale x 8 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = fmul <vscale x 2 x double> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = fmul <vscale x 4 x double> poison, poison ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = fmul <vscale x 4 x half> undef, undef - %V8F16 = fmul <vscale x 8 x half> undef, undef - %V16F16 = fmul <vscale x 16 x half> undef, undef + %V4F16 = fmul <vscale x 4 x half> poison, poison + %V8F16 = fmul <vscale x 8 x half> poison, poison + %V16F16 = fmul <vscale x 16 x half> poison, poison - %V2F32 = fmul <vscale x 2 x float> undef, undef - %V4F32 = fmul <vscale x 4 x float> undef, undef - %V8F32 = fmul <vscale x 8 x float> undef, undef + %V2F32 = fmul <vscale x 2 x float> poison, poison + %V4F32 = fmul <vscale x 4 x float> poison, poison + %V8F32 = fmul <vscale x 8 x float> poison, poison - %V2F64 = fmul <vscale x 2 x double> undef, undef - %V4F64 = fmul <vscale x 4 x double> undef, undef + %V2F64 = fmul <vscale x 2 x double> poison, poison + %V4F64 = fmul <vscale x 4 x double> poison, poison ret void } define void @fdiv() { ; CHECK-LABEL: 'fdiv' -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V4F16 = fdiv <vscale x 4 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V8F16 = fdiv <vscale x 8 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of 4 for: %V16F16 = fdiv <vscale x 16 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V2F32 = fdiv <vscale x 2 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V4F32 = fdiv <vscale x 4 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of 4 for: %V8F32 = fdiv <vscale x 8 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V2F64 = fdiv <vscale x 2 x double> undef, undef -; CHECK-NEXT: Cost Model: Found costs of 4 for: %V4F64 = fdiv <vscale x 4 x double> undef, undef +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V4F16 = fdiv <vscale x 4 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V8F16 = fdiv <vscale x 8 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of 4 for: %V16F16 = fdiv <vscale x 16 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V2F32 = fdiv <vscale x 2 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V4F32 = fdiv <vscale x 4 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of 4 for: %V8F32 = fdiv <vscale x 8 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:4 Lat:4 SizeLat:4 for: %V2F64 = fdiv <vscale x 2 x double> poison, poison +; CHECK-NEXT: Cost Model: Found costs of 4 for: %V4F64 = fdiv <vscale x 4 x double> poison, poison ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = fdiv <vscale x 4 x half> undef, undef - %V8F16 = fdiv <vscale x 8 x half> undef, undef - %V16F16 = fdiv <vscale x 16 x half> undef, undef + %V4F16 = fdiv <vscale x 4 x half> poison, poison + %V8F16 = fdiv <vscale x 8 x half> poison, poison + %V16F16 = fdiv <vscale x 16 x half> poison, poison - %V2F32 = fdiv <vscale x 2 x float> undef, undef - %V4F32 = fdiv <vscale x 4 x float> undef, undef - %V8F32 = fdiv <vscale x 8 x float> undef, undef + %V2F32 = fdiv <vscale x 2 x float> poison, poison + %V4F32 = fdiv <vscale x 4 x float> poison, poison + %V8F32 = fdiv <vscale x 8 x float> poison, poison - %V2F64 = fdiv <vscale x 2 x double> undef, undef - %V4F64 = fdiv <vscale x 4 x double> undef, undef + %V2F64 = fdiv <vscale x 2 x double> poison, poison + %V4F64 = fdiv <vscale x 4 x double> poison, poison ret void } define void @frem() { ; CHECK-LABEL: 'frem' -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V4F16 = frem <vscale x 4 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V8F16 = frem <vscale x 8 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V16F16 = frem <vscale x 16 x half> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V2F32 = frem <vscale x 2 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V4F32 = frem <vscale x 4 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V8F32 = frem <vscale x 8 x float> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V2F64 = frem <vscale x 2 x double> undef, undef -; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V4F64 = frem <vscale x 4 x double> undef, undef +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V4F16 = frem <vscale x 4 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V8F16 = frem <vscale x 8 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V16F16 = frem <vscale x 16 x half> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V2F32 = frem <vscale x 2 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V4F32 = frem <vscale x 4 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V8F32 = frem <vscale x 8 x float> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V2F64 = frem <vscale x 2 x double> poison, poison +; CHECK-NEXT: Cost Model: Found costs of RThru:Invalid CodeSize:4 Lat:4 SizeLat:4 for: %V4F64 = frem <vscale x 4 x double> poison, poison ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = frem <vscale x 4 x half> undef, undef - %V8F16 = frem <vscale x 8 x half> undef, undef - %V16F16 = frem <vscale x 16 x half> undef, undef + %V4F16 = frem <vscale x 4 x half> poison, poison + %V8F16 = frem <vscale x 8 x half> poison, poison + %V16F16 = frem <vscale x 16 x half> poison, poison - %V2F32 = frem <vscale x 2 x float> undef, undef - %V4F32 = frem <vscale x 4 x float> undef, undef - %V8F32 = frem <vscale x 8 x float> undef, undef + %V2F32 = frem <vscale x 2 x float> poison, poison + %V4F32 = frem <vscale x 4 x float> poison, poison + %V8F32 = frem <vscale x 8 x float> poison, poison - %V2F64 = frem <vscale x 2 x double> undef, undef - %V4F64 = frem <vscale x 4 x double> undef, undef + %V2F64 = frem <vscale x 2 x double> poison, poison + %V4F64 = frem <vscale x 4 x double> poison, poison ret void } define void @fma() { ; CHECK-LABEL: 'fma' -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = call <vscale x 4 x half> @llvm.fma.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = call <vscale x 8 x half> @llvm.fma.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x half> undef, <vscale x 8 x half> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = call <vscale x 16 x half> @llvm.fma.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x half> undef, <vscale x 16 x half> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = call <vscale x 2 x float> @llvm.fma.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = call <vscale x 4 x float> @llvm.fma.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = call <vscale x 8 x float> @llvm.fma.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x float> undef, <vscale x 8 x float> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = call <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = call <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = call <vscale x 4 x half> @llvm.fma.nxv4f16(<vscale x 4 x half> poison, <vscale x 4 x half> poison, <vscale x 4 x half> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = call <vscale x 8 x half> @llvm.fma.nxv8f16(<vscale x 8 x half> poison, <vscale x 8 x half> poison, <vscale x 8 x half> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = call <vscale x 16 x half> @llvm.fma.nxv16f16(<vscale x 16 x half> poison, <vscale x 16 x half> poison, <vscale x 16 x half> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = call <vscale x 2 x float> @llvm.fma.nxv2f32(<vscale x 2 x float> poison, <vscale x 2 x float> poison, <vscale x 2 x float> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = call <vscale x 4 x float> @llvm.fma.nxv4f32(<vscale x 4 x float> poison, <vscale x 4 x float> poison, <vscale x 4 x float> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = call <vscale x 8 x float> @llvm.fma.nxv8f32(<vscale x 8 x float> poison, <vscale x 8 x float> poison, <vscale x 8 x float> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = call <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double> poison, <vscale x 2 x double> poison, <vscale x 2 x double> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = call <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double> poison, <vscale x 4 x double> poison, <vscale x 4 x double> poison) ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = call <vscale x 4 x half> @llvm.fma.v4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef) - %V8F16 = call <vscale x 8 x half> @llvm.fma.v8f16(<vscale x 8 x half> undef, <vscale x 8 x half> undef, <vscale x 8 x half> undef) - %V16F16 = call <vscale x 16 x half> @llvm.fma.v16f16(<vscale x 16 x half> undef, <vscale x 16 x half> undef, <vscale x 16 x half> undef) + %V4F16 = call <vscale x 4 x half> @llvm.fma.v4f16(<vscale x 4 x half> poison, <vscale x 4 x half> poison, <vscale x 4 x half> poison) + %V8F16 = call <vscale x 8 x half> @llvm.fma.v8f16(<vscale x 8 x half> poison, <vscale x 8 x half> poison, <vscale x 8 x half> poison) + %V16F16 = call <vscale x 16 x half> @llvm.fma.v16f16(<vscale x 16 x half> poison, <vscale x 16 x half> poison, <vscale x 16 x half> poison) - %V2F32 = call <vscale x 2 x float> @llvm.fma.v2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef) - %V4F32 = call <vscale x 4 x float> @llvm.fma.v4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef) - %V8F32 = call <vscale x 8 x float> @llvm.fma.v8f32(<vscale x 8 x float> undef, <vscale x 8 x float> undef, <vscale x 8 x float> undef) + %V2F32 = call <vscale x 2 x float> @llvm.fma.v2f32(<vscale x 2 x float> poison, <vscale x 2 x float> poison, <vscale x 2 x float> poison) + %V4F32 = call <vscale x 4 x float> @llvm.fma.v4f32(<vscale x 4 x float> poison, <vscale x 4 x float> poison, <vscale x 4 x float> poison) + %V8F32 = call <vscale x 8 x float> @llvm.fma.v8f32(<vscale x 8 x float> poison, <vscale x 8 x float> poison, <vscale x 8 x float> poison) - %V2F64 = call <vscale x 2 x double> @llvm.fma.v2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef) - %V4F64 = call <vscale x 4 x double> @llvm.fma.v4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef) + %V2F64 = call <vscale x 2 x double> @llvm.fma.v2f64(<vscale x 2 x double> poison, <vscale x 2 x double> poison, <vscale x 2 x double> poison) + %V4F64 = call <vscale x 4 x double> @llvm.fma.v4f64(<vscale x 4 x double> poison, <vscale x 4 x double> poison, <vscale x 4 x double> poison) ret void } define void @fmuladd() { ; CHECK-LABEL: 'fmuladd' -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = call <vscale x 4 x half> @llvm.fmuladd.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = call <vscale x 8 x half> @llvm.fmuladd.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x half> undef, <vscale x 8 x half> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = call <vscale x 16 x half> @llvm.fmuladd.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x half> undef, <vscale x 16 x half> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = call <vscale x 2 x float> @llvm.fmuladd.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = call <vscale x 4 x float> @llvm.fmuladd.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = call <vscale x 8 x float> @llvm.fmuladd.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x float> undef, <vscale x 8 x float> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = call <vscale x 2 x double> @llvm.fmuladd.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef) -; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = call <vscale x 4 x double> @llvm.fmuladd.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F16 = call <vscale x 4 x half> @llvm.fmuladd.nxv4f16(<vscale x 4 x half> poison, <vscale x 4 x half> poison, <vscale x 4 x half> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V8F16 = call <vscale x 8 x half> @llvm.fmuladd.nxv8f16(<vscale x 8 x half> poison, <vscale x 8 x half> poison, <vscale x 8 x half> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V16F16 = call <vscale x 16 x half> @llvm.fmuladd.nxv16f16(<vscale x 16 x half> poison, <vscale x 16 x half> poison, <vscale x 16 x half> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F32 = call <vscale x 2 x float> @llvm.fmuladd.nxv2f32(<vscale x 2 x float> poison, <vscale x 2 x float> poison, <vscale x 2 x float> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V4F32 = call <vscale x 4 x float> @llvm.fmuladd.nxv4f32(<vscale x 4 x float> poison, <vscale x 4 x float> poison, <vscale x 4 x float> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V8F32 = call <vscale x 8 x float> @llvm.fmuladd.nxv8f32(<vscale x 8 x float> poison, <vscale x 8 x float> poison, <vscale x 8 x float> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:2 CodeSize:1 Lat:3 SizeLat:1 for: %V2F64 = call <vscale x 2 x double> @llvm.fmuladd.nxv2f64(<vscale x 2 x double> poison, <vscale x 2 x double> poison, <vscale x 2 x double> poison) +; CHECK-NEXT: Cost Model: Found costs of RThru:4 CodeSize:1 Lat:3 SizeLat:1 for: %V4F64 = call <vscale x 4 x double> @llvm.fmuladd.nxv4f64(<vscale x 4 x double> poison, <vscale x 4 x double> poison, <vscale x 4 x double> poison) ; CHECK-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void ; - %V4F16 = call <vscale x 4 x half> @llvm.fmuladd.v4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef) - %V8F16 = call <vscale x 8 x half> @llvm.fmuladd.v8f16(<vscale x 8 x half> undef, <vscale x 8 x half> undef, <vscale x 8 x half> undef) - %V16F16 = call <vscale x 16 x half> @llvm.fmuladd.v16f16(<vscale x 16 x half> undef, <vscale x 16 x half> undef, <vscale x 16 x half> undef) + %V4F16 = call <vscale x 4 x half> @llvm.fmuladd.v4f16(<vscale x 4 x half> poison, <vscale x 4 x half> poison, <vscale x 4 x half> poison) + %V8F16 = call <vscale x 8 x half> @llvm.fmuladd.v8f16(<vscale x 8 x half> poison, <vscale x 8 x half> poison, <vscale x 8 x half> poison) + %V16F16 = call <vscale x 16 x half> @llvm.fmuladd.v16f16(<vscale x 16 x half> poison, <vscale x 16 x half> poison, <vscale x 16 x half> poison) - %V2F32 = call <vscale x 2 x float> @llvm.fmuladd.v2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef) - %V4F32 = call <vscale x 4 x float> @llvm.fmuladd.v4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef) - %V8F32 = call <vscale x 8 x float> @llvm.fmuladd.v8f32(<vscale x 8 x float> undef, <vscale x 8 x float> undef, <vscale x 8 x float> undef) + %V2F32 = call <vscale x 2 x float> @llvm.fmuladd.v2f32(<vscale x 2 x float> poison, <vscale x 2 x float> poison, <vscale x 2 x float> poison) + %V4F32 = call <vscale x 4 x float> @llvm.fmuladd.v4f32(<vscale x 4 x float> poison, <vscale x 4 x float> poison, <vscale x 4 x float> poison) + %V8F32 = call <vscale x 8 x float> @llvm.fmuladd.v8f32(<vscale x 8 x float> poison, <vscale x 8 x float> poison, <vscale x 8 x float> poison) - %V2F64 = call <vscale x 2 x double> @llvm.fmuladd.v2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef) - %V4F64 = call <vscale x 4 x double> @llvm.fmuladd.v4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef) + %V2F64 = call <vscale x 2 x double> @llvm.fmuladd.v2f64(<vscale x 2 x double> poison, <vscale x 2 x double> poison, <vscale x 2 x double> poison) + %V4F64 = call <vscale x 4 x double> @llvm.fmuladd.v4f64(<vscale x 4 x double> poison, <vscale x 4 x double> poison, <vscale x 4 x double> poison) ret void } diff --git a/llvm/test/Analysis/DependenceAnalysis/monotonicity-no-wrap-flags.ll b/llvm/test/Analysis/DependenceAnalysis/monotonicity-no-wrap-flags.ll index 7411dc9..df42c75 100644 --- a/llvm/test/Analysis/DependenceAnalysis/monotonicity-no-wrap-flags.ll +++ b/llvm/test/Analysis/DependenceAnalysis/monotonicity-no-wrap-flags.ll @@ -298,7 +298,8 @@ exit: } ; The value of step reccurence is not invariant with respect to the outer most -; loop (the i-loop). +; loop (the i-loop). It is theoretically multivariate monotonic by definition, +; but we cannot handle non-affine addrec for now. ; ; offset_i = 0; ; for (int i = 0; i < 100; i++) { @@ -312,7 +313,8 @@ define void @step_is_variant(ptr %a) { ; CHECK-NEXT: Monotonicity check: ; CHECK-NEXT: Inst: store i8 0, ptr %idx, align 1 ; CHECK-NEXT: Expr: {%offset.i,+,1}<nuw><nsw><%loop.j> -; CHECK-NEXT: Monotonicity: MultivariateSignedMonotonic +; CHECK-NEXT: Monotonicity: Unknown +; CHECK-NEXT: Reason: %offset.i ; CHECK-EMPTY: ; CHECK-NEXT: Src: store i8 0, ptr %idx, align 1 --> Dst: store i8 0, ptr %idx, align 1 ; CHECK-NEXT: da analyze - confused! @@ -346,6 +348,56 @@ exit: ret void } +; The value of step reccurence is not invariant with respect to the outer most +; loop (the i-loop). Actually, `offset_i` is not monotonic. +; +; offset_i = 0; +; for (int i = 0; i < 100; i++) { +; for (int j = 0; j < 100; j++) +; a[offset_i + j] = 0; +; offset_i += (i % 2 == 0) ? -1 : 3; +; } +; +define void @step_is_variant2(ptr %a) { +; CHECK-LABEL: 'step_is_variant2' +; CHECK-NEXT: Monotonicity check: +; CHECK-NEXT: Inst: store i8 0, ptr %idx, align 1 +; CHECK-NEXT: Expr: {%offset.i,+,1}<nsw><%loop.j> +; CHECK-NEXT: Monotonicity: Unknown +; CHECK-NEXT: Reason: %offset.i +; CHECK-EMPTY: +; CHECK-NEXT: Src: store i8 0, ptr %idx, align 1 --> Dst: store i8 0, ptr %idx, align 1 +; CHECK-NEXT: da analyze - confused! +; +entry: + br label %loop.i.header + +loop.i.header: + %i = phi i64 [ 0, %entry ], [ %i.inc, %loop.i.latch ] + %offset.i = phi i64 [ 0, %entry ], [ %offset.i.next, %loop.i.latch ] + %step.i.0 = phi i64 [ -1, %entry ], [ %step.i.1, %loop.i.latch ] + %step.i.1 = phi i64 [ 3, %entry ], [ %step.i.0, %loop.i.latch ] + br label %loop.j + +loop.j: + %j = phi i64 [ 0, %loop.i.header ], [ %j.inc, %loop.j ] + %offset = add nsw i64 %offset.i, %j + %idx = getelementptr inbounds i8, ptr %a, i64 %offset + store i8 0, ptr %idx + %j.inc = add nsw i64 %j, 1 + %exitcond.j = icmp eq i64 %j.inc, 100 + br i1 %exitcond.j, label %loop.i.latch, label %loop.j + +loop.i.latch: + %i.inc = add nsw i64 %i, 1 + %offset.i.next = add nsw i64 %offset.i, %step.i.0 + %exitcond.i = icmp eq i64 %i.inc, 100 + br i1 %exitcond.i, label %exit, label %loop.i.header + +exit: + ret void +} + ; The AddRec doesn't have nsw flag for the j-loop, since the store may not be ; executed. ; diff --git a/llvm/test/CMakeLists.txt b/llvm/test/CMakeLists.txt index f01422e..e547c34 100644 --- a/llvm/test/CMakeLists.txt +++ b/llvm/test/CMakeLists.txt @@ -30,6 +30,7 @@ llvm_canonicalize_cmake_booleans( LLVM_INCLUDE_SPIRV_TOOLS_TESTS LLVM_APPEND_VC_REV LLVM_HAS_LOGF128 + LLVM_ENABLE_ONDISK_CAS ) configure_lit_site_cfg( @@ -81,6 +82,7 @@ set(LLVM_TEST_DEPENDS llvm-bcanalyzer llvm-bitcode-strip llvm-c-test + llvm-cas llvm-cat llvm-cfi-verify llvm-cgdata diff --git a/llvm/test/CodeGen/AArch64/aarch64-matmul-fp16.ll b/llvm/test/CodeGen/AArch64/aarch64-matmul-fp16.ll new file mode 100644 index 0000000..8d1abdd --- /dev/null +++ b/llvm/test/CodeGen/AArch64/aarch64-matmul-fp16.ll @@ -0,0 +1,14 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6 +; RUN: llc -mtriple aarch64-none-linux-gnu -mattr=+neon,+f8f16mm < %s | FileCheck %s +; RUN: llc -mtriple aarch64-none-linux-gnu -mattr=+neon,+f8f16mm -global-isel < %s | FileCheck %s + +define <8 x half> @fmmla.v8f16.v16i8(<8 x half> %r, <16 x i8> %a, <16 x i8> %b) { +; CHECK-LABEL: fmmla.v8f16.v16i8: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fmmla v0.8h, v1.16b, v2.16b +; CHECK-NEXT: ret +entry: + %vfmmla1.i = tail call <8 x half> @llvm.aarch64.neon.fmmla.v8f16.v16i8(<8 x half> %r, <16 x i8> %a, <16 x i8> %b) #3 + ret <8 x half> %vfmmla1.i +} + diff --git a/llvm/test/CodeGen/AArch64/aarch64-matmul-fp32.ll b/llvm/test/CodeGen/AArch64/aarch64-matmul-fp32.ll new file mode 100644 index 0000000..4c33567 --- /dev/null +++ b/llvm/test/CodeGen/AArch64/aarch64-matmul-fp32.ll @@ -0,0 +1,13 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6 +; RUN: llc -mtriple aarch64-none-linux-gnu -mattr=+neon,+f8f32mm < %s | FileCheck %s +; RUN: llc -mtriple aarch64-none-linux-gnu -mattr=+neon,+f8f32mm -global-isel < %s | FileCheck %s + +define <4 x float> @fmmla.v4f32.v16i8(<4 x float> %r, <16 x i8> %a, <16 x i8> %b) { +; CHECK-LABEL: fmmla.v4f32.v16i8: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fmmla v0.4s, v1.16b, v2.16b +; CHECK-NEXT: ret +entry: + %vfmmla1.i = tail call <4 x float> @llvm.aarch64.neon.fmmla.v4f32.v16i8(<4 x float> %r, <16 x i8> %a, <16 x i8> %b) #3 + ret <4 x float> %vfmmla1.i +} diff --git a/llvm/test/CodeGen/AArch64/sve2p1-fdot.ll b/llvm/test/CodeGen/AArch64/sve2p1-fdot.ll new file mode 100644 index 0000000..9dbe096 --- /dev/null +++ b/llvm/test/CodeGen/AArch64/sve2p1-fdot.ll @@ -0,0 +1,93 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6 +; RUN: llc -mattr=+sve2 < %s | FileCheck %s --check-prefixes=CHECK,SVE2 +; RUN: llc -global-isel -global-isel-abort=2 -mattr=+sve2 < %s | FileCheck %s --check-prefixes=CHECK,SVE2 +; RUN: llc -mattr=+sve2p1 < %s | FileCheck %s --check-prefixes=CHECK,SVE2P1 +; RUN: llc -global-isel -global-isel-abort=2 -mattr=+sve2p1 < %s | FileCheck %s --check-prefixes=CHECK,SVE2P1 + +target triple = "aarch64-linux-gnu" + +define <vscale x 4 x float> @fdot_wide_nxv4f32(<vscale x 4 x float> %acc, <vscale x 8 x half> %a, <vscale x 8 x half> %b) { +; SVE2-LABEL: fdot_wide_nxv4f32: +; SVE2: // %bb.0: // %entry +; SVE2-NEXT: uunpklo z3.s, z1.h +; SVE2-NEXT: uunpklo z4.s, z2.h +; SVE2-NEXT: ptrue p0.s +; SVE2-NEXT: uunpkhi z1.s, z1.h +; SVE2-NEXT: uunpkhi z2.s, z2.h +; SVE2-NEXT: fcvt z3.s, p0/m, z3.h +; SVE2-NEXT: fcvt z4.s, p0/m, z4.h +; SVE2-NEXT: fcvt z1.s, p0/m, z1.h +; SVE2-NEXT: fcvt z2.s, p0/m, z2.h +; SVE2-NEXT: fmul z3.s, z3.s, z4.s +; SVE2-NEXT: fmul z1.s, z1.s, z2.s +; SVE2-NEXT: fadd z0.s, z0.s, z3.s +; SVE2-NEXT: fadd z0.s, z0.s, z1.s +; SVE2-NEXT: ret +; +; SVE2P1-LABEL: fdot_wide_nxv4f32: +; SVE2P1: // %bb.0: // %entry +; SVE2P1-NEXT: fdot z0.s, z1.h, z2.h +; SVE2P1-NEXT: ret +entry: + %a.wide = fpext <vscale x 8 x half> %a to <vscale x 8 x float> + %b.wide = fpext <vscale x 8 x half> %b to <vscale x 8 x float> + %mult = fmul <vscale x 8 x float> %a.wide, %b.wide + %partial.reduce = call <vscale x 4 x float> @llvm.vector.partial.reduce.fadd(<vscale x 4 x float> %acc, <vscale x 8 x float> %mult) + ret <vscale x 4 x float> %partial.reduce +} + +define <vscale x 4 x float> @fdot_splat_nxv4f32(<vscale x 4 x float> %acc, <vscale x 8 x half> %a) { +; SVE2-LABEL: fdot_splat_nxv4f32: +; SVE2: // %bb.0: // %entry +; SVE2-NEXT: uunpklo z2.s, z1.h +; SVE2-NEXT: ptrue p0.s +; SVE2-NEXT: uunpkhi z1.s, z1.h +; SVE2-NEXT: fcvt z2.s, p0/m, z2.h +; SVE2-NEXT: fcvt z1.s, p0/m, z1.h +; SVE2-NEXT: fadd z0.s, z0.s, z2.s +; SVE2-NEXT: fadd z0.s, z0.s, z1.s +; SVE2-NEXT: ret +; +; SVE2P1-LABEL: fdot_splat_nxv4f32: +; SVE2P1: // %bb.0: // %entry +; SVE2P1-NEXT: fmov z2.h, #1.00000000 +; SVE2P1-NEXT: fdot z0.s, z1.h, z2.h +; SVE2P1-NEXT: ret +entry: + %a.wide = fpext <vscale x 8 x half> %a to <vscale x 8 x float> + %partial.reduce = call <vscale x 4 x float> @llvm.vector.partial.reduce.fadd(<vscale x 4 x float> %acc, <vscale x 8 x float> %a.wide) + ret <vscale x 4 x float> %partial.reduce +} + +define <vscale x 8 x half> @partial_reduce_nxv8f16(<vscale x 8 x half> %acc, <vscale x 16 x half> %a) { +; CHECK-LABEL: partial_reduce_nxv8f16: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fadd z0.h, z0.h, z1.h +; CHECK-NEXT: fadd z0.h, z0.h, z2.h +; CHECK-NEXT: ret +entry: + %partial.reduce = call <vscale x 8 x half> @llvm.vector.partial.reduce.fadd(<vscale x 8 x half> %acc, <vscale x 16 x half> %a) + ret <vscale x 8 x half> %partial.reduce +} + +define <vscale x 4 x float> @partial_reduce_nxv4f32(<vscale x 4 x float> %acc, <vscale x 8 x float> %a) { +; CHECK-LABEL: partial_reduce_nxv4f32: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fadd z0.s, z0.s, z1.s +; CHECK-NEXT: fadd z0.s, z0.s, z2.s +; CHECK-NEXT: ret +entry: + %partial.reduce = call <vscale x 4 x float> @llvm.vector.partial.reduce.fadd(<vscale x 4 x float> %acc, <vscale x 8 x float> %a) + ret <vscale x 4 x float> %partial.reduce +} + +define <vscale x 2 x double> @partial_reduce_nxv2f64(<vscale x 2 x double> %acc, <vscale x 4 x double> %a) { +; CHECK-LABEL: partial_reduce_nxv2f64: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fadd z0.d, z0.d, z1.d +; CHECK-NEXT: fadd z0.d, z0.d, z2.d +; CHECK-NEXT: ret +entry: + %partial.reduce = call <vscale x 2 x double> @llvm.vector.partial.reduce.fadd(<vscale x 2 x double> %acc, <vscale x 4 x double> %a) + ret <vscale x 2 x double> %partial.reduce +} diff --git a/llvm/test/CodeGen/AArch64/sve2p1-fixed-length-fdot.ll b/llvm/test/CodeGen/AArch64/sve2p1-fixed-length-fdot.ll new file mode 100644 index 0000000..89216ce --- /dev/null +++ b/llvm/test/CodeGen/AArch64/sve2p1-fixed-length-fdot.ll @@ -0,0 +1,230 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6 +; RUN: llc -mattr=+sve2 < %s | FileCheck %s --check-prefixes=CHECK,SVE2 +; RUN: llc -mattr=+sve2p1 < %s | FileCheck %s --check-prefixes=CHECK,SVE2P1 + +target triple = "aarch64-linux-gnu" + +define void @fdot_wide_v8f32(ptr %accptr, ptr %aptr, ptr %bptr) vscale_range(2,0) { +; SVE2-LABEL: fdot_wide_v8f32: +; SVE2: // %bb.0: // %entry +; SVE2-NEXT: ptrue p0.s, vl8 +; SVE2-NEXT: mov x8, #8 // =0x8 +; SVE2-NEXT: ld1h { z0.s }, p0/z, [x1] +; SVE2-NEXT: ld1h { z1.s }, p0/z, [x2] +; SVE2-NEXT: ld1h { z2.s }, p0/z, [x1, x8, lsl #1] +; SVE2-NEXT: ld1h { z3.s }, p0/z, [x2, x8, lsl #1] +; SVE2-NEXT: fcvt z0.s, p0/m, z0.h +; SVE2-NEXT: fcvt z1.s, p0/m, z1.h +; SVE2-NEXT: fcvt z2.s, p0/m, z2.h +; SVE2-NEXT: fcvt z3.s, p0/m, z3.h +; SVE2-NEXT: fmul z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: ld1w { z1.s }, p0/z, [x0] +; SVE2-NEXT: fmul z2.s, p0/m, z2.s, z3.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z2.s +; SVE2-NEXT: st1w { z0.s }, p0, [x0] +; SVE2-NEXT: ret +; +; SVE2P1-LABEL: fdot_wide_v8f32: +; SVE2P1: // %bb.0: // %entry +; SVE2P1-NEXT: ptrue p0.s, vl8 +; SVE2P1-NEXT: ptrue p1.h, vl16 +; SVE2P1-NEXT: ld1w { z0.s }, p0/z, [x0] +; SVE2P1-NEXT: ld1h { z1.h }, p1/z, [x1] +; SVE2P1-NEXT: ld1h { z2.h }, p1/z, [x2] +; SVE2P1-NEXT: fdot z0.s, z1.h, z2.h +; SVE2P1-NEXT: st1w { z0.s }, p0, [x0] +; SVE2P1-NEXT: ret +entry: + %acc = load <8 x float>, ptr %accptr + %a = load <16 x half>, ptr %aptr + %b = load <16 x half>, ptr %bptr + %a.wide = fpext <16 x half> %a to <16 x float> + %b.wide = fpext <16 x half> %b to <16 x float> + %mult = fmul <16 x float> %a.wide, %b.wide + %partial.reduce = call <8 x float> @llvm.vector.partial.reduce.fadd(<8 x float> %acc, <16 x float> %mult) + store <8 x float> %partial.reduce, ptr %accptr + ret void +} + +define void @fdot_wide_v16f32(ptr %accptr, ptr %aptr, ptr %bptr) vscale_range(4,0) { +; SVE2-LABEL: fdot_wide_v16f32: +; SVE2: // %bb.0: // %entry +; SVE2-NEXT: ptrue p0.s, vl16 +; SVE2-NEXT: mov x8, #16 // =0x10 +; SVE2-NEXT: ld1h { z0.s }, p0/z, [x1] +; SVE2-NEXT: ld1h { z1.s }, p0/z, [x2] +; SVE2-NEXT: ld1h { z2.s }, p0/z, [x1, x8, lsl #1] +; SVE2-NEXT: ld1h { z3.s }, p0/z, [x2, x8, lsl #1] +; SVE2-NEXT: fcvt z0.s, p0/m, z0.h +; SVE2-NEXT: fcvt z1.s, p0/m, z1.h +; SVE2-NEXT: fcvt z2.s, p0/m, z2.h +; SVE2-NEXT: fcvt z3.s, p0/m, z3.h +; SVE2-NEXT: fmul z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: ld1w { z1.s }, p0/z, [x0] +; SVE2-NEXT: fmul z2.s, p0/m, z2.s, z3.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z2.s +; SVE2-NEXT: st1w { z0.s }, p0, [x0] +; SVE2-NEXT: ret +; +; SVE2P1-LABEL: fdot_wide_v16f32: +; SVE2P1: // %bb.0: // %entry +; SVE2P1-NEXT: ptrue p0.s, vl16 +; SVE2P1-NEXT: ptrue p1.h, vl32 +; SVE2P1-NEXT: ld1w { z0.s }, p0/z, [x0] +; SVE2P1-NEXT: ld1h { z1.h }, p1/z, [x1] +; SVE2P1-NEXT: ld1h { z2.h }, p1/z, [x2] +; SVE2P1-NEXT: fdot z0.s, z1.h, z2.h +; SVE2P1-NEXT: st1w { z0.s }, p0, [x0] +; SVE2P1-NEXT: ret +entry: + %acc = load <16 x float>, ptr %accptr + %a = load <32 x half>, ptr %aptr + %b = load <32 x half>, ptr %bptr + %a.wide = fpext <32 x half> %a to <32 x float> + %b.wide = fpext <32 x half> %b to <32 x float> + %mult = fmul <32 x float> %a.wide, %b.wide + %partial.reduce = call <16 x float> @llvm.vector.partial.reduce.fadd(<16 x float> %acc, <32 x float> %mult) + store <16 x float> %partial.reduce, ptr %accptr + ret void +} + +define void @fdot_wide_v32f32(ptr %accptr, ptr %aptr, ptr %bptr) vscale_range(8,0) { +; SVE2-LABEL: fdot_wide_v32f32: +; SVE2: // %bb.0: // %entry +; SVE2-NEXT: ptrue p0.s, vl32 +; SVE2-NEXT: mov x8, #32 // =0x20 +; SVE2-NEXT: ld1h { z0.s }, p0/z, [x1] +; SVE2-NEXT: ld1h { z1.s }, p0/z, [x2] +; SVE2-NEXT: ld1h { z2.s }, p0/z, [x1, x8, lsl #1] +; SVE2-NEXT: ld1h { z3.s }, p0/z, [x2, x8, lsl #1] +; SVE2-NEXT: fcvt z0.s, p0/m, z0.h +; SVE2-NEXT: fcvt z1.s, p0/m, z1.h +; SVE2-NEXT: fcvt z2.s, p0/m, z2.h +; SVE2-NEXT: fcvt z3.s, p0/m, z3.h +; SVE2-NEXT: fmul z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: ld1w { z1.s }, p0/z, [x0] +; SVE2-NEXT: fmul z2.s, p0/m, z2.s, z3.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z2.s +; SVE2-NEXT: st1w { z0.s }, p0, [x0] +; SVE2-NEXT: ret +; +; SVE2P1-LABEL: fdot_wide_v32f32: +; SVE2P1: // %bb.0: // %entry +; SVE2P1-NEXT: ptrue p0.s, vl32 +; SVE2P1-NEXT: ptrue p1.h, vl64 +; SVE2P1-NEXT: ld1w { z0.s }, p0/z, [x0] +; SVE2P1-NEXT: ld1h { z1.h }, p1/z, [x1] +; SVE2P1-NEXT: ld1h { z2.h }, p1/z, [x2] +; SVE2P1-NEXT: fdot z0.s, z1.h, z2.h +; SVE2P1-NEXT: st1w { z0.s }, p0, [x0] +; SVE2P1-NEXT: ret +entry: + %acc = load <32 x float>, ptr %accptr + %a = load <64 x half>, ptr %aptr + %b = load <64 x half>, ptr %bptr + %a.wide = fpext <64 x half> %a to <64 x float> + %b.wide = fpext <64 x half> %b to <64 x float> + %mult = fmul <64 x float> %a.wide, %b.wide + %partial.reduce = call <32 x float> @llvm.vector.partial.reduce.fadd(<32 x float> %acc, <64 x float> %mult) + store <32 x float> %partial.reduce, ptr %accptr + ret void +} + +define void @fdot_wide_v64f32(ptr %accptr, ptr %aptr, ptr %bptr) vscale_range(16,0) { +; SVE2-LABEL: fdot_wide_v64f32: +; SVE2: // %bb.0: // %entry +; SVE2-NEXT: ptrue p0.s, vl64 +; SVE2-NEXT: mov x8, #64 // =0x40 +; SVE2-NEXT: ld1h { z0.s }, p0/z, [x1] +; SVE2-NEXT: ld1h { z1.s }, p0/z, [x2] +; SVE2-NEXT: ld1h { z2.s }, p0/z, [x1, x8, lsl #1] +; SVE2-NEXT: ld1h { z3.s }, p0/z, [x2, x8, lsl #1] +; SVE2-NEXT: fcvt z0.s, p0/m, z0.h +; SVE2-NEXT: fcvt z1.s, p0/m, z1.h +; SVE2-NEXT: fcvt z2.s, p0/m, z2.h +; SVE2-NEXT: fcvt z3.s, p0/m, z3.h +; SVE2-NEXT: fmul z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: ld1w { z1.s }, p0/z, [x0] +; SVE2-NEXT: fmul z2.s, p0/m, z2.s, z3.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z1.s +; SVE2-NEXT: fadd z0.s, p0/m, z0.s, z2.s +; SVE2-NEXT: st1w { z0.s }, p0, [x0] +; SVE2-NEXT: ret +; +; SVE2P1-LABEL: fdot_wide_v64f32: +; SVE2P1: // %bb.0: // %entry +; SVE2P1-NEXT: ptrue p0.s, vl64 +; SVE2P1-NEXT: ptrue p1.h, vl128 +; SVE2P1-NEXT: ld1w { z0.s }, p0/z, [x0] +; SVE2P1-NEXT: ld1h { z1.h }, p1/z, [x1] +; SVE2P1-NEXT: ld1h { z2.h }, p1/z, [x2] +; SVE2P1-NEXT: fdot z0.s, z1.h, z2.h +; SVE2P1-NEXT: st1w { z0.s }, p0, [x0] +; SVE2P1-NEXT: ret +entry: + %acc = load <64 x float>, ptr %accptr + %a = load <128 x half>, ptr %aptr + %b = load <128 x half>, ptr %bptr + %a.wide = fpext <128 x half> %a to <128 x float> + %b.wide = fpext <128 x half> %b to <128 x float> + %mult = fmul <128 x float> %a.wide, %b.wide + %partial.reduce = call <64 x float> @llvm.vector.partial.reduce.fadd(<64 x float> %acc, <128 x float> %mult) + store <64 x float> %partial.reduce, ptr %accptr + ret void +} + +define <4 x float> @fixed_fdot_wide(<4 x float> %acc, <8 x half> %a, <8 x half> %b) { +; CHECK-LABEL: fixed_fdot_wide: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fcvtl v3.4s, v1.4h +; CHECK-NEXT: fcvtl v4.4s, v2.4h +; CHECK-NEXT: fcvtl2 v1.4s, v1.8h +; CHECK-NEXT: fcvtl2 v2.4s, v2.8h +; CHECK-NEXT: fmul v3.4s, v3.4s, v4.4s +; CHECK-NEXT: fmul v1.4s, v1.4s, v2.4s +; CHECK-NEXT: fadd v0.4s, v0.4s, v3.4s +; CHECK-NEXT: fadd v0.4s, v0.4s, v1.4s +; CHECK-NEXT: ret +entry: + %a.wide = fpext <8 x half> %a to <8 x float> + %b.wide = fpext <8 x half> %b to <8 x float> + %mult = fmul <8 x float> %a.wide, %b.wide + %partial.reduce = call <4 x float> @llvm.vector.partial.reduce.fadd(<4 x float> %acc, <8 x float> %mult) + ret <4 x float> %partial.reduce +} + +define <8 x half> @partial_reduce_half(<8 x half> %acc, <16 x half> %a) { +; CHECK-LABEL: partial_reduce_half: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fadd v0.8h, v0.8h, v1.8h +; CHECK-NEXT: fadd v0.8h, v0.8h, v2.8h +; CHECK-NEXT: ret +entry: + %partial.reduce = call <8 x half> @llvm.vector.partial.reduce.fadd(<8 x half> %acc, <16 x half> %a) + ret <8 x half> %partial.reduce +} + +define <4 x float> @partial_reduce_float(<4 x float> %acc, <8 x float> %a) { +; CHECK-LABEL: partial_reduce_float: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fadd v0.4s, v0.4s, v1.4s +; CHECK-NEXT: fadd v0.4s, v0.4s, v2.4s +; CHECK-NEXT: ret +entry: + %partial.reduce = call <4 x float> @llvm.vector.partial.reduce.fadd(<4 x float> %acc, <8 x float> %a) + ret <4 x float> %partial.reduce +} + +define <2 x double> @partial_reduce_double(<2 x double> %acc, <4 x double> %a) { +; CHECK-LABEL: partial_reduce_double: +; CHECK: // %bb.0: // %entry +; CHECK-NEXT: fadd v0.2d, v0.2d, v1.2d +; CHECK-NEXT: fadd v0.2d, v0.2d, v2.2d +; CHECK-NEXT: ret +entry: + %partial.reduce = call <2 x double> @llvm.vector.partial.reduce.fadd(<2 x double> %acc, <4 x double> %a) + ret <2 x double> %partial.reduce +} diff --git a/llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-idiv.ll b/llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-idiv.ll index 51df8c3..54b1554 100644 --- a/llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-idiv.ll +++ b/llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-idiv.ll @@ -7772,7 +7772,6 @@ define amdgpu_kernel void @sdiv_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0xd ; GFX6-NEXT: s_load_dwordx4 s[4:7], s[4:5], 0x9 ; GFX6-NEXT: s_mov_b32 s3, 0xf000 -; GFX6-NEXT: s_mov_b32 s2, -1 ; GFX6-NEXT: s_waitcnt lgkmcnt(0) ; GFX6-NEXT: s_lshl_b64 s[0:1], 0x1000, s0 ; GFX6-NEXT: s_ashr_i32 s8, s1, 31 @@ -7782,8 +7781,8 @@ define amdgpu_kernel void @sdiv_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_xor_b64 s[10:11], s[0:1], s[8:9] ; GFX6-NEXT: v_cvt_f32_u32_e32 v0, s10 ; GFX6-NEXT: v_cvt_f32_u32_e32 v1, s11 -; GFX6-NEXT: s_sub_u32 s12, 0, s10 -; GFX6-NEXT: s_subb_u32 s13, 0, s11 +; GFX6-NEXT: s_sub_u32 s0, 0, s10 +; GFX6-NEXT: s_subb_u32 s1, 0, s11 ; GFX6-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GFX6-NEXT: v_rcp_f32_e32 v0, v0 ; GFX6-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -7792,128 +7791,121 @@ define amdgpu_kernel void @sdiv_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GFX6-NEXT: v_mul_hi_u32 v2, s12, v0 -; GFX6-NEXT: v_readfirstlane_b32 s14, v1 -; GFX6-NEXT: v_readfirstlane_b32 s0, v0 -; GFX6-NEXT: s_mul_i32 s1, s12, s14 -; GFX6-NEXT: v_readfirstlane_b32 s17, v2 -; GFX6-NEXT: s_mul_i32 s15, s13, s0 -; GFX6-NEXT: s_mul_i32 s16, s12, s0 -; GFX6-NEXT: s_add_i32 s1, s17, s1 -; GFX6-NEXT: v_mul_hi_u32 v3, v0, s16 -; GFX6-NEXT: s_add_i32 s1, s1, s15 -; GFX6-NEXT: v_mul_hi_u32 v0, v0, s1 -; GFX6-NEXT: v_mul_hi_u32 v4, v1, s16 -; GFX6-NEXT: v_readfirstlane_b32 s15, v3 -; GFX6-NEXT: s_mul_i32 s17, s0, s1 -; GFX6-NEXT: v_mul_hi_u32 v1, v1, s1 -; GFX6-NEXT: s_add_u32 s15, s15, s17 -; GFX6-NEXT: v_readfirstlane_b32 s17, v0 -; GFX6-NEXT: s_addc_u32 s17, 0, s17 -; GFX6-NEXT: s_mul_i32 s16, s14, s16 -; GFX6-NEXT: v_readfirstlane_b32 s18, v4 -; GFX6-NEXT: s_add_u32 s15, s15, s16 -; GFX6-NEXT: s_addc_u32 s15, s17, s18 -; GFX6-NEXT: v_readfirstlane_b32 s16, v1 -; GFX6-NEXT: s_addc_u32 s16, s16, 0 -; GFX6-NEXT: s_mul_i32 s1, s14, s1 -; GFX6-NEXT: s_add_u32 s1, s15, s1 -; GFX6-NEXT: s_addc_u32 s15, 0, s16 -; GFX6-NEXT: s_add_u32 s16, s0, s1 -; GFX6-NEXT: v_mov_b32_e32 v0, s16 -; GFX6-NEXT: v_mul_hi_u32 v0, s12, v0 -; GFX6-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GFX6-NEXT: s_or_b32 s0, s0, s1 -; GFX6-NEXT: s_addc_u32 s14, s14, s15 -; GFX6-NEXT: s_mul_i32 s0, s12, s14 -; GFX6-NEXT: v_readfirstlane_b32 s1, v0 -; GFX6-NEXT: s_add_i32 s0, s1, s0 -; GFX6-NEXT: s_mul_i32 s13, s13, s16 -; GFX6-NEXT: s_mul_i32 s1, s12, s16 -; GFX6-NEXT: s_add_i32 s0, s0, s13 -; GFX6-NEXT: v_mov_b32_e32 v2, s1 -; GFX6-NEXT: v_mov_b32_e32 v0, s0 -; GFX6-NEXT: v_mul_hi_u32 v3, s14, v2 -; GFX6-NEXT: v_mul_hi_u32 v2, s16, v2 -; GFX6-NEXT: v_mul_hi_u32 v1, s14, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, s16, v0 -; GFX6-NEXT: s_mul_i32 s13, s16, s0 -; GFX6-NEXT: v_readfirstlane_b32 s17, v2 -; GFX6-NEXT: s_add_u32 s13, s17, s13 -; GFX6-NEXT: v_readfirstlane_b32 s15, v0 -; GFX6-NEXT: s_mul_i32 s1, s14, s1 -; GFX6-NEXT: s_addc_u32 s15, 0, s15 -; GFX6-NEXT: v_readfirstlane_b32 s12, v3 -; GFX6-NEXT: s_add_u32 s1, s13, s1 -; GFX6-NEXT: s_addc_u32 s1, s15, s12 +; GFX6-NEXT: v_mul_hi_u32 v2, s0, v0 ; GFX6-NEXT: v_readfirstlane_b32 s12, v1 -; GFX6-NEXT: s_addc_u32 s12, s12, 0 -; GFX6-NEXT: s_mul_i32 s0, s14, s0 -; GFX6-NEXT: s_add_u32 s0, s1, s0 -; GFX6-NEXT: s_addc_u32 s12, 0, s12 -; GFX6-NEXT: s_add_u32 s15, s16, s0 -; GFX6-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GFX6-NEXT: s_or_b32 s0, s0, s1 -; GFX6-NEXT: s_addc_u32 s14, s14, s12 +; GFX6-NEXT: v_readfirstlane_b32 s2, v0 +; GFX6-NEXT: s_mul_i32 s13, s0, s12 +; GFX6-NEXT: v_readfirstlane_b32 s16, v2 +; GFX6-NEXT: s_mul_i32 s14, s1, s2 +; GFX6-NEXT: s_mul_i32 s15, s0, s2 +; GFX6-NEXT: s_add_i32 s13, s16, s13 +; GFX6-NEXT: v_mul_hi_u32 v3, v0, s15 +; GFX6-NEXT: s_add_i32 s13, s13, s14 +; GFX6-NEXT: v_mul_hi_u32 v0, v0, s13 +; GFX6-NEXT: v_mul_hi_u32 v4, v1, s15 +; GFX6-NEXT: v_readfirstlane_b32 s14, v3 +; GFX6-NEXT: s_mul_i32 s16, s2, s13 +; GFX6-NEXT: v_mul_hi_u32 v1, v1, s13 +; GFX6-NEXT: s_add_u32 s14, s14, s16 +; GFX6-NEXT: v_readfirstlane_b32 s16, v0 +; GFX6-NEXT: s_mul_i32 s15, s12, s15 +; GFX6-NEXT: s_addc_u32 s16, 0, s16 +; GFX6-NEXT: v_readfirstlane_b32 s17, v4 +; GFX6-NEXT: s_add_u32 s14, s14, s15 +; GFX6-NEXT: s_addc_u32 s14, s16, s17 +; GFX6-NEXT: v_readfirstlane_b32 s15, v1 +; GFX6-NEXT: s_addc_u32 s15, s15, 0 +; GFX6-NEXT: s_mul_i32 s13, s12, s13 +; GFX6-NEXT: s_add_u32 s13, s14, s13 +; GFX6-NEXT: s_addc_u32 s14, 0, s15 +; GFX6-NEXT: s_add_u32 s13, s2, s13 +; GFX6-NEXT: v_mov_b32_e32 v0, s13 +; GFX6-NEXT: v_mul_hi_u32 v0, s0, v0 +; GFX6-NEXT: s_addc_u32 s12, s12, s14 +; GFX6-NEXT: s_mul_i32 s14, s0, s12 +; GFX6-NEXT: s_mul_i32 s1, s1, s13 +; GFX6-NEXT: v_readfirstlane_b32 s15, v0 +; GFX6-NEXT: s_add_i32 s14, s15, s14 +; GFX6-NEXT: s_mul_i32 s0, s0, s13 +; GFX6-NEXT: s_add_i32 s1, s14, s1 +; GFX6-NEXT: v_mov_b32_e32 v2, s0 +; GFX6-NEXT: v_mov_b32_e32 v0, s1 +; GFX6-NEXT: v_mul_hi_u32 v3, s12, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s13, v2 +; GFX6-NEXT: v_mul_hi_u32 v1, s12, v0 +; GFX6-NEXT: v_mul_hi_u32 v0, s13, v0 +; GFX6-NEXT: s_mul_i32 s15, s13, s1 +; GFX6-NEXT: v_readfirstlane_b32 s17, v2 +; GFX6-NEXT: s_add_u32 s15, s17, s15 +; GFX6-NEXT: v_readfirstlane_b32 s16, v0 +; GFX6-NEXT: s_mul_i32 s0, s12, s0 +; GFX6-NEXT: s_addc_u32 s16, 0, s16 +; GFX6-NEXT: v_readfirstlane_b32 s14, v3 +; GFX6-NEXT: s_add_u32 s0, s15, s0 +; GFX6-NEXT: s_addc_u32 s0, s16, s14 +; GFX6-NEXT: v_readfirstlane_b32 s14, v1 +; GFX6-NEXT: s_addc_u32 s14, s14, 0 +; GFX6-NEXT: s_mul_i32 s1, s12, s1 +; GFX6-NEXT: s_add_u32 s0, s0, s1 +; GFX6-NEXT: s_addc_u32 s1, 0, s14 +; GFX6-NEXT: s_add_u32 s14, s13, s0 +; GFX6-NEXT: s_addc_u32 s15, s12, s1 ; GFX6-NEXT: s_ashr_i32 s12, s7, 31 ; GFX6-NEXT: s_add_u32 s0, s6, s12 ; GFX6-NEXT: s_mov_b32 s13, s12 ; GFX6-NEXT: s_addc_u32 s1, s7, s12 ; GFX6-NEXT: s_xor_b64 s[6:7], s[0:1], s[12:13] -; GFX6-NEXT: v_mov_b32_e32 v0, s14 +; GFX6-NEXT: v_mov_b32_e32 v0, s15 ; GFX6-NEXT: v_mul_hi_u32 v1, s6, v0 -; GFX6-NEXT: v_mov_b32_e32 v2, s15 +; GFX6-NEXT: v_mov_b32_e32 v2, s14 ; GFX6-NEXT: v_mul_hi_u32 v3, s6, v2 ; GFX6-NEXT: s_mov_b32 s0, s4 ; GFX6-NEXT: v_readfirstlane_b32 s4, v1 ; GFX6-NEXT: v_mul_hi_u32 v1, s7, v2 -; GFX6-NEXT: s_mul_i32 s1, s6, s14 +; GFX6-NEXT: s_mul_i32 s1, s6, s15 ; GFX6-NEXT: v_readfirstlane_b32 s16, v3 ; GFX6-NEXT: v_mul_hi_u32 v0, s7, v0 ; GFX6-NEXT: s_add_u32 s1, s16, s1 ; GFX6-NEXT: s_addc_u32 s4, 0, s4 -; GFX6-NEXT: s_mul_i32 s15, s7, s15 +; GFX6-NEXT: s_mul_i32 s14, s7, s14 ; GFX6-NEXT: v_readfirstlane_b32 s16, v1 -; GFX6-NEXT: s_add_u32 s1, s1, s15 +; GFX6-NEXT: s_add_u32 s1, s1, s14 ; GFX6-NEXT: s_addc_u32 s1, s4, s16 ; GFX6-NEXT: v_readfirstlane_b32 s4, v0 ; GFX6-NEXT: s_addc_u32 s4, s4, 0 -; GFX6-NEXT: s_mul_i32 s14, s7, s14 -; GFX6-NEXT: s_add_u32 s16, s1, s14 -; GFX6-NEXT: v_mov_b32_e32 v0, s16 +; GFX6-NEXT: s_mul_i32 s14, s7, s15 +; GFX6-NEXT: s_add_u32 s14, s1, s14 +; GFX6-NEXT: v_mov_b32_e32 v0, s14 ; GFX6-NEXT: v_mul_hi_u32 v0, s10, v0 -; GFX6-NEXT: s_addc_u32 s17, 0, s4 +; GFX6-NEXT: s_addc_u32 s15, 0, s4 ; GFX6-NEXT: s_mov_b32 s1, s5 -; GFX6-NEXT: s_mul_i32 s4, s10, s17 +; GFX6-NEXT: s_mul_i32 s4, s10, s15 ; GFX6-NEXT: v_readfirstlane_b32 s5, v0 ; GFX6-NEXT: s_add_i32 s4, s5, s4 -; GFX6-NEXT: s_mul_i32 s5, s11, s16 -; GFX6-NEXT: s_add_i32 s18, s4, s5 -; GFX6-NEXT: s_sub_i32 s14, s7, s18 -; GFX6-NEXT: s_mul_i32 s4, s10, s16 +; GFX6-NEXT: s_mul_i32 s5, s11, s14 +; GFX6-NEXT: s_add_i32 s16, s4, s5 +; GFX6-NEXT: s_sub_i32 s17, s7, s16 +; GFX6-NEXT: s_mul_i32 s4, s10, s14 ; GFX6-NEXT: s_sub_u32 s6, s6, s4 ; GFX6-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GFX6-NEXT: s_or_b32 s15, s4, s5 -; GFX6-NEXT: s_subb_u32 s19, s14, s11 -; GFX6-NEXT: s_sub_u32 s20, s6, s10 -; GFX6-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GFX6-NEXT: s_or_b32 s14, s14, s15 -; GFX6-NEXT: s_subb_u32 s14, s19, 0 -; GFX6-NEXT: s_cmp_ge_u32 s14, s11 -; GFX6-NEXT: s_cselect_b32 s15, -1, 0 -; GFX6-NEXT: s_cmp_ge_u32 s20, s10 +; GFX6-NEXT: s_subb_u32 s17, s17, s11 +; GFX6-NEXT: s_sub_u32 s18, s6, s10 +; GFX6-NEXT: s_subb_u32 s17, s17, 0 +; GFX6-NEXT: s_cmp_ge_u32 s17, s11 ; GFX6-NEXT: s_cselect_b32 s19, -1, 0 -; GFX6-NEXT: s_cmp_eq_u32 s14, s11 -; GFX6-NEXT: s_cselect_b32 s14, s19, s15 -; GFX6-NEXT: s_add_u32 s15, s16, 1 -; GFX6-NEXT: s_addc_u32 s19, s17, 0 -; GFX6-NEXT: s_add_u32 s20, s16, 2 -; GFX6-NEXT: s_addc_u32 s21, s17, 0 -; GFX6-NEXT: s_cmp_lg_u32 s14, 0 -; GFX6-NEXT: s_cselect_b32 s14, s20, s15 -; GFX6-NEXT: s_cselect_b32 s15, s21, s19 +; GFX6-NEXT: s_cmp_ge_u32 s18, s10 +; GFX6-NEXT: s_cselect_b32 s18, -1, 0 +; GFX6-NEXT: s_cmp_eq_u32 s17, s11 +; GFX6-NEXT: s_cselect_b32 s17, s18, s19 +; GFX6-NEXT: s_add_u32 s18, s14, 1 +; GFX6-NEXT: s_addc_u32 s19, s15, 0 +; GFX6-NEXT: s_add_u32 s20, s14, 2 +; GFX6-NEXT: s_addc_u32 s21, s15, 0 +; GFX6-NEXT: s_cmp_lg_u32 s17, 0 +; GFX6-NEXT: s_cselect_b32 s17, s20, s18 +; GFX6-NEXT: s_cselect_b32 s18, s21, s19 ; GFX6-NEXT: s_or_b32 s4, s4, s5 -; GFX6-NEXT: s_subb_u32 s4, s7, s18 +; GFX6-NEXT: s_subb_u32 s4, s7, s16 ; GFX6-NEXT: s_cmp_ge_u32 s4, s11 ; GFX6-NEXT: s_cselect_b32 s5, -1, 0 ; GFX6-NEXT: s_cmp_ge_u32 s6, s10 @@ -7921,13 +7913,14 @@ define amdgpu_kernel void @sdiv_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_cmp_eq_u32 s4, s11 ; GFX6-NEXT: s_cselect_b32 s4, s6, s5 ; GFX6-NEXT: s_cmp_lg_u32 s4, 0 -; GFX6-NEXT: s_cselect_b32 s5, s15, s17 -; GFX6-NEXT: s_cselect_b32 s4, s14, s16 +; GFX6-NEXT: s_cselect_b32 s5, s18, s15 +; GFX6-NEXT: s_cselect_b32 s4, s17, s14 ; GFX6-NEXT: s_xor_b64 s[6:7], s[12:13], s[8:9] ; GFX6-NEXT: s_xor_b64 s[4:5], s[4:5], s[6:7] ; GFX6-NEXT: s_sub_u32 s4, s4, s6 ; GFX6-NEXT: s_subb_u32 s5, s5, s7 ; GFX6-NEXT: v_mov_b32_e32 v0, s4 +; GFX6-NEXT: s_mov_b32 s2, -1 ; GFX6-NEXT: v_mov_b32_e32 v1, s5 ; GFX6-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0 ; GFX6-NEXT: s_endpgm @@ -8278,8 +8271,8 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_xor_b64 s[6:7], s[6:7], s[2:3] ; GFX6-NEXT: v_cvt_f32_u32_e32 v0, s6 ; GFX6-NEXT: v_cvt_f32_u32_e32 v1, s7 -; GFX6-NEXT: s_sub_u32 s14, 0, s6 -; GFX6-NEXT: s_subb_u32 s15, 0, s7 +; GFX6-NEXT: s_sub_u32 s12, 0, s6 +; GFX6-NEXT: s_subb_u32 s13, 0, s7 ; GFX6-NEXT: v_mac_f32_e32 v0, 0x4f800000, v1 ; GFX6-NEXT: v_rcp_f32_e32 v0, v0 ; GFX6-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -8288,69 +8281,65 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: v_mac_f32_e32 v0, 0xcf800000, v1 ; GFX6-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GFX6-NEXT: v_mul_hi_u32 v2, s14, v0 -; GFX6-NEXT: v_readfirstlane_b32 s16, v1 -; GFX6-NEXT: v_readfirstlane_b32 s12, v0 -; GFX6-NEXT: s_mul_i32 s13, s14, s16 +; GFX6-NEXT: v_mul_hi_u32 v2, s12, v0 +; GFX6-NEXT: v_readfirstlane_b32 s14, v1 +; GFX6-NEXT: v_readfirstlane_b32 s15, v0 +; GFX6-NEXT: s_mul_i32 s16, s12, s14 ; GFX6-NEXT: v_readfirstlane_b32 s19, v2 -; GFX6-NEXT: s_mul_i32 s17, s15, s12 -; GFX6-NEXT: s_mul_i32 s18, s14, s12 -; GFX6-NEXT: s_add_i32 s13, s19, s13 +; GFX6-NEXT: s_mul_i32 s17, s13, s15 +; GFX6-NEXT: s_mul_i32 s18, s12, s15 +; GFX6-NEXT: s_add_i32 s16, s19, s16 ; GFX6-NEXT: v_mul_hi_u32 v3, v0, s18 -; GFX6-NEXT: s_add_i32 s13, s13, s17 -; GFX6-NEXT: v_mul_hi_u32 v0, v0, s13 +; GFX6-NEXT: s_add_i32 s16, s16, s17 +; GFX6-NEXT: v_mul_hi_u32 v0, v0, s16 ; GFX6-NEXT: v_mul_hi_u32 v4, v1, s18 ; GFX6-NEXT: v_readfirstlane_b32 s17, v3 -; GFX6-NEXT: s_mul_i32 s20, s12, s13 -; GFX6-NEXT: v_mul_hi_u32 v1, v1, s13 +; GFX6-NEXT: s_mul_i32 s20, s15, s16 +; GFX6-NEXT: v_mul_hi_u32 v1, v1, s16 ; GFX6-NEXT: s_add_u32 s17, s17, s20 ; GFX6-NEXT: v_readfirstlane_b32 s20, v0 -; GFX6-NEXT: s_mul_i32 s18, s16, s18 +; GFX6-NEXT: s_mul_i32 s18, s14, s18 ; GFX6-NEXT: s_addc_u32 s20, 0, s20 ; GFX6-NEXT: v_readfirstlane_b32 s19, v4 ; GFX6-NEXT: s_add_u32 s17, s17, s18 ; GFX6-NEXT: s_addc_u32 s17, s20, s19 ; GFX6-NEXT: v_readfirstlane_b32 s18, v1 ; GFX6-NEXT: s_addc_u32 s18, s18, 0 -; GFX6-NEXT: s_mul_i32 s13, s16, s13 -; GFX6-NEXT: s_add_u32 s13, s17, s13 +; GFX6-NEXT: s_mul_i32 s16, s14, s16 +; GFX6-NEXT: s_add_u32 s16, s17, s16 ; GFX6-NEXT: s_addc_u32 s17, 0, s18 -; GFX6-NEXT: s_add_u32 s18, s12, s13 -; GFX6-NEXT: v_mov_b32_e32 v0, s18 -; GFX6-NEXT: v_mul_hi_u32 v0, s14, v0 -; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_addc_u32 s16, s16, s17 -; GFX6-NEXT: s_mul_i32 s12, s14, s16 -; GFX6-NEXT: v_readfirstlane_b32 s13, v0 -; GFX6-NEXT: s_add_i32 s12, s13, s12 -; GFX6-NEXT: s_mul_i32 s15, s15, s18 -; GFX6-NEXT: s_mul_i32 s13, s14, s18 -; GFX6-NEXT: s_add_i32 s12, s12, s15 -; GFX6-NEXT: v_mov_b32_e32 v2, s13 -; GFX6-NEXT: v_mov_b32_e32 v0, s12 -; GFX6-NEXT: v_mul_hi_u32 v3, s16, v2 -; GFX6-NEXT: v_mul_hi_u32 v2, s18, v2 -; GFX6-NEXT: v_mul_hi_u32 v1, s16, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, s18, v0 -; GFX6-NEXT: s_mul_i32 s15, s18, s12 -; GFX6-NEXT: v_readfirstlane_b32 s19, v2 -; GFX6-NEXT: s_add_u32 s15, s19, s15 +; GFX6-NEXT: s_add_u32 s15, s15, s16 +; GFX6-NEXT: v_mov_b32_e32 v0, s15 +; GFX6-NEXT: v_mul_hi_u32 v0, s12, v0 +; GFX6-NEXT: s_addc_u32 s14, s14, s17 +; GFX6-NEXT: s_mul_i32 s16, s12, s14 +; GFX6-NEXT: s_mul_i32 s13, s13, s15 ; GFX6-NEXT: v_readfirstlane_b32 s17, v0 -; GFX6-NEXT: s_mul_i32 s13, s16, s13 -; GFX6-NEXT: s_addc_u32 s17, 0, s17 -; GFX6-NEXT: v_readfirstlane_b32 s14, v3 -; GFX6-NEXT: s_add_u32 s13, s15, s13 -; GFX6-NEXT: s_addc_u32 s13, s17, s14 -; GFX6-NEXT: v_readfirstlane_b32 s14, v1 -; GFX6-NEXT: s_addc_u32 s14, s14, 0 -; GFX6-NEXT: s_mul_i32 s12, s16, s12 -; GFX6-NEXT: s_add_u32 s12, s13, s12 -; GFX6-NEXT: s_addc_u32 s14, 0, s14 -; GFX6-NEXT: s_add_u32 s15, s18, s12 -; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_addc_u32 s14, s16, s14 +; GFX6-NEXT: s_add_i32 s16, s17, s16 +; GFX6-NEXT: s_mul_i32 s12, s12, s15 +; GFX6-NEXT: s_add_i32 s13, s16, s13 +; GFX6-NEXT: v_mov_b32_e32 v2, s12 +; GFX6-NEXT: v_mov_b32_e32 v0, s13 +; GFX6-NEXT: v_mul_hi_u32 v3, s14, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s15, v2 +; GFX6-NEXT: v_mul_hi_u32 v1, s14, v0 +; GFX6-NEXT: v_mul_hi_u32 v0, s15, v0 +; GFX6-NEXT: s_mul_i32 s17, s15, s13 +; GFX6-NEXT: v_readfirstlane_b32 s19, v2 +; GFX6-NEXT: s_add_u32 s17, s19, s17 +; GFX6-NEXT: v_readfirstlane_b32 s18, v0 +; GFX6-NEXT: s_mul_i32 s12, s14, s12 +; GFX6-NEXT: s_addc_u32 s18, 0, s18 +; GFX6-NEXT: v_readfirstlane_b32 s16, v3 +; GFX6-NEXT: s_add_u32 s12, s17, s12 +; GFX6-NEXT: s_addc_u32 s12, s18, s16 +; GFX6-NEXT: v_readfirstlane_b32 s16, v1 +; GFX6-NEXT: s_addc_u32 s16, s16, 0 +; GFX6-NEXT: s_mul_i32 s13, s14, s13 +; GFX6-NEXT: s_add_u32 s12, s12, s13 +; GFX6-NEXT: s_addc_u32 s13, 0, s16 +; GFX6-NEXT: s_add_u32 s15, s15, s12 +; GFX6-NEXT: s_addc_u32 s14, s14, s13 ; GFX6-NEXT: s_ashr_i32 s12, s9, 31 ; GFX6-NEXT: s_add_u32 s8, s8, s12 ; GFX6-NEXT: s_mov_b32 s13, s12 @@ -8374,40 +8363,37 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: v_readfirstlane_b32 s16, v0 ; GFX6-NEXT: s_addc_u32 s16, s16, 0 ; GFX6-NEXT: s_mul_i32 s14, s9, s14 -; GFX6-NEXT: s_add_u32 s18, s15, s14 -; GFX6-NEXT: v_mov_b32_e32 v0, s18 +; GFX6-NEXT: s_add_u32 s17, s15, s14 +; GFX6-NEXT: v_mov_b32_e32 v0, s17 ; GFX6-NEXT: v_mul_hi_u32 v0, s6, v0 -; GFX6-NEXT: s_addc_u32 s19, 0, s16 -; GFX6-NEXT: s_mul_i32 s14, s6, s19 +; GFX6-NEXT: s_addc_u32 s16, 0, s16 +; GFX6-NEXT: s_mul_i32 s14, s6, s16 ; GFX6-NEXT: v_readfirstlane_b32 s15, v0 ; GFX6-NEXT: s_add_i32 s14, s15, s14 -; GFX6-NEXT: s_mul_i32 s15, s7, s18 -; GFX6-NEXT: s_add_i32 s20, s14, s15 -; GFX6-NEXT: s_sub_i32 s16, s9, s20 -; GFX6-NEXT: s_mul_i32 s14, s6, s18 +; GFX6-NEXT: s_mul_i32 s15, s7, s17 +; GFX6-NEXT: s_add_i32 s18, s14, s15 +; GFX6-NEXT: s_sub_i32 s19, s9, s18 +; GFX6-NEXT: s_mul_i32 s14, s6, s17 ; GFX6-NEXT: s_sub_u32 s8, s8, s14 ; GFX6-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GFX6-NEXT: s_or_b32 s17, s14, s15 -; GFX6-NEXT: s_subb_u32 s21, s16, s7 -; GFX6-NEXT: s_sub_u32 s22, s8, s6 -; GFX6-NEXT: s_cselect_b64 s[16:17], -1, 0 -; GFX6-NEXT: s_or_b32 s16, s16, s17 -; GFX6-NEXT: s_subb_u32 s16, s21, 0 -; GFX6-NEXT: s_cmp_ge_u32 s16, s7 -; GFX6-NEXT: s_cselect_b32 s17, -1, 0 -; GFX6-NEXT: s_cmp_ge_u32 s22, s6 +; GFX6-NEXT: s_subb_u32 s19, s19, s7 +; GFX6-NEXT: s_sub_u32 s20, s8, s6 +; GFX6-NEXT: s_subb_u32 s19, s19, 0 +; GFX6-NEXT: s_cmp_ge_u32 s19, s7 ; GFX6-NEXT: s_cselect_b32 s21, -1, 0 -; GFX6-NEXT: s_cmp_eq_u32 s16, s7 -; GFX6-NEXT: s_cselect_b32 s16, s21, s17 -; GFX6-NEXT: s_add_u32 s17, s18, 1 -; GFX6-NEXT: s_addc_u32 s21, s19, 0 -; GFX6-NEXT: s_add_u32 s22, s18, 2 -; GFX6-NEXT: s_addc_u32 s23, s19, 0 -; GFX6-NEXT: s_cmp_lg_u32 s16, 0 -; GFX6-NEXT: s_cselect_b32 s16, s22, s17 -; GFX6-NEXT: s_cselect_b32 s17, s23, s21 +; GFX6-NEXT: s_cmp_ge_u32 s20, s6 +; GFX6-NEXT: s_cselect_b32 s20, -1, 0 +; GFX6-NEXT: s_cmp_eq_u32 s19, s7 +; GFX6-NEXT: s_cselect_b32 s19, s20, s21 +; GFX6-NEXT: s_add_u32 s20, s17, 1 +; GFX6-NEXT: s_addc_u32 s21, s16, 0 +; GFX6-NEXT: s_add_u32 s22, s17, 2 +; GFX6-NEXT: s_addc_u32 s23, s16, 0 +; GFX6-NEXT: s_cmp_lg_u32 s19, 0 +; GFX6-NEXT: s_cselect_b32 s19, s22, s20 +; GFX6-NEXT: s_cselect_b32 s20, s23, s21 ; GFX6-NEXT: s_or_b32 s14, s14, s15 -; GFX6-NEXT: s_subb_u32 s9, s9, s20 +; GFX6-NEXT: s_subb_u32 s9, s9, s18 ; GFX6-NEXT: s_cmp_ge_u32 s9, s7 ; GFX6-NEXT: s_cselect_b32 s14, -1, 0 ; GFX6-NEXT: s_cmp_ge_u32 s8, s6 @@ -8415,12 +8401,12 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_cmp_eq_u32 s9, s7 ; GFX6-NEXT: s_cselect_b32 s6, s6, s14 ; GFX6-NEXT: s_cmp_lg_u32 s6, 0 -; GFX6-NEXT: s_cselect_b32 s7, s17, s19 -; GFX6-NEXT: s_cselect_b32 s6, s16, s18 +; GFX6-NEXT: s_cselect_b32 s7, s20, s16 +; GFX6-NEXT: s_cselect_b32 s6, s19, s17 ; GFX6-NEXT: s_xor_b64 s[2:3], s[12:13], s[2:3] ; GFX6-NEXT: s_xor_b64 s[6:7], s[6:7], s[2:3] -; GFX6-NEXT: s_sub_u32 s16, s6, s2 -; GFX6-NEXT: s_subb_u32 s17, s7, s3 +; GFX6-NEXT: s_sub_u32 s14, s6, s2 +; GFX6-NEXT: s_subb_u32 s15, s7, s3 ; GFX6-NEXT: s_ashr_i32 s6, s1, 31 ; GFX6-NEXT: s_add_u32 s0, s0, s6 ; GFX6-NEXT: s_mov_b32 s7, s6 @@ -8428,8 +8414,8 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_xor_b64 s[8:9], s[0:1], s[6:7] ; GFX6-NEXT: v_cvt_f32_u32_e32 v0, s8 ; GFX6-NEXT: v_cvt_f32_u32_e32 v1, s9 -; GFX6-NEXT: s_sub_u32 s12, 0, s8 -; GFX6-NEXT: s_subb_u32 s13, 0, s9 +; GFX6-NEXT: s_sub_u32 s2, 0, s8 +; GFX6-NEXT: s_subb_u32 s3, 0, s9 ; GFX6-NEXT: v_mac_f32_e32 v0, 0x4f800000, v1 ; GFX6-NEXT: v_rcp_f32_e32 v0, v0 ; GFX6-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -8438,128 +8424,121 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: v_mac_f32_e32 v0, 0xcf800000, v1 ; GFX6-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GFX6-NEXT: v_mul_hi_u32 v2, s12, v0 -; GFX6-NEXT: v_readfirstlane_b32 s14, v1 -; GFX6-NEXT: v_readfirstlane_b32 s2, v0 -; GFX6-NEXT: s_mul_i32 s1, s12, s14 -; GFX6-NEXT: v_readfirstlane_b32 s3, v2 -; GFX6-NEXT: s_mul_i32 s0, s13, s2 -; GFX6-NEXT: s_add_i32 s1, s3, s1 -; GFX6-NEXT: s_add_i32 s3, s1, s0 -; GFX6-NEXT: s_mul_i32 s15, s12, s2 -; GFX6-NEXT: v_mul_hi_u32 v2, v0, s3 -; GFX6-NEXT: v_mul_hi_u32 v0, v0, s15 -; GFX6-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9 -; GFX6-NEXT: s_mul_i32 s4, s2, s3 -; GFX6-NEXT: v_readfirstlane_b32 s5, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s2, v0 +; GFX6-NEXT: v_readfirstlane_b32 s12, v1 +; GFX6-NEXT: v_readfirstlane_b32 s0, v0 +; GFX6-NEXT: s_mul_i32 s13, s2, s12 +; GFX6-NEXT: v_readfirstlane_b32 s16, v2 +; GFX6-NEXT: s_mul_i32 s1, s3, s0 +; GFX6-NEXT: s_add_i32 s13, s16, s13 +; GFX6-NEXT: s_add_i32 s13, s13, s1 +; GFX6-NEXT: s_mul_i32 s1, s2, s0 +; GFX6-NEXT: v_mul_hi_u32 v2, v0, s13 +; GFX6-NEXT: v_mul_hi_u32 v0, v0, s1 +; GFX6-NEXT: s_mul_i32 s16, s0, s13 +; GFX6-NEXT: v_readfirstlane_b32 s17, v2 ; GFX6-NEXT: v_readfirstlane_b32 s18, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, v1, s15 -; GFX6-NEXT: v_mul_hi_u32 v1, v1, s3 -; GFX6-NEXT: s_add_u32 s4, s18, s4 -; GFX6-NEXT: s_addc_u32 s5, 0, s5 -; GFX6-NEXT: s_mul_i32 s15, s14, s15 +; GFX6-NEXT: v_mul_hi_u32 v0, v1, s1 +; GFX6-NEXT: v_mul_hi_u32 v1, v1, s13 +; GFX6-NEXT: s_add_u32 s16, s18, s16 +; GFX6-NEXT: s_addc_u32 s17, 0, s17 +; GFX6-NEXT: s_mul_i32 s1, s12, s1 ; GFX6-NEXT: v_readfirstlane_b32 s18, v0 -; GFX6-NEXT: s_add_u32 s4, s4, s15 -; GFX6-NEXT: s_addc_u32 s4, s5, s18 -; GFX6-NEXT: v_readfirstlane_b32 s5, v1 -; GFX6-NEXT: s_addc_u32 s5, s5, 0 -; GFX6-NEXT: s_mul_i32 s3, s14, s3 -; GFX6-NEXT: s_add_u32 s3, s4, s3 -; GFX6-NEXT: s_addc_u32 s4, 0, s5 -; GFX6-NEXT: s_add_u32 s5, s2, s3 -; GFX6-NEXT: v_mov_b32_e32 v0, s5 -; GFX6-NEXT: v_mul_hi_u32 v0, s12, v0 -; GFX6-NEXT: s_cselect_b64 s[2:3], -1, 0 -; GFX6-NEXT: s_or_b32 s2, s2, s3 -; GFX6-NEXT: s_addc_u32 s4, s14, s4 -; GFX6-NEXT: s_mul_i32 s2, s12, s4 -; GFX6-NEXT: v_readfirstlane_b32 s3, v0 -; GFX6-NEXT: s_add_i32 s2, s3, s2 -; GFX6-NEXT: s_mul_i32 s13, s13, s5 -; GFX6-NEXT: s_mul_i32 s3, s12, s5 -; GFX6-NEXT: s_add_i32 s2, s2, s13 -; GFX6-NEXT: v_mov_b32_e32 v2, s3 -; GFX6-NEXT: v_mov_b32_e32 v0, s2 +; GFX6-NEXT: s_add_u32 s1, s16, s1 +; GFX6-NEXT: s_addc_u32 s1, s17, s18 +; GFX6-NEXT: v_readfirstlane_b32 s16, v1 +; GFX6-NEXT: s_addc_u32 s16, s16, 0 +; GFX6-NEXT: s_mul_i32 s13, s12, s13 +; GFX6-NEXT: s_add_u32 s1, s1, s13 +; GFX6-NEXT: s_addc_u32 s13, 0, s16 +; GFX6-NEXT: s_add_u32 s16, s0, s1 +; GFX6-NEXT: v_mov_b32_e32 v0, s16 +; GFX6-NEXT: v_mul_hi_u32 v0, s2, v0 +; GFX6-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9 +; GFX6-NEXT: s_addc_u32 s4, s12, s13 +; GFX6-NEXT: s_mul_i32 s5, s2, s4 +; GFX6-NEXT: v_readfirstlane_b32 s12, v0 +; GFX6-NEXT: s_add_i32 s5, s12, s5 +; GFX6-NEXT: s_mul_i32 s3, s3, s16 +; GFX6-NEXT: s_mul_i32 s2, s2, s16 +; GFX6-NEXT: s_add_i32 s3, s5, s3 +; GFX6-NEXT: v_mov_b32_e32 v2, s2 +; GFX6-NEXT: v_mov_b32_e32 v0, s3 ; GFX6-NEXT: v_mul_hi_u32 v3, s4, v2 -; GFX6-NEXT: v_mul_hi_u32 v2, s5, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s16, v2 ; GFX6-NEXT: v_mul_hi_u32 v1, s4, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, s5, v0 -; GFX6-NEXT: s_mul_i32 s13, s5, s2 -; GFX6-NEXT: v_readfirstlane_b32 s15, v2 -; GFX6-NEXT: s_add_u32 s13, s15, s13 -; GFX6-NEXT: v_readfirstlane_b32 s14, v0 -; GFX6-NEXT: s_mul_i32 s3, s4, s3 -; GFX6-NEXT: s_addc_u32 s14, 0, s14 -; GFX6-NEXT: v_readfirstlane_b32 s12, v3 -; GFX6-NEXT: s_add_u32 s3, s13, s3 -; GFX6-NEXT: s_addc_u32 s3, s14, s12 -; GFX6-NEXT: v_readfirstlane_b32 s12, v1 -; GFX6-NEXT: s_addc_u32 s12, s12, 0 +; GFX6-NEXT: v_mul_hi_u32 v0, s16, v0 +; GFX6-NEXT: s_mul_i32 s12, s16, s3 +; GFX6-NEXT: v_readfirstlane_b32 s17, v2 +; GFX6-NEXT: s_add_u32 s12, s17, s12 +; GFX6-NEXT: v_readfirstlane_b32 s13, v0 ; GFX6-NEXT: s_mul_i32 s2, s4, s2 -; GFX6-NEXT: s_add_u32 s2, s3, s2 -; GFX6-NEXT: s_addc_u32 s12, 0, s12 -; GFX6-NEXT: s_add_u32 s13, s5, s2 -; GFX6-NEXT: s_cselect_b64 s[2:3], -1, 0 -; GFX6-NEXT: s_or_b32 s2, s2, s3 -; GFX6-NEXT: s_addc_u32 s12, s4, s12 +; GFX6-NEXT: s_addc_u32 s13, 0, s13 +; GFX6-NEXT: v_readfirstlane_b32 s5, v3 +; GFX6-NEXT: s_add_u32 s2, s12, s2 +; GFX6-NEXT: s_addc_u32 s2, s13, s5 +; GFX6-NEXT: v_readfirstlane_b32 s5, v1 +; GFX6-NEXT: s_addc_u32 s5, s5, 0 +; GFX6-NEXT: s_mul_i32 s3, s4, s3 +; GFX6-NEXT: s_add_u32 s2, s2, s3 +; GFX6-NEXT: s_addc_u32 s3, 0, s5 +; GFX6-NEXT: s_add_u32 s12, s16, s2 +; GFX6-NEXT: s_addc_u32 s13, s4, s3 ; GFX6-NEXT: s_ashr_i32 s4, s11, 31 ; GFX6-NEXT: s_add_u32 s2, s10, s4 ; GFX6-NEXT: s_mov_b32 s5, s4 ; GFX6-NEXT: s_addc_u32 s3, s11, s4 ; GFX6-NEXT: s_xor_b64 s[10:11], s[2:3], s[4:5] -; GFX6-NEXT: v_mov_b32_e32 v0, s12 +; GFX6-NEXT: v_mov_b32_e32 v0, s13 ; GFX6-NEXT: v_mul_hi_u32 v1, s10, v0 -; GFX6-NEXT: v_mov_b32_e32 v2, s13 +; GFX6-NEXT: v_mov_b32_e32 v2, s12 ; GFX6-NEXT: v_mul_hi_u32 v3, s10, v2 -; GFX6-NEXT: s_mul_i32 s2, s10, s12 -; GFX6-NEXT: v_readfirstlane_b32 s14, v1 +; GFX6-NEXT: s_mul_i32 s2, s10, s13 +; GFX6-NEXT: v_readfirstlane_b32 s16, v1 ; GFX6-NEXT: v_mul_hi_u32 v1, s11, v2 -; GFX6-NEXT: v_readfirstlane_b32 s15, v3 +; GFX6-NEXT: v_readfirstlane_b32 s17, v3 ; GFX6-NEXT: v_mul_hi_u32 v0, s11, v0 -; GFX6-NEXT: s_add_u32 s2, s15, s2 -; GFX6-NEXT: s_addc_u32 s14, 0, s14 -; GFX6-NEXT: s_mul_i32 s13, s11, s13 -; GFX6-NEXT: v_readfirstlane_b32 s15, v1 -; GFX6-NEXT: s_add_u32 s2, s2, s13 -; GFX6-NEXT: s_addc_u32 s2, s14, s15 -; GFX6-NEXT: v_readfirstlane_b32 s13, v0 -; GFX6-NEXT: s_addc_u32 s13, s13, 0 +; GFX6-NEXT: s_add_u32 s2, s17, s2 +; GFX6-NEXT: s_addc_u32 s16, 0, s16 ; GFX6-NEXT: s_mul_i32 s12, s11, s12 -; GFX6-NEXT: s_add_u32 s18, s2, s12 -; GFX6-NEXT: v_mov_b32_e32 v0, s18 +; GFX6-NEXT: v_readfirstlane_b32 s17, v1 +; GFX6-NEXT: s_add_u32 s2, s2, s12 +; GFX6-NEXT: s_addc_u32 s2, s16, s17 +; GFX6-NEXT: v_readfirstlane_b32 s12, v0 +; GFX6-NEXT: s_addc_u32 s12, s12, 0 +; GFX6-NEXT: s_mul_i32 s13, s11, s13 +; GFX6-NEXT: s_add_u32 s16, s2, s13 +; GFX6-NEXT: v_mov_b32_e32 v0, s16 ; GFX6-NEXT: v_mul_hi_u32 v0, s8, v0 -; GFX6-NEXT: s_addc_u32 s19, 0, s13 -; GFX6-NEXT: s_mul_i32 s12, s8, s19 +; GFX6-NEXT: s_addc_u32 s17, 0, s12 +; GFX6-NEXT: s_mul_i32 s12, s8, s17 ; GFX6-NEXT: s_mov_b32 s3, 0xf000 ; GFX6-NEXT: v_readfirstlane_b32 s13, v0 ; GFX6-NEXT: s_add_i32 s12, s13, s12 -; GFX6-NEXT: s_mul_i32 s13, s9, s18 -; GFX6-NEXT: s_add_i32 s20, s12, s13 -; GFX6-NEXT: s_sub_i32 s14, s11, s20 -; GFX6-NEXT: s_mul_i32 s12, s8, s18 +; GFX6-NEXT: s_mul_i32 s13, s9, s16 +; GFX6-NEXT: s_add_i32 s18, s12, s13 +; GFX6-NEXT: s_sub_i32 s19, s11, s18 +; GFX6-NEXT: s_mul_i32 s12, s8, s16 ; GFX6-NEXT: s_sub_u32 s10, s10, s12 ; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s15, s12, s13 -; GFX6-NEXT: s_subb_u32 s21, s14, s9 -; GFX6-NEXT: s_sub_u32 s22, s10, s8 -; GFX6-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GFX6-NEXT: s_or_b32 s14, s14, s15 -; GFX6-NEXT: s_subb_u32 s14, s21, 0 -; GFX6-NEXT: s_cmp_ge_u32 s14, s9 -; GFX6-NEXT: s_cselect_b32 s15, -1, 0 -; GFX6-NEXT: s_cmp_ge_u32 s22, s8 +; GFX6-NEXT: s_subb_u32 s19, s19, s9 +; GFX6-NEXT: s_sub_u32 s20, s10, s8 +; GFX6-NEXT: s_subb_u32 s19, s19, 0 +; GFX6-NEXT: s_cmp_ge_u32 s19, s9 ; GFX6-NEXT: s_cselect_b32 s21, -1, 0 -; GFX6-NEXT: s_cmp_eq_u32 s14, s9 -; GFX6-NEXT: s_cselect_b32 s14, s21, s15 -; GFX6-NEXT: s_add_u32 s15, s18, 1 -; GFX6-NEXT: s_addc_u32 s21, s19, 0 -; GFX6-NEXT: s_add_u32 s22, s18, 2 -; GFX6-NEXT: s_addc_u32 s23, s19, 0 -; GFX6-NEXT: s_cmp_lg_u32 s14, 0 -; GFX6-NEXT: s_cselect_b32 s14, s22, s15 -; GFX6-NEXT: s_cselect_b32 s15, s23, s21 +; GFX6-NEXT: s_cmp_ge_u32 s20, s8 +; GFX6-NEXT: s_cselect_b32 s20, -1, 0 +; GFX6-NEXT: s_cmp_eq_u32 s19, s9 +; GFX6-NEXT: s_cselect_b32 s19, s20, s21 +; GFX6-NEXT: s_add_u32 s20, s16, 1 +; GFX6-NEXT: s_addc_u32 s21, s17, 0 +; GFX6-NEXT: s_add_u32 s22, s16, 2 +; GFX6-NEXT: s_addc_u32 s23, s17, 0 +; GFX6-NEXT: s_cmp_lg_u32 s19, 0 +; GFX6-NEXT: s_cselect_b32 s19, s22, s20 +; GFX6-NEXT: s_cselect_b32 s20, s23, s21 ; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_subb_u32 s11, s11, s20 +; GFX6-NEXT: s_subb_u32 s11, s11, s18 ; GFX6-NEXT: s_cmp_ge_u32 s11, s9 ; GFX6-NEXT: s_cselect_b32 s12, -1, 0 ; GFX6-NEXT: s_cmp_ge_u32 s10, s8 @@ -8567,15 +8546,15 @@ define amdgpu_kernel void @sdiv_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_cmp_eq_u32 s11, s9 ; GFX6-NEXT: s_cselect_b32 s8, s8, s12 ; GFX6-NEXT: s_cmp_lg_u32 s8, 0 -; GFX6-NEXT: s_cselect_b32 s9, s15, s19 -; GFX6-NEXT: s_cselect_b32 s8, s14, s18 +; GFX6-NEXT: s_cselect_b32 s9, s20, s17 +; GFX6-NEXT: s_cselect_b32 s8, s19, s16 ; GFX6-NEXT: s_xor_b64 s[4:5], s[4:5], s[6:7] ; GFX6-NEXT: s_xor_b64 s[6:7], s[8:9], s[4:5] ; GFX6-NEXT: s_sub_u32 s4, s6, s4 ; GFX6-NEXT: s_subb_u32 s5, s7, s5 ; GFX6-NEXT: s_mov_b32 s2, -1 -; GFX6-NEXT: v_mov_b32_e32 v0, s16 -; GFX6-NEXT: v_mov_b32_e32 v1, s17 +; GFX6-NEXT: v_mov_b32_e32 v0, s14 +; GFX6-NEXT: v_mov_b32_e32 v1, s15 ; GFX6-NEXT: v_mov_b32_e32 v2, s4 ; GFX6-NEXT: v_mov_b32_e32 v3, s5 ; GFX6-NEXT: s_waitcnt lgkmcnt(0) @@ -9015,105 +8994,100 @@ define amdgpu_kernel void @srem_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_xor_b64 s[8:9], s[0:1], s[2:3] ; GFX6-NEXT: v_cvt_f32_u32_e32 v0, s8 ; GFX6-NEXT: v_cvt_f32_u32_e32 v1, s9 -; GFX6-NEXT: s_sub_u32 s10, 0, s8 -; GFX6-NEXT: s_subb_u32 s11, 0, s9 +; GFX6-NEXT: s_sub_u32 s0, 0, s8 +; GFX6-NEXT: s_subb_u32 s1, 0, s9 ; GFX6-NEXT: s_mov_b32 s3, 0xf000 ; GFX6-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GFX6-NEXT: v_rcp_f32_e32 v0, v0 -; GFX6-NEXT: s_mov_b32 s2, -1 ; GFX6-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 ; GFX6-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0 ; GFX6-NEXT: v_trunc_f32_e32 v1, v1 ; GFX6-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GFX6-NEXT: v_mul_hi_u32 v2, s10, v0 -; GFX6-NEXT: v_readfirstlane_b32 s12, v1 -; GFX6-NEXT: v_readfirstlane_b32 s0, v0 -; GFX6-NEXT: s_mul_i32 s1, s10, s12 -; GFX6-NEXT: v_readfirstlane_b32 s15, v2 -; GFX6-NEXT: s_mul_i32 s13, s11, s0 -; GFX6-NEXT: s_mul_i32 s14, s10, s0 -; GFX6-NEXT: s_add_i32 s1, s15, s1 -; GFX6-NEXT: v_mul_hi_u32 v3, v0, s14 -; GFX6-NEXT: s_add_i32 s1, s1, s13 -; GFX6-NEXT: v_mul_hi_u32 v0, v0, s1 -; GFX6-NEXT: v_mul_hi_u32 v4, v1, s14 -; GFX6-NEXT: v_readfirstlane_b32 s13, v3 -; GFX6-NEXT: s_mul_i32 s15, s0, s1 -; GFX6-NEXT: v_mul_hi_u32 v1, v1, s1 -; GFX6-NEXT: s_add_u32 s13, s13, s15 -; GFX6-NEXT: v_readfirstlane_b32 s15, v0 -; GFX6-NEXT: s_addc_u32 s15, 0, s15 -; GFX6-NEXT: s_mul_i32 s14, s12, s14 -; GFX6-NEXT: v_readfirstlane_b32 s16, v4 -; GFX6-NEXT: s_add_u32 s13, s13, s14 -; GFX6-NEXT: s_addc_u32 s13, s15, s16 -; GFX6-NEXT: v_readfirstlane_b32 s14, v1 -; GFX6-NEXT: s_addc_u32 s14, s14, 0 -; GFX6-NEXT: s_mul_i32 s1, s12, s1 -; GFX6-NEXT: s_add_u32 s1, s13, s1 -; GFX6-NEXT: s_addc_u32 s13, 0, s14 -; GFX6-NEXT: s_add_u32 s14, s0, s1 -; GFX6-NEXT: v_mov_b32_e32 v0, s14 -; GFX6-NEXT: v_mul_hi_u32 v0, s10, v0 -; GFX6-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GFX6-NEXT: s_or_b32 s0, s0, s1 -; GFX6-NEXT: s_addc_u32 s12, s12, s13 -; GFX6-NEXT: s_mul_i32 s0, s10, s12 -; GFX6-NEXT: v_readfirstlane_b32 s1, v0 -; GFX6-NEXT: s_add_i32 s0, s1, s0 -; GFX6-NEXT: s_mul_i32 s11, s11, s14 -; GFX6-NEXT: s_mul_i32 s1, s10, s14 -; GFX6-NEXT: s_add_i32 s0, s0, s11 -; GFX6-NEXT: v_mov_b32_e32 v2, s1 -; GFX6-NEXT: v_mov_b32_e32 v0, s0 -; GFX6-NEXT: v_mul_hi_u32 v3, s12, v2 -; GFX6-NEXT: v_mul_hi_u32 v2, s14, v2 -; GFX6-NEXT: v_mul_hi_u32 v1, s12, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, s14, v0 -; GFX6-NEXT: s_mul_i32 s11, s14, s0 -; GFX6-NEXT: v_readfirstlane_b32 s15, v2 -; GFX6-NEXT: s_add_u32 s11, s15, s11 -; GFX6-NEXT: v_readfirstlane_b32 s13, v0 -; GFX6-NEXT: s_mul_i32 s1, s12, s1 -; GFX6-NEXT: s_addc_u32 s13, 0, s13 -; GFX6-NEXT: v_readfirstlane_b32 s10, v3 -; GFX6-NEXT: s_add_u32 s1, s11, s1 -; GFX6-NEXT: s_addc_u32 s1, s13, s10 +; GFX6-NEXT: v_mul_hi_u32 v2, s0, v0 ; GFX6-NEXT: v_readfirstlane_b32 s10, v1 -; GFX6-NEXT: s_addc_u32 s10, s10, 0 -; GFX6-NEXT: s_mul_i32 s0, s12, s0 -; GFX6-NEXT: s_add_u32 s0, s1, s0 -; GFX6-NEXT: s_addc_u32 s10, 0, s10 -; GFX6-NEXT: s_add_u32 s13, s14, s0 -; GFX6-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GFX6-NEXT: s_or_b32 s0, s0, s1 -; GFX6-NEXT: s_addc_u32 s12, s12, s10 +; GFX6-NEXT: v_readfirstlane_b32 s2, v0 +; GFX6-NEXT: s_mul_i32 s11, s0, s10 +; GFX6-NEXT: v_readfirstlane_b32 s14, v2 +; GFX6-NEXT: s_mul_i32 s12, s1, s2 +; GFX6-NEXT: s_mul_i32 s13, s0, s2 +; GFX6-NEXT: s_add_i32 s11, s14, s11 +; GFX6-NEXT: v_mul_hi_u32 v3, v0, s13 +; GFX6-NEXT: s_add_i32 s11, s11, s12 +; GFX6-NEXT: v_mul_hi_u32 v0, v0, s11 +; GFX6-NEXT: v_mul_hi_u32 v4, v1, s13 +; GFX6-NEXT: v_readfirstlane_b32 s12, v3 +; GFX6-NEXT: s_mul_i32 s14, s2, s11 +; GFX6-NEXT: v_mul_hi_u32 v1, v1, s11 +; GFX6-NEXT: s_add_u32 s12, s12, s14 +; GFX6-NEXT: v_readfirstlane_b32 s14, v0 +; GFX6-NEXT: s_mul_i32 s13, s10, s13 +; GFX6-NEXT: s_addc_u32 s14, 0, s14 +; GFX6-NEXT: v_readfirstlane_b32 s15, v4 +; GFX6-NEXT: s_add_u32 s12, s12, s13 +; GFX6-NEXT: s_addc_u32 s12, s14, s15 +; GFX6-NEXT: v_readfirstlane_b32 s13, v1 +; GFX6-NEXT: s_addc_u32 s13, s13, 0 +; GFX6-NEXT: s_mul_i32 s11, s10, s11 +; GFX6-NEXT: s_add_u32 s11, s12, s11 +; GFX6-NEXT: s_addc_u32 s12, 0, s13 +; GFX6-NEXT: s_add_u32 s11, s2, s11 +; GFX6-NEXT: v_mov_b32_e32 v0, s11 +; GFX6-NEXT: v_mul_hi_u32 v0, s0, v0 +; GFX6-NEXT: s_addc_u32 s10, s10, s12 +; GFX6-NEXT: s_mul_i32 s12, s0, s10 +; GFX6-NEXT: s_mul_i32 s1, s1, s11 +; GFX6-NEXT: v_readfirstlane_b32 s13, v0 +; GFX6-NEXT: s_add_i32 s12, s13, s12 +; GFX6-NEXT: s_mul_i32 s0, s0, s11 +; GFX6-NEXT: s_add_i32 s1, s12, s1 +; GFX6-NEXT: v_mov_b32_e32 v2, s0 +; GFX6-NEXT: v_mov_b32_e32 v0, s1 +; GFX6-NEXT: v_mul_hi_u32 v3, s10, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s11, v2 +; GFX6-NEXT: v_mul_hi_u32 v1, s10, v0 +; GFX6-NEXT: v_mul_hi_u32 v0, s11, v0 +; GFX6-NEXT: s_mul_i32 s13, s11, s1 +; GFX6-NEXT: v_readfirstlane_b32 s15, v2 +; GFX6-NEXT: s_add_u32 s13, s15, s13 +; GFX6-NEXT: v_readfirstlane_b32 s14, v0 +; GFX6-NEXT: s_mul_i32 s0, s10, s0 +; GFX6-NEXT: s_addc_u32 s14, 0, s14 +; GFX6-NEXT: v_readfirstlane_b32 s12, v3 +; GFX6-NEXT: s_add_u32 s0, s13, s0 +; GFX6-NEXT: s_addc_u32 s0, s14, s12 +; GFX6-NEXT: v_readfirstlane_b32 s12, v1 +; GFX6-NEXT: s_addc_u32 s12, s12, 0 +; GFX6-NEXT: s_mul_i32 s1, s10, s1 +; GFX6-NEXT: s_add_u32 s0, s0, s1 +; GFX6-NEXT: s_addc_u32 s1, 0, s12 +; GFX6-NEXT: s_add_u32 s12, s11, s0 +; GFX6-NEXT: s_addc_u32 s13, s10, s1 ; GFX6-NEXT: s_ashr_i32 s10, s7, 31 ; GFX6-NEXT: s_add_u32 s0, s6, s10 ; GFX6-NEXT: s_mov_b32 s11, s10 ; GFX6-NEXT: s_addc_u32 s1, s7, s10 ; GFX6-NEXT: s_xor_b64 s[6:7], s[0:1], s[10:11] -; GFX6-NEXT: v_mov_b32_e32 v0, s12 +; GFX6-NEXT: v_mov_b32_e32 v0, s13 ; GFX6-NEXT: v_mul_hi_u32 v1, s6, v0 -; GFX6-NEXT: v_mov_b32_e32 v2, s13 +; GFX6-NEXT: v_mov_b32_e32 v2, s12 ; GFX6-NEXT: v_mul_hi_u32 v3, s6, v2 ; GFX6-NEXT: s_mov_b32 s0, s4 ; GFX6-NEXT: v_readfirstlane_b32 s4, v1 ; GFX6-NEXT: v_mul_hi_u32 v1, s7, v2 -; GFX6-NEXT: s_mul_i32 s1, s6, s12 +; GFX6-NEXT: s_mul_i32 s1, s6, s13 ; GFX6-NEXT: v_readfirstlane_b32 s14, v3 ; GFX6-NEXT: v_mul_hi_u32 v0, s7, v0 ; GFX6-NEXT: s_add_u32 s1, s14, s1 ; GFX6-NEXT: s_addc_u32 s4, 0, s4 -; GFX6-NEXT: s_mul_i32 s13, s7, s13 +; GFX6-NEXT: s_mul_i32 s12, s7, s12 ; GFX6-NEXT: v_readfirstlane_b32 s14, v1 -; GFX6-NEXT: s_add_u32 s1, s1, s13 +; GFX6-NEXT: s_add_u32 s1, s1, s12 ; GFX6-NEXT: s_addc_u32 s1, s4, s14 ; GFX6-NEXT: v_readfirstlane_b32 s4, v0 ; GFX6-NEXT: s_addc_u32 s4, s4, 0 -; GFX6-NEXT: s_mul_i32 s12, s7, s12 +; GFX6-NEXT: s_mul_i32 s12, s7, s13 ; GFX6-NEXT: s_add_u32 s12, s1, s12 ; GFX6-NEXT: v_mov_b32_e32 v0, s12 ; GFX6-NEXT: v_mul_hi_u32 v0, s8, v0 @@ -9128,11 +9102,9 @@ define amdgpu_kernel void @srem_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_mul_i32 s4, s8, s12 ; GFX6-NEXT: s_sub_u32 s6, s6, s4 ; GFX6-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GFX6-NEXT: s_or_b32 s12, s4, s5 ; GFX6-NEXT: s_subb_u32 s15, s13, s9 ; GFX6-NEXT: s_sub_u32 s16, s6, s8 ; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s17, s12, s13 ; GFX6-NEXT: s_subb_u32 s17, s15, 0 ; GFX6-NEXT: s_cmp_ge_u32 s17, s9 ; GFX6-NEXT: s_cselect_b32 s18, -1, 0 @@ -9141,13 +9113,11 @@ define amdgpu_kernel void @srem_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_cmp_eq_u32 s17, s9 ; GFX6-NEXT: s_cselect_b32 s18, s19, s18 ; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_subb_u32 s15, s15, s9 -; GFX6-NEXT: s_sub_u32 s19, s16, s8 -; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_subb_u32 s12, s15, 0 +; GFX6-NEXT: s_subb_u32 s12, s15, s9 +; GFX6-NEXT: s_sub_u32 s13, s16, s8 +; GFX6-NEXT: s_subb_u32 s12, s12, 0 ; GFX6-NEXT: s_cmp_lg_u32 s18, 0 -; GFX6-NEXT: s_cselect_b32 s13, s19, s16 +; GFX6-NEXT: s_cselect_b32 s13, s13, s16 ; GFX6-NEXT: s_cselect_b32 s12, s12, s17 ; GFX6-NEXT: s_or_b32 s4, s4, s5 ; GFX6-NEXT: s_subb_u32 s4, s7, s14 @@ -9164,6 +9134,7 @@ define amdgpu_kernel void @srem_i64_pow2_shl_denom(ptr addrspace(1) %out, i64 %x ; GFX6-NEXT: s_sub_u32 s4, s4, s10 ; GFX6-NEXT: s_subb_u32 s5, s5, s10 ; GFX6-NEXT: v_mov_b32_e32 v0, s4 +; GFX6-NEXT: s_mov_b32 s2, -1 ; GFX6-NEXT: v_mov_b32_e32 v1, s5 ; GFX6-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0 ; GFX6-NEXT: s_endpgm @@ -9405,8 +9376,8 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_xor_b64 s[2:3], s[2:3], s[6:7] ; GFX6-NEXT: v_cvt_f32_u32_e32 v0, s2 ; GFX6-NEXT: v_cvt_f32_u32_e32 v1, s3 -; GFX6-NEXT: s_sub_u32 s12, 0, s2 -; GFX6-NEXT: s_subb_u32 s13, 0, s3 +; GFX6-NEXT: s_sub_u32 s6, 0, s2 +; GFX6-NEXT: s_subb_u32 s7, 0, s3 ; GFX6-NEXT: v_mac_f32_e32 v0, 0x4f800000, v1 ; GFX6-NEXT: v_rcp_f32_e32 v0, v0 ; GFX6-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -9415,69 +9386,65 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: v_mac_f32_e32 v0, 0xcf800000, v1 ; GFX6-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GFX6-NEXT: v_mul_hi_u32 v2, s12, v0 -; GFX6-NEXT: v_readfirstlane_b32 s14, v1 -; GFX6-NEXT: v_readfirstlane_b32 s6, v0 -; GFX6-NEXT: s_mul_i32 s7, s12, s14 +; GFX6-NEXT: v_mul_hi_u32 v2, s6, v0 +; GFX6-NEXT: v_readfirstlane_b32 s12, v1 +; GFX6-NEXT: v_readfirstlane_b32 s13, v0 +; GFX6-NEXT: s_mul_i32 s14, s6, s12 ; GFX6-NEXT: v_readfirstlane_b32 s17, v2 -; GFX6-NEXT: s_mul_i32 s15, s13, s6 -; GFX6-NEXT: s_mul_i32 s16, s12, s6 -; GFX6-NEXT: s_add_i32 s7, s17, s7 +; GFX6-NEXT: s_mul_i32 s15, s7, s13 +; GFX6-NEXT: s_mul_i32 s16, s6, s13 +; GFX6-NEXT: s_add_i32 s14, s17, s14 ; GFX6-NEXT: v_mul_hi_u32 v3, v0, s16 -; GFX6-NEXT: s_add_i32 s7, s7, s15 -; GFX6-NEXT: v_mul_hi_u32 v0, v0, s7 +; GFX6-NEXT: s_add_i32 s14, s14, s15 +; GFX6-NEXT: v_mul_hi_u32 v0, v0, s14 ; GFX6-NEXT: v_mul_hi_u32 v4, v1, s16 ; GFX6-NEXT: v_readfirstlane_b32 s15, v3 -; GFX6-NEXT: s_mul_i32 s18, s6, s7 -; GFX6-NEXT: v_mul_hi_u32 v1, v1, s7 +; GFX6-NEXT: s_mul_i32 s18, s13, s14 +; GFX6-NEXT: v_mul_hi_u32 v1, v1, s14 ; GFX6-NEXT: s_add_u32 s15, s15, s18 ; GFX6-NEXT: v_readfirstlane_b32 s18, v0 -; GFX6-NEXT: s_mul_i32 s16, s14, s16 +; GFX6-NEXT: s_mul_i32 s16, s12, s16 ; GFX6-NEXT: s_addc_u32 s18, 0, s18 ; GFX6-NEXT: v_readfirstlane_b32 s17, v4 ; GFX6-NEXT: s_add_u32 s15, s15, s16 ; GFX6-NEXT: s_addc_u32 s15, s18, s17 ; GFX6-NEXT: v_readfirstlane_b32 s16, v1 ; GFX6-NEXT: s_addc_u32 s16, s16, 0 -; GFX6-NEXT: s_mul_i32 s7, s14, s7 -; GFX6-NEXT: s_add_u32 s7, s15, s7 +; GFX6-NEXT: s_mul_i32 s14, s12, s14 +; GFX6-NEXT: s_add_u32 s14, s15, s14 ; GFX6-NEXT: s_addc_u32 s15, 0, s16 -; GFX6-NEXT: s_add_u32 s16, s6, s7 -; GFX6-NEXT: v_mov_b32_e32 v0, s16 -; GFX6-NEXT: v_mul_hi_u32 v0, s12, v0 -; GFX6-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GFX6-NEXT: s_or_b32 s6, s6, s7 -; GFX6-NEXT: s_addc_u32 s14, s14, s15 -; GFX6-NEXT: s_mul_i32 s6, s12, s14 -; GFX6-NEXT: v_readfirstlane_b32 s7, v0 -; GFX6-NEXT: s_add_i32 s6, s7, s6 -; GFX6-NEXT: s_mul_i32 s13, s13, s16 -; GFX6-NEXT: s_mul_i32 s7, s12, s16 -; GFX6-NEXT: s_add_i32 s6, s6, s13 -; GFX6-NEXT: v_mov_b32_e32 v2, s7 -; GFX6-NEXT: v_mov_b32_e32 v0, s6 -; GFX6-NEXT: v_mul_hi_u32 v3, s14, v2 -; GFX6-NEXT: v_mul_hi_u32 v2, s16, v2 -; GFX6-NEXT: v_mul_hi_u32 v1, s14, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, s16, v0 -; GFX6-NEXT: s_mul_i32 s13, s16, s6 -; GFX6-NEXT: v_readfirstlane_b32 s17, v2 -; GFX6-NEXT: s_add_u32 s13, s17, s13 +; GFX6-NEXT: s_add_u32 s13, s13, s14 +; GFX6-NEXT: v_mov_b32_e32 v0, s13 +; GFX6-NEXT: v_mul_hi_u32 v0, s6, v0 +; GFX6-NEXT: s_addc_u32 s12, s12, s15 +; GFX6-NEXT: s_mul_i32 s14, s6, s12 +; GFX6-NEXT: s_mul_i32 s7, s7, s13 ; GFX6-NEXT: v_readfirstlane_b32 s15, v0 -; GFX6-NEXT: s_mul_i32 s7, s14, s7 -; GFX6-NEXT: s_addc_u32 s15, 0, s15 -; GFX6-NEXT: v_readfirstlane_b32 s12, v3 -; GFX6-NEXT: s_add_u32 s7, s13, s7 -; GFX6-NEXT: s_addc_u32 s7, s15, s12 -; GFX6-NEXT: v_readfirstlane_b32 s12, v1 -; GFX6-NEXT: s_addc_u32 s12, s12, 0 -; GFX6-NEXT: s_mul_i32 s6, s14, s6 -; GFX6-NEXT: s_add_u32 s6, s7, s6 -; GFX6-NEXT: s_addc_u32 s12, 0, s12 -; GFX6-NEXT: s_add_u32 s13, s16, s6 -; GFX6-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GFX6-NEXT: s_or_b32 s6, s6, s7 -; GFX6-NEXT: s_addc_u32 s12, s14, s12 +; GFX6-NEXT: s_add_i32 s14, s15, s14 +; GFX6-NEXT: s_mul_i32 s6, s6, s13 +; GFX6-NEXT: s_add_i32 s7, s14, s7 +; GFX6-NEXT: v_mov_b32_e32 v2, s6 +; GFX6-NEXT: v_mov_b32_e32 v0, s7 +; GFX6-NEXT: v_mul_hi_u32 v3, s12, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s13, v2 +; GFX6-NEXT: v_mul_hi_u32 v1, s12, v0 +; GFX6-NEXT: v_mul_hi_u32 v0, s13, v0 +; GFX6-NEXT: s_mul_i32 s15, s13, s7 +; GFX6-NEXT: v_readfirstlane_b32 s17, v2 +; GFX6-NEXT: s_add_u32 s15, s17, s15 +; GFX6-NEXT: v_readfirstlane_b32 s16, v0 +; GFX6-NEXT: s_mul_i32 s6, s12, s6 +; GFX6-NEXT: s_addc_u32 s16, 0, s16 +; GFX6-NEXT: v_readfirstlane_b32 s14, v3 +; GFX6-NEXT: s_add_u32 s6, s15, s6 +; GFX6-NEXT: s_addc_u32 s6, s16, s14 +; GFX6-NEXT: v_readfirstlane_b32 s14, v1 +; GFX6-NEXT: s_addc_u32 s14, s14, 0 +; GFX6-NEXT: s_mul_i32 s7, s12, s7 +; GFX6-NEXT: s_add_u32 s6, s6, s7 +; GFX6-NEXT: s_addc_u32 s7, 0, s14 +; GFX6-NEXT: s_add_u32 s13, s13, s6 +; GFX6-NEXT: s_addc_u32 s12, s12, s7 ; GFX6-NEXT: s_ashr_i32 s6, s9, 31 ; GFX6-NEXT: s_add_u32 s8, s8, s6 ; GFX6-NEXT: s_mov_b32 s7, s6 @@ -9514,11 +9481,9 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_mul_i32 s12, s2, s12 ; GFX6-NEXT: s_sub_u32 s8, s8, s12 ; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s15, s12, s13 ; GFX6-NEXT: s_subb_u32 s17, s14, s3 ; GFX6-NEXT: s_sub_u32 s18, s8, s2 ; GFX6-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GFX6-NEXT: s_or_b32 s19, s14, s15 ; GFX6-NEXT: s_subb_u32 s19, s17, 0 ; GFX6-NEXT: s_cmp_ge_u32 s19, s3 ; GFX6-NEXT: s_cselect_b32 s20, -1, 0 @@ -9527,13 +9492,11 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_cmp_eq_u32 s19, s3 ; GFX6-NEXT: s_cselect_b32 s20, s21, s20 ; GFX6-NEXT: s_or_b32 s14, s14, s15 -; GFX6-NEXT: s_subb_u32 s17, s17, s3 -; GFX6-NEXT: s_sub_u32 s21, s18, s2 -; GFX6-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GFX6-NEXT: s_or_b32 s14, s14, s15 -; GFX6-NEXT: s_subb_u32 s14, s17, 0 +; GFX6-NEXT: s_subb_u32 s14, s17, s3 +; GFX6-NEXT: s_sub_u32 s15, s18, s2 +; GFX6-NEXT: s_subb_u32 s14, s14, 0 ; GFX6-NEXT: s_cmp_lg_u32 s20, 0 -; GFX6-NEXT: s_cselect_b32 s15, s21, s18 +; GFX6-NEXT: s_cselect_b32 s15, s15, s18 ; GFX6-NEXT: s_cselect_b32 s14, s14, s19 ; GFX6-NEXT: s_or_b32 s12, s12, s13 ; GFX6-NEXT: s_subb_u32 s9, s9, s16 @@ -9556,8 +9519,8 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_xor_b64 s[6:7], s[0:1], s[2:3] ; GFX6-NEXT: v_cvt_f32_u32_e32 v0, s6 ; GFX6-NEXT: v_cvt_f32_u32_e32 v1, s7 -; GFX6-NEXT: s_sub_u32 s8, 0, s6 -; GFX6-NEXT: s_subb_u32 s9, 0, s7 +; GFX6-NEXT: s_sub_u32 s2, 0, s6 +; GFX6-NEXT: s_subb_u32 s3, 0, s7 ; GFX6-NEXT: v_mac_f32_e32 v0, 0x4f800000, v1 ; GFX6-NEXT: v_rcp_f32_e32 v0, v0 ; GFX6-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -9566,70 +9529,66 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: v_mac_f32_e32 v0, 0xcf800000, v1 ; GFX6-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GFX6-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GFX6-NEXT: v_mul_hi_u32 v2, s8, v0 -; GFX6-NEXT: v_readfirstlane_b32 s12, v1 -; GFX6-NEXT: v_readfirstlane_b32 s2, v0 -; GFX6-NEXT: s_mul_i32 s1, s8, s12 -; GFX6-NEXT: v_readfirstlane_b32 s3, v2 -; GFX6-NEXT: s_mul_i32 s0, s9, s2 -; GFX6-NEXT: s_add_i32 s1, s3, s1 -; GFX6-NEXT: s_add_i32 s3, s1, s0 -; GFX6-NEXT: s_mul_i32 s13, s8, s2 -; GFX6-NEXT: v_mul_hi_u32 v2, v0, s3 -; GFX6-NEXT: v_mul_hi_u32 v0, v0, s13 -; GFX6-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9 -; GFX6-NEXT: s_mul_i32 s4, s2, s3 -; GFX6-NEXT: v_readfirstlane_b32 s5, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s2, v0 +; GFX6-NEXT: v_readfirstlane_b32 s8, v1 +; GFX6-NEXT: v_readfirstlane_b32 s0, v0 +; GFX6-NEXT: s_mul_i32 s9, s2, s8 +; GFX6-NEXT: v_readfirstlane_b32 s12, v2 +; GFX6-NEXT: s_mul_i32 s1, s3, s0 +; GFX6-NEXT: s_add_i32 s9, s12, s9 +; GFX6-NEXT: s_add_i32 s9, s9, s1 +; GFX6-NEXT: s_mul_i32 s1, s2, s0 +; GFX6-NEXT: v_mul_hi_u32 v2, v0, s9 +; GFX6-NEXT: v_mul_hi_u32 v0, v0, s1 +; GFX6-NEXT: s_mul_i32 s12, s0, s9 +; GFX6-NEXT: v_readfirstlane_b32 s13, v2 ; GFX6-NEXT: v_readfirstlane_b32 s16, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, v1, s13 -; GFX6-NEXT: v_mul_hi_u32 v1, v1, s3 -; GFX6-NEXT: s_add_u32 s4, s16, s4 -; GFX6-NEXT: s_addc_u32 s5, 0, s5 -; GFX6-NEXT: s_mul_i32 s13, s12, s13 +; GFX6-NEXT: v_mul_hi_u32 v0, v1, s1 +; GFX6-NEXT: v_mul_hi_u32 v1, v1, s9 +; GFX6-NEXT: s_add_u32 s12, s16, s12 +; GFX6-NEXT: s_addc_u32 s13, 0, s13 +; GFX6-NEXT: s_mul_i32 s1, s8, s1 ; GFX6-NEXT: v_readfirstlane_b32 s16, v0 -; GFX6-NEXT: s_add_u32 s4, s4, s13 -; GFX6-NEXT: s_addc_u32 s4, s5, s16 -; GFX6-NEXT: v_readfirstlane_b32 s5, v1 -; GFX6-NEXT: s_addc_u32 s5, s5, 0 -; GFX6-NEXT: s_mul_i32 s3, s12, s3 -; GFX6-NEXT: s_add_u32 s3, s4, s3 -; GFX6-NEXT: s_addc_u32 s4, 0, s5 -; GFX6-NEXT: s_add_u32 s5, s2, s3 -; GFX6-NEXT: v_mov_b32_e32 v0, s5 -; GFX6-NEXT: v_mul_hi_u32 v0, s8, v0 -; GFX6-NEXT: s_cselect_b64 s[2:3], -1, 0 -; GFX6-NEXT: s_or_b32 s2, s2, s3 -; GFX6-NEXT: s_addc_u32 s4, s12, s4 -; GFX6-NEXT: s_mul_i32 s2, s8, s4 -; GFX6-NEXT: v_readfirstlane_b32 s3, v0 -; GFX6-NEXT: s_add_i32 s2, s3, s2 -; GFX6-NEXT: s_mul_i32 s9, s9, s5 -; GFX6-NEXT: s_mul_i32 s3, s8, s5 -; GFX6-NEXT: s_add_i32 s2, s2, s9 -; GFX6-NEXT: v_mov_b32_e32 v2, s3 -; GFX6-NEXT: v_mov_b32_e32 v0, s2 +; GFX6-NEXT: s_add_u32 s1, s12, s1 +; GFX6-NEXT: s_addc_u32 s1, s13, s16 +; GFX6-NEXT: v_readfirstlane_b32 s12, v1 +; GFX6-NEXT: s_addc_u32 s12, s12, 0 +; GFX6-NEXT: s_mul_i32 s9, s8, s9 +; GFX6-NEXT: s_add_u32 s1, s1, s9 +; GFX6-NEXT: s_addc_u32 s9, 0, s12 +; GFX6-NEXT: s_add_u32 s12, s0, s1 +; GFX6-NEXT: v_mov_b32_e32 v0, s12 +; GFX6-NEXT: v_mul_hi_u32 v0, s2, v0 +; GFX6-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9 +; GFX6-NEXT: s_addc_u32 s4, s8, s9 +; GFX6-NEXT: s_mul_i32 s5, s2, s4 +; GFX6-NEXT: v_readfirstlane_b32 s8, v0 +; GFX6-NEXT: s_add_i32 s5, s8, s5 +; GFX6-NEXT: s_mul_i32 s3, s3, s12 +; GFX6-NEXT: s_mul_i32 s2, s2, s12 +; GFX6-NEXT: s_add_i32 s3, s5, s3 +; GFX6-NEXT: v_mov_b32_e32 v2, s2 +; GFX6-NEXT: v_mov_b32_e32 v0, s3 ; GFX6-NEXT: v_mul_hi_u32 v3, s4, v2 -; GFX6-NEXT: v_mul_hi_u32 v2, s5, v2 +; GFX6-NEXT: v_mul_hi_u32 v2, s12, v2 ; GFX6-NEXT: v_mul_hi_u32 v1, s4, v0 -; GFX6-NEXT: v_mul_hi_u32 v0, s5, v0 -; GFX6-NEXT: s_mul_i32 s9, s5, s2 +; GFX6-NEXT: v_mul_hi_u32 v0, s12, v0 +; GFX6-NEXT: s_mul_i32 s8, s12, s3 ; GFX6-NEXT: v_readfirstlane_b32 s13, v2 -; GFX6-NEXT: s_add_u32 s9, s13, s9 -; GFX6-NEXT: v_readfirstlane_b32 s12, v0 -; GFX6-NEXT: s_mul_i32 s3, s4, s3 -; GFX6-NEXT: s_addc_u32 s12, 0, s12 -; GFX6-NEXT: v_readfirstlane_b32 s8, v3 -; GFX6-NEXT: s_add_u32 s3, s9, s3 -; GFX6-NEXT: s_addc_u32 s3, s12, s8 -; GFX6-NEXT: v_readfirstlane_b32 s8, v1 -; GFX6-NEXT: s_addc_u32 s8, s8, 0 +; GFX6-NEXT: s_add_u32 s8, s13, s8 +; GFX6-NEXT: v_readfirstlane_b32 s9, v0 ; GFX6-NEXT: s_mul_i32 s2, s4, s2 -; GFX6-NEXT: s_add_u32 s2, s3, s2 -; GFX6-NEXT: s_addc_u32 s8, 0, s8 -; GFX6-NEXT: s_add_u32 s12, s5, s2 -; GFX6-NEXT: s_cselect_b64 s[2:3], -1, 0 -; GFX6-NEXT: s_or_b32 s2, s2, s3 -; GFX6-NEXT: s_addc_u32 s13, s4, s8 +; GFX6-NEXT: s_addc_u32 s9, 0, s9 +; GFX6-NEXT: v_readfirstlane_b32 s5, v3 +; GFX6-NEXT: s_add_u32 s2, s8, s2 +; GFX6-NEXT: s_addc_u32 s2, s9, s5 +; GFX6-NEXT: v_readfirstlane_b32 s5, v1 +; GFX6-NEXT: s_addc_u32 s5, s5, 0 +; GFX6-NEXT: s_mul_i32 s3, s4, s3 +; GFX6-NEXT: s_add_u32 s2, s2, s3 +; GFX6-NEXT: s_addc_u32 s3, 0, s5 +; GFX6-NEXT: s_add_u32 s12, s12, s2 +; GFX6-NEXT: s_addc_u32 s13, s4, s3 ; GFX6-NEXT: s_ashr_i32 s4, s11, 31 ; GFX6-NEXT: s_add_u32 s2, s10, s4 ; GFX6-NEXT: s_mov_b32 s5, s4 @@ -9667,11 +9626,9 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_mul_i32 s10, s6, s11 ; GFX6-NEXT: s_sub_u32 s8, s8, s10 ; GFX6-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GFX6-NEXT: s_or_b32 s13, s10, s11 ; GFX6-NEXT: s_subb_u32 s17, s12, s7 ; GFX6-NEXT: s_sub_u32 s18, s8, s6 ; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s19, s12, s13 ; GFX6-NEXT: s_subb_u32 s19, s17, 0 ; GFX6-NEXT: s_cmp_ge_u32 s19, s7 ; GFX6-NEXT: s_cselect_b32 s20, -1, 0 @@ -9680,13 +9637,11 @@ define amdgpu_kernel void @srem_v2i64_pow2_shl_denom(ptr addrspace(1) %out, <2 x ; GFX6-NEXT: s_cmp_eq_u32 s19, s7 ; GFX6-NEXT: s_cselect_b32 s20, s21, s20 ; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_subb_u32 s17, s17, s7 -; GFX6-NEXT: s_sub_u32 s21, s18, s6 -; GFX6-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GFX6-NEXT: s_or_b32 s12, s12, s13 -; GFX6-NEXT: s_subb_u32 s12, s17, 0 +; GFX6-NEXT: s_subb_u32 s12, s17, s7 +; GFX6-NEXT: s_sub_u32 s13, s18, s6 +; GFX6-NEXT: s_subb_u32 s12, s12, 0 ; GFX6-NEXT: s_cmp_lg_u32 s20, 0 -; GFX6-NEXT: s_cselect_b32 s13, s21, s18 +; GFX6-NEXT: s_cselect_b32 s13, s13, s18 ; GFX6-NEXT: s_cselect_b32 s12, s12, s19 ; GFX6-NEXT: s_or_b32 s10, s10, s11 ; GFX6-NEXT: s_subb_u32 s9, s9, s16 diff --git a/llvm/test/CodeGen/AMDGPU/amdgpu-simplify-uniform-waterfall.ll b/llvm/test/CodeGen/AMDGPU/amdgpu-simplify-uniform-waterfall.ll index c962c05..5d79696 100644 --- a/llvm/test/CodeGen/AMDGPU/amdgpu-simplify-uniform-waterfall.ll +++ b/llvm/test/CodeGen/AMDGPU/amdgpu-simplify-uniform-waterfall.ll @@ -239,7 +239,8 @@ define protected amdgpu_kernel void @trivial_uniform_waterfall(ptr addrspace(1) ; PASS-CHECK-NEXT: [[IS_DONE:%.*]] = icmp eq i64 [[BALLOT]], 0 ; PASS-CHECK-NEXT: br i1 [[TMP0]], label %[[EXIT:.*]], label %[[IF:.*]] ; PASS-CHECK: [[IF]]: -; PASS-CHECK-NEXT: [[IS_FIRST_ACTIVE_ID:%.*]] = icmp eq i32 0, 0 +; PASS-CHECK-NEXT: [[FIRST_ACTIVE_ID:%.*]] = tail call noundef i32 @llvm.amdgcn.readfirstlane.i32(i32 0) +; PASS-CHECK-NEXT: [[IS_FIRST_ACTIVE_ID:%.*]] = icmp eq i32 0, [[FIRST_ACTIVE_ID]] ; PASS-CHECK-NEXT: br i1 [[IS_FIRST_ACTIVE_ID]], label %[[WORK:.*]], label %[[TAIL]] ; PASS-CHECK: [[WORK]]: ; PASS-CHECK-NEXT: store i32 5, ptr addrspace(1) [[OUT]], align 4 @@ -308,7 +309,8 @@ define protected amdgpu_kernel void @uniform_waterfall(ptr addrspace(1) %out, i3 ; PASS-CHECK-NEXT: [[IS_DONE:%.*]] = icmp eq i64 [[BALLOT]], 0 ; PASS-CHECK-NEXT: br i1 [[TMP0]], label %[[EXIT:.*]], label %[[IF:.*]] ; PASS-CHECK: [[IF]]: -; PASS-CHECK-NEXT: [[IS_FIRST_ACTIVE_ID:%.*]] = icmp eq i32 [[MYMASK]], [[MYMASK]] +; PASS-CHECK-NEXT: [[FIRST_ACTIVE_ID:%.*]] = tail call noundef i32 @llvm.amdgcn.readfirstlane.i32(i32 [[MYMASK]]) +; PASS-CHECK-NEXT: [[IS_FIRST_ACTIVE_ID:%.*]] = icmp eq i32 [[MYMASK]], [[FIRST_ACTIVE_ID]] ; PASS-CHECK-NEXT: br i1 [[IS_FIRST_ACTIVE_ID]], label %[[WORK:.*]], label %[[TAIL]] ; PASS-CHECK: [[WORK]]: ; PASS-CHECK-NEXT: store i32 5, ptr addrspace(1) [[OUT]], align 4 diff --git a/llvm/test/CodeGen/AMDGPU/amdgpu-uniform-intrinsic-combine.ll b/llvm/test/CodeGen/AMDGPU/amdgpu-uniform-intrinsic-combine.ll index a7e828c..402ccd9 100644 --- a/llvm/test/CodeGen/AMDGPU/amdgpu-uniform-intrinsic-combine.ll +++ b/llvm/test/CodeGen/AMDGPU/amdgpu-uniform-intrinsic-combine.ll @@ -248,12 +248,14 @@ define amdgpu_kernel void @readfirstlane_constant(ptr addrspace(1) %out) { ; ; PASS-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_constant( ; PASS-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]]) #[[ATTR0]] { -; PASS-CHECK-NEXT: store i32 7, ptr addrspace(1) [[OUT]], align 4 +; PASS-CHECK-NEXT: [[V:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 7) +; PASS-CHECK-NEXT: store i32 [[V]], ptr addrspace(1) [[OUT]], align 4 ; PASS-CHECK-NEXT: ret void ; ; DCE-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_constant( ; DCE-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]]) #[[ATTR0]] { -; DCE-CHECK-NEXT: store i32 7, ptr addrspace(1) [[OUT]], align 4 +; DCE-CHECK-NEXT: [[V:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 7) +; DCE-CHECK-NEXT: store i32 [[V]], ptr addrspace(1) [[OUT]], align 4 ; DCE-CHECK-NEXT: ret void ; %v = call i32 @llvm.amdgcn.readfirstlane(i32 7) @@ -269,12 +271,14 @@ define amdgpu_kernel void @readfirstlane_with_argument(ptr addrspace(1) %out, i3 ; ; PASS-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_with_argument( ; PASS-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]], i32 [[SRC0:%.*]]) #[[ATTR0]] { -; PASS-CHECK-NEXT: store i32 [[SRC0]], ptr addrspace(1) [[OUT]], align 4 +; PASS-CHECK-NEXT: [[V:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[SRC0]]) +; PASS-CHECK-NEXT: store i32 [[V]], ptr addrspace(1) [[OUT]], align 4 ; PASS-CHECK-NEXT: ret void ; ; DCE-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_with_argument( ; DCE-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]], i32 [[SRC0:%.*]]) #[[ATTR0]] { -; DCE-CHECK-NEXT: store i32 [[SRC0]], ptr addrspace(1) [[OUT]], align 4 +; DCE-CHECK-NEXT: [[V:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[SRC0]]) +; DCE-CHECK-NEXT: store i32 [[V]], ptr addrspace(1) [[OUT]], align 4 ; DCE-CHECK-NEXT: ret void ; %v = call i32 @llvm.amdgcn.readfirstlane(i32 %src0) @@ -360,12 +364,16 @@ define amdgpu_kernel void @readfirstlane_with_readfirstlane(ptr addrspace(1) %ou ; ; PASS-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_with_readfirstlane( ; PASS-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]]) #[[ATTR0]] { -; PASS-CHECK-NEXT: store i32 5, ptr addrspace(1) [[OUT]], align 4 +; PASS-CHECK-NEXT: [[V1:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 5) +; PASS-CHECK-NEXT: [[V2:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[V1]]) +; PASS-CHECK-NEXT: store i32 [[V2]], ptr addrspace(1) [[OUT]], align 4 ; PASS-CHECK-NEXT: ret void ; ; DCE-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_with_readfirstlane( ; DCE-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]]) #[[ATTR0]] { -; DCE-CHECK-NEXT: store i32 5, ptr addrspace(1) [[OUT]], align 4 +; DCE-CHECK-NEXT: [[V1:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 5) +; DCE-CHECK-NEXT: [[V2:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[V1]]) +; DCE-CHECK-NEXT: store i32 [[V2]], ptr addrspace(1) [[OUT]], align 4 ; DCE-CHECK-NEXT: ret void ; %v1 = call i32 @llvm.amdgcn.readfirstlane(i32 5) @@ -388,7 +396,8 @@ define amdgpu_kernel void @readfirstlane_with_readlane(ptr addrspace(1) %out) { ; PASS-CHECK-NEXT: [[TIDX:%.*]] = call i32 @llvm.amdgcn.workitem.id.x() ; PASS-CHECK-NEXT: [[TIDY:%.*]] = call i32 @llvm.amdgcn.workitem.id.y() ; PASS-CHECK-NEXT: [[V1:%.*]] = call i32 @llvm.amdgcn.readlane.i32(i32 [[TIDX]], i32 [[TIDY]]) -; PASS-CHECK-NEXT: store i32 [[V1]], ptr addrspace(1) [[OUT]], align 4 +; PASS-CHECK-NEXT: [[V2:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[V1]]) +; PASS-CHECK-NEXT: store i32 [[V2]], ptr addrspace(1) [[OUT]], align 4 ; PASS-CHECK-NEXT: ret void ; ; DCE-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_with_readlane( @@ -396,7 +405,8 @@ define amdgpu_kernel void @readfirstlane_with_readlane(ptr addrspace(1) %out) { ; DCE-CHECK-NEXT: [[TIDX:%.*]] = call i32 @llvm.amdgcn.workitem.id.x() ; DCE-CHECK-NEXT: [[TIDY:%.*]] = call i32 @llvm.amdgcn.workitem.id.y() ; DCE-CHECK-NEXT: [[V1:%.*]] = call i32 @llvm.amdgcn.readlane.i32(i32 [[TIDX]], i32 [[TIDY]]) -; DCE-CHECK-NEXT: store i32 [[V1]], ptr addrspace(1) [[OUT]], align 4 +; DCE-CHECK-NEXT: [[V2:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[V1]]) +; DCE-CHECK-NEXT: store i32 [[V2]], ptr addrspace(1) [[OUT]], align 4 ; DCE-CHECK-NEXT: ret void ; %tidx = call i32 @llvm.amdgcn.workitem.id.x() @@ -537,13 +547,15 @@ define amdgpu_kernel void @readfirstlane_random(ptr addrspace(1) %out) { ; PASS-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_random( ; PASS-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]]) #[[ATTR0]] { ; PASS-CHECK-NEXT: [[RANDOM:%.*]] = xor i32 123, 456 -; PASS-CHECK-NEXT: store i32 [[RANDOM]], ptr addrspace(1) [[OUT]], align 4 +; PASS-CHECK-NEXT: [[V:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[RANDOM]]) +; PASS-CHECK-NEXT: store i32 [[V]], ptr addrspace(1) [[OUT]], align 4 ; PASS-CHECK-NEXT: ret void ; ; DCE-CHECK-LABEL: define amdgpu_kernel void @readfirstlane_random( ; DCE-CHECK-SAME: ptr addrspace(1) [[OUT:%.*]]) #[[ATTR0]] { ; DCE-CHECK-NEXT: [[RANDOM:%.*]] = xor i32 123, 456 -; DCE-CHECK-NEXT: store i32 [[RANDOM]], ptr addrspace(1) [[OUT]], align 4 +; DCE-CHECK-NEXT: [[V:%.*]] = call i32 @llvm.amdgcn.readfirstlane.i32(i32 [[RANDOM]]) +; DCE-CHECK-NEXT: store i32 [[V]], ptr addrspace(1) [[OUT]], align 4 ; DCE-CHECK-NEXT: ret void ; %random = xor i32 123, 456 diff --git a/llvm/test/CodeGen/AMDGPU/call-args-inreg-bfloat.ll b/llvm/test/CodeGen/AMDGPU/call-args-inreg-bfloat.ll new file mode 100644 index 0000000..04e4724 --- /dev/null +++ b/llvm/test/CodeGen/AMDGPU/call-args-inreg-bfloat.ll @@ -0,0 +1,130 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4 +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefix=GFX9 %s +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 < %s | FileCheck -enable-var-scope -check-prefix=GFX11 %s + +; We've separated this file from call-args-inreg.ll since GlobalISel does not support the bfloat type. +; Ideally, we should merge the two files once that support lands. + +declare hidden void @external_void_func_bf16_inreg(bfloat inreg) #0 +declare hidden void @external_void_func_v2bf16_inreg(<2 x bfloat> inreg) #0 + +define void @test_call_external_void_func_bf16_inreg(bfloat inreg %arg) #0 { +; GFX9-LABEL: test_call_external_void_func_bf16_inreg: +; GFX9: ; %bb.0: +; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-NEXT: s_mov_b32 s17, s33 +; GFX9-NEXT: s_mov_b32 s33, s32 +; GFX9-NEXT: s_or_saveexec_b64 s[18:19], -1 +; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill +; GFX9-NEXT: s_mov_b64 exec, s[18:19] +; GFX9-NEXT: v_writelane_b32 v40, s17, 2 +; GFX9-NEXT: s_addk_i32 s32, 0x400 +; GFX9-NEXT: v_writelane_b32 v40, s30, 0 +; GFX9-NEXT: s_getpc_b64 s[18:19] +; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_bf16_inreg@rel32@lo+4 +; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_bf16_inreg@rel32@hi+12 +; GFX9-NEXT: s_mov_b32 s0, s16 +; GFX9-NEXT: v_writelane_b32 v40, s31, 1 +; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GFX9-NEXT: v_readlane_b32 s31, v40, 1 +; GFX9-NEXT: v_readlane_b32 s30, v40, 0 +; GFX9-NEXT: s_mov_b32 s32, s33 +; GFX9-NEXT: v_readlane_b32 s4, v40, 2 +; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 +; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload +; GFX9-NEXT: s_mov_b64 exec, s[6:7] +; GFX9-NEXT: s_mov_b32 s33, s4 +; GFX9-NEXT: s_waitcnt vmcnt(0) +; GFX9-NEXT: s_setpc_b64 s[30:31] +; +; GFX11-LABEL: test_call_external_void_func_bf16_inreg: +; GFX11: ; %bb.0: +; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX11-NEXT: s_mov_b32 s1, s33 +; GFX11-NEXT: s_mov_b32 s33, s32 +; GFX11-NEXT: s_or_saveexec_b32 s2, -1 +; GFX11-NEXT: scratch_store_b32 off, v40, s33 ; 4-byte Folded Spill +; GFX11-NEXT: s_mov_b32 exec_lo, s2 +; GFX11-NEXT: v_writelane_b32 v40, s1, 2 +; GFX11-NEXT: s_add_i32 s32, s32, 16 +; GFX11-NEXT: s_getpc_b64 s[2:3] +; GFX11-NEXT: s_add_u32 s2, s2, external_void_func_bf16_inreg@rel32@lo+4 +; GFX11-NEXT: s_addc_u32 s3, s3, external_void_func_bf16_inreg@rel32@hi+12 +; GFX11-NEXT: v_writelane_b32 v40, s30, 0 +; GFX11-NEXT: v_writelane_b32 v40, s31, 1 +; GFX11-NEXT: s_swappc_b64 s[30:31], s[2:3] +; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) +; GFX11-NEXT: v_readlane_b32 s31, v40, 1 +; GFX11-NEXT: v_readlane_b32 s30, v40, 0 +; GFX11-NEXT: s_mov_b32 s32, s33 +; GFX11-NEXT: v_readlane_b32 s0, v40, 2 +; GFX11-NEXT: s_or_saveexec_b32 s1, -1 +; GFX11-NEXT: scratch_load_b32 v40, off, s33 ; 4-byte Folded Reload +; GFX11-NEXT: s_mov_b32 exec_lo, s1 +; GFX11-NEXT: s_mov_b32 s33, s0 +; GFX11-NEXT: s_waitcnt vmcnt(0) +; GFX11-NEXT: s_setpc_b64 s[30:31] + call void @external_void_func_bf16_inreg(bfloat inreg %arg) + ret void +} + +define void @test_call_external_void_func_v2bf16_inreg(<2 x bfloat> inreg %arg) #0 { +; GFX9-LABEL: test_call_external_void_func_v2bf16_inreg: +; GFX9: ; %bb.0: +; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-NEXT: s_mov_b32 s17, s33 +; GFX9-NEXT: s_mov_b32 s33, s32 +; GFX9-NEXT: s_or_saveexec_b64 s[18:19], -1 +; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill +; GFX9-NEXT: s_mov_b64 exec, s[18:19] +; GFX9-NEXT: v_writelane_b32 v40, s17, 2 +; GFX9-NEXT: s_addk_i32 s32, 0x400 +; GFX9-NEXT: v_writelane_b32 v40, s30, 0 +; GFX9-NEXT: s_getpc_b64 s[18:19] +; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_v2bf16_inreg@rel32@lo+4 +; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_v2bf16_inreg@rel32@hi+12 +; GFX9-NEXT: s_mov_b32 s0, s16 +; GFX9-NEXT: v_writelane_b32 v40, s31, 1 +; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GFX9-NEXT: v_readlane_b32 s31, v40, 1 +; GFX9-NEXT: v_readlane_b32 s30, v40, 0 +; GFX9-NEXT: s_mov_b32 s32, s33 +; GFX9-NEXT: v_readlane_b32 s4, v40, 2 +; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 +; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload +; GFX9-NEXT: s_mov_b64 exec, s[6:7] +; GFX9-NEXT: s_mov_b32 s33, s4 +; GFX9-NEXT: s_waitcnt vmcnt(0) +; GFX9-NEXT: s_setpc_b64 s[30:31] +; +; GFX11-LABEL: test_call_external_void_func_v2bf16_inreg: +; GFX11: ; %bb.0: +; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX11-NEXT: s_mov_b32 s1, s33 +; GFX11-NEXT: s_mov_b32 s33, s32 +; GFX11-NEXT: s_or_saveexec_b32 s2, -1 +; GFX11-NEXT: scratch_store_b32 off, v40, s33 ; 4-byte Folded Spill +; GFX11-NEXT: s_mov_b32 exec_lo, s2 +; GFX11-NEXT: v_writelane_b32 v40, s1, 2 +; GFX11-NEXT: s_add_i32 s32, s32, 16 +; GFX11-NEXT: s_getpc_b64 s[2:3] +; GFX11-NEXT: s_add_u32 s2, s2, external_void_func_v2bf16_inreg@rel32@lo+4 +; GFX11-NEXT: s_addc_u32 s3, s3, external_void_func_v2bf16_inreg@rel32@hi+12 +; GFX11-NEXT: v_writelane_b32 v40, s30, 0 +; GFX11-NEXT: v_writelane_b32 v40, s31, 1 +; GFX11-NEXT: s_swappc_b64 s[30:31], s[2:3] +; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) +; GFX11-NEXT: v_readlane_b32 s31, v40, 1 +; GFX11-NEXT: v_readlane_b32 s30, v40, 0 +; GFX11-NEXT: s_mov_b32 s32, s33 +; GFX11-NEXT: v_readlane_b32 s0, v40, 2 +; GFX11-NEXT: s_or_saveexec_b32 s1, -1 +; GFX11-NEXT: scratch_load_b32 v40, off, s33 ; 4-byte Folded Reload +; GFX11-NEXT: s_mov_b32 exec_lo, s1 +; GFX11-NEXT: s_mov_b32 s33, s0 +; GFX11-NEXT: s_waitcnt vmcnt(0) +; GFX11-NEXT: s_setpc_b64 s[30:31] + call void @external_void_func_v2bf16_inreg(<2 x bfloat> inreg %arg) + ret void +} + diff --git a/llvm/test/CodeGen/AMDGPU/call-args-inreg.ll b/llvm/test/CodeGen/AMDGPU/call-args-inreg.ll index d1cede6..f96007a 100644 --- a/llvm/test/CodeGen/AMDGPU/call-args-inreg.ll +++ b/llvm/test/CodeGen/AMDGPU/call-args-inreg.ll @@ -1,6 +1,8 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4 -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefix=GFX9 %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 < %s | FileCheck -enable-var-scope -check-prefix=GFX11 %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefixes=GFX9,SDAG %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 < %s | FileCheck -enable-var-scope -check-prefixes=GFX11,SDAG %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefixes=GFX9,GISEL %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 < %s | FileCheck -enable-var-scope -check-prefixes=GFX11,GISEL %s declare hidden void @external_void_func_i8_inreg(i8 inreg) #0 declare hidden void @external_void_func_i16_inreg(i32 inreg) #0 @@ -12,11 +14,9 @@ declare hidden void @external_void_func_v4i32_inreg(<4 x i32> inreg) #0 declare hidden void @external_void_func_v8i32_inreg(<8 x i32> inreg) #0 declare hidden void @external_void_func_v16i32_inreg(<16 x i32> inreg) #0 declare hidden void @external_void_func_f16_inreg(half inreg) #0 -declare hidden void @external_void_func_bf16_inreg(bfloat inreg) #0 declare hidden void @external_void_func_f32_inreg(float inreg) #0 declare hidden void @external_void_func_f64_inreg(double inreg) #0 declare hidden void @external_void_func_v2f16_inreg(<2 x half> inreg) #0 -declare hidden void @external_void_func_v2bf16_inreg(<2 x bfloat> inreg) #0 declare hidden void @external_void_func_v3f16_inreg(<3 x half> inreg) #0 declare hidden void @external_void_func_v4f16_inreg(<4 x half> inreg) #0 @@ -212,35 +212,6 @@ define void @test_call_external_void_func_i32_inreg(i32 inreg %arg) #0 { } define void @test_call_external_void_func_i64_inreg(i64 inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_i64_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_i64_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_i64_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_i64_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -273,35 +244,6 @@ define void @test_call_external_void_func_i64_inreg(i64 inreg %arg) #0 { } define void @test_call_external_void_func_v2i32_inreg(<2 x i32> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v2i32_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_v2i32_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_v2i32_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v2i32_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -334,36 +276,6 @@ define void @test_call_external_void_func_v2i32_inreg(<2 x i32> inreg %arg) #0 { } define void @test_call_external_void_func_v3i32_inreg(<3 x i32> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v3i32_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s19, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s19, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[20:21] -; GFX9-NEXT: s_add_u32 s20, s20, external_void_func_v3i32_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s21, s21, external_void_func_v3i32_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s2, s18 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[20:21] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v3i32_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -396,37 +308,6 @@ define void @test_call_external_void_func_v3i32_inreg(<3 x i32> inreg %arg) #0 { } define void @test_call_external_void_func_v4i32_inreg(<4 x i32> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v4i32_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s20, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[22:23], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[22:23] -; GFX9-NEXT: v_writelane_b32 v40, s20, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[20:21] -; GFX9-NEXT: s_add_u32 s20, s20, external_void_func_v4i32_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s21, s21, external_void_func_v4i32_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s3, s19 -; GFX9-NEXT: s_mov_b32 s2, s18 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[20:21] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v4i32_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -459,41 +340,6 @@ define void @test_call_external_void_func_v4i32_inreg(<4 x i32> inreg %arg) #0 { } define void @test_call_external_void_func_v8i32_inreg(<8 x i32> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v8i32_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s24, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[26:27], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[26:27] -; GFX9-NEXT: v_writelane_b32 v40, s24, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[24:25] -; GFX9-NEXT: s_add_u32 s24, s24, external_void_func_v8i32_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s25, s25, external_void_func_v8i32_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s3, s19 -; GFX9-NEXT: s_mov_b32 s2, s18 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: s_mov_b32 s16, s20 -; GFX9-NEXT: s_mov_b32 s17, s21 -; GFX9-NEXT: s_mov_b32 s18, s22 -; GFX9-NEXT: s_mov_b32 s19, s23 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[24:25] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v8i32_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -585,66 +431,6 @@ define void @test_call_external_void_func_f16_inreg(half inreg %arg) #0 { ret void } -define void @test_call_external_void_func_bf16_inreg(bfloat inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_bf16_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s17, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[18:19], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[18:19] -; GFX9-NEXT: v_writelane_b32 v40, s17, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_bf16_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_bf16_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; -; GFX11-LABEL: test_call_external_void_func_bf16_inreg: -; GFX11: ; %bb.0: -; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX11-NEXT: s_mov_b32 s1, s33 -; GFX11-NEXT: s_mov_b32 s33, s32 -; GFX11-NEXT: s_or_saveexec_b32 s2, -1 -; GFX11-NEXT: scratch_store_b32 off, v40, s33 ; 4-byte Folded Spill -; GFX11-NEXT: s_mov_b32 exec_lo, s2 -; GFX11-NEXT: v_writelane_b32 v40, s1, 2 -; GFX11-NEXT: s_add_i32 s32, s32, 16 -; GFX11-NEXT: s_getpc_b64 s[2:3] -; GFX11-NEXT: s_add_u32 s2, s2, external_void_func_bf16_inreg@rel32@lo+4 -; GFX11-NEXT: s_addc_u32 s3, s3, external_void_func_bf16_inreg@rel32@hi+12 -; GFX11-NEXT: v_writelane_b32 v40, s30, 0 -; GFX11-NEXT: v_writelane_b32 v40, s31, 1 -; GFX11-NEXT: s_swappc_b64 s[30:31], s[2:3] -; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) -; GFX11-NEXT: v_readlane_b32 s31, v40, 1 -; GFX11-NEXT: v_readlane_b32 s30, v40, 0 -; GFX11-NEXT: s_mov_b32 s32, s33 -; GFX11-NEXT: v_readlane_b32 s0, v40, 2 -; GFX11-NEXT: s_or_saveexec_b32 s1, -1 -; GFX11-NEXT: scratch_load_b32 v40, off, s33 ; 4-byte Folded Reload -; GFX11-NEXT: s_mov_b32 exec_lo, s1 -; GFX11-NEXT: s_mov_b32 s33, s0 -; GFX11-NEXT: s_waitcnt vmcnt(0) -; GFX11-NEXT: s_setpc_b64 s[30:31] - call void @external_void_func_bf16_inreg(bfloat inreg %arg) - ret void -} - define void @test_call_external_void_func_f32_inreg(float inreg %arg) #0 { ; GFX9-LABEL: test_call_external_void_func_f32_inreg: ; GFX9: ; %bb.0: @@ -706,35 +492,6 @@ define void @test_call_external_void_func_f32_inreg(float inreg %arg) #0 { } define void @test_call_external_void_func_f64_inreg(double inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_f64_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_f64_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_f64_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_f64_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -826,97 +583,7 @@ define void @test_call_external_void_func_v2f16_inreg(<2 x half> inreg %arg) #0 ret void } - -define void @test_call_external_void_func_v2bf16_inreg(<2 x bfloat> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v2bf16_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s17, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[18:19], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[18:19] -; GFX9-NEXT: v_writelane_b32 v40, s17, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_v2bf16_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_v2bf16_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; -; GFX11-LABEL: test_call_external_void_func_v2bf16_inreg: -; GFX11: ; %bb.0: -; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX11-NEXT: s_mov_b32 s1, s33 -; GFX11-NEXT: s_mov_b32 s33, s32 -; GFX11-NEXT: s_or_saveexec_b32 s2, -1 -; GFX11-NEXT: scratch_store_b32 off, v40, s33 ; 4-byte Folded Spill -; GFX11-NEXT: s_mov_b32 exec_lo, s2 -; GFX11-NEXT: v_writelane_b32 v40, s1, 2 -; GFX11-NEXT: s_add_i32 s32, s32, 16 -; GFX11-NEXT: s_getpc_b64 s[2:3] -; GFX11-NEXT: s_add_u32 s2, s2, external_void_func_v2bf16_inreg@rel32@lo+4 -; GFX11-NEXT: s_addc_u32 s3, s3, external_void_func_v2bf16_inreg@rel32@hi+12 -; GFX11-NEXT: v_writelane_b32 v40, s30, 0 -; GFX11-NEXT: v_writelane_b32 v40, s31, 1 -; GFX11-NEXT: s_swappc_b64 s[30:31], s[2:3] -; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) -; GFX11-NEXT: v_readlane_b32 s31, v40, 1 -; GFX11-NEXT: v_readlane_b32 s30, v40, 0 -; GFX11-NEXT: s_mov_b32 s32, s33 -; GFX11-NEXT: v_readlane_b32 s0, v40, 2 -; GFX11-NEXT: s_or_saveexec_b32 s1, -1 -; GFX11-NEXT: scratch_load_b32 v40, off, s33 ; 4-byte Folded Reload -; GFX11-NEXT: s_mov_b32 exec_lo, s1 -; GFX11-NEXT: s_mov_b32 s33, s0 -; GFX11-NEXT: s_waitcnt vmcnt(0) -; GFX11-NEXT: s_setpc_b64 s[30:31] - call void @external_void_func_v2bf16_inreg(<2 x bfloat> inreg %arg) - ret void -} - define void @test_call_external_void_func_v3f16_inreg(<3 x half> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v3f16_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_v3f16_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_v3f16_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v3f16_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -949,35 +616,6 @@ define void @test_call_external_void_func_v3f16_inreg(<3 x half> inreg %arg) #0 } define void @test_call_external_void_func_v4f16_inreg(<4 x half> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v4f16_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_v4f16_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_v4f16_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v4f16_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1010,35 +648,6 @@ define void @test_call_external_void_func_v4f16_inreg(<4 x half> inreg %arg) #0 } define void @test_call_external_void_func_p0_inreg(ptr inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_p0_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_p0_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_p0_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_p0_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1071,35 +680,6 @@ define void @test_call_external_void_func_p0_inreg(ptr inreg %arg) #0 { } define void @test_call_external_void_func_p1_inreg(ptr addrspace(1) inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_p1_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_p1_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_p1_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_p1_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1192,37 +772,6 @@ define void @test_call_external_void_func_p3_inreg(ptr addrspace(3) inreg %arg) } define void @test_call_external_void_func_v2p1_inreg(<2 x ptr addrspace(1)> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v2p1_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s20, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[22:23], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[22:23] -; GFX9-NEXT: v_writelane_b32 v40, s20, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[20:21] -; GFX9-NEXT: s_add_u32 s20, s20, external_void_func_v2p1_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s21, s21, external_void_func_v2p1_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s3, s19 -; GFX9-NEXT: s_mov_b32 s2, s18 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[20:21] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v2p1_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1255,35 +804,6 @@ define void @test_call_external_void_func_v2p1_inreg(<2 x ptr addrspace(1)> inre } define void @test_call_external_void_func_v2p5_inreg(<2 x ptr addrspace(5)> inreg %arg) #0 { -; GFX9-LABEL: test_call_external_void_func_v2p5_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s18, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[20:21], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[20:21] -; GFX9-NEXT: v_writelane_b32 v40, s18, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[18:19] -; GFX9-NEXT: s_add_u32 s18, s18, external_void_func_v2p5_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s19, s19, external_void_func_v2p5_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[18:19] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_v2p5_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1316,38 +836,6 @@ define void @test_call_external_void_func_v2p5_inreg(<2 x ptr addrspace(5)> inre } define void @test_call_external_void_func_i64_inreg_i32_inreg_i64_inreg(i64 inreg %arg0, i32 inreg %arg1, i64 inreg %arg2) #0 { -; GFX9-LABEL: test_call_external_void_func_i64_inreg_i32_inreg_i64_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s21, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[22:23], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[22:23] -; GFX9-NEXT: v_writelane_b32 v40, s21, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[22:23] -; GFX9-NEXT: s_add_u32 s22, s22, external_void_func_i64_inreg_i32_inreg_i64_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s23, s23, external_void_func_i64_inreg_i32_inreg_i64_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s3, s19 -; GFX9-NEXT: s_mov_b32 s2, s18 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: s_mov_b32 s16, s20 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[22:23] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_i64_inreg_i32_inreg_i64_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1380,46 +868,6 @@ define void @test_call_external_void_func_i64_inreg_i32_inreg_i64_inreg(i64 inre } define void @test_call_external_void_func_a15i32_inreg([13 x i32] inreg %arg0) #0 { -; GFX9-LABEL: test_call_external_void_func_a15i32_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s29, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[40:41], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[40:41] -; GFX9-NEXT: v_writelane_b32 v40, s29, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[40:41] -; GFX9-NEXT: s_add_u32 s40, s40, external_void_func_a15i32_inreg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s41, s41, external_void_func_a15i32_inreg@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s3, s19 -; GFX9-NEXT: s_mov_b32 s2, s18 -; GFX9-NEXT: s_mov_b32 s1, s17 -; GFX9-NEXT: s_mov_b32 s0, s16 -; GFX9-NEXT: s_mov_b32 s16, s20 -; GFX9-NEXT: s_mov_b32 s17, s21 -; GFX9-NEXT: s_mov_b32 s18, s22 -; GFX9-NEXT: s_mov_b32 s19, s23 -; GFX9-NEXT: s_mov_b32 s20, s24 -; GFX9-NEXT: s_mov_b32 s21, s25 -; GFX9-NEXT: s_mov_b32 s22, s26 -; GFX9-NEXT: s_mov_b32 s23, s27 -; GFX9-NEXT: s_mov_b32 s24, s28 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[40:41] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_a15i32_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1454,47 +902,6 @@ define void @test_call_external_void_func_a15i32_inreg([13 x i32] inreg %arg0) # ; FIXME: This should also fail define void @test_call_external_void_func_a15i32_inreg_i32_inreg([13 x i32] inreg %arg0, i32 inreg %arg1) #1 { -; GFX9-LABEL: test_call_external_void_func_a15i32_inreg_i32_inreg: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: s_mov_b32 s21, s33 -; GFX9-NEXT: s_mov_b32 s33, s32 -; GFX9-NEXT: s_or_saveexec_b64 s[22:23], -1 -; GFX9-NEXT: buffer_store_dword v40, off, s[0:3], s33 ; 4-byte Folded Spill -; GFX9-NEXT: s_mov_b64 exec, s[22:23] -; GFX9-NEXT: v_writelane_b32 v40, s21, 2 -; GFX9-NEXT: s_addk_i32 s32, 0x400 -; GFX9-NEXT: v_writelane_b32 v40, s30, 0 -; GFX9-NEXT: s_getpc_b64 s[22:23] -; GFX9-NEXT: s_add_u32 s22, s22, external_void_func_a15i32_inreg_i32_inreg__noimplicit@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s23, s23, external_void_func_a15i32_inreg_i32_inreg__noimplicit@rel32@hi+12 -; GFX9-NEXT: s_mov_b32 s3, s7 -; GFX9-NEXT: s_mov_b32 s2, s6 -; GFX9-NEXT: s_mov_b32 s1, s5 -; GFX9-NEXT: s_mov_b32 s0, s4 -; GFX9-NEXT: s_mov_b32 s4, s8 -; GFX9-NEXT: s_mov_b32 s5, s9 -; GFX9-NEXT: s_mov_b32 s6, s10 -; GFX9-NEXT: s_mov_b32 s7, s11 -; GFX9-NEXT: s_mov_b32 s8, s15 -; GFX9-NEXT: s_mov_b32 s9, s16 -; GFX9-NEXT: s_mov_b32 s10, s17 -; GFX9-NEXT: s_mov_b32 s11, s18 -; GFX9-NEXT: s_mov_b32 s15, s19 -; GFX9-NEXT: s_mov_b32 s16, s20 -; GFX9-NEXT: v_writelane_b32 v40, s31, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[22:23] -; GFX9-NEXT: v_readlane_b32 s31, v40, 1 -; GFX9-NEXT: v_readlane_b32 s30, v40, 0 -; GFX9-NEXT: s_mov_b32 s32, s33 -; GFX9-NEXT: v_readlane_b32 s4, v40, 2 -; GFX9-NEXT: s_or_saveexec_b64 s[6:7], -1 -; GFX9-NEXT: buffer_load_dword v40, off, s[0:3], s33 ; 4-byte Folded Reload -; GFX9-NEXT: s_mov_b64 exec, s[6:7] -; GFX9-NEXT: s_mov_b32 s33, s4 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_setpc_b64 s[30:31] -; ; GFX11-LABEL: test_call_external_void_func_a15i32_inreg_i32_inreg: ; GFX11: ; %bb.0: ; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -1529,3 +936,6 @@ define void @test_call_external_void_func_a15i32_inreg_i32_inreg([13 x i32] inre attributes #0 = { nounwind } attributes #1 = { nounwind "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-queue-ptr" "amdgpu-no-work-group-id-x" "amdgpu-no-work-group-id-y" "amdgpu-no-work-group-id-z" "amdgpu-no-work-item-id-x" "amdgpu-no-work-item-id-y" "amdgpu-no-work-item-id-z" } +;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line: +; GISEL: {{.*}} +; SDAG: {{.*}} diff --git a/llvm/test/CodeGen/AMDGPU/call-argument-types.ll b/llvm/test/CodeGen/AMDGPU/call-argument-types.ll index 832e43f..c407f76 100644 --- a/llvm/test/CodeGen/AMDGPU/call-argument-types.ll +++ b/llvm/test/CodeGen/AMDGPU/call-argument-types.ll @@ -1,10 +1,11 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2 ; RUN: llc -mtriple=amdgcn -mcpu=fiji -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=VI %s ; RUN: llc -mtriple=amdgcn -mcpu=hawaii -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=CI %s -; RUN: llc -mtriple=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=GFX9 %s +; RUN: llc -mtriple=amdgcn -global-isel=0 -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=GFX9,SDAG %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16 -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=GFX11,GFX11-TRUE16 %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -mattr=-real-true16 -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=GFX11,GFX11-FAKE16 %s ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=HSA %s +; RUN: llc -mtriple=amdgcn -global-isel=1 -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 < %s | FileCheck -enable-var-scope -check-prefixes=GFX9,GISEL %s declare hidden void @external_void_func_i1(i1) #0 declare hidden void @external_void_func_i1_signext(i1 signext) #0 @@ -100,24 +101,24 @@ define amdgpu_kernel void @test_call_external_void_func_i1_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i1_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i1@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i1@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i1_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i1@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i1@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i1_imm: ; GFX11: ; %bb.0: @@ -145,6 +146,25 @@ define amdgpu_kernel void @test_call_external_void_func_i1_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i1_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i1@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i1@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_i1(i1 true) ret void } @@ -196,28 +216,28 @@ define amdgpu_kernel void @test_call_external_void_func_i1_signext(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i1_signext: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i1_signext@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i1_signext@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: v_bfe_i32 v0, v0, 0, 1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i1_signext: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i1_signext@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i1_signext@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: v_bfe_i32 v0, v0, 0, 1 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i1_signext: ; GFX11: ; %bb.0: @@ -253,6 +273,29 @@ define amdgpu_kernel void @test_call_external_void_func_i1_signext(i32) #0 { ; HSA-NEXT: v_bfe_i32 v0, v0, 0, 1 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i1_signext: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i1_signext@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i1_signext@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: v_bfe_i32 v0, v0, 0, 1 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %var = load volatile i1, ptr addrspace(1) poison call void @external_void_func_i1_signext(i1 signext %var) ret void @@ -306,28 +349,28 @@ define amdgpu_kernel void @test_call_external_void_func_i1_zeroext(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i1_zeroext: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i1_zeroext@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i1_zeroext@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: v_and_b32_e32 v0, 1, v0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i1_zeroext: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i1_zeroext@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i1_zeroext@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: v_and_b32_e32 v0, 1, v0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i1_zeroext: ; GFX11: ; %bb.0: @@ -363,6 +406,29 @@ define amdgpu_kernel void @test_call_external_void_func_i1_zeroext(i32) #0 { ; HSA-NEXT: v_and_b32_e32 v0, 1, v0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i1_zeroext: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i1_zeroext@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i1_zeroext@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: v_and_b32_e32 v0, 1, v0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %var = load volatile i1, ptr addrspace(1) poison call void @external_void_func_i1_zeroext(i1 zeroext %var) ret void @@ -407,24 +473,24 @@ define amdgpu_kernel void @test_call_external_void_func_i8_imm(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i8_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i8@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i8@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x7b -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i8_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i8@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i8@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x7b +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-TRUE16-LABEL: test_call_external_void_func_i8_imm: ; GFX11-TRUE16: ; %bb.0: @@ -463,6 +529,25 @@ define amdgpu_kernel void @test_call_external_void_func_i8_imm(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i8_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i8@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i8@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x7b +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_i8(i8 123) ret void } @@ -513,27 +598,27 @@ define amdgpu_kernel void @test_call_external_void_func_i8_signext(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i8_signext: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_sbyte v0, off, s[0:3], 0 glc -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i8_signext@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i8_signext@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i8_signext: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_sbyte v0, off, s[0:3], 0 glc +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i8_signext@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i8_signext@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i8_signext: ; GFX11: ; %bb.0: @@ -567,6 +652,28 @@ define amdgpu_kernel void @test_call_external_void_func_i8_signext(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i8_signext: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: buffer_load_sbyte v0, off, s[0:3], 0 glc +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i8_signext@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i8_signext@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %var = load volatile i8, ptr addrspace(1) poison call void @external_void_func_i8_signext(i8 signext %var) ret void @@ -617,27 +724,27 @@ define amdgpu_kernel void @test_call_external_void_func_i8_zeroext(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i8_zeroext: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i8_zeroext@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i8_zeroext@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i8_zeroext: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i8_zeroext@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i8_zeroext@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i8_zeroext: ; GFX11: ; %bb.0: @@ -671,6 +778,28 @@ define amdgpu_kernel void @test_call_external_void_func_i8_zeroext(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i8_zeroext: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 glc +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i8_zeroext@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i8_zeroext@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %var = load volatile i8, ptr addrspace(1) poison call void @external_void_func_i8_zeroext(i8 zeroext %var) ret void @@ -715,24 +844,24 @@ define amdgpu_kernel void @test_call_external_void_func_i16_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i16_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x7b -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i16_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x7b +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-TRUE16-LABEL: test_call_external_void_func_i16_imm: ; GFX11-TRUE16: ; %bb.0: @@ -771,6 +900,25 @@ define amdgpu_kernel void @test_call_external_void_func_i16_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i16_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i16@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x7b +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_i16(i16 123) ret void } @@ -820,27 +968,27 @@ define amdgpu_kernel void @test_call_external_void_func_i16_signext(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i16_signext: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_sshort v0, off, s[0:3], 0 glc -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i16_signext@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i16_signext@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i16_signext: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_sshort v0, off, s[0:3], 0 glc +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i16_signext@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i16_signext@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i16_signext: ; GFX11: ; %bb.0: @@ -874,6 +1022,28 @@ define amdgpu_kernel void @test_call_external_void_func_i16_signext(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i16_signext: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: buffer_load_sshort v0, off, s[0:3], 0 glc +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i16_signext@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i16_signext@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %var = load volatile i16, ptr addrspace(1) poison call void @external_void_func_i16_signext(i16 signext %var) ret void @@ -924,27 +1094,27 @@ define amdgpu_kernel void @test_call_external_void_func_i16_zeroext(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i16_zeroext: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_ushort v0, off, s[0:3], 0 glc -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i16_zeroext@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i16_zeroext@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i16_zeroext: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_ushort v0, off, s[0:3], 0 glc +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i16_zeroext@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i16_zeroext@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i16_zeroext: ; GFX11: ; %bb.0: @@ -978,6 +1148,28 @@ define amdgpu_kernel void @test_call_external_void_func_i16_zeroext(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i16_zeroext: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: buffer_load_ushort v0, off, s[0:3], 0 glc +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i16_zeroext@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i16_zeroext@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %var = load volatile i16, ptr addrspace(1) poison call void @external_void_func_i16_zeroext(i16 zeroext %var) ret void @@ -1022,24 +1214,24 @@ define amdgpu_kernel void @test_call_external_void_func_i32_imm(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 42 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 42 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i32_imm: ; GFX11: ; %bb.0: @@ -1067,6 +1259,25 @@ define amdgpu_kernel void @test_call_external_void_func_i32_imm(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 42 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_i32(i32 42) ret void } @@ -1112,25 +1323,25 @@ define amdgpu_kernel void @test_call_external_void_func_i64_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_i64_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_i64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_i64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x7b -; GFX9-NEXT: v_mov_b32_e32 v1, 0 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_i64_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_i64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_i64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x7b +; SDAG-NEXT: v_mov_b32_e32 v1, 0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_i64_imm: ; GFX11: ; %bb.0: @@ -1159,6 +1370,26 @@ define amdgpu_kernel void @test_call_external_void_func_i64_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_i64_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_i64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_i64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x7b +; GISEL-NEXT: v_mov_b32_e32 v1, 0 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_i64(i64 123) ret void } @@ -1208,27 +1439,27 @@ define amdgpu_kernel void @test_call_external_void_func_v2i64() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2i64: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], 0 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2i64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2i64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2i64: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], 0 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2i64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2i64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2i64: ; GFX11: ; %bb.0: @@ -1262,6 +1493,31 @@ define amdgpu_kernel void @test_call_external_void_func_v2i64() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2i64: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], 0 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x0 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2i64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2i64@rel32@hi+12 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: v_mov_b32_e32 v2, s2 +; GISEL-NEXT: v_mov_b32_e32 v3, s3 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <2 x i64>, ptr addrspace(1) null call void @external_void_func_v2i64(<2 x i64> %val) ret void @@ -1312,27 +1568,27 @@ define amdgpu_kernel void @test_call_external_void_func_v2i64_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2i64_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2i64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2i64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1 -; GFX9-NEXT: v_mov_b32_e32 v1, 2 -; GFX9-NEXT: v_mov_b32_e32 v2, 3 -; GFX9-NEXT: v_mov_b32_e32 v3, 4 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2i64_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2i64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2i64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1 +; SDAG-NEXT: v_mov_b32_e32 v1, 2 +; SDAG-NEXT: v_mov_b32_e32 v2, 3 +; SDAG-NEXT: v_mov_b32_e32 v3, 4 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2i64_imm: ; GFX11: ; %bb.0: @@ -1364,6 +1620,28 @@ define amdgpu_kernel void @test_call_external_void_func_v2i64_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2i64_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2i64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2i64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1 +; GISEL-NEXT: v_mov_b32_e32 v1, 2 +; GISEL-NEXT: v_mov_b32_e32 v2, 3 +; GISEL-NEXT: v_mov_b32_e32 v3, 4 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v2i64(<2 x i64> <i64 8589934593, i64 17179869187>) ret void } @@ -1417,29 +1695,29 @@ define amdgpu_kernel void @test_call_external_void_func_v3i64() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3i64: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], 0 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3i64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3i64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v4, 1 -; GFX9-NEXT: v_mov_b32_e32 v5, 2 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3i64: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], 0 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3i64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3i64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v4, 1 +; SDAG-NEXT: v_mov_b32_e32 v5, 2 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3i64: ; GFX11: ; %bb.0: @@ -1476,6 +1754,33 @@ define amdgpu_kernel void @test_call_external_void_func_v3i64() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3i64: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], 0 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x0 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3i64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3i64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v4, 1 +; GISEL-NEXT: v_mov_b32_e32 v5, 2 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: v_mov_b32_e32 v2, s2 +; GISEL-NEXT: v_mov_b32_e32 v3, s3 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %load = load <2 x i64>, ptr addrspace(1) null %val = shufflevector <2 x i64> %load, <2 x i64> <i64 8589934593, i64 poison>, <3 x i32> <i32 0, i32 1, i32 2> @@ -1536,31 +1841,31 @@ define amdgpu_kernel void @test_call_external_void_func_v4i64() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v4i64: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], 0 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v4i64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v4i64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v4, 1 -; GFX9-NEXT: v_mov_b32_e32 v5, 2 -; GFX9-NEXT: v_mov_b32_e32 v6, 3 -; GFX9-NEXT: v_mov_b32_e32 v7, 4 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v4i64: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], 0 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v4i64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v4i64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v4, 1 +; SDAG-NEXT: v_mov_b32_e32 v5, 2 +; SDAG-NEXT: v_mov_b32_e32 v6, 3 +; SDAG-NEXT: v_mov_b32_e32 v7, 4 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v4i64: ; GFX11: ; %bb.0: @@ -1600,6 +1905,35 @@ define amdgpu_kernel void @test_call_external_void_func_v4i64() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v4i64: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], 0 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x0 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v4i64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v4i64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v4, 1 +; GISEL-NEXT: v_mov_b32_e32 v5, 2 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: v_mov_b32_e32 v2, s2 +; GISEL-NEXT: v_mov_b32_e32 v3, s3 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: v_mov_b32_e32 v6, 3 +; GISEL-NEXT: v_mov_b32_e32 v7, 4 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %load = load <2 x i64>, ptr addrspace(1) null %val = shufflevector <2 x i64> %load, <2 x i64> <i64 8589934593, i64 17179869187>, <4 x i32> <i32 0, i32 1, i32 2, i32 3> call void @external_void_func_v4i64(<4 x i64> %val) @@ -1645,24 +1979,24 @@ define amdgpu_kernel void @test_call_external_void_func_f16_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_f16_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_f16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_f16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x4400 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_f16_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_f16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_f16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x4400 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-TRUE16-LABEL: test_call_external_void_func_f16_imm: ; GFX11-TRUE16: ; %bb.0: @@ -1701,6 +2035,25 @@ define amdgpu_kernel void @test_call_external_void_func_f16_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_f16_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_f16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_f16@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x4400 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_f16(half 4.0) ret void } @@ -1744,24 +2097,24 @@ define amdgpu_kernel void @test_call_external_void_func_f32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_f32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_f32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_f32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 4.0 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_f32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_f32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_f32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 4.0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_f32_imm: ; GFX11: ; %bb.0: @@ -1789,6 +2142,25 @@ define amdgpu_kernel void @test_call_external_void_func_f32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_f32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_f32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_f32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 4.0 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_f32(float 4.0) ret void } @@ -1834,25 +2206,25 @@ define amdgpu_kernel void @test_call_external_void_func_v2f32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2f32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2f32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2f32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1.0 -; GFX9-NEXT: v_mov_b32_e32 v1, 2.0 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2f32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2f32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2f32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1.0 +; SDAG-NEXT: v_mov_b32_e32 v1, 2.0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2f32_imm: ; GFX11: ; %bb.0: @@ -1881,6 +2253,26 @@ define amdgpu_kernel void @test_call_external_void_func_v2f32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2f32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2f32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2f32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1.0 +; GISEL-NEXT: v_mov_b32_e32 v1, 2.0 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v2f32(<2 x float> <float 1.0, float 2.0>) ret void } @@ -1928,26 +2320,26 @@ define amdgpu_kernel void @test_call_external_void_func_v3f32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3f32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3f32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3f32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1.0 -; GFX9-NEXT: v_mov_b32_e32 v1, 2.0 -; GFX9-NEXT: v_mov_b32_e32 v2, 4.0 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3f32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3f32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3f32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1.0 +; SDAG-NEXT: v_mov_b32_e32 v1, 2.0 +; SDAG-NEXT: v_mov_b32_e32 v2, 4.0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3f32_imm: ; GFX11: ; %bb.0: @@ -1978,6 +2370,27 @@ define amdgpu_kernel void @test_call_external_void_func_v3f32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3f32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3f32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3f32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1.0 +; GISEL-NEXT: v_mov_b32_e32 v1, 2.0 +; GISEL-NEXT: v_mov_b32_e32 v2, 4.0 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v3f32(<3 x float> <float 1.0, float 2.0, float 4.0>) ret void } @@ -2029,28 +2442,28 @@ define amdgpu_kernel void @test_call_external_void_func_v5f32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v5f32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v5f32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v5f32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1.0 -; GFX9-NEXT: v_mov_b32_e32 v1, 2.0 -; GFX9-NEXT: v_mov_b32_e32 v2, 4.0 -; GFX9-NEXT: v_mov_b32_e32 v3, -1.0 -; GFX9-NEXT: v_mov_b32_e32 v4, 0.5 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v5f32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v5f32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v5f32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1.0 +; SDAG-NEXT: v_mov_b32_e32 v1, 2.0 +; SDAG-NEXT: v_mov_b32_e32 v2, 4.0 +; SDAG-NEXT: v_mov_b32_e32 v3, -1.0 +; SDAG-NEXT: v_mov_b32_e32 v4, 0.5 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v5f32_imm: ; GFX11: ; %bb.0: @@ -2084,6 +2497,29 @@ define amdgpu_kernel void @test_call_external_void_func_v5f32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v5f32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v5f32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v5f32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1.0 +; GISEL-NEXT: v_mov_b32_e32 v1, 2.0 +; GISEL-NEXT: v_mov_b32_e32 v2, 4.0 +; GISEL-NEXT: v_mov_b32_e32 v3, -1.0 +; GISEL-NEXT: v_mov_b32_e32 v4, 0.5 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v5f32(<5 x float> <float 1.0, float 2.0, float 4.0, float -1.0, float 0.5>) ret void } @@ -2129,25 +2565,25 @@ define amdgpu_kernel void @test_call_external_void_func_f64_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_f64_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_f64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_f64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0 -; GFX9-NEXT: v_mov_b32_e32 v1, 0x40100000 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_f64_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_f64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_f64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: v_mov_b32_e32 v1, 0x40100000 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_f64_imm: ; GFX11: ; %bb.0: @@ -2176,6 +2612,26 @@ define amdgpu_kernel void @test_call_external_void_func_f64_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_f64_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_f64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_f64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: v_mov_b32_e32 v1, 0x40100000 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_f64(double 4.0) ret void } @@ -2225,27 +2681,27 @@ define amdgpu_kernel void @test_call_external_void_func_v2f64_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2f64_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2f64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2f64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0 -; GFX9-NEXT: v_mov_b32_e32 v1, 2.0 -; GFX9-NEXT: v_mov_b32_e32 v2, 0 -; GFX9-NEXT: v_mov_b32_e32 v3, 0x40100000 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2f64_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2f64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2f64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: v_mov_b32_e32 v1, 2.0 +; SDAG-NEXT: v_mov_b32_e32 v2, 0 +; SDAG-NEXT: v_mov_b32_e32 v3, 0x40100000 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2f64_imm: ; GFX11: ; %bb.0: @@ -2277,6 +2733,28 @@ define amdgpu_kernel void @test_call_external_void_func_v2f64_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2f64_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2f64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2f64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: v_mov_b32_e32 v1, 2.0 +; GISEL-NEXT: v_mov_b32_e32 v2, 0 +; GISEL-NEXT: v_mov_b32_e32 v3, 0x40100000 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v2f64(<2 x double> <double 2.0, double 4.0>) ret void } @@ -2330,29 +2808,29 @@ define amdgpu_kernel void @test_call_external_void_func_v3f64_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3f64_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3f64@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3f64@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0 -; GFX9-NEXT: v_mov_b32_e32 v1, 2.0 -; GFX9-NEXT: v_mov_b32_e32 v2, 0 -; GFX9-NEXT: v_mov_b32_e32 v3, 0x40100000 -; GFX9-NEXT: v_mov_b32_e32 v4, 0 -; GFX9-NEXT: v_mov_b32_e32 v5, 0x40200000 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3f64_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3f64@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3f64@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: v_mov_b32_e32 v1, 2.0 +; SDAG-NEXT: v_mov_b32_e32 v2, 0 +; SDAG-NEXT: v_mov_b32_e32 v3, 0x40100000 +; SDAG-NEXT: v_mov_b32_e32 v4, 0 +; SDAG-NEXT: v_mov_b32_e32 v5, 0x40200000 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3f64_imm: ; GFX11: ; %bb.0: @@ -2387,6 +2865,30 @@ define amdgpu_kernel void @test_call_external_void_func_v3f64_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3f64_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3f64@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3f64@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: v_mov_b32_e32 v1, 2.0 +; GISEL-NEXT: v_mov_b32_e32 v2, 0 +; GISEL-NEXT: v_mov_b32_e32 v3, 0x40100000 +; GISEL-NEXT: v_mov_b32_e32 v4, 0 +; GISEL-NEXT: v_mov_b32_e32 v5, 0x40200000 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v3f64(<3 x double> <double 2.0, double 4.0, double 8.0>) ret void } @@ -2436,26 +2938,26 @@ define amdgpu_kernel void @test_call_external_void_func_v2i16() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2i16: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dword v0, off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2i16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2i16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2i16: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dword v0, off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2i16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2i16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2i16: ; GFX11: ; %bb.0: @@ -2487,6 +2989,27 @@ define amdgpu_kernel void @test_call_external_void_func_v2i16() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2i16: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_load_dword s8, s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2i16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2i16@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s8 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <2 x i16>, ptr addrspace(1) poison call void @external_void_func_v2i16(<2 x i16> %val) ret void @@ -2539,26 +3062,26 @@ define amdgpu_kernel void @test_call_external_void_func_v3i16() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3i16: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3i16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3i16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3i16: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3i16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3i16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3i16: ; GFX11: ; %bb.0: @@ -2590,6 +3113,28 @@ define amdgpu_kernel void @test_call_external_void_func_v3i16() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3i16: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3i16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3i16@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <3 x i16>, ptr addrspace(1) poison call void @external_void_func_v3i16(<3 x i16> %val) ret void @@ -2643,26 +3188,26 @@ define amdgpu_kernel void @test_call_external_void_func_v3f16() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3f16: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3f16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3f16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3f16: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3f16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3f16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3f16: ; GFX11: ; %bb.0: @@ -2694,6 +3239,28 @@ define amdgpu_kernel void @test_call_external_void_func_v3f16() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3f16: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3f16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3f16@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <3 x half>, ptr addrspace(1) poison call void @external_void_func_v3f16(<3 x half> %val) ret void @@ -2741,25 +3308,25 @@ define amdgpu_kernel void @test_call_external_void_func_v3i16_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3i16_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3i16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3i16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x20001 -; GFX9-NEXT: v_mov_b32_e32 v1, 3 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3i16_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3i16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3i16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x20001 +; SDAG-NEXT: v_mov_b32_e32 v1, 3 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3i16_imm: ; GFX11: ; %bb.0: @@ -2788,6 +3355,26 @@ define amdgpu_kernel void @test_call_external_void_func_v3i16_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3i16_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3i16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3i16@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x20001 +; GISEL-NEXT: v_mov_b32_e32 v1, 3 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v3i16(<3 x i16> <i16 1, i16 2, i16 3>) ret void } @@ -2834,25 +3421,25 @@ define amdgpu_kernel void @test_call_external_void_func_v3f16_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3f16_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3f16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3f16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x40003c00 -; GFX9-NEXT: v_mov_b32_e32 v1, 0x4400 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3f16_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3f16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3f16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x40003c00 +; SDAG-NEXT: v_mov_b32_e32 v1, 0x4400 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3f16_imm: ; GFX11: ; %bb.0: @@ -2882,6 +3469,26 @@ define amdgpu_kernel void @test_call_external_void_func_v3f16_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3f16_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3f16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3f16@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x40003c00 +; GISEL-NEXT: v_mov_b32_e32 v1, 0x4400 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v3f16(<3 x half> <half 1.0, half 2.0, half 4.0>) ret void } @@ -2934,26 +3541,26 @@ define amdgpu_kernel void @test_call_external_void_func_v4i16() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v4i16: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v4i16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v4i16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v4i16: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v4i16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v4i16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v4i16: ; GFX11: ; %bb.0: @@ -2985,6 +3592,28 @@ define amdgpu_kernel void @test_call_external_void_func_v4i16() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v4i16: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v4i16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v4i16@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <4 x i16>, ptr addrspace(1) poison call void @external_void_func_v4i16(<4 x i16> %val) ret void @@ -3033,25 +3662,25 @@ define amdgpu_kernel void @test_call_external_void_func_v4i16_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v4i16_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v4i16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v4i16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 0x20001 -; GFX9-NEXT: v_mov_b32_e32 v1, 0x40003 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v4i16_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v4i16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v4i16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 0x20001 +; SDAG-NEXT: v_mov_b32_e32 v1, 0x40003 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v4i16_imm: ; GFX11: ; %bb.0: @@ -3081,6 +3710,26 @@ define amdgpu_kernel void @test_call_external_void_func_v4i16_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v4i16_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v4i16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v4i16@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0x20001 +; GISEL-NEXT: v_mov_b32_e32 v1, 0x40003 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v4i16(<4 x i16> <i16 1, i16 2, i16 3, i16 4>) ret void } @@ -3132,26 +3781,26 @@ define amdgpu_kernel void @test_call_external_void_func_v2f16() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2f16: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dword v0, off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2f16@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2f16@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2f16: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dword v0, off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2f16@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2f16@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2f16: ; GFX11: ; %bb.0: @@ -3183,6 +3832,27 @@ define amdgpu_kernel void @test_call_external_void_func_v2f16() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2f16: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_load_dword s8, s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2f16@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2f16@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s8 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <2 x half>, ptr addrspace(1) poison call void @external_void_func_v2f16(<2 x half> %val) ret void @@ -3231,26 +3901,26 @@ define amdgpu_kernel void @test_call_external_void_func_v2i32() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx2 v[0:1], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2i32: ; GFX11: ; %bb.0: @@ -3282,6 +3952,28 @@ define amdgpu_kernel void @test_call_external_void_func_v2i32() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2i32@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <2 x i32>, ptr addrspace(1) poison call void @external_void_func_v2i32(<2 x i32> %val) ret void @@ -3328,25 +4020,25 @@ define amdgpu_kernel void @test_call_external_void_func_v2i32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v2i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v2i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v2i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1 -; GFX9-NEXT: v_mov_b32_e32 v1, 2 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v2i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v2i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v2i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1 +; SDAG-NEXT: v_mov_b32_e32 v1, 2 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v2i32_imm: ; GFX11: ; %bb.0: @@ -3375,6 +4067,26 @@ define amdgpu_kernel void @test_call_external_void_func_v2i32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v2i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v2i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v2i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1 +; GISEL-NEXT: v_mov_b32_e32 v1, 2 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v2i32(<2 x i32> <i32 1, i32 2>) ret void } @@ -3422,26 +4134,26 @@ define amdgpu_kernel void @test_call_external_void_func_v3i32_imm(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 3 -; GFX9-NEXT: v_mov_b32_e32 v1, 4 -; GFX9-NEXT: v_mov_b32_e32 v2, 5 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 3 +; SDAG-NEXT: v_mov_b32_e32 v1, 4 +; SDAG-NEXT: v_mov_b32_e32 v2, 5 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3i32_imm: ; GFX11: ; %bb.0: @@ -3472,6 +4184,27 @@ define amdgpu_kernel void @test_call_external_void_func_v3i32_imm(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 3 +; GISEL-NEXT: v_mov_b32_e32 v1, 4 +; GISEL-NEXT: v_mov_b32_e32 v2, 5 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v3i32(<3 x i32> <i32 3, i32 4, i32 5>) ret void } @@ -3521,27 +4254,27 @@ define amdgpu_kernel void @test_call_external_void_func_v3i32_i32(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v3i32_i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v3i32_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v3i32_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 3 -; GFX9-NEXT: v_mov_b32_e32 v1, 4 -; GFX9-NEXT: v_mov_b32_e32 v2, 5 -; GFX9-NEXT: v_mov_b32_e32 v3, 6 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v3i32_i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v3i32_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v3i32_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 3 +; SDAG-NEXT: v_mov_b32_e32 v1, 4 +; SDAG-NEXT: v_mov_b32_e32 v2, 5 +; SDAG-NEXT: v_mov_b32_e32 v3, 6 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v3i32_i32: ; GFX11: ; %bb.0: @@ -3573,6 +4306,28 @@ define amdgpu_kernel void @test_call_external_void_func_v3i32_i32(i32) #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v3i32_i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v3i32_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v3i32_i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 3 +; GISEL-NEXT: v_mov_b32_e32 v1, 4 +; GISEL-NEXT: v_mov_b32_e32 v2, 5 +; GISEL-NEXT: v_mov_b32_e32 v3, 6 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v3i32_i32(<3 x i32> <i32 3, i32 4, i32 5>, i32 6) ret void } @@ -3620,26 +4375,26 @@ define amdgpu_kernel void @test_call_external_void_func_v4i32() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v4i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v4i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v4i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v4i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v4i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v4i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v4i32: ; GFX11: ; %bb.0: @@ -3671,6 +4426,30 @@ define amdgpu_kernel void @test_call_external_void_func_v4i32() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v4i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x0 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v4i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v4i32@rel32@hi+12 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v1, s1 +; GISEL-NEXT: v_mov_b32_e32 v2, s2 +; GISEL-NEXT: v_mov_b32_e32 v3, s3 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = load <4 x i32>, ptr addrspace(1) poison call void @external_void_func_v4i32(<4 x i32> %val) ret void @@ -3721,27 +4500,27 @@ define amdgpu_kernel void @test_call_external_void_func_v4i32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v4i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v4i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v4i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1 -; GFX9-NEXT: v_mov_b32_e32 v1, 2 -; GFX9-NEXT: v_mov_b32_e32 v2, 3 -; GFX9-NEXT: v_mov_b32_e32 v3, 4 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v4i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v4i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v4i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1 +; SDAG-NEXT: v_mov_b32_e32 v1, 2 +; SDAG-NEXT: v_mov_b32_e32 v2, 3 +; SDAG-NEXT: v_mov_b32_e32 v3, 4 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v4i32_imm: ; GFX11: ; %bb.0: @@ -3773,6 +4552,28 @@ define amdgpu_kernel void @test_call_external_void_func_v4i32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v4i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v4i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v4i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1 +; GISEL-NEXT: v_mov_b32_e32 v1, 2 +; GISEL-NEXT: v_mov_b32_e32 v2, 3 +; GISEL-NEXT: v_mov_b32_e32 v3, 4 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v4i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>) ret void } @@ -3824,28 +4625,28 @@ define amdgpu_kernel void @test_call_external_void_func_v5i32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v5i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v5i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v5i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1 -; GFX9-NEXT: v_mov_b32_e32 v1, 2 -; GFX9-NEXT: v_mov_b32_e32 v2, 3 -; GFX9-NEXT: v_mov_b32_e32 v3, 4 -; GFX9-NEXT: v_mov_b32_e32 v4, 5 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v5i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v5i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v5i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1 +; SDAG-NEXT: v_mov_b32_e32 v1, 2 +; SDAG-NEXT: v_mov_b32_e32 v2, 3 +; SDAG-NEXT: v_mov_b32_e32 v3, 4 +; SDAG-NEXT: v_mov_b32_e32 v4, 5 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v5i32_imm: ; GFX11: ; %bb.0: @@ -3879,6 +4680,29 @@ define amdgpu_kernel void @test_call_external_void_func_v5i32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v5i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v5i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v5i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1 +; GISEL-NEXT: v_mov_b32_e32 v1, 2 +; GISEL-NEXT: v_mov_b32_e32 v2, 3 +; GISEL-NEXT: v_mov_b32_e32 v3, 4 +; GISEL-NEXT: v_mov_b32_e32 v4, 5 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v5i32(<5 x i32> <i32 1, i32 2, i32 3, i32 4, i32 5>) ret void } @@ -3932,29 +4756,29 @@ define amdgpu_kernel void @test_call_external_void_func_v8i32() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v8i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: buffer_load_dwordx4 v[4:7], off, s[0:3], 0 offset:16 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v8i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v8i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v8i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: buffer_load_dwordx4 v[4:7], off, s[0:3], 0 offset:16 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v8i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v8i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v8i32: ; GFX11: ; %bb.0: @@ -3993,6 +4817,36 @@ define amdgpu_kernel void @test_call_external_void_func_v8i32() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v8i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_load_dwordx8 s[8:15], s[0:1], 0x0 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v8i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v8i32@rel32@hi+12 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s8 +; GISEL-NEXT: v_mov_b32_e32 v1, s9 +; GISEL-NEXT: v_mov_b32_e32 v2, s10 +; GISEL-NEXT: v_mov_b32_e32 v3, s11 +; GISEL-NEXT: v_mov_b32_e32 v4, s12 +; GISEL-NEXT: v_mov_b32_e32 v5, s13 +; GISEL-NEXT: v_mov_b32_e32 v6, s14 +; GISEL-NEXT: v_mov_b32_e32 v7, s15 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %ptr = load ptr addrspace(1), ptr addrspace(4) poison %val = load <8 x i32>, ptr addrspace(1) %ptr call void @external_void_func_v8i32(<8 x i32> %val) @@ -4052,31 +4906,31 @@ define amdgpu_kernel void @test_call_external_void_func_v8i32_imm() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v8i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v8i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v8i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: v_mov_b32_e32 v0, 1 -; GFX9-NEXT: v_mov_b32_e32 v1, 2 -; GFX9-NEXT: v_mov_b32_e32 v2, 3 -; GFX9-NEXT: v_mov_b32_e32 v3, 4 -; GFX9-NEXT: v_mov_b32_e32 v4, 5 -; GFX9-NEXT: v_mov_b32_e32 v5, 6 -; GFX9-NEXT: v_mov_b32_e32 v6, 7 -; GFX9-NEXT: v_mov_b32_e32 v7, 8 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v8i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v8i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v8i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: v_mov_b32_e32 v0, 1 +; SDAG-NEXT: v_mov_b32_e32 v1, 2 +; SDAG-NEXT: v_mov_b32_e32 v2, 3 +; SDAG-NEXT: v_mov_b32_e32 v3, 4 +; SDAG-NEXT: v_mov_b32_e32 v4, 5 +; SDAG-NEXT: v_mov_b32_e32 v5, 6 +; SDAG-NEXT: v_mov_b32_e32 v6, 7 +; SDAG-NEXT: v_mov_b32_e32 v7, 8 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v8i32_imm: ; GFX11: ; %bb.0: @@ -4114,6 +4968,32 @@ define amdgpu_kernel void @test_call_external_void_func_v8i32_imm() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v8i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v8i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v8i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 1 +; GISEL-NEXT: v_mov_b32_e32 v1, 2 +; GISEL-NEXT: v_mov_b32_e32 v2, 3 +; GISEL-NEXT: v_mov_b32_e32 v3, 4 +; GISEL-NEXT: v_mov_b32_e32 v4, 5 +; GISEL-NEXT: v_mov_b32_e32 v5, 6 +; GISEL-NEXT: v_mov_b32_e32 v6, 7 +; GISEL-NEXT: v_mov_b32_e32 v7, 8 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm call void @external_void_func_v8i32(<8 x i32> <i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8>) ret void } @@ -4171,31 +5051,31 @@ define amdgpu_kernel void @test_call_external_void_func_v16i32() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v16i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: buffer_load_dwordx4 v[4:7], off, s[0:3], 0 offset:16 -; GFX9-NEXT: buffer_load_dwordx4 v[8:11], off, s[0:3], 0 offset:32 -; GFX9-NEXT: buffer_load_dwordx4 v[12:15], off, s[0:3], 0 offset:48 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v16i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v16i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v16i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: buffer_load_dwordx4 v[4:7], off, s[0:3], 0 offset:16 +; SDAG-NEXT: buffer_load_dwordx4 v[8:11], off, s[0:3], 0 offset:32 +; SDAG-NEXT: buffer_load_dwordx4 v[12:15], off, s[0:3], 0 offset:48 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v16i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v16i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v16i32: ; GFX11: ; %bb.0: @@ -4238,6 +5118,44 @@ define amdgpu_kernel void @test_call_external_void_func_v16i32() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v16i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_load_dwordx16 s[8:23], s[0:1], 0x0 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v16i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v16i32@rel32@hi+12 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s8 +; GISEL-NEXT: v_mov_b32_e32 v1, s9 +; GISEL-NEXT: v_mov_b32_e32 v2, s10 +; GISEL-NEXT: v_mov_b32_e32 v3, s11 +; GISEL-NEXT: v_mov_b32_e32 v4, s12 +; GISEL-NEXT: v_mov_b32_e32 v5, s13 +; GISEL-NEXT: v_mov_b32_e32 v6, s14 +; GISEL-NEXT: v_mov_b32_e32 v7, s15 +; GISEL-NEXT: v_mov_b32_e32 v8, s16 +; GISEL-NEXT: v_mov_b32_e32 v9, s17 +; GISEL-NEXT: v_mov_b32_e32 v10, s18 +; GISEL-NEXT: v_mov_b32_e32 v11, s19 +; GISEL-NEXT: v_mov_b32_e32 v12, s20 +; GISEL-NEXT: v_mov_b32_e32 v13, s21 +; GISEL-NEXT: v_mov_b32_e32 v14, s22 +; GISEL-NEXT: v_mov_b32_e32 v15, s23 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %ptr = load ptr addrspace(1), ptr addrspace(4) poison %val = load <16 x i32>, ptr addrspace(1) %ptr call void @external_void_func_v16i32(<16 x i32> %val) @@ -4309,37 +5227,37 @@ define amdgpu_kernel void @test_call_external_void_func_v32i32() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[8:9] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v32i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x0 -; GFX9-NEXT: s_mov_b32 s7, 0xf000 -; GFX9-NEXT: s_mov_b32 s6, -1 -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: buffer_load_dwordx4 v[24:27], off, s[4:7], 0 offset:96 -; GFX9-NEXT: buffer_load_dwordx4 v[28:31], off, s[4:7], 0 offset:112 -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[4:7], 0 -; GFX9-NEXT: buffer_load_dwordx4 v[4:7], off, s[4:7], 0 offset:16 -; GFX9-NEXT: buffer_load_dwordx4 v[8:11], off, s[4:7], 0 offset:32 -; GFX9-NEXT: buffer_load_dwordx4 v[12:15], off, s[4:7], 0 offset:48 -; GFX9-NEXT: buffer_load_dwordx4 v[16:19], off, s[4:7], 0 offset:64 -; GFX9-NEXT: buffer_load_dwordx4 v[20:23], off, s[4:7], 0 offset:80 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_getpc_b64 s[8:9] -; GFX9-NEXT: s_add_u32 s8, s8, external_void_func_v32i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s9, s9, external_void_func_v32i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_waitcnt vmcnt(6) -; GFX9-NEXT: buffer_store_dword v31, off, s[36:39], s32 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[8:9] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v32i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x0 +; SDAG-NEXT: s_mov_b32 s7, 0xf000 +; SDAG-NEXT: s_mov_b32 s6, -1 +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: buffer_load_dwordx4 v[24:27], off, s[4:7], 0 offset:96 +; SDAG-NEXT: buffer_load_dwordx4 v[28:31], off, s[4:7], 0 offset:112 +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[4:7], 0 +; SDAG-NEXT: buffer_load_dwordx4 v[4:7], off, s[4:7], 0 offset:16 +; SDAG-NEXT: buffer_load_dwordx4 v[8:11], off, s[4:7], 0 offset:32 +; SDAG-NEXT: buffer_load_dwordx4 v[12:15], off, s[4:7], 0 offset:48 +; SDAG-NEXT: buffer_load_dwordx4 v[16:19], off, s[4:7], 0 offset:64 +; SDAG-NEXT: buffer_load_dwordx4 v[20:23], off, s[4:7], 0 offset:80 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_getpc_b64 s[8:9] +; SDAG-NEXT: s_add_u32 s8, s8, external_void_func_v32i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s9, s9, external_void_func_v32i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_waitcnt vmcnt(6) +; SDAG-NEXT: buffer_store_dword v31, off, s[36:39], s32 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[8:9] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v32i32: ; GFX11: ; %bb.0: @@ -4394,6 +5312,62 @@ define amdgpu_kernel void @test_call_external_void_func_v32i32() #0 { ; HSA-NEXT: buffer_store_dword v31, off, s[0:3], s32 ; HSA-NEXT: s_swappc_b64 s[30:31], s[12:13] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v32i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s52, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s53, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s54, -1 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_load_dwordx16 s[8:23], s[0:1], 0x40 +; GISEL-NEXT: s_load_dwordx16 s[36:51], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s55, 0xe00000 +; GISEL-NEXT: s_add_u32 s52, s52, s3 +; GISEL-NEXT: s_addc_u32 s53, s53, 0 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s23 +; GISEL-NEXT: s_mov_b64 s[0:1], s[52:53] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v32i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v32i32@rel32@hi+12 +; GISEL-NEXT: buffer_store_dword v0, off, s[52:55], s32 +; GISEL-NEXT: v_mov_b32_e32 v0, s36 +; GISEL-NEXT: v_mov_b32_e32 v1, s37 +; GISEL-NEXT: v_mov_b32_e32 v2, s38 +; GISEL-NEXT: v_mov_b32_e32 v3, s39 +; GISEL-NEXT: v_mov_b32_e32 v4, s40 +; GISEL-NEXT: v_mov_b32_e32 v5, s41 +; GISEL-NEXT: v_mov_b32_e32 v6, s42 +; GISEL-NEXT: v_mov_b32_e32 v7, s43 +; GISEL-NEXT: v_mov_b32_e32 v8, s44 +; GISEL-NEXT: v_mov_b32_e32 v9, s45 +; GISEL-NEXT: v_mov_b32_e32 v10, s46 +; GISEL-NEXT: v_mov_b32_e32 v11, s47 +; GISEL-NEXT: v_mov_b32_e32 v12, s48 +; GISEL-NEXT: v_mov_b32_e32 v13, s49 +; GISEL-NEXT: v_mov_b32_e32 v14, s50 +; GISEL-NEXT: v_mov_b32_e32 v15, s51 +; GISEL-NEXT: v_mov_b32_e32 v16, s8 +; GISEL-NEXT: v_mov_b32_e32 v17, s9 +; GISEL-NEXT: v_mov_b32_e32 v18, s10 +; GISEL-NEXT: v_mov_b32_e32 v19, s11 +; GISEL-NEXT: s_mov_b64 s[2:3], s[54:55] +; GISEL-NEXT: v_mov_b32_e32 v20, s12 +; GISEL-NEXT: v_mov_b32_e32 v21, s13 +; GISEL-NEXT: v_mov_b32_e32 v22, s14 +; GISEL-NEXT: v_mov_b32_e32 v23, s15 +; GISEL-NEXT: v_mov_b32_e32 v24, s16 +; GISEL-NEXT: v_mov_b32_e32 v25, s17 +; GISEL-NEXT: v_mov_b32_e32 v26, s18 +; GISEL-NEXT: v_mov_b32_e32 v27, s19 +; GISEL-NEXT: v_mov_b32_e32 v28, s20 +; GISEL-NEXT: v_mov_b32_e32 v29, s21 +; GISEL-NEXT: v_mov_b32_e32 v30, s22 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %ptr = load ptr addrspace(1), ptr addrspace(4) poison %val = load <32 x i32>, ptr addrspace(1) %ptr call void @external_void_func_v32i32(<32 x i32> %val) @@ -4471,40 +5445,40 @@ define amdgpu_kernel void @test_call_external_void_func_v32i32_i32(i32) #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v32i32_i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x0 -; GFX9-NEXT: s_mov_b32 s7, 0xf000 -; GFX9-NEXT: s_mov_b32 s6, -1 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: buffer_load_dword v32, off, s[4:7], 0 -; GFX9-NEXT: buffer_load_dwordx4 v[28:31], off, s[4:7], 0 offset:112 -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[4:7], 0 -; GFX9-NEXT: buffer_load_dwordx4 v[4:7], off, s[4:7], 0 offset:16 -; GFX9-NEXT: buffer_load_dwordx4 v[8:11], off, s[4:7], 0 offset:32 -; GFX9-NEXT: buffer_load_dwordx4 v[12:15], off, s[4:7], 0 offset:48 -; GFX9-NEXT: buffer_load_dwordx4 v[16:19], off, s[4:7], 0 offset:64 -; GFX9-NEXT: buffer_load_dwordx4 v[20:23], off, s[4:7], 0 offset:80 -; GFX9-NEXT: buffer_load_dwordx4 v[24:27], off, s[4:7], 0 offset:96 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v32i32_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v32i32_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_waitcnt vmcnt(8) -; GFX9-NEXT: buffer_store_dword v32, off, s[36:39], s32 offset:4 -; GFX9-NEXT: s_waitcnt vmcnt(8) -; GFX9-NEXT: buffer_store_dword v31, off, s[36:39], s32 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v32i32_i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x0 +; SDAG-NEXT: s_mov_b32 s7, 0xf000 +; SDAG-NEXT: s_mov_b32 s6, -1 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: buffer_load_dword v32, off, s[4:7], 0 +; SDAG-NEXT: buffer_load_dwordx4 v[28:31], off, s[4:7], 0 offset:112 +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[4:7], 0 +; SDAG-NEXT: buffer_load_dwordx4 v[4:7], off, s[4:7], 0 offset:16 +; SDAG-NEXT: buffer_load_dwordx4 v[8:11], off, s[4:7], 0 offset:32 +; SDAG-NEXT: buffer_load_dwordx4 v[12:15], off, s[4:7], 0 offset:48 +; SDAG-NEXT: buffer_load_dwordx4 v[16:19], off, s[4:7], 0 offset:64 +; SDAG-NEXT: buffer_load_dwordx4 v[20:23], off, s[4:7], 0 offset:80 +; SDAG-NEXT: buffer_load_dwordx4 v[24:27], off, s[4:7], 0 offset:96 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v32i32_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v32i32_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_waitcnt vmcnt(8) +; SDAG-NEXT: buffer_store_dword v32, off, s[36:39], s32 offset:4 +; SDAG-NEXT: s_waitcnt vmcnt(8) +; SDAG-NEXT: buffer_store_dword v31, off, s[36:39], s32 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v32i32_i32: ; GFX11: ; %bb.0: @@ -4566,6 +5540,67 @@ define amdgpu_kernel void @test_call_external_void_func_v32i32_i32(i32) #0 { ; HSA-NEXT: buffer_store_dword v31, off, s[0:3], s32 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v32i32_i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s52, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s53, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s54, -1 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_load_dwordx16 s[8:23], s[0:1], 0x40 +; GISEL-NEXT: s_load_dwordx16 s[36:51], s[0:1], 0x0 +; GISEL-NEXT: s_load_dword s2, s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s55, 0xe00000 +; GISEL-NEXT: s_add_u32 s52, s52, s5 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_addc_u32 s53, s53, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s2 +; GISEL-NEXT: ; kill: killed $sgpr0_sgpr1 +; GISEL-NEXT: ; kill: killed $sgpr0_sgpr1 +; GISEL-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:4 +; GISEL-NEXT: v_mov_b32_e32 v0, s23 +; GISEL-NEXT: s_mov_b64 s[0:1], s[52:53] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v32i32_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v32i32_i32@rel32@hi+12 +; GISEL-NEXT: buffer_store_dword v0, off, s[52:55], s32 +; GISEL-NEXT: v_mov_b32_e32 v0, s36 +; GISEL-NEXT: v_mov_b32_e32 v1, s37 +; GISEL-NEXT: v_mov_b32_e32 v2, s38 +; GISEL-NEXT: v_mov_b32_e32 v3, s39 +; GISEL-NEXT: v_mov_b32_e32 v4, s40 +; GISEL-NEXT: v_mov_b32_e32 v5, s41 +; GISEL-NEXT: v_mov_b32_e32 v6, s42 +; GISEL-NEXT: v_mov_b32_e32 v7, s43 +; GISEL-NEXT: v_mov_b32_e32 v8, s44 +; GISEL-NEXT: v_mov_b32_e32 v9, s45 +; GISEL-NEXT: v_mov_b32_e32 v10, s46 +; GISEL-NEXT: v_mov_b32_e32 v11, s47 +; GISEL-NEXT: v_mov_b32_e32 v12, s48 +; GISEL-NEXT: v_mov_b32_e32 v13, s49 +; GISEL-NEXT: v_mov_b32_e32 v14, s50 +; GISEL-NEXT: v_mov_b32_e32 v15, s51 +; GISEL-NEXT: v_mov_b32_e32 v16, s8 +; GISEL-NEXT: v_mov_b32_e32 v17, s9 +; GISEL-NEXT: v_mov_b32_e32 v18, s10 +; GISEL-NEXT: v_mov_b32_e32 v19, s11 +; GISEL-NEXT: s_mov_b64 s[2:3], s[54:55] +; GISEL-NEXT: v_mov_b32_e32 v20, s12 +; GISEL-NEXT: v_mov_b32_e32 v21, s13 +; GISEL-NEXT: v_mov_b32_e32 v22, s14 +; GISEL-NEXT: v_mov_b32_e32 v23, s15 +; GISEL-NEXT: v_mov_b32_e32 v24, s16 +; GISEL-NEXT: v_mov_b32_e32 v25, s17 +; GISEL-NEXT: v_mov_b32_e32 v26, s18 +; GISEL-NEXT: v_mov_b32_e32 v27, s19 +; GISEL-NEXT: v_mov_b32_e32 v28, s20 +; GISEL-NEXT: v_mov_b32_e32 v29, s21 +; GISEL-NEXT: v_mov_b32_e32 v30, s22 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %ptr0 = load ptr addrspace(1), ptr addrspace(4) poison %val0 = load <32 x i32>, ptr addrspace(1) %ptr0 %val1 = load i32, ptr addrspace(1) poison @@ -4622,29 +5657,29 @@ define amdgpu_kernel void @test_call_external_i32_func_i32_imm(ptr addrspace(1) ; CI-NEXT: s_waitcnt vmcnt(0) ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_i32_func_i32_imm: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s48, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s49, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s50, -1 -; GFX9-NEXT: s_mov_b32 s51, 0xe00000 -; GFX9-NEXT: s_add_u32 s48, s48, s5 -; GFX9-NEXT: s_load_dwordx2 s[36:37], s[2:3], 0x24 -; GFX9-NEXT: s_addc_u32 s49, s49, 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[48:49] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_i32_func_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_i32_func_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[50:51] -; GFX9-NEXT: v_mov_b32_e32 v0, 42 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_mov_b32 s39, 0xf000 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: buffer_store_dword v0, off, s[36:39], 0 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_i32_func_i32_imm: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s48, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s49, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s50, -1 +; SDAG-NEXT: s_mov_b32 s51, 0xe00000 +; SDAG-NEXT: s_add_u32 s48, s48, s5 +; SDAG-NEXT: s_load_dwordx2 s[36:37], s[2:3], 0x24 +; SDAG-NEXT: s_addc_u32 s49, s49, 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[48:49] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_i32_func_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_i32_func_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[50:51] +; SDAG-NEXT: v_mov_b32_e32 v0, 42 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 s39, 0xf000 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: buffer_store_dword v0, off, s[36:39], 0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_i32_func_i32_imm: ; GFX11: ; %bb.0: @@ -4682,6 +5717,30 @@ define amdgpu_kernel void @test_call_external_i32_func_i32_imm(ptr addrspace(1) ; HSA-NEXT: buffer_store_dword v0, off, s[36:39], 0 ; HSA-NEXT: s_waitcnt vmcnt(0) ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_i32_func_i32_imm: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s48, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s49, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s50, -1 +; GISEL-NEXT: s_mov_b32 s51, 0xe00000 +; GISEL-NEXT: s_add_u32 s48, s48, s5 +; GISEL-NEXT: s_load_dwordx2 s[36:37], s[2:3], 0x24 +; GISEL-NEXT: s_addc_u32 s49, s49, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[48:49] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_i32_func_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_i32_func_i32@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 42 +; GISEL-NEXT: s_mov_b64 s[2:3], s[50:51] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xf000 +; GISEL-NEXT: buffer_store_dword v0, off, s[36:39], 0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = call i32 @external_i32_func_i32(i32 42) store volatile i32 %val, ptr addrspace(1) %out ret void @@ -4736,29 +5795,29 @@ define amdgpu_kernel void @test_call_external_void_func_struct_i8_i32() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_struct_i8_i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 -; GFX9-NEXT: buffer_load_dword v1, off, s[0:3], 0 offset:4 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_struct_i8_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_struct_i8_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_struct_i8_i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: buffer_load_ubyte v0, off, s[0:3], 0 +; SDAG-NEXT: buffer_load_dword v1, off, s[0:3], 0 offset:4 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_struct_i8_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_struct_i8_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_struct_i8_i32: ; GFX11: ; %bb.0: @@ -4797,6 +5856,30 @@ define amdgpu_kernel void @test_call_external_void_func_struct_i8_i32() #0 { ; HSA-NEXT: s_mov_b32 s32, 0 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_struct_i8_i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_load_dwordx2 s[4:5], s[0:1], 0x0 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[8:9] +; GISEL-NEXT: s_add_u32 s8, s8, external_void_func_struct_i8_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s9, s9, external_void_func_struct_i8_i32@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s4 +; GISEL-NEXT: v_mov_b32_e32 v1, s5 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[8:9] +; GISEL-NEXT: s_endpgm %ptr0 = load ptr addrspace(1), ptr addrspace(4) poison %val = load { i8, i32 }, ptr addrspace(1) %ptr0 call void @external_void_func_struct_i8_i32({ i8, i32 } %val) @@ -4860,34 +5943,34 @@ define amdgpu_kernel void @test_call_external_void_func_byval_struct_i8_i32() #0 ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_byval_struct_i8_i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: v_mov_b32_e32 v0, 3 -; GFX9-NEXT: buffer_store_byte v0, off, s[36:39], 0 -; GFX9-NEXT: v_mov_b32_e32 v0, 8 -; GFX9-NEXT: buffer_store_dword v0, off, s[36:39], 0 offset:4 -; GFX9-NEXT: buffer_load_dword v0, off, s[36:39], 0 offset:4 -; GFX9-NEXT: s_nop 0 -; GFX9-NEXT: buffer_load_dword v1, off, s[36:39], 0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_movk_i32 s32, 0x400 -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_byval_struct_i8_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_byval_struct_i8_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_dword v0, off, s[36:39], s32 offset:4 -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_dword v1, off, s[36:39], s32 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_byval_struct_i8_i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: v_mov_b32_e32 v0, 3 +; SDAG-NEXT: buffer_store_byte v0, off, s[36:39], 0 +; SDAG-NEXT: v_mov_b32_e32 v0, 8 +; SDAG-NEXT: buffer_store_dword v0, off, s[36:39], 0 offset:4 +; SDAG-NEXT: buffer_load_dword v0, off, s[36:39], 0 offset:4 +; SDAG-NEXT: s_nop 0 +; SDAG-NEXT: buffer_load_dword v1, off, s[36:39], 0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_movk_i32 s32, 0x400 +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_byval_struct_i8_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_byval_struct_i8_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_dword v0, off, s[36:39], s32 offset:4 +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_dword v1, off, s[36:39], s32 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-TRUE16-LABEL: test_call_external_void_func_byval_struct_i8_i32: ; GFX11-TRUE16: ; %bb.0: @@ -4948,6 +6031,35 @@ define amdgpu_kernel void @test_call_external_void_func_byval_struct_i8_i32() #0 ; HSA-NEXT: buffer_store_dword v1, off, s[0:3], s32 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_byval_struct_i8_i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: v_mov_b32_e32 v0, 3 +; GISEL-NEXT: buffer_store_byte v0, off, s[36:39], 0 +; GISEL-NEXT: v_mov_b32_e32 v0, 8 +; GISEL-NEXT: buffer_store_dword v0, off, s[36:39], 0 offset:4 +; GISEL-NEXT: buffer_load_dword v0, off, s[36:39], 0 +; GISEL-NEXT: s_nop 0 +; GISEL-NEXT: buffer_load_dword v1, off, s[36:39], 0 offset:4 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_movk_i32 s32, 0x400 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_byval_struct_i8_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_byval_struct_i8_i32@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_waitcnt vmcnt(1) +; GISEL-NEXT: buffer_store_dword v0, off, s[36:39], s32 +; GISEL-NEXT: s_waitcnt vmcnt(1) +; GISEL-NEXT: buffer_store_dword v1, off, s[36:39], s32 offset:4 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %val = alloca { i8, i32 }, align 8, addrspace(5) %gep0 = getelementptr inbounds { i8, i32 }, ptr addrspace(5) %val, i32 0, i32 0 %gep1 = getelementptr inbounds { i8, i32 }, ptr addrspace(5) %val, i32 0, i32 1 @@ -5034,44 +6146,44 @@ define amdgpu_kernel void @test_call_external_void_func_sret_struct_i8_i32_byval ; CI-NEXT: s_waitcnt vmcnt(0) ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_sret_struct_i8_i32_byval_struct_i8_i32: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s5 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: v_mov_b32_e32 v0, 3 -; GFX9-NEXT: buffer_store_byte v0, off, s[36:39], 0 -; GFX9-NEXT: v_mov_b32_e32 v0, 8 -; GFX9-NEXT: buffer_store_dword v0, off, s[36:39], 0 offset:4 -; GFX9-NEXT: buffer_load_dword v0, off, s[36:39], 0 offset:4 -; GFX9-NEXT: s_nop 0 -; GFX9-NEXT: buffer_load_dword v1, off, s[36:39], 0 -; GFX9-NEXT: s_movk_i32 s32, 0x800 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_sret_struct_i8_i32_byval_struct_i8_i32@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_sret_struct_i8_i32_byval_struct_i8_i32@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_dword v0, off, s[36:39], s32 offset:4 -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_dword v1, off, s[36:39], s32 -; GFX9-NEXT: v_mov_b32_e32 v0, 8 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: buffer_load_ubyte v0, off, s[36:39], 0 offset:8 -; GFX9-NEXT: buffer_load_dword v1, off, s[36:39], 0 offset:12 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_byte v0, off, s[0:3], 0 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: buffer_store_dword v1, off, s[0:3], 0 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_sret_struct_i8_i32_byval_struct_i8_i32: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s5 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: v_mov_b32_e32 v0, 3 +; SDAG-NEXT: buffer_store_byte v0, off, s[36:39], 0 +; SDAG-NEXT: v_mov_b32_e32 v0, 8 +; SDAG-NEXT: buffer_store_dword v0, off, s[36:39], 0 offset:4 +; SDAG-NEXT: buffer_load_dword v0, off, s[36:39], 0 offset:4 +; SDAG-NEXT: s_nop 0 +; SDAG-NEXT: buffer_load_dword v1, off, s[36:39], 0 +; SDAG-NEXT: s_movk_i32 s32, 0x800 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_sret_struct_i8_i32_byval_struct_i8_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_sret_struct_i8_i32_byval_struct_i8_i32@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_dword v0, off, s[36:39], s32 offset:4 +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_dword v1, off, s[36:39], s32 +; SDAG-NEXT: v_mov_b32_e32 v0, 8 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: buffer_load_ubyte v0, off, s[36:39], 0 offset:8 +; SDAG-NEXT: buffer_load_dword v1, off, s[36:39], 0 offset:12 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_byte v0, off, s[0:3], 0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: buffer_store_dword v1, off, s[0:3], 0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm ; ; GFX11-TRUE16-LABEL: test_call_external_void_func_sret_struct_i8_i32_byval_struct_i8_i32: ; GFX11-TRUE16: ; %bb.0: @@ -5170,6 +6282,45 @@ define amdgpu_kernel void @test_call_external_void_func_sret_struct_i8_i32_byval ; HSA-NEXT: buffer_store_dword v1, off, s[4:7], 0 ; HSA-NEXT: s_waitcnt vmcnt(0) ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_sret_struct_i8_i32_byval_struct_i8_i32: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s5 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: v_mov_b32_e32 v0, 3 +; GISEL-NEXT: buffer_store_byte v0, off, s[36:39], 0 +; GISEL-NEXT: v_mov_b32_e32 v0, 8 +; GISEL-NEXT: buffer_store_dword v0, off, s[36:39], 0 offset:4 +; GISEL-NEXT: buffer_load_dword v0, off, s[36:39], 0 +; GISEL-NEXT: s_nop 0 +; GISEL-NEXT: buffer_load_dword v1, off, s[36:39], 0 offset:4 +; GISEL-NEXT: s_movk_i32 s32, 0x800 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_sret_struct_i8_i32_byval_struct_i8_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_sret_struct_i8_i32_byval_struct_i8_i32@rel32@hi+12 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_waitcnt vmcnt(1) +; GISEL-NEXT: buffer_store_dword v0, off, s[36:39], s32 +; GISEL-NEXT: s_waitcnt vmcnt(1) +; GISEL-NEXT: buffer_store_dword v1, off, s[36:39], s32 offset:4 +; GISEL-NEXT: v_mov_b32_e32 v0, 8 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: buffer_load_ubyte v0, off, s[36:39], 0 offset:8 +; GISEL-NEXT: buffer_load_dword v1, off, s[36:39], 0 offset:12 +; GISEL-NEXT: s_mov_b32 s2, -1 +; GISEL-NEXT: s_mov_b32 s3, 0xf000 +; GISEL-NEXT: s_waitcnt vmcnt(1) +; GISEL-NEXT: buffer_store_byte v0, off, s[0:3], 0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: buffer_store_dword v1, off, s[0:3], 0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %in.val = alloca { i8, i32 }, align 8, addrspace(5) %out.val = alloca { i8, i32 }, align 8, addrspace(5) %in.gep0 = getelementptr inbounds { i8, i32 }, ptr addrspace(5) %in.val, i32 0, i32 0 @@ -5272,47 +6423,47 @@ define amdgpu_kernel void @test_call_external_void_func_v16i8() #0 { ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: test_call_external_void_func_v16i8: -; GFX9: ; %bb.0: -; GFX9-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 -; GFX9-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s38, -1 -; GFX9-NEXT: s_mov_b32 s39, 0xe00000 -; GFX9-NEXT: s_add_u32 s36, s36, s3 -; GFX9-NEXT: s_mov_b32 s3, 0xf000 -; GFX9-NEXT: s_mov_b32 s2, -1 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 -; GFX9-NEXT: s_addc_u32 s37, s37, 0 -; GFX9-NEXT: s_mov_b64 s[0:1], s[36:37] -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, external_void_func_v16i8@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, external_void_func_v16i8@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[38:39] -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_waitcnt vmcnt(0) -; GFX9-NEXT: v_lshrrev_b32_e32 v16, 8, v0 -; GFX9-NEXT: v_lshrrev_b32_e32 v17, 16, v0 -; GFX9-NEXT: v_lshrrev_b32_e32 v18, 24, v0 -; GFX9-NEXT: v_lshrrev_b32_e32 v5, 8, v1 -; GFX9-NEXT: v_lshrrev_b32_e32 v6, 16, v1 -; GFX9-NEXT: v_lshrrev_b32_e32 v7, 24, v1 -; GFX9-NEXT: v_lshrrev_b32_e32 v9, 8, v2 -; GFX9-NEXT: v_lshrrev_b32_e32 v10, 16, v2 -; GFX9-NEXT: v_lshrrev_b32_e32 v11, 24, v2 -; GFX9-NEXT: v_lshrrev_b32_e32 v13, 8, v3 -; GFX9-NEXT: v_lshrrev_b32_e32 v14, 16, v3 -; GFX9-NEXT: v_lshrrev_b32_e32 v15, 24, v3 -; GFX9-NEXT: v_mov_b32_e32 v4, v1 -; GFX9-NEXT: v_mov_b32_e32 v8, v2 -; GFX9-NEXT: v_mov_b32_e32 v12, v3 -; GFX9-NEXT: v_mov_b32_e32 v1, v16 -; GFX9-NEXT: v_mov_b32_e32 v2, v17 -; GFX9-NEXT: v_mov_b32_e32 v3, v18 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: test_call_external_void_func_v16i8: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; SDAG-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s38, -1 +; SDAG-NEXT: s_mov_b32 s39, 0xe00000 +; SDAG-NEXT: s_add_u32 s36, s36, s3 +; SDAG-NEXT: s_mov_b32 s3, 0xf000 +; SDAG-NEXT: s_mov_b32 s2, -1 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: buffer_load_dwordx4 v[0:3], off, s[0:3], 0 +; SDAG-NEXT: s_addc_u32 s37, s37, 0 +; SDAG-NEXT: s_mov_b64 s[0:1], s[36:37] +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, external_void_func_v16i8@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, external_void_func_v16i8@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[38:39] +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: v_lshrrev_b32_e32 v16, 8, v0 +; SDAG-NEXT: v_lshrrev_b32_e32 v17, 16, v0 +; SDAG-NEXT: v_lshrrev_b32_e32 v18, 24, v0 +; SDAG-NEXT: v_lshrrev_b32_e32 v5, 8, v1 +; SDAG-NEXT: v_lshrrev_b32_e32 v6, 16, v1 +; SDAG-NEXT: v_lshrrev_b32_e32 v7, 24, v1 +; SDAG-NEXT: v_lshrrev_b32_e32 v9, 8, v2 +; SDAG-NEXT: v_lshrrev_b32_e32 v10, 16, v2 +; SDAG-NEXT: v_lshrrev_b32_e32 v11, 24, v2 +; SDAG-NEXT: v_lshrrev_b32_e32 v13, 8, v3 +; SDAG-NEXT: v_lshrrev_b32_e32 v14, 16, v3 +; SDAG-NEXT: v_lshrrev_b32_e32 v15, 24, v3 +; SDAG-NEXT: v_mov_b32_e32 v4, v1 +; SDAG-NEXT: v_mov_b32_e32 v8, v2 +; SDAG-NEXT: v_mov_b32_e32 v12, v3 +; SDAG-NEXT: v_mov_b32_e32 v1, v16 +; SDAG-NEXT: v_mov_b32_e32 v2, v17 +; SDAG-NEXT: v_mov_b32_e32 v3, v18 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: test_call_external_void_func_v16i8: ; GFX11: ; %bb.0: @@ -5384,6 +6535,56 @@ define amdgpu_kernel void @test_call_external_void_func_v16i8() #0 { ; HSA-NEXT: v_mov_b32_e32 v3, v18 ; HSA-NEXT: s_swappc_b64 s[30:31], s[8:9] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: test_call_external_void_func_v16i8: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[0:1], 0x0 +; GISEL-NEXT: s_mov_b32 s36, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s37, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s38, -1 +; GISEL-NEXT: s_mov_b32 s39, 0xe00000 +; GISEL-NEXT: s_add_u32 s36, s36, s3 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_load_dwordx4 s[0:3], s[0:1], 0x0 +; GISEL-NEXT: s_addc_u32 s37, s37, 0 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_lshr_b32 s8, s0, 8 +; GISEL-NEXT: s_lshr_b32 s9, s0, 16 +; GISEL-NEXT: s_lshr_b32 s10, s0, 24 +; GISEL-NEXT: s_lshr_b32 s11, s1, 8 +; GISEL-NEXT: s_lshr_b32 s12, s1, 16 +; GISEL-NEXT: s_lshr_b32 s13, s1, 24 +; GISEL-NEXT: s_lshr_b32 s14, s2, 8 +; GISEL-NEXT: s_lshr_b32 s15, s2, 16 +; GISEL-NEXT: s_lshr_b32 s16, s2, 24 +; GISEL-NEXT: s_lshr_b32 s17, s3, 8 +; GISEL-NEXT: s_lshr_b32 s18, s3, 16 +; GISEL-NEXT: s_lshr_b32 s19, s3, 24 +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: v_mov_b32_e32 v4, s1 +; GISEL-NEXT: v_mov_b32_e32 v8, s2 +; GISEL-NEXT: v_mov_b32_e32 v12, s3 +; GISEL-NEXT: s_mov_b64 s[0:1], s[36:37] +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, external_void_func_v16i8@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, external_void_func_v16i8@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v1, s8 +; GISEL-NEXT: v_mov_b32_e32 v2, s9 +; GISEL-NEXT: v_mov_b32_e32 v3, s10 +; GISEL-NEXT: v_mov_b32_e32 v5, s11 +; GISEL-NEXT: v_mov_b32_e32 v6, s12 +; GISEL-NEXT: v_mov_b32_e32 v7, s13 +; GISEL-NEXT: v_mov_b32_e32 v9, s14 +; GISEL-NEXT: v_mov_b32_e32 v10, s15 +; GISEL-NEXT: v_mov_b32_e32 v11, s16 +; GISEL-NEXT: v_mov_b32_e32 v13, s17 +; GISEL-NEXT: v_mov_b32_e32 v14, s18 +; GISEL-NEXT: v_mov_b32_e32 v15, s19 +; GISEL-NEXT: s_mov_b64 s[2:3], s[38:39] +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm %ptr = load ptr addrspace(1), ptr addrspace(4) poison %val = load <16 x i8>, ptr addrspace(1) %ptr call void @external_void_func_v16i8(<16 x i8> %val) @@ -5509,64 +6710,64 @@ define amdgpu_kernel void @stack_passed_arg_alignment_v32i32_f64(<32 x i32> %val ; CI-NEXT: s_swappc_b64 s[30:31], s[4:5] ; CI-NEXT: s_endpgm ; -; GFX9-LABEL: stack_passed_arg_alignment_v32i32_f64: -; GFX9: ; %bb.0: ; %entry -; GFX9-NEXT: s_mov_b32 s52, SCRATCH_RSRC_DWORD0 -; GFX9-NEXT: s_mov_b32 s53, SCRATCH_RSRC_DWORD1 -; GFX9-NEXT: s_mov_b32 s54, -1 -; GFX9-NEXT: s_mov_b32 s55, 0xe00000 -; GFX9-NEXT: s_add_u32 s52, s52, s5 -; GFX9-NEXT: s_load_dwordx16 s[8:23], s[2:3], 0x64 -; GFX9-NEXT: s_load_dwordx2 s[4:5], s[2:3], 0xa4 -; GFX9-NEXT: s_load_dwordx16 s[36:51], s[2:3], 0x24 -; GFX9-NEXT: s_mov_b32 s32, 0 -; GFX9-NEXT: s_addc_u32 s53, s53, 0 -; GFX9-NEXT: s_waitcnt lgkmcnt(0) -; GFX9-NEXT: v_mov_b32_e32 v0, s23 -; GFX9-NEXT: buffer_store_dword v0, off, s[52:55], s32 -; GFX9-NEXT: v_mov_b32_e32 v0, s4 -; GFX9-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:4 -; GFX9-NEXT: v_mov_b32_e32 v0, s5 -; GFX9-NEXT: s_mov_b64 s[6:7], s[0:1] -; GFX9-NEXT: s_mov_b64 s[0:1], s[52:53] -; GFX9-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:8 -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, stack_passed_f64_arg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, stack_passed_f64_arg@rel32@hi+12 -; GFX9-NEXT: s_mov_b64 s[2:3], s[54:55] -; GFX9-NEXT: v_mov_b32_e32 v0, s36 -; GFX9-NEXT: v_mov_b32_e32 v1, s37 -; GFX9-NEXT: v_mov_b32_e32 v2, s38 -; GFX9-NEXT: v_mov_b32_e32 v3, s39 -; GFX9-NEXT: v_mov_b32_e32 v4, s40 -; GFX9-NEXT: v_mov_b32_e32 v5, s41 -; GFX9-NEXT: v_mov_b32_e32 v6, s42 -; GFX9-NEXT: v_mov_b32_e32 v7, s43 -; GFX9-NEXT: v_mov_b32_e32 v8, s44 -; GFX9-NEXT: v_mov_b32_e32 v9, s45 -; GFX9-NEXT: v_mov_b32_e32 v10, s46 -; GFX9-NEXT: v_mov_b32_e32 v11, s47 -; GFX9-NEXT: v_mov_b32_e32 v12, s48 -; GFX9-NEXT: v_mov_b32_e32 v13, s49 -; GFX9-NEXT: v_mov_b32_e32 v14, s50 -; GFX9-NEXT: v_mov_b32_e32 v15, s51 -; GFX9-NEXT: v_mov_b32_e32 v16, s8 -; GFX9-NEXT: v_mov_b32_e32 v17, s9 -; GFX9-NEXT: v_mov_b32_e32 v18, s10 -; GFX9-NEXT: v_mov_b32_e32 v19, s11 -; GFX9-NEXT: v_mov_b32_e32 v20, s12 -; GFX9-NEXT: v_mov_b32_e32 v21, s13 -; GFX9-NEXT: v_mov_b32_e32 v22, s14 -; GFX9-NEXT: v_mov_b32_e32 v23, s15 -; GFX9-NEXT: v_mov_b32_e32 v24, s16 -; GFX9-NEXT: v_mov_b32_e32 v25, s17 -; GFX9-NEXT: v_mov_b32_e32 v26, s18 -; GFX9-NEXT: v_mov_b32_e32 v27, s19 -; GFX9-NEXT: v_mov_b32_e32 v28, s20 -; GFX9-NEXT: v_mov_b32_e32 v29, s21 -; GFX9-NEXT: v_mov_b32_e32 v30, s22 -; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5] -; GFX9-NEXT: s_endpgm +; SDAG-LABEL: stack_passed_arg_alignment_v32i32_f64: +; SDAG: ; %bb.0: ; %entry +; SDAG-NEXT: s_mov_b32 s52, SCRATCH_RSRC_DWORD0 +; SDAG-NEXT: s_mov_b32 s53, SCRATCH_RSRC_DWORD1 +; SDAG-NEXT: s_mov_b32 s54, -1 +; SDAG-NEXT: s_mov_b32 s55, 0xe00000 +; SDAG-NEXT: s_add_u32 s52, s52, s5 +; SDAG-NEXT: s_load_dwordx16 s[8:23], s[2:3], 0x64 +; SDAG-NEXT: s_load_dwordx2 s[4:5], s[2:3], 0xa4 +; SDAG-NEXT: s_load_dwordx16 s[36:51], s[2:3], 0x24 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_addc_u32 s53, s53, 0 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: v_mov_b32_e32 v0, s23 +; SDAG-NEXT: buffer_store_dword v0, off, s[52:55], s32 +; SDAG-NEXT: v_mov_b32_e32 v0, s4 +; SDAG-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:4 +; SDAG-NEXT: v_mov_b32_e32 v0, s5 +; SDAG-NEXT: s_mov_b64 s[6:7], s[0:1] +; SDAG-NEXT: s_mov_b64 s[0:1], s[52:53] +; SDAG-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:8 +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, stack_passed_f64_arg@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, stack_passed_f64_arg@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[2:3], s[54:55] +; SDAG-NEXT: v_mov_b32_e32 v0, s36 +; SDAG-NEXT: v_mov_b32_e32 v1, s37 +; SDAG-NEXT: v_mov_b32_e32 v2, s38 +; SDAG-NEXT: v_mov_b32_e32 v3, s39 +; SDAG-NEXT: v_mov_b32_e32 v4, s40 +; SDAG-NEXT: v_mov_b32_e32 v5, s41 +; SDAG-NEXT: v_mov_b32_e32 v6, s42 +; SDAG-NEXT: v_mov_b32_e32 v7, s43 +; SDAG-NEXT: v_mov_b32_e32 v8, s44 +; SDAG-NEXT: v_mov_b32_e32 v9, s45 +; SDAG-NEXT: v_mov_b32_e32 v10, s46 +; SDAG-NEXT: v_mov_b32_e32 v11, s47 +; SDAG-NEXT: v_mov_b32_e32 v12, s48 +; SDAG-NEXT: v_mov_b32_e32 v13, s49 +; SDAG-NEXT: v_mov_b32_e32 v14, s50 +; SDAG-NEXT: v_mov_b32_e32 v15, s51 +; SDAG-NEXT: v_mov_b32_e32 v16, s8 +; SDAG-NEXT: v_mov_b32_e32 v17, s9 +; SDAG-NEXT: v_mov_b32_e32 v18, s10 +; SDAG-NEXT: v_mov_b32_e32 v19, s11 +; SDAG-NEXT: v_mov_b32_e32 v20, s12 +; SDAG-NEXT: v_mov_b32_e32 v21, s13 +; SDAG-NEXT: v_mov_b32_e32 v22, s14 +; SDAG-NEXT: v_mov_b32_e32 v23, s15 +; SDAG-NEXT: v_mov_b32_e32 v24, s16 +; SDAG-NEXT: v_mov_b32_e32 v25, s17 +; SDAG-NEXT: v_mov_b32_e32 v26, s18 +; SDAG-NEXT: v_mov_b32_e32 v27, s19 +; SDAG-NEXT: v_mov_b32_e32 v28, s20 +; SDAG-NEXT: v_mov_b32_e32 v29, s21 +; SDAG-NEXT: v_mov_b32_e32 v30, s22 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[4:5] +; SDAG-NEXT: s_endpgm ; ; GFX11-LABEL: stack_passed_arg_alignment_v32i32_f64: ; GFX11: ; %bb.0: ; %entry @@ -5662,6 +6863,65 @@ define amdgpu_kernel void @stack_passed_arg_alignment_v32i32_f64(<32 x i32> %val ; HSA-NEXT: v_mov_b32_e32 v30, s22 ; HSA-NEXT: s_swappc_b64 s[30:31], s[24:25] ; HSA-NEXT: s_endpgm +; +; GISEL-LABEL: stack_passed_arg_alignment_v32i32_f64: +; GISEL: ; %bb.0: ; %entry +; GISEL-NEXT: s_mov_b32 s52, SCRATCH_RSRC_DWORD0 +; GISEL-NEXT: s_mov_b32 s53, SCRATCH_RSRC_DWORD1 +; GISEL-NEXT: s_mov_b32 s54, -1 +; GISEL-NEXT: s_mov_b64 s[6:7], s[0:1] +; GISEL-NEXT: s_load_dwordx16 s[8:23], s[2:3], 0x64 +; GISEL-NEXT: s_load_dwordx2 s[0:1], s[2:3], 0xa4 +; GISEL-NEXT: s_load_dwordx16 s[36:51], s[2:3], 0x24 +; GISEL-NEXT: s_mov_b32 s55, 0xe00000 +; GISEL-NEXT: s_add_u32 s52, s52, s5 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_addc_u32 s53, s53, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: v_mov_b32_e32 v0, s23 +; GISEL-NEXT: buffer_store_dword v0, off, s[52:55], s32 +; GISEL-NEXT: v_mov_b32_e32 v0, s0 +; GISEL-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:4 +; GISEL-NEXT: v_mov_b32_e32 v0, s1 +; GISEL-NEXT: s_mov_b64 s[0:1], s[52:53] +; GISEL-NEXT: buffer_store_dword v0, off, s[52:55], s32 offset:8 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, stack_passed_f64_arg@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, stack_passed_f64_arg@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, s36 +; GISEL-NEXT: v_mov_b32_e32 v1, s37 +; GISEL-NEXT: v_mov_b32_e32 v2, s38 +; GISEL-NEXT: v_mov_b32_e32 v3, s39 +; GISEL-NEXT: v_mov_b32_e32 v4, s40 +; GISEL-NEXT: v_mov_b32_e32 v5, s41 +; GISEL-NEXT: v_mov_b32_e32 v6, s42 +; GISEL-NEXT: v_mov_b32_e32 v7, s43 +; GISEL-NEXT: v_mov_b32_e32 v8, s44 +; GISEL-NEXT: v_mov_b32_e32 v9, s45 +; GISEL-NEXT: v_mov_b32_e32 v10, s46 +; GISEL-NEXT: v_mov_b32_e32 v11, s47 +; GISEL-NEXT: v_mov_b32_e32 v12, s48 +; GISEL-NEXT: v_mov_b32_e32 v13, s49 +; GISEL-NEXT: v_mov_b32_e32 v14, s50 +; GISEL-NEXT: v_mov_b32_e32 v15, s51 +; GISEL-NEXT: v_mov_b32_e32 v16, s8 +; GISEL-NEXT: v_mov_b32_e32 v17, s9 +; GISEL-NEXT: v_mov_b32_e32 v18, s10 +; GISEL-NEXT: v_mov_b32_e32 v19, s11 +; GISEL-NEXT: s_mov_b64 s[2:3], s[54:55] +; GISEL-NEXT: v_mov_b32_e32 v20, s12 +; GISEL-NEXT: v_mov_b32_e32 v21, s13 +; GISEL-NEXT: v_mov_b32_e32 v22, s14 +; GISEL-NEXT: v_mov_b32_e32 v23, s15 +; GISEL-NEXT: v_mov_b32_e32 v24, s16 +; GISEL-NEXT: v_mov_b32_e32 v25, s17 +; GISEL-NEXT: v_mov_b32_e32 v26, s18 +; GISEL-NEXT: v_mov_b32_e32 v27, s19 +; GISEL-NEXT: v_mov_b32_e32 v28, s20 +; GISEL-NEXT: v_mov_b32_e32 v29, s21 +; GISEL-NEXT: v_mov_b32_e32 v30, s22 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[4:5] +; GISEL-NEXT: s_endpgm entry: call void @stack_passed_f64_arg(<32 x i32> %val, double %tmp) ret void @@ -5702,22 +6962,22 @@ define void @tail_call_byval_align16(<32 x i32> %val, double %tmp) #0 { ; CI-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:16 ; CI-NEXT: s_setpc_b64 s[4:5] ; -; GFX9-LABEL: tail_call_byval_align16: -; GFX9: ; %bb.0: ; %entry -; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) -; GFX9-NEXT: buffer_load_dword v31, off, s[0:3], s32 offset:28 -; GFX9-NEXT: buffer_load_dword v32, off, s[0:3], s32 -; GFX9-NEXT: s_getpc_b64 s[4:5] -; GFX9-NEXT: s_add_u32 s4, s4, byval_align16_f64_arg@rel32@lo+4 -; GFX9-NEXT: s_addc_u32 s5, s5, byval_align16_f64_arg@rel32@hi+12 -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:20 -; GFX9-NEXT: buffer_load_dword v31, off, s[0:3], s32 offset:24 -; GFX9-NEXT: s_waitcnt vmcnt(2) -; GFX9-NEXT: buffer_store_dword v32, off, s[0:3], s32 -; GFX9-NEXT: s_waitcnt vmcnt(1) -; GFX9-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:16 -; GFX9-NEXT: s_setpc_b64 s[4:5] +; SDAG-LABEL: tail_call_byval_align16: +; SDAG: ; %bb.0: ; %entry +; SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; SDAG-NEXT: buffer_load_dword v31, off, s[0:3], s32 offset:28 +; SDAG-NEXT: buffer_load_dword v32, off, s[0:3], s32 +; SDAG-NEXT: s_getpc_b64 s[4:5] +; SDAG-NEXT: s_add_u32 s4, s4, byval_align16_f64_arg@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s5, s5, byval_align16_f64_arg@rel32@hi+12 +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:20 +; SDAG-NEXT: buffer_load_dword v31, off, s[0:3], s32 offset:24 +; SDAG-NEXT: s_waitcnt vmcnt(2) +; SDAG-NEXT: buffer_store_dword v32, off, s[0:3], s32 +; SDAG-NEXT: s_waitcnt vmcnt(1) +; SDAG-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:16 +; SDAG-NEXT: s_setpc_b64 s[4:5] ; ; GFX11-LABEL: tail_call_byval_align16: ; GFX11: ; %bb.0: ; %entry @@ -5749,6 +7009,23 @@ define void @tail_call_byval_align16(<32 x i32> %val, double %tmp) #0 { ; HSA-NEXT: s_waitcnt vmcnt(1) ; HSA-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:16 ; HSA-NEXT: s_setpc_b64 s[4:5] +; +; GISEL-LABEL: tail_call_byval_align16: +; GISEL: ; %bb.0: ; %entry +; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GISEL-NEXT: buffer_load_dword v31, off, s[0:3], s32 +; GISEL-NEXT: s_getpc_b64 s[4:5] +; GISEL-NEXT: s_add_u32 s4, s4, byval_align16_f64_arg@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s5, s5, byval_align16_f64_arg@rel32@hi+12 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: buffer_store_dword v31, off, s[0:3], s32 +; GISEL-NEXT: buffer_load_dword v31, off, s[0:3], s32 offset:28 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:20 +; GISEL-NEXT: buffer_load_dword v31, off, s[0:3], s32 offset:24 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: buffer_store_dword v31, off, s[0:3], s32 offset:16 +; GISEL-NEXT: s_setpc_b64 s[4:5] entry: %alloca = alloca double, align 8, addrspace(5) tail call void @byval_align16_f64_arg(<32 x i32> %val, ptr addrspace(5) byval(double) align 16 %alloca) diff --git a/llvm/test/CodeGen/AMDGPU/call-c-function.ll b/llvm/test/CodeGen/AMDGPU/call-c-function.ll index e1bb3ea..4fbc727 100644 --- a/llvm/test/CodeGen/AMDGPU/call-c-function.ll +++ b/llvm/test/CodeGen/AMDGPU/call-c-function.ll @@ -1,21 +1,68 @@ ; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py -; RUN: llc -global-isel=0 -stop-after=finalize-isel -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -o - %s | FileCheck -enable-var-scope %s +; RUN: llc -global-isel=0 -stop-after=finalize-isel -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -o - %s | FileCheck -check-prefix=SDAG -enable-var-scope %s +; RUN: llc -global-isel=1 -stop-after=finalize-isel -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -o - %s | FileCheck -check-prefix=GISEL -enable-var-scope %s ; Test that we don't explode on calls from shaders to functions with the C calling convention. define amdgpu_ps void @amdgpu_ps_call_default_cc() { - ; CHECK-LABEL: name: amdgpu_ps_call_default_cc - ; CHECK: bb.0.main_body: - ; CHECK-NEXT: S_ENDPGM 0 + ; SDAG-LABEL: name: amdgpu_ps_call_default_cc + ; SDAG: bb.0.main_body: + ; SDAG-NEXT: S_ENDPGM 0 + ; + ; GISEL-LABEL: name: amdgpu_ps_call_default_cc + ; GISEL: bb.1.main_body: + ; GISEL-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32 + ; GISEL-NEXT: [[DEF:%[0-9]+]]:sreg_64 = IMPLICIT_DEF + ; GISEL-NEXT: [[DEF1:%[0-9]+]]:sreg_64 = IMPLICIT_DEF + ; GISEL-NEXT: [[DEF2:%[0-9]+]]:sreg_32 = IMPLICIT_DEF + ; GISEL-NEXT: [[COPY:%[0-9]+]]:sgpr_128 = COPY $sgpr96_sgpr97_sgpr98_sgpr99 + ; GISEL-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY]] + ; GISEL-NEXT: $sgpr4_sgpr5 = COPY [[DEF]] + ; GISEL-NEXT: $sgpr6_sgpr7 = COPY [[DEF]] + ; GISEL-NEXT: [[S_MOV_B:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO 0 + ; GISEL-NEXT: $sgpr8_sgpr9 = COPY [[S_MOV_B]] + ; GISEL-NEXT: $sgpr10_sgpr11 = COPY [[DEF1]] + ; GISEL-NEXT: $sgpr12 = COPY [[DEF2]] + ; GISEL-NEXT: $sgpr13 = COPY [[DEF2]] + ; GISEL-NEXT: $sgpr14 = COPY [[DEF2]] + ; GISEL-NEXT: $sgpr15 = COPY [[DEF2]] + ; GISEL-NEXT: $vgpr31 = COPY [[DEF2]] + ; GISEL-NEXT: [[S_MOV_B1:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO 0 + ; GISEL-NEXT: $sgpr30_sgpr31 = noconvergent SI_CALL [[S_MOV_B1]], 0, csr_amdgpu, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31 + ; GISEL-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32 + ; GISEL-NEXT: S_ENDPGM 0 main_body: call void null() ret void } define amdgpu_gfx void @amdgpu_gfx_call_default_cc() { - ; CHECK-LABEL: name: amdgpu_gfx_call_default_cc - ; CHECK: bb.0.main_body: - ; CHECK-NEXT: SI_RETURN + ; SDAG-LABEL: name: amdgpu_gfx_call_default_cc + ; SDAG: bb.0.main_body: + ; SDAG-NEXT: SI_RETURN + ; + ; GISEL-LABEL: name: amdgpu_gfx_call_default_cc + ; GISEL: bb.1.main_body: + ; GISEL-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32 + ; GISEL-NEXT: [[DEF:%[0-9]+]]:sreg_64 = IMPLICIT_DEF + ; GISEL-NEXT: [[DEF1:%[0-9]+]]:sreg_64 = IMPLICIT_DEF + ; GISEL-NEXT: [[DEF2:%[0-9]+]]:sreg_32 = IMPLICIT_DEF + ; GISEL-NEXT: [[COPY:%[0-9]+]]:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3 + ; GISEL-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY]] + ; GISEL-NEXT: $sgpr4_sgpr5 = COPY [[DEF]] + ; GISEL-NEXT: $sgpr6_sgpr7 = COPY [[DEF]] + ; GISEL-NEXT: [[S_MOV_B:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO 0 + ; GISEL-NEXT: $sgpr8_sgpr9 = COPY [[S_MOV_B]] + ; GISEL-NEXT: $sgpr10_sgpr11 = COPY [[DEF1]] + ; GISEL-NEXT: $sgpr12 = COPY [[DEF2]] + ; GISEL-NEXT: $sgpr13 = COPY [[DEF2]] + ; GISEL-NEXT: $sgpr14 = COPY [[DEF2]] + ; GISEL-NEXT: $sgpr15 = COPY [[DEF2]] + ; GISEL-NEXT: $vgpr31 = COPY [[DEF2]] + ; GISEL-NEXT: [[S_MOV_B1:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO 0 + ; GISEL-NEXT: $sgpr30_sgpr31 = noconvergent SI_CALL [[S_MOV_B1]], 0, csr_amdgpu, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31 + ; GISEL-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32 + ; GISEL-NEXT: SI_RETURN main_body: call void null() ret void diff --git a/llvm/test/CodeGen/AMDGPU/call-constexpr.ll b/llvm/test/CodeGen/AMDGPU/call-constexpr.ll index 5f324df..fe0b018 100644 --- a/llvm/test/CodeGen/AMDGPU/call-constexpr.ll +++ b/llvm/test/CodeGen/AMDGPU/call-constexpr.ll @@ -1,84 +1,341 @@ -; RUN: llc -mtriple=amdgcn-amd-amdhsa < %s | FileCheck -check-prefix=GCN %s +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4 +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa < %s | FileCheck -check-prefixes=GCN,SDAG %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa < %s | FileCheck -check-prefixes=GCN,GISEL %s -; GCN-LABEL: {{^}}test_bitcast_return_type_noinline: -; GCN: s_getpc_b64 -; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, ret_i32_noinline@rel32@lo+4 -; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, ret_i32_noinline@rel32@hi+12 -; GCN: s_swappc_b64 define amdgpu_kernel void @test_bitcast_return_type_noinline() #0 { +; SDAG-LABEL: test_bitcast_return_type_noinline: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, ret_i32_noinline@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, ret_i32_noinline@rel32@hi+12 +; SDAG-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; SDAG-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; SDAG-NEXT: v_or_b32_e32 v0, v0, v1 +; SDAG-NEXT: v_or_b32_e32 v31, v0, v2 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: v_add_f32_e32 v0, 1.0, v0 +; SDAG-NEXT: flat_store_dword v[0:1], v0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: test_bitcast_return_type_noinline: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; GISEL-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; GISEL-NEXT: v_or_b32_e32 v0, v0, v1 +; GISEL-NEXT: v_or_b32_e32 v31, v0, v2 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, ret_i32_noinline@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, ret_i32_noinline@rel32@hi+12 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: v_add_f32_e32 v0, 1.0, v0 +; GISEL-NEXT: flat_store_dword v[0:1], v0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = call float @ret_i32_noinline() %op = fadd float %val, 1.0 store volatile float %op, ptr addrspace(1) poison ret void } -; GCN-LABEL: {{^}}test_bitcast_return_type_alwaysinline: -; GCN: s_swappc_b64 define amdgpu_kernel void @test_bitcast_return_type_alwaysinline() #0 { +; SDAG-LABEL: test_bitcast_return_type_alwaysinline: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, ret_i32_alwaysinline@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, ret_i32_alwaysinline@rel32@hi+12 +; SDAG-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; SDAG-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; SDAG-NEXT: v_or_b32_e32 v0, v0, v1 +; SDAG-NEXT: v_or_b32_e32 v31, v0, v2 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: v_add_f32_e32 v0, 1.0, v0 +; SDAG-NEXT: flat_store_dword v[0:1], v0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: test_bitcast_return_type_alwaysinline: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; GISEL-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; GISEL-NEXT: v_or_b32_e32 v0, v0, v1 +; GISEL-NEXT: v_or_b32_e32 v31, v0, v2 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, ret_i32_alwaysinline@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, ret_i32_alwaysinline@rel32@hi+12 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: v_add_f32_e32 v0, 1.0, v0 +; GISEL-NEXT: flat_store_dword v[0:1], v0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = call float @ret_i32_alwaysinline() %op = fadd float %val, 1.0 store volatile float %op, ptr addrspace(1) poison ret void } -; GCN-LABEL: {{^}}test_bitcast_argument_type: -; GCN: s_getpc_b64 -; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, ident_i32@rel32@lo+4 -; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, ident_i32@rel32@hi+12 -; GCN: s_swappc_b64 define amdgpu_kernel void @test_bitcast_argument_type() #0 { +; SDAG-LABEL: test_bitcast_argument_type: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, ident_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, ident_i32@rel32@hi+12 +; SDAG-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; SDAG-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; SDAG-NEXT: v_or_b32_e32 v0, v0, v1 +; SDAG-NEXT: v_or_b32_e32 v31, v0, v2 +; SDAG-NEXT: v_mov_b32_e32 v0, 2.0 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: v_add_i32_e32 v0, vcc, 1, v0 +; SDAG-NEXT: flat_store_dword v[0:1], v0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: test_bitcast_argument_type: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; GISEL-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, ident_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, ident_i32@rel32@hi+12 +; GISEL-NEXT: v_or_b32_e32 v0, v0, v1 +; GISEL-NEXT: v_or_b32_e32 v31, v0, v2 +; GISEL-NEXT: v_mov_b32_e32 v0, 2.0 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: v_add_i32_e32 v0, vcc, 1, v0 +; GISEL-NEXT: flat_store_dword v[0:1], v0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = call i32 @ident_i32(float 2.0) %op = add i32 %val, 1 store volatile i32 %op, ptr addrspace(1) poison ret void } -; GCN-LABEL: {{^}}test_bitcast_argument_and_return_types: -; GCN: s_getpc_b64 -; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, ident_i32@rel32@lo+4 -; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, ident_i32@rel32@hi+12 -; GCN: s_swappc_b64 define amdgpu_kernel void @test_bitcast_argument_and_return_types() #0 { +; SDAG-LABEL: test_bitcast_argument_and_return_types: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, ident_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, ident_i32@rel32@hi+12 +; SDAG-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; SDAG-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; SDAG-NEXT: v_or_b32_e32 v0, v0, v1 +; SDAG-NEXT: v_or_b32_e32 v31, v0, v2 +; SDAG-NEXT: v_mov_b32_e32 v0, 2.0 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: v_add_f32_e32 v0, 1.0, v0 +; SDAG-NEXT: flat_store_dword v[0:1], v0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: test_bitcast_argument_and_return_types: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; GISEL-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, ident_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, ident_i32@rel32@hi+12 +; GISEL-NEXT: v_or_b32_e32 v0, v0, v1 +; GISEL-NEXT: v_or_b32_e32 v31, v0, v2 +; GISEL-NEXT: v_mov_b32_e32 v0, 2.0 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: v_add_f32_e32 v0, 1.0, v0 +; GISEL-NEXT: flat_store_dword v[0:1], v0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = call float @ident_i32(float 2.0) %op = fadd float %val, 1.0 store volatile float %op, ptr addrspace(1) poison ret void } -; GCN-LABEL: {{^}}use_workitem_id_x: -; GCN: s_waitcnt -; GCN-NEXT: v_and_b32_e32 [[TMP:v[0-9]+]], 0x3ff, v31 -; GCN-NEXT: v_add_i32_e32 v0, vcc, [[TMP]], v0 -; GCN-NEXT: s_setpc_b64 define hidden i32 @use_workitem_id_x(i32 %arg0) #3 { +; GCN-LABEL: use_workitem_id_x: +; GCN: ; %bb.0: +; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GCN-NEXT: v_and_b32_e32 v1, 0x3ff, v31 +; GCN-NEXT: v_add_i32_e32 v0, vcc, v1, v0 +; GCN-NEXT: s_setpc_b64 s[30:31] %id = call i32 @llvm.amdgcn.workitem.id.x() %op = add i32 %id, %arg0 ret i32 %op } -; GCN-LABEL: {{^}}test_bitcast_use_workitem_id_x: -; GCN: v_mov_b32_e32 v31, v0 -; GCN: s_getpc_b64 -; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, use_workitem_id_x@rel32@lo+4 -; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, use_workitem_id_x@rel32@hi+12 -; GCN: v_mov_b32_e32 v0, 9 -; GCN: s_swappc_b64 -; GCN: v_add_f32_e32 define amdgpu_kernel void @test_bitcast_use_workitem_id_x() #3 { +; SDAG-LABEL: test_bitcast_use_workitem_id_x: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: v_mov_b32_e32 v31, v0 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, use_workitem_id_x@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, use_workitem_id_x@rel32@hi+12 +; SDAG-NEXT: v_mov_b32_e32 v0, 9 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: v_add_f32_e32 v0, 1.0, v0 +; SDAG-NEXT: flat_store_dword v[0:1], v0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: test_bitcast_use_workitem_id_x: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: v_mov_b32_e32 v31, v0 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, use_workitem_id_x@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, use_workitem_id_x@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 9 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: v_add_f32_e32 v0, 1.0, v0 +; GISEL-NEXT: flat_store_dword v[0:1], v0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = call float @use_workitem_id_x(i32 9) %op = fadd float %val, 1.0 store volatile float %op, ptr addrspace(1) poison ret void } -; GCN-LABEL: {{^}}test_invoke: -; GCN: s_getpc_b64 -; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, ident_i32@rel32@lo+4 -; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, ident_i32@rel32@hi+12 -; GCN: s_swappc_b64 @_ZTIi = external global ptr declare i32 @__gxx_personality_v0(...) define amdgpu_kernel void @test_invoke() #0 personality ptr @__gxx_personality_v0 { +; SDAG-LABEL: test_invoke: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, ident_i32@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, ident_i32@rel32@hi+12 +; SDAG-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; SDAG-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; SDAG-NEXT: v_or_b32_e32 v0, v0, v1 +; SDAG-NEXT: v_or_b32_e32 v31, v0, v2 +; SDAG-NEXT: v_mov_b32_e32 v0, 2.0 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: v_add_f32_e32 v0, 1.0, v0 +; SDAG-NEXT: flat_store_dword v[0:1], v0 +; SDAG-NEXT: s_waitcnt vmcnt(0) +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: test_invoke: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; GISEL-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, ident_i32@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, ident_i32@rel32@hi+12 +; GISEL-NEXT: v_or_b32_e32 v0, v0, v1 +; GISEL-NEXT: v_or_b32_e32 v31, v0, v2 +; GISEL-NEXT: v_mov_b32_e32 v0, 2.0 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: v_add_f32_e32 v0, 1.0, v0 +; GISEL-NEXT: flat_store_dword v[0:1], v0 +; GISEL-NEXT: s_waitcnt vmcnt(0) +; GISEL-NEXT: s_endpgm %val = invoke float @ident_i32(float 2.0) to label %continue unwind label %broken @@ -96,14 +353,28 @@ continue: ; arguments before we lower any calls to them. define hidden i32 @ret_i32_noinline() #0 { +; GCN-LABEL: ret_i32_noinline: +; GCN: ; %bb.0: +; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GCN-NEXT: v_mov_b32_e32 v0, 4 +; GCN-NEXT: s_setpc_b64 s[30:31] ret i32 4 } define hidden i32 @ret_i32_alwaysinline() #1 { +; GCN-LABEL: ret_i32_alwaysinline: +; GCN: ; %bb.0: +; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GCN-NEXT: v_mov_b32_e32 v0, 4 +; GCN-NEXT: s_setpc_b64 s[30:31] ret i32 4 } define hidden i32 @ident_i32(i32 %i) #0 { +; GCN-LABEL: ident_i32: +; GCN: ; %bb.0: +; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GCN-NEXT: s_setpc_b64 s[30:31] ret i32 %i } diff --git a/llvm/test/CodeGen/AMDGPU/call-defs-mode-register.ll b/llvm/test/CodeGen/AMDGPU/call-defs-mode-register.ll index ffe536d..4b5a49f 100644 --- a/llvm/test/CodeGen/AMDGPU/call-defs-mode-register.ll +++ b/llvm/test/CodeGen/AMDGPU/call-defs-mode-register.ll @@ -1,5 +1,6 @@ ; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5 -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -simplify-mir -stop-after=finalize-isel < %s | FileCheck %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -simplify-mir -stop-after=finalize-isel < %s | FileCheck -check-prefixes=SDAG %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -simplify-mir -stop-after=finalize-isel < %s | FileCheck -check-prefixes=GISEL %s ; Check that call / asm get an implicit-def $mode added to them in ; strictfp functions. @@ -7,46 +8,80 @@ declare protected void @maybe_defs_mode() #0 define float @call_changes_mode(float %x, float %y) #0 { - ; CHECK-LABEL: name: call_changes_mode - ; CHECK: bb.0 (%ir-block.0): - ; CHECK-NEXT: liveins: $vgpr0, $vgpr1 - ; CHECK-NEXT: {{ $}} - ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1 - ; CHECK-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0 - ; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $scc, implicit-def $sgpr32, implicit $sgpr32 - ; CHECK-NEXT: [[SI_PC_ADD_REL_OFFSET:%[0-9]+]]:sreg_64 = SI_PC_ADD_REL_OFFSET target-flags(amdgpu-rel32-lo) @maybe_defs_mode, target-flags(amdgpu-rel32-hi) @maybe_defs_mode, implicit-def dead $scc - ; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3 - ; CHECK-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY2]] - ; CHECK-NEXT: $sgpr30_sgpr31 = SI_CALL killed [[SI_PC_ADD_REL_OFFSET]], @maybe_defs_mode, csr_amdgpu, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit-def $mode - ; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $scc, implicit-def $sgpr32, implicit $sgpr32 - ; CHECK-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64 0, [[COPY1]], 0, [[COPY]], 0, 0, implicit $mode, implicit $exec - ; CHECK-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]] - ; CHECK-NEXT: SI_RETURN implicit $vgpr0 + ; SDAG-LABEL: name: call_changes_mode + ; SDAG: bb.0 (%ir-block.0): + ; SDAG-NEXT: liveins: $vgpr0, $vgpr1 + ; SDAG-NEXT: {{ $}} + ; SDAG-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1 + ; SDAG-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0 + ; SDAG-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $scc, implicit-def $sgpr32, implicit $sgpr32 + ; SDAG-NEXT: [[SI_PC_ADD_REL_OFFSET:%[0-9]+]]:sreg_64 = SI_PC_ADD_REL_OFFSET target-flags(amdgpu-rel32-lo) @maybe_defs_mode, target-flags(amdgpu-rel32-hi) @maybe_defs_mode, implicit-def dead $scc + ; SDAG-NEXT: [[COPY2:%[0-9]+]]:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3 + ; SDAG-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY2]] + ; SDAG-NEXT: $sgpr30_sgpr31 = SI_CALL killed [[SI_PC_ADD_REL_OFFSET]], @maybe_defs_mode, csr_amdgpu, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit-def $mode + ; SDAG-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $scc, implicit-def $sgpr32, implicit $sgpr32 + ; SDAG-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64 0, [[COPY1]], 0, [[COPY]], 0, 0, implicit $mode, implicit $exec + ; SDAG-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]] + ; SDAG-NEXT: SI_RETURN implicit $vgpr0 + ; + ; GISEL-LABEL: name: call_changes_mode + ; GISEL: bb.1 (%ir-block.0): + ; GISEL-NEXT: liveins: $vgpr0, $vgpr1 + ; GISEL-NEXT: {{ $}} + ; GISEL-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0 + ; GISEL-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1 + ; GISEL-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32 + ; GISEL-NEXT: [[COPY2:%[0-9]+]]:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3 + ; GISEL-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY2]] + ; GISEL-NEXT: [[SI_PC_ADD_REL_OFFSET:%[0-9]+]]:sreg_64 = SI_PC_ADD_REL_OFFSET target-flags(amdgpu-rel32-lo) @maybe_defs_mode, target-flags(amdgpu-rel32-hi) @maybe_defs_mode, implicit-def $scc + ; GISEL-NEXT: $sgpr30_sgpr31 = noconvergent SI_CALL [[SI_PC_ADD_REL_OFFSET]], @maybe_defs_mode, csr_amdgpu, implicit $sgpr0_sgpr1_sgpr2_sgpr3 + ; GISEL-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32 + ; GISEL-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64 0, [[COPY]], 0, [[COPY1]], 0, 0, implicit $mode, implicit $exec + ; GISEL-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]] + ; GISEL-NEXT: SI_RETURN implicit $vgpr0 call void @maybe_defs_mode() %val = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") ret float %val } define void @tail_call_changes_mode() #0 { - ; CHECK-LABEL: name: tail_call_changes_mode - ; CHECK: bb.0 (%ir-block.0): - ; CHECK-NEXT: [[SI_PC_ADD_REL_OFFSET:%[0-9]+]]:ccr_sgpr_64 = SI_PC_ADD_REL_OFFSET target-flags(amdgpu-rel32-lo) @maybe_defs_mode, target-flags(amdgpu-rel32-hi) @maybe_defs_mode, implicit-def dead $scc - ; CHECK-NEXT: SI_TCRETURN killed [[SI_PC_ADD_REL_OFFSET]], @maybe_defs_mode, 0, csr_amdgpu, implicit-def $mode + ; SDAG-LABEL: name: tail_call_changes_mode + ; SDAG: bb.0 (%ir-block.0): + ; SDAG-NEXT: [[SI_PC_ADD_REL_OFFSET:%[0-9]+]]:ccr_sgpr_64 = SI_PC_ADD_REL_OFFSET target-flags(amdgpu-rel32-lo) @maybe_defs_mode, target-flags(amdgpu-rel32-hi) @maybe_defs_mode, implicit-def dead $scc + ; SDAG-NEXT: SI_TCRETURN killed [[SI_PC_ADD_REL_OFFSET]], @maybe_defs_mode, 0, csr_amdgpu, implicit-def $mode + ; + ; GISEL-LABEL: name: tail_call_changes_mode + ; GISEL: bb.1 (%ir-block.0): + ; GISEL-NEXT: [[COPY:%[0-9]+]]:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3 + ; GISEL-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY]] + ; GISEL-NEXT: [[SI_PC_ADD_REL_OFFSET:%[0-9]+]]:ccr_sgpr_64 = SI_PC_ADD_REL_OFFSET target-flags(amdgpu-rel32-lo) @maybe_defs_mode, target-flags(amdgpu-rel32-hi) @maybe_defs_mode, implicit-def $scc + ; GISEL-NEXT: SI_TCRETURN [[SI_PC_ADD_REL_OFFSET]], @maybe_defs_mode, 0, csr_amdgpu, implicit $sgpr0_sgpr1_sgpr2_sgpr3 tail call void @maybe_defs_mode() ret void } define float @asm_changes_mode(float %x, float %y) #0 { - ; CHECK-LABEL: name: asm_changes_mode - ; CHECK: bb.0 (%ir-block.0): - ; CHECK-NEXT: liveins: $vgpr0, $vgpr1 - ; CHECK-NEXT: {{ $}} - ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1 - ; CHECK-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0 - ; CHECK-NEXT: INLINEASM &"; maybe defs mode", 1 /* sideeffect attdialect */, implicit-def $mode - ; CHECK-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64 0, [[COPY1]], 0, [[COPY]], 0, 0, implicit $mode, implicit $exec - ; CHECK-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]] - ; CHECK-NEXT: SI_RETURN implicit $vgpr0 + ; SDAG-LABEL: name: asm_changes_mode + ; SDAG: bb.0 (%ir-block.0): + ; SDAG-NEXT: liveins: $vgpr0, $vgpr1 + ; SDAG-NEXT: {{ $}} + ; SDAG-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1 + ; SDAG-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0 + ; SDAG-NEXT: INLINEASM &"; maybe defs mode", 1 /* sideeffect attdialect */, implicit-def $mode + ; SDAG-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64 0, [[COPY1]], 0, [[COPY]], 0, 0, implicit $mode, implicit $exec + ; SDAG-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]] + ; SDAG-NEXT: SI_RETURN implicit $vgpr0 + ; + ; GISEL-LABEL: name: asm_changes_mode + ; GISEL: bb.1 (%ir-block.0): + ; GISEL-NEXT: liveins: $vgpr0, $vgpr1 + ; GISEL-NEXT: {{ $}} + ; GISEL-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0 + ; GISEL-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1 + ; GISEL-NEXT: INLINEASM &"; maybe defs mode", 1 /* sideeffect attdialect */, implicit-def $mode + ; GISEL-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64 0, [[COPY]], 0, [[COPY1]], 0, 0, implicit $mode, implicit $exec + ; GISEL-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]] + ; GISEL-NEXT: SI_RETURN implicit $vgpr0 call void asm sideeffect "; maybe defs mode", ""() %val = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") ret float %val diff --git a/llvm/test/CodeGen/AMDGPU/call-encoding.ll b/llvm/test/CodeGen/AMDGPU/call-encoding.ll index 6954c34..6c36c242 100644 --- a/llvm/test/CodeGen/AMDGPU/call-encoding.ll +++ b/llvm/test/CodeGen/AMDGPU/call-encoding.ll @@ -1,5 +1,7 @@ -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=fiji -d - | FileCheck --check-prefix=GCN %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=gfx900 -d - | FileCheck --check-prefix=GCN %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=fiji -d - | FileCheck --check-prefix=GCN %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=gfx900 -d - | FileCheck --check-prefix=GCN %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=fiji -d - | FileCheck --check-prefix=GCN %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=gfx900 -d - | FileCheck --check-prefix=GCN %s ; XUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=hawaii -d - | FileCheck --check-prefixes=GCN,CI %s ; GCN: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) diff --git a/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll b/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll index 4df1049..b250227 100644 --- a/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll +++ b/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll @@ -1,8 +1,13 @@ -; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN,CI %s -; RUN: sed 's/CODE_OBJECT_VERSION/500/g' %s | llc -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN-V5 %s -; RUN: sed 's/CODE_OBJECT_VERSION/600/g' %s | llc -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN-V5 %s -; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-ipra=0 | FileCheck -check-prefixes=GCN,VI,VI-NOBUG %s -; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -mtriple=amdgcn-amd-amdhsa -mcpu=iceland -enable-ipra=0 | FileCheck -check-prefixes=GCN,VI,VI-BUG %s +; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN,CI %s +; RUN: sed 's/CODE_OBJECT_VERSION/500/g' %s | llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN-V5 %s +; RUN: sed 's/CODE_OBJECT_VERSION/600/g' %s | llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN-V5 %s +; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-ipra=0 | FileCheck -check-prefixes=GCN,VI,VI-NOBUG %s +; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=iceland -enable-ipra=0 | FileCheck -check-prefixes=GCN,VI,VI-BUG %s +; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN,CI %s +; RUN: sed 's/CODE_OBJECT_VERSION/500/g' %s | llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN-V5 %s +; RUN: sed 's/CODE_OBJECT_VERSION/600/g' %s | llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -enable-ipra=0 | FileCheck -check-prefixes=GCN-V5 %s +; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-ipra=0 | FileCheck -check-prefixes=GCN,VI,VI-NOBUG %s +; RUN: sed 's/CODE_OBJECT_VERSION/400/g' %s | llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=iceland -enable-ipra=0 | FileCheck -check-prefixes=GCN,VI,VI-BUG %s ; Make sure to run a GPU with the SGPR allocation bug. diff --git a/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll b/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll index 61a195f..aed1079 100644 --- a/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll +++ b/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll @@ -1,8 +1,12 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5 -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-ipra=0 -mattr=+enable-flat-scratch < %s | FileCheck -check-prefixes=GCN,FLATSCR %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF,SDAG %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF,SDAG %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF,SDAG %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-ipra=0 -mattr=+enable-flat-scratch < %s | FileCheck -check-prefixes=GCN,FLATSCR,SDAG %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF,GISEL %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF,GISEL %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-ipra=0 < %s | FileCheck -check-prefixes=GCN,MUBUF,GISEL %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-ipra=0 -mattr=+enable-flat-scratch < %s | FileCheck -check-prefixes=GCN,FLATSCR,GISEL %s declare hidden void @external_void_func_void() #3 @@ -223,41 +227,6 @@ define hidden void @void_func_void_clobber_vcc() #2 { } define amdgpu_kernel void @test_call_void_func_void_clobber_vcc(ptr addrspace(1) %out) #0 { -; FLATSCR-LABEL: test_call_void_func_void_clobber_vcc: -; FLATSCR: ; %bb.0: -; FLATSCR-NEXT: s_add_u32 flat_scratch_lo, s8, s13 -; FLATSCR-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; FLATSCR-NEXT: s_add_u32 s8, s4, 8 -; FLATSCR-NEXT: s_addc_u32 s9, s5, 0 -; FLATSCR-NEXT: v_lshlrev_b32_e32 v2, 20, v2 -; FLATSCR-NEXT: v_lshlrev_b32_e32 v1, 10, v1 -; FLATSCR-NEXT: s_mov_b32 s14, s12 -; FLATSCR-NEXT: s_mov_b32 s13, s11 -; FLATSCR-NEXT: s_mov_b32 s12, s10 -; FLATSCR-NEXT: s_mov_b64 s[10:11], s[6:7] -; FLATSCR-NEXT: s_getpc_b64 s[16:17] -; FLATSCR-NEXT: s_add_u32 s16, s16, void_func_void_clobber_vcc@rel32@lo+4 -; FLATSCR-NEXT: s_addc_u32 s17, s17, void_func_void_clobber_vcc@rel32@hi+12 -; FLATSCR-NEXT: v_or3_b32 v31, v0, v1, v2 -; FLATSCR-NEXT: s_mov_b64 s[4:5], s[0:1] -; FLATSCR-NEXT: s_mov_b64 s[6:7], s[2:3] -; FLATSCR-NEXT: s_mov_b32 s32, 0 -; FLATSCR-NEXT: ;;#ASMSTART -; FLATSCR-NEXT: ; def vcc -; FLATSCR-NEXT: ;;#ASMEND -; FLATSCR-NEXT: s_mov_b64 s[34:35], vcc -; FLATSCR-NEXT: s_swappc_b64 s[30:31], s[16:17] -; FLATSCR-NEXT: global_load_dword v0, v[0:1], off glc -; FLATSCR-NEXT: s_waitcnt vmcnt(0) -; FLATSCR-NEXT: s_mov_b64 vcc, s[34:35] -; FLATSCR-NEXT: global_load_dword v0, v[0:1], off glc -; FLATSCR-NEXT: s_waitcnt vmcnt(0) -; FLATSCR-NEXT: ; kill: killed $vgpr0_vgpr1 -; FLATSCR-NEXT: ; kill: killed $vgpr0_vgpr1 -; FLATSCR-NEXT: ;;#ASMSTART -; FLATSCR-NEXT: ; use vcc -; FLATSCR-NEXT: ;;#ASMEND -; FLATSCR-NEXT: s_endpgm %vcc = call i64 asm sideeffect "; def $0", "={vcc}"() call void @void_func_void_clobber_vcc() %val0 = load volatile i32, ptr addrspace(1) poison @@ -463,51 +432,11 @@ define hidden void @void_func_void_clobber_s34() #2 { } define amdgpu_kernel void @test_call_void_func_void_clobber_s33() #0 { -; FLATSCR-LABEL: test_call_void_func_void_clobber_s33: -; FLATSCR: ; %bb.0: -; FLATSCR-NEXT: s_add_u32 flat_scratch_lo, s8, s13 -; FLATSCR-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; FLATSCR-NEXT: v_lshlrev_b32_e32 v2, 20, v2 -; FLATSCR-NEXT: v_lshlrev_b32_e32 v1, 10, v1 -; FLATSCR-NEXT: s_mov_b32 s14, s12 -; FLATSCR-NEXT: s_mov_b32 s13, s11 -; FLATSCR-NEXT: s_mov_b32 s12, s10 -; FLATSCR-NEXT: s_mov_b64 s[10:11], s[6:7] -; FLATSCR-NEXT: s_mov_b64 s[8:9], s[4:5] -; FLATSCR-NEXT: s_getpc_b64 s[16:17] -; FLATSCR-NEXT: s_add_u32 s16, s16, void_func_void_clobber_s33@rel32@lo+4 -; FLATSCR-NEXT: s_addc_u32 s17, s17, void_func_void_clobber_s33@rel32@hi+12 -; FLATSCR-NEXT: v_or3_b32 v31, v0, v1, v2 -; FLATSCR-NEXT: s_mov_b64 s[4:5], s[0:1] -; FLATSCR-NEXT: s_mov_b64 s[6:7], s[2:3] -; FLATSCR-NEXT: s_mov_b32 s32, 0 -; FLATSCR-NEXT: s_swappc_b64 s[30:31], s[16:17] -; FLATSCR-NEXT: s_endpgm call void @void_func_void_clobber_s33() ret void } define amdgpu_kernel void @test_call_void_func_void_clobber_s34() #0 { -; FLATSCR-LABEL: test_call_void_func_void_clobber_s34: -; FLATSCR: ; %bb.0: -; FLATSCR-NEXT: s_add_u32 flat_scratch_lo, s8, s13 -; FLATSCR-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; FLATSCR-NEXT: v_lshlrev_b32_e32 v2, 20, v2 -; FLATSCR-NEXT: v_lshlrev_b32_e32 v1, 10, v1 -; FLATSCR-NEXT: s_mov_b32 s14, s12 -; FLATSCR-NEXT: s_mov_b32 s13, s11 -; FLATSCR-NEXT: s_mov_b32 s12, s10 -; FLATSCR-NEXT: s_mov_b64 s[10:11], s[6:7] -; FLATSCR-NEXT: s_mov_b64 s[8:9], s[4:5] -; FLATSCR-NEXT: s_getpc_b64 s[16:17] -; FLATSCR-NEXT: s_add_u32 s16, s16, void_func_void_clobber_s34@rel32@lo+4 -; FLATSCR-NEXT: s_addc_u32 s17, s17, void_func_void_clobber_s34@rel32@hi+12 -; FLATSCR-NEXT: v_or3_b32 v31, v0, v1, v2 -; FLATSCR-NEXT: s_mov_b64 s[4:5], s[0:1] -; FLATSCR-NEXT: s_mov_b64 s[6:7], s[2:3] -; FLATSCR-NEXT: s_mov_b32 s32, 0 -; FLATSCR-NEXT: s_swappc_b64 s[30:31], s[16:17] -; FLATSCR-NEXT: s_endpgm call void @void_func_void_clobber_s34() ret void } @@ -748,3 +677,6 @@ attributes #0 = { nounwind } attributes #1 = { nounwind readnone } attributes #2 = { nounwind noinline } attributes #3 = { nounwind "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-cluster-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-cluster-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-cluster-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" } +;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line: +; GISEL: {{.*}} +; SDAG: {{.*}} diff --git a/llvm/test/CodeGen/AMDGPU/call-return-types.ll b/llvm/test/CodeGen/AMDGPU/call-return-types.ll index c0f74fd..21c3696 100644 --- a/llvm/test/CodeGen/AMDGPU/call-return-types.ll +++ b/llvm/test/CodeGen/AMDGPU/call-return-types.ll @@ -1,7 +1,12 @@ -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX89 %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii < %s | FileCheck -check-prefixes=GCN,GFX7 %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX89 %s -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX11 %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX89 %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii < %s | FileCheck -check-prefixes=GCN,GFX7 %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX89 %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX11 %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX89 %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii < %s | FileCheck -check-prefixes=GCN,GFX7 %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=GCN,GFX89 %s + +; Ideally, we would also like to test GlobalISel with gfx11 but we are currently blocked on llvm-project#166501. declare void @external_void_func_void() #0 diff --git a/llvm/test/CodeGen/AMDGPU/call-skip.ll b/llvm/test/CodeGen/AMDGPU/call-skip.ll index ea2bba1..e2ca278 100644 --- a/llvm/test/CodeGen/AMDGPU/call-skip.ll +++ b/llvm/test/CodeGen/AMDGPU/call-skip.ll @@ -1,4 +1,6 @@ -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii < %s | FileCheck -enable-var-scope -check-prefix=GCN %s +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc -mtriple=amdgcn-amd-amdhsa -global-isel=0 -mcpu=hawaii < %s | FileCheck -enable-var-scope -check-prefixes=GCN,SDAG %s +; RUN: llc -mtriple=amdgcn-amd-amdhsa -global-isel=1 -mcpu=hawaii < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GISEL %s ; A call should be skipped if all lanes are zero, since we don't know ; what side effects should be avoided inside the call. @@ -6,12 +8,37 @@ define hidden void @func() #1 { ret void } -; GCN-LABEL: {{^}}if_call: -; GCN: s_and_saveexec_b64 -; GCN-NEXT: s_cbranch_execz [[END:.LBB[0-9]+_[0-9]+]] -; GCN: s_swappc_b64 -; GCN: [[END]]: define void @if_call(i32 %flag) #0 { +; GCN-LABEL: if_call: +; GCN: ; %bb.0: +; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GCN-NEXT: s_mov_b32 s20, s33 +; GCN-NEXT: s_mov_b32 s33, s32 +; GCN-NEXT: s_xor_saveexec_b64 s[16:17], -1 +; GCN-NEXT: buffer_store_dword v1, off, s[0:3], s33 ; 4-byte Folded Spill +; GCN-NEXT: s_mov_b64 exec, s[16:17] +; GCN-NEXT: v_writelane_b32 v1, s30, 0 +; GCN-NEXT: v_cmp_eq_u32_e32 vcc, 0, v0 +; GCN-NEXT: s_addk_i32 s32, 0x400 +; GCN-NEXT: v_writelane_b32 v1, s31, 1 +; GCN-NEXT: s_and_saveexec_b64 s[16:17], vcc +; GCN-NEXT: s_cbranch_execz .LBB1_2 +; GCN-NEXT: ; %bb.1: ; %call +; GCN-NEXT: s_getpc_b64 s[18:19] +; GCN-NEXT: s_add_u32 s18, s18, func@rel32@lo+4 +; GCN-NEXT: s_addc_u32 s19, s19, func@rel32@hi+12 +; GCN-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GCN-NEXT: .LBB1_2: ; %end +; GCN-NEXT: s_or_b64 exec, exec, s[16:17] +; GCN-NEXT: v_readlane_b32 s31, v1, 1 +; GCN-NEXT: v_readlane_b32 s30, v1, 0 +; GCN-NEXT: s_mov_b32 s32, s33 +; GCN-NEXT: s_xor_saveexec_b64 s[4:5], -1 +; GCN-NEXT: buffer_load_dword v1, off, s[0:3], s33 ; 4-byte Folded Reload +; GCN-NEXT: s_mov_b64 exec, s[4:5] +; GCN-NEXT: s_mov_b32 s33, s20 +; GCN-NEXT: s_waitcnt vmcnt(0) +; GCN-NEXT: s_setpc_b64 s[30:31] %cc = icmp eq i32 %flag, 0 br i1 %cc, label %call, label %end @@ -23,12 +50,20 @@ end: ret void } -; GCN-LABEL: {{^}}if_asm: -; GCN: s_and_saveexec_b64 -; GCN-NEXT: s_cbranch_execz [[END:.LBB[0-9]+_[0-9]+]] -; GCN: ; sample asm -; GCN: [[END]]: define void @if_asm(i32 %flag) #0 { +; GCN-LABEL: if_asm: +; GCN: ; %bb.0: +; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GCN-NEXT: v_cmp_eq_u32_e32 vcc, 0, v0 +; GCN-NEXT: s_and_saveexec_b64 s[4:5], vcc +; GCN-NEXT: s_cbranch_execz .LBB2_2 +; GCN-NEXT: ; %bb.1: ; %call +; GCN-NEXT: ;;#ASMSTART +; GCN-NEXT: ; sample asm +; GCN-NEXT: ;;#ASMEND +; GCN-NEXT: .LBB2_2: ; %end +; GCN-NEXT: s_or_b64 exec, exec, s[4:5] +; GCN-NEXT: s_setpc_b64 s[30:31] %cc = icmp eq i32 %flag, 0 br i1 %cc, label %call, label %end @@ -40,11 +75,58 @@ end: ret void } -; GCN-LABEL: {{^}}if_call_kernel: -; GCN: s_and_saveexec_b64 -; GCN-NEXT: s_cbranch_execz .LBB3_2 -; GCN: s_swappc_b64 define amdgpu_kernel void @if_call_kernel() #0 { +; SDAG-LABEL: if_call_kernel: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_add_i32 s12, s12, s17 +; SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; SDAG-NEXT: s_add_u32 s0, s0, s17 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: v_cmp_eq_u32_e32 vcc, 0, v0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; SDAG-NEXT: s_and_saveexec_b64 s[12:13], vcc +; SDAG-NEXT: s_cbranch_execz .LBB3_2 +; SDAG-NEXT: ; %bb.1: ; %call +; SDAG-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; SDAG-NEXT: v_lshlrev_b32_e32 v2, 20, v2 +; SDAG-NEXT: v_or_b32_e32 v0, v0, v1 +; SDAG-NEXT: s_getpc_b64 s[18:19] +; SDAG-NEXT: s_add_u32 s18, s18, func@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s19, s19, func@rel32@hi+12 +; SDAG-NEXT: v_or_b32_e32 v31, v0, v2 +; SDAG-NEXT: s_mov_b32 s12, s14 +; SDAG-NEXT: s_mov_b32 s13, s15 +; SDAG-NEXT: s_mov_b32 s14, s16 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[18:19] +; SDAG-NEXT: .LBB3_2: ; %end +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: if_call_kernel: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_add_i32 s12, s12, s17 +; GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; GISEL-NEXT: s_add_u32 s0, s0, s17 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v0 +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; GISEL-NEXT: s_and_saveexec_b64 s[12:13], vcc +; GISEL-NEXT: s_cbranch_execz .LBB3_2 +; GISEL-NEXT: ; %bb.1: ; %call +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 10, v1 +; GISEL-NEXT: v_or_b32_e32 v0, v0, v1 +; GISEL-NEXT: v_lshlrev_b32_e32 v1, 20, v2 +; GISEL-NEXT: s_getpc_b64 s[18:19] +; GISEL-NEXT: s_add_u32 s18, s18, func@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s19, s19, func@rel32@hi+12 +; GISEL-NEXT: v_or_b32_e32 v31, v0, v1 +; GISEL-NEXT: s_mov_b32 s12, s14 +; GISEL-NEXT: s_mov_b32 s13, s15 +; GISEL-NEXT: s_mov_b32 s14, s16 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[18:19] +; GISEL-NEXT: .LBB3_2: ; %end +; GISEL-NEXT: s_endpgm %id = call i32 @llvm.amdgcn.workitem.id.x() %cc = icmp eq i32 %id, 0 br i1 %cc, label %call, label %end diff --git a/llvm/test/CodeGen/AMDGPU/call-waitcnt.ll b/llvm/test/CodeGen/AMDGPU/call-waitcnt.ll index 675acd0..a52942c 100644 --- a/llvm/test/CodeGen/AMDGPU/call-waitcnt.ll +++ b/llvm/test/CodeGen/AMDGPU/call-waitcnt.ll @@ -1,5 +1,6 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py -; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefix=GCN %s +; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefixes=GCN,SDAG %s +; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GISEL %s ; Load argument depends on waitcnt which should be skipped. define amdgpu_kernel void @call_memory_arg_load(ptr addrspace(3) %ptr, i32) #0 { @@ -27,24 +28,43 @@ define amdgpu_kernel void @call_memory_arg_load(ptr addrspace(3) %ptr, i32) #0 { ; Memory waitcnt with no register dependence on the call define amdgpu_kernel void @call_memory_no_dep(ptr addrspace(1) %ptr, i32) #0 { -; GCN-LABEL: call_memory_no_dep: -; GCN: ; %bb.0: -; GCN-NEXT: s_load_dwordx2 s[6:7], s[6:7], 0x0 -; GCN-NEXT: s_add_u32 flat_scratch_lo, s8, s11 -; GCN-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; GCN-NEXT: s_add_u32 s0, s0, s11 -; GCN-NEXT: s_addc_u32 s1, s1, 0 -; GCN-NEXT: v_mov_b32_e32 v0, 0 -; GCN-NEXT: s_getpc_b64 s[8:9] -; GCN-NEXT: s_add_u32 s8, s8, func@rel32@lo+4 -; GCN-NEXT: s_addc_u32 s9, s9, func@rel32@hi+12 -; GCN-NEXT: s_waitcnt lgkmcnt(0) -; GCN-NEXT: global_store_dword v0, v0, s[6:7] -; GCN-NEXT: s_mov_b64 s[6:7], s[4:5] -; GCN-NEXT: v_mov_b32_e32 v0, 0 -; GCN-NEXT: s_mov_b32 s32, 0 -; GCN-NEXT: s_swappc_b64 s[30:31], s[8:9] -; GCN-NEXT: s_endpgm +; SDAG-LABEL: call_memory_no_dep: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_load_dwordx2 s[6:7], s[6:7], 0x0 +; SDAG-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; SDAG-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; SDAG-NEXT: s_add_u32 s0, s0, s11 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: s_getpc_b64 s[8:9] +; SDAG-NEXT: s_add_u32 s8, s8, func@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s9, s9, func@rel32@hi+12 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: global_store_dword v0, v0, s[6:7] +; SDAG-NEXT: s_mov_b64 s[6:7], s[4:5] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[8:9] +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: call_memory_no_dep: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_load_dwordx2 s[6:7], s[6:7], 0x0 +; GISEL-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; GISEL-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; GISEL-NEXT: s_add_u32 s0, s0, s11 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: s_getpc_b64 s[8:9] +; GISEL-NEXT: s_add_u32 s8, s8, func@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s9, s9, func@rel32@hi+12 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: global_store_dword v0, v0, s[6:7] +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[4:5] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[8:9] +; GISEL-NEXT: s_endpgm store i32 0, ptr addrspace(1) %ptr call void @func(i32 0) ret void @@ -52,46 +72,82 @@ define amdgpu_kernel void @call_memory_no_dep(ptr addrspace(1) %ptr, i32) #0 { ; Should not wait after the call before memory define amdgpu_kernel void @call_no_wait_after_call(ptr addrspace(1) %ptr, i32) #0 { -; GCN-LABEL: call_no_wait_after_call: -; GCN: ; %bb.0: -; GCN-NEXT: s_add_u32 flat_scratch_lo, s8, s11 -; GCN-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; GCN-NEXT: s_load_dwordx2 s[34:35], s[6:7], 0x0 -; GCN-NEXT: s_add_u32 s0, s0, s11 -; GCN-NEXT: s_addc_u32 s1, s1, 0 -; GCN-NEXT: s_getpc_b64 s[8:9] -; GCN-NEXT: s_add_u32 s8, s8, func@rel32@lo+4 -; GCN-NEXT: s_addc_u32 s9, s9, func@rel32@hi+12 -; GCN-NEXT: s_mov_b64 s[6:7], s[4:5] -; GCN-NEXT: v_mov_b32_e32 v0, 0 -; GCN-NEXT: s_mov_b32 s32, 0 -; GCN-NEXT: v_mov_b32_e32 v40, 0 -; GCN-NEXT: s_swappc_b64 s[30:31], s[8:9] -; GCN-NEXT: global_store_dword v40, v40, s[34:35] -; GCN-NEXT: s_endpgm +; SDAG-LABEL: call_no_wait_after_call: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; SDAG-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; SDAG-NEXT: s_load_dwordx2 s[34:35], s[6:7], 0x0 +; SDAG-NEXT: s_add_u32 s0, s0, s11 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_getpc_b64 s[8:9] +; SDAG-NEXT: s_add_u32 s8, s8, func@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s9, s9, func@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[6:7], s[4:5] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: v_mov_b32_e32 v40, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[8:9] +; SDAG-NEXT: global_store_dword v40, v40, s[34:35] +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: call_no_wait_after_call: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; GISEL-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; GISEL-NEXT: s_load_dwordx2 s[34:35], s[6:7], 0x0 +; GISEL-NEXT: s_add_u32 s0, s0, s11 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_getpc_b64 s[8:9] +; GISEL-NEXT: s_add_u32 s8, s8, func@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s9, s9, func@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[4:5] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[8:9] +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: global_store_dword v0, v0, s[34:35] +; GISEL-NEXT: s_endpgm call void @func(i32 0) store i32 0, ptr addrspace(1) %ptr ret void } define amdgpu_kernel void @call_no_wait_after_call_return_val(ptr addrspace(1) %ptr, i32) #0 { -; GCN-LABEL: call_no_wait_after_call_return_val: -; GCN: ; %bb.0: -; GCN-NEXT: s_add_u32 flat_scratch_lo, s8, s11 -; GCN-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; GCN-NEXT: s_load_dwordx2 s[34:35], s[6:7], 0x0 -; GCN-NEXT: s_add_u32 s0, s0, s11 -; GCN-NEXT: s_addc_u32 s1, s1, 0 -; GCN-NEXT: s_getpc_b64 s[8:9] -; GCN-NEXT: s_add_u32 s8, s8, func.return@rel32@lo+4 -; GCN-NEXT: s_addc_u32 s9, s9, func.return@rel32@hi+12 -; GCN-NEXT: s_mov_b64 s[6:7], s[4:5] -; GCN-NEXT: v_mov_b32_e32 v0, 0 -; GCN-NEXT: s_mov_b32 s32, 0 -; GCN-NEXT: v_mov_b32_e32 v40, 0 -; GCN-NEXT: s_swappc_b64 s[30:31], s[8:9] -; GCN-NEXT: global_store_dword v40, v0, s[34:35] -; GCN-NEXT: s_endpgm +; SDAG-LABEL: call_no_wait_after_call_return_val: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; SDAG-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; SDAG-NEXT: s_load_dwordx2 s[34:35], s[6:7], 0x0 +; SDAG-NEXT: s_add_u32 s0, s0, s11 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_getpc_b64 s[8:9] +; SDAG-NEXT: s_add_u32 s8, s8, func.return@rel32@lo+4 +; SDAG-NEXT: s_addc_u32 s9, s9, func.return@rel32@hi+12 +; SDAG-NEXT: s_mov_b64 s[6:7], s[4:5] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: v_mov_b32_e32 v40, 0 +; SDAG-NEXT: s_swappc_b64 s[30:31], s[8:9] +; SDAG-NEXT: global_store_dword v40, v0, s[34:35] +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: call_no_wait_after_call_return_val: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; GISEL-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; GISEL-NEXT: s_load_dwordx2 s[34:35], s[6:7], 0x0 +; GISEL-NEXT: s_add_u32 s0, s0, s11 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_getpc_b64 s[8:9] +; GISEL-NEXT: s_add_u32 s8, s8, func.return@rel32@lo+4 +; GISEL-NEXT: s_addc_u32 s9, s9, func.return@rel32@hi+12 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[4:5] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_swappc_b64 s[30:31], s[8:9] +; GISEL-NEXT: v_mov_b32_e32 v1, 0 +; GISEL-NEXT: global_store_dword v1, v0, s[34:35] +; GISEL-NEXT: s_endpgm %rv = call i32 @func.return(i32 0) store i32 %rv, ptr addrspace(1) %ptr ret void @@ -99,22 +155,39 @@ define amdgpu_kernel void @call_no_wait_after_call_return_val(ptr addrspace(1) % ; Need to wait for the address dependency define amdgpu_kernel void @call_got_load(ptr addrspace(1) %ptr, i32) #0 { -; GCN-LABEL: call_got_load: -; GCN: ; %bb.0: -; GCN-NEXT: s_add_u32 flat_scratch_lo, s8, s11 -; GCN-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 -; GCN-NEXT: s_add_u32 s0, s0, s11 -; GCN-NEXT: s_addc_u32 s1, s1, 0 -; GCN-NEXT: s_getpc_b64 s[6:7] -; GCN-NEXT: s_add_u32 s6, s6, got.func@gotpcrel32@lo+4 -; GCN-NEXT: s_addc_u32 s7, s7, got.func@gotpcrel32@hi+12 -; GCN-NEXT: s_load_dwordx2 s[8:9], s[6:7], 0x0 -; GCN-NEXT: s_mov_b64 s[6:7], s[4:5] -; GCN-NEXT: v_mov_b32_e32 v0, 0 -; GCN-NEXT: s_mov_b32 s32, 0 -; GCN-NEXT: s_waitcnt lgkmcnt(0) -; GCN-NEXT: s_swappc_b64 s[30:31], s[8:9] -; GCN-NEXT: s_endpgm +; SDAG-LABEL: call_got_load: +; SDAG: ; %bb.0: +; SDAG-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; SDAG-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; SDAG-NEXT: s_add_u32 s0, s0, s11 +; SDAG-NEXT: s_addc_u32 s1, s1, 0 +; SDAG-NEXT: s_getpc_b64 s[6:7] +; SDAG-NEXT: s_add_u32 s6, s6, got.func@gotpcrel32@lo+4 +; SDAG-NEXT: s_addc_u32 s7, s7, got.func@gotpcrel32@hi+12 +; SDAG-NEXT: s_load_dwordx2 s[8:9], s[6:7], 0x0 +; SDAG-NEXT: s_mov_b64 s[6:7], s[4:5] +; SDAG-NEXT: v_mov_b32_e32 v0, 0 +; SDAG-NEXT: s_mov_b32 s32, 0 +; SDAG-NEXT: s_waitcnt lgkmcnt(0) +; SDAG-NEXT: s_swappc_b64 s[30:31], s[8:9] +; SDAG-NEXT: s_endpgm +; +; GISEL-LABEL: call_got_load: +; GISEL: ; %bb.0: +; GISEL-NEXT: s_add_u32 flat_scratch_lo, s8, s11 +; GISEL-NEXT: s_addc_u32 flat_scratch_hi, s9, 0 +; GISEL-NEXT: s_add_u32 s0, s0, s11 +; GISEL-NEXT: s_addc_u32 s1, s1, 0 +; GISEL-NEXT: s_getpc_b64 s[6:7] +; GISEL-NEXT: s_add_u32 s6, s6, got.func@gotpcrel32@lo+4 +; GISEL-NEXT: s_addc_u32 s7, s7, got.func@gotpcrel32@hi+12 +; GISEL-NEXT: s_load_dwordx2 s[8:9], s[6:7], 0x0 +; GISEL-NEXT: v_mov_b32_e32 v0, 0 +; GISEL-NEXT: s_mov_b64 s[6:7], s[4:5] +; GISEL-NEXT: s_mov_b32 s32, 0 +; GISEL-NEXT: s_waitcnt lgkmcnt(0) +; GISEL-NEXT: s_swappc_b64 s[30:31], s[8:9] +; GISEL-NEXT: s_endpgm call void @got.func(i32 0) ret void } diff --git a/llvm/test/CodeGen/AMDGPU/carryout-selection.ll b/llvm/test/CodeGen/AMDGPU/carryout-selection.ll index b96de17..8d05317 100644 --- a/llvm/test/CodeGen/AMDGPU/carryout-selection.ll +++ b/llvm/test/CodeGen/AMDGPU/carryout-selection.ll @@ -702,8 +702,6 @@ define amdgpu_kernel void @suaddo64(ptr addrspace(1) %out, ptr addrspace(1) %car ; CISI-NEXT: s_mov_b32 s10, -1 ; CISI-NEXT: s_waitcnt lgkmcnt(0) ; CISI-NEXT: s_add_u32 s4, s4, s6 -; CISI-NEXT: s_cselect_b64 s[12:13], -1, 0 -; CISI-NEXT: s_or_b32 s6, s12, s13 ; CISI-NEXT: s_addc_u32 s5, s5, s7 ; CISI-NEXT: s_mov_b32 s8, s0 ; CISI-NEXT: s_mov_b32 s9, s1 @@ -1674,8 +1672,6 @@ define amdgpu_kernel void @susubo64(ptr addrspace(1) %out, ptr addrspace(1) %car ; CISI-NEXT: s_mov_b32 s10, -1 ; CISI-NEXT: s_waitcnt lgkmcnt(0) ; CISI-NEXT: s_sub_u32 s4, s4, s6 -; CISI-NEXT: s_cselect_b64 s[12:13], -1, 0 -; CISI-NEXT: s_or_b32 s6, s12, s13 ; CISI-NEXT: s_subb_u32 s5, s5, s7 ; CISI-NEXT: s_mov_b32 s8, s0 ; CISI-NEXT: s_mov_b32 s9, s1 diff --git a/llvm/test/CodeGen/AMDGPU/expand-scalar-carry-out-select-user.ll b/llvm/test/CodeGen/AMDGPU/expand-scalar-carry-out-select-user.ll index dbdea8e..71af21a1 100644 --- a/llvm/test/CodeGen/AMDGPU/expand-scalar-carry-out-select-user.ll +++ b/llvm/test/CodeGen/AMDGPU/expand-scalar-carry-out-select-user.ll @@ -12,8 +12,6 @@ define i32 @s_add_co_select_user() { ; GFX7-NEXT: s_load_dword s6, s[4:5], 0x0 ; GFX7-NEXT: s_waitcnt lgkmcnt(0) ; GFX7-NEXT: s_add_u32 s7, s6, s6 -; GFX7-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GFX7-NEXT: s_or_b32 s4, s4, s5 ; GFX7-NEXT: s_addc_u32 s8, s6, 0 ; GFX7-NEXT: s_cselect_b64 s[4:5], -1, 0 ; GFX7-NEXT: s_and_b64 s[4:5], s[4:5], exec @@ -88,15 +86,13 @@ bb: define amdgpu_kernel void @s_add_co_br_user(i32 %i) { ; GFX7-LABEL: s_add_co_br_user: ; GFX7: ; %bb.0: ; %bb -; GFX7-NEXT: s_load_dword s2, s[8:9], 0x0 +; GFX7-NEXT: s_load_dword s0, s[8:9], 0x0 ; GFX7-NEXT: s_add_i32 s12, s12, s17 ; GFX7-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 ; GFX7-NEXT: s_mov_b32 flat_scratch_lo, s13 ; GFX7-NEXT: s_waitcnt lgkmcnt(0) -; GFX7-NEXT: s_add_u32 s0, s2, s2 -; GFX7-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GFX7-NEXT: s_or_b32 s0, s0, s1 -; GFX7-NEXT: s_addc_u32 s0, s2, 0 +; GFX7-NEXT: s_add_u32 s1, s0, s0 +; GFX7-NEXT: s_addc_u32 s0, s0, 0 ; GFX7-NEXT: s_cselect_b64 s[0:1], -1, 0 ; GFX7-NEXT: s_andn2_b64 vcc, exec, s[0:1] ; GFX7-NEXT: s_cbranch_vccnz .LBB1_2 diff --git a/llvm/test/CodeGen/AMDGPU/fix-sgpr-copies-wwm.ll b/llvm/test/CodeGen/AMDGPU/fix-sgpr-copies-wwm.ll index b8f084d..db32135 100644 --- a/llvm/test/CodeGen/AMDGPU/fix-sgpr-copies-wwm.ll +++ b/llvm/test/CodeGen/AMDGPU/fix-sgpr-copies-wwm.ll @@ -4,14 +4,24 @@ define amdgpu_gs i32 @main() { ; CHECK-LABEL: main: ; CHECK: ; %bb.0: ; %bb +; CHECK-NEXT: s_bitcmp1_b32 0, 0 ; CHECK-NEXT: s_mov_b32 s0, 0 -; CHECK-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(SALU_CYCLE_1) +; CHECK-NEXT: s_cselect_b32 s1, -1, 0 +; CHECK-NEXT: s_or_saveexec_b32 s2, -1 +; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, 1, s1 +; CHECK-NEXT: s_delay_alu instid0(VALU_DEP_1) +; CHECK-NEXT: v_readfirstlane_b32 s1, v0 +; CHECK-NEXT: s_mov_b32 exec_lo, s2 +; CHECK-NEXT: s_or_b32 s0, s0, s1 +; CHECK-NEXT: s_wait_alu 0xfffe ; CHECK-NEXT: s_bitcmp1_b32 s0, 0 ; CHECK-NEXT: s_cselect_b32 s0, -1, 0 +; CHECK-NEXT: s_wait_alu 0xfffe ; CHECK-NEXT: s_xor_b32 s0, s0, -1 -; CHECK-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) -; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, 1, s0 -; CHECK-NEXT: v_readfirstlane_b32 s0, v0 +; CHECK-NEXT: s_wait_alu 0xfffe +; CHECK-NEXT: v_cndmask_b32_e64 v1, 0, 1, s0 +; CHECK-NEXT: s_delay_alu instid0(VALU_DEP_1) +; CHECK-NEXT: v_readfirstlane_b32 s0, v1 ; CHECK-NEXT: s_wait_alu 0xf1ff ; CHECK-NEXT: ; return to shader part epilog bb: diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.readfirstlane.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.readfirstlane.ll index 02d2990..d1ba892 100644 --- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.readfirstlane.ll +++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.readfirstlane.ll @@ -396,7 +396,8 @@ define amdgpu_kernel void @test_readfirstlane_imm_f64(ptr addrspace(1) %out) { ; ; CHECK-GISEL-LABEL: test_readfirstlane_imm_f64: ; CHECK-GISEL: ; %bb.0: -; CHECK-GISEL-NEXT: s_mov_b64 s[0:1], 0x4040000000000000 +; CHECK-GISEL-NEXT: s_mov_b32 s0, 0 +; CHECK-GISEL-NEXT: s_mov_b32 s1, 0x40400000 ; CHECK-GISEL-NEXT: ;;#ASMSTART ; CHECK-GISEL-NEXT: ; use s[0:1] ; CHECK-GISEL-NEXT: ;;#ASMEND @@ -455,13 +456,14 @@ define amdgpu_kernel void @test_readfirstlane_imm_fold_i64(ptr addrspace(1) %out ; CHECK-GISEL-LABEL: test_readfirstlane_imm_fold_i64: ; CHECK-GISEL: ; %bb.0: ; CHECK-GISEL-NEXT: s_load_dwordx2 s[0:1], s[8:9], 0x0 +; CHECK-GISEL-NEXT: s_mov_b64 s[2:3], 32 ; CHECK-GISEL-NEXT: s_add_i32 s12, s12, s17 -; CHECK-GISEL-NEXT: v_mov_b32_e32 v0, 32 +; CHECK-GISEL-NEXT: v_mov_b32_e32 v0, s2 ; CHECK-GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 -; CHECK-GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 ; CHECK-GISEL-NEXT: s_waitcnt lgkmcnt(0) ; CHECK-GISEL-NEXT: v_mov_b32_e32 v3, s1 -; CHECK-GISEL-NEXT: v_mov_b32_e32 v1, 0 +; CHECK-GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; CHECK-GISEL-NEXT: v_mov_b32_e32 v1, s3 ; CHECK-GISEL-NEXT: v_mov_b32_e32 v2, s0 ; CHECK-GISEL-NEXT: flat_store_dwordx2 v[2:3], v[0:1] ; CHECK-GISEL-NEXT: s_endpgm @@ -488,13 +490,15 @@ define amdgpu_kernel void @test_readfirstlane_imm_fold_f64(ptr addrspace(1) %out ; CHECK-GISEL-LABEL: test_readfirstlane_imm_fold_f64: ; CHECK-GISEL: ; %bb.0: ; CHECK-GISEL-NEXT: s_load_dwordx2 s[0:1], s[8:9], 0x0 +; CHECK-GISEL-NEXT: s_mov_b32 s2, 0 ; CHECK-GISEL-NEXT: s_add_i32 s12, s12, s17 -; CHECK-GISEL-NEXT: v_mov_b32_e32 v0, 0 -; CHECK-GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 -; CHECK-GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; CHECK-GISEL-NEXT: s_mov_b32 s3, 0x40400000 +; CHECK-GISEL-NEXT: v_mov_b32_e32 v0, s2 ; CHECK-GISEL-NEXT: s_waitcnt lgkmcnt(0) ; CHECK-GISEL-NEXT: v_mov_b32_e32 v3, s1 -; CHECK-GISEL-NEXT: v_mov_b32_e32 v1, 0x40400000 +; CHECK-GISEL-NEXT: s_mov_b32 flat_scratch_lo, s13 +; CHECK-GISEL-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 +; CHECK-GISEL-NEXT: v_mov_b32_e32 v1, s3 ; CHECK-GISEL-NEXT: v_mov_b32_e32 v2, s0 ; CHECK-GISEL-NEXT: flat_store_dwordx2 v[2:3], v[0:1] ; CHECK-GISEL-NEXT: s_endpgm @@ -584,17 +588,17 @@ define amdgpu_kernel void @test_readfirstlane_copy_from_sgpr_i64(ptr addrspace(1 ; CHECK-SDAG: ; %bb.0: ; CHECK-SDAG-NEXT: s_load_dwordx2 s[0:1], s[8:9], 0x0 ; CHECK-SDAG-NEXT: s_add_i32 s12, s12, s17 +; CHECK-SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; CHECK-SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 ; CHECK-SDAG-NEXT: ;;#ASMSTART ; CHECK-SDAG-NEXT: s_mov_b64 s[2:3], 0 ; CHECK-SDAG-NEXT: ;;#ASMEND -; CHECK-SDAG-NEXT: v_mov_b32_e32 v2, s2 -; CHECK-SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 ; CHECK-SDAG-NEXT: s_waitcnt lgkmcnt(0) -; CHECK-SDAG-NEXT: v_mov_b32_e32 v0, s0 -; CHECK-SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 -; CHECK-SDAG-NEXT: v_mov_b32_e32 v1, s1 -; CHECK-SDAG-NEXT: v_mov_b32_e32 v3, s3 -; CHECK-SDAG-NEXT: flat_store_dwordx2 v[0:1], v[2:3] +; CHECK-SDAG-NEXT: v_mov_b32_e32 v3, s1 +; CHECK-SDAG-NEXT: v_mov_b32_e32 v0, s2 +; CHECK-SDAG-NEXT: v_mov_b32_e32 v1, s3 +; CHECK-SDAG-NEXT: v_mov_b32_e32 v2, s0 +; CHECK-SDAG-NEXT: flat_store_dwordx2 v[2:3], v[0:1] ; CHECK-SDAG-NEXT: s_endpgm ; ; CHECK-GISEL-LABEL: test_readfirstlane_copy_from_sgpr_i64: @@ -624,17 +628,17 @@ define amdgpu_kernel void @test_readfirstlane_copy_from_sgpr_f64(ptr addrspace(1 ; CHECK-SDAG: ; %bb.0: ; CHECK-SDAG-NEXT: s_load_dwordx2 s[0:1], s[8:9], 0x0 ; CHECK-SDAG-NEXT: s_add_i32 s12, s12, s17 +; CHECK-SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 +; CHECK-SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 ; CHECK-SDAG-NEXT: ;;#ASMSTART ; CHECK-SDAG-NEXT: s_mov_b64 s[2:3], 0 ; CHECK-SDAG-NEXT: ;;#ASMEND -; CHECK-SDAG-NEXT: v_mov_b32_e32 v2, s2 -; CHECK-SDAG-NEXT: s_mov_b32 flat_scratch_lo, s13 ; CHECK-SDAG-NEXT: s_waitcnt lgkmcnt(0) -; CHECK-SDAG-NEXT: v_mov_b32_e32 v0, s0 -; CHECK-SDAG-NEXT: s_lshr_b32 flat_scratch_hi, s12, 8 -; CHECK-SDAG-NEXT: v_mov_b32_e32 v1, s1 -; CHECK-SDAG-NEXT: v_mov_b32_e32 v3, s3 -; CHECK-SDAG-NEXT: flat_store_dwordx2 v[0:1], v[2:3] +; CHECK-SDAG-NEXT: v_mov_b32_e32 v3, s1 +; CHECK-SDAG-NEXT: v_mov_b32_e32 v0, s2 +; CHECK-SDAG-NEXT: v_mov_b32_e32 v1, s3 +; CHECK-SDAG-NEXT: v_mov_b32_e32 v2, s0 +; CHECK-SDAG-NEXT: flat_store_dwordx2 v[2:3], v[0:1] ; CHECK-SDAG-NEXT: s_endpgm ; ; CHECK-GISEL-LABEL: test_readfirstlane_copy_from_sgpr_f64: diff --git a/llvm/test/CodeGen/AMDGPU/optimize-compare.mir b/llvm/test/CodeGen/AMDGPU/optimize-compare.mir index fba42c4..fa452f3 100644 --- a/llvm/test/CodeGen/AMDGPU/optimize-compare.mir +++ b/llvm/test/CodeGen/AMDGPU/optimize-compare.mir @@ -2277,3 +2277,181 @@ body: | S_ENDPGM 0 ... + +--- +name: s_cselect_b64_s_or_b32_s_cmp_lg_u32_0x00000000 +body: | + ; GCN-LABEL: name: s_cselect_b64_s_or_b32_s_cmp_lg_u32_0x00000000 + ; GCN: bb.0: + ; GCN-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000) + ; GCN-NEXT: liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF + ; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY [[DEF]] + ; GCN-NEXT: S_CMP_LG_U32 [[COPY]], 0, implicit-def $scc + ; GCN-NEXT: [[S_CSELECT_B64_:%[0-9]+]]:sreg_64_xexec = S_CSELECT_B64 -1, 0, implicit $scc + ; GCN-NEXT: [[COPY1:%[0-9]+]]:sreg_32_xm0_xexec = COPY [[S_CSELECT_B64_]].sub0 + ; GCN-NEXT: [[COPY2:%[0-9]+]]:sreg_32_xm0_xexec = COPY [[S_CSELECT_B64_]].sub1 + ; GCN-NEXT: S_CBRANCH_SCC0 %bb.2, implicit $scc + ; GCN-NEXT: S_BRANCH %bb.1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.1: + ; GCN-NEXT: successors: %bb.2(0x80000000) + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.2: + ; GCN-NEXT: S_ENDPGM 0 + bb.0: + successors: %bb.1(0x40000000), %bb.2(0x40000000) + liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + %0:vgpr_32 = IMPLICIT_DEF + %2:sreg_32 = COPY %0 + S_CMP_LG_U32 %2, 0, implicit-def $scc + %31:sreg_64_xexec = S_CSELECT_B64 -1, 0, implicit $scc + %40:sreg_32_xm0_xexec = COPY %31.sub0:sreg_64_xexec + %41:sreg_32_xm0_xexec = COPY %31.sub1:sreg_64_xexec + %sgpr4:sreg_32 = S_OR_B32 %40:sreg_32_xm0_xexec, %41:sreg_32_xm0_xexec, implicit-def $scc + S_CMP_LG_U32 %sgpr4, 0, implicit-def $scc + S_CBRANCH_SCC0 %bb.2, implicit $scc + S_BRANCH %bb.1 + + bb.1: + successors: %bb.2(0x80000000) + + bb.2: + S_ENDPGM 0 + +... +--- +# Do not delete s_or_b32 because of intervening def of scc +name: s_cselect_b64_s_or_b32_s_cmp_lg_u32_0x00000000_cant_optimize_intervening +body: | + ; GCN-LABEL: name: s_cselect_b64_s_or_b32_s_cmp_lg_u32_0x00000000_cant_optimize_intervening + ; GCN: bb.0: + ; GCN-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000) + ; GCN-NEXT: liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF + ; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY [[DEF]] + ; GCN-NEXT: S_CMP_LG_U32 [[COPY]], 0, implicit-def $scc + ; GCN-NEXT: [[S_CSELECT_B64_:%[0-9]+]]:sreg_64_xexec = S_CSELECT_B64 -1, 0, implicit $scc + ; GCN-NEXT: S_CMP_LG_U32 [[COPY]], 0, implicit-def $scc + ; GCN-NEXT: [[COPY1:%[0-9]+]]:sreg_32_xm0_xexec = COPY [[S_CSELECT_B64_]].sub0 + ; GCN-NEXT: [[COPY2:%[0-9]+]]:sreg_32_xm0_xexec = COPY [[S_CSELECT_B64_]].sub1 + ; GCN-NEXT: %sgpr4:sreg_32 = S_OR_B32 [[COPY1]], [[COPY2]], implicit-def $scc + ; GCN-NEXT: S_CBRANCH_SCC0 %bb.2, implicit $scc + ; GCN-NEXT: S_BRANCH %bb.1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.1: + ; GCN-NEXT: successors: %bb.2(0x80000000) + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.2: + ; GCN-NEXT: S_ENDPGM 0 + bb.0: + successors: %bb.1(0x40000000), %bb.2(0x40000000) + liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + %0:vgpr_32 = IMPLICIT_DEF + %2:sreg_32 = COPY %0 + S_CMP_LG_U32 %2, 0, implicit-def $scc + %31:sreg_64_xexec = S_CSELECT_B64 -1, 0, implicit $scc + S_CMP_LG_U32 %2, 0, implicit-def $scc + %40:sreg_32_xm0_xexec = COPY %31.sub0:sreg_64_xexec + %41:sreg_32_xm0_xexec = COPY %31.sub1:sreg_64_xexec + %sgpr4:sreg_32 = S_OR_B32 %40:sreg_32_xm0_xexec, %41:sreg_32_xm0_xexec, implicit-def $scc + S_CMP_LG_U32 %sgpr4, 0, implicit-def $scc + S_CBRANCH_SCC0 %bb.2, implicit $scc + S_BRANCH %bb.1 + + bb.1: + successors: %bb.2(0x80000000) + + bb.2: + S_ENDPGM 0 + +... + +--- +# Do not delete s_or_b32 since both operands are sub1. +name: s_cselect_b64_s_or_b32_s_cmp_lg_u32_0x00000000_cant_optimize +body: | + ; GCN-LABEL: name: s_cselect_b64_s_or_b32_s_cmp_lg_u32_0x00000000_cant_optimize + ; GCN: bb.0: + ; GCN-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000) + ; GCN-NEXT: liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF + ; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY [[DEF]] + ; GCN-NEXT: S_CMP_LG_U32 [[COPY]], 0, implicit-def $scc + ; GCN-NEXT: [[S_CSELECT_B64_:%[0-9]+]]:sreg_64_xexec = S_CSELECT_B64 1, 0, implicit $scc + ; GCN-NEXT: [[COPY1:%[0-9]+]]:sreg_32_xm0_xexec = COPY [[S_CSELECT_B64_]].sub1 + ; GCN-NEXT: [[COPY2:%[0-9]+]]:sreg_32 = COPY [[S_CSELECT_B64_]].sub1 + ; GCN-NEXT: %sgpr4:sreg_32 = S_OR_B32 [[COPY1]], [[COPY2]], implicit-def $scc + ; GCN-NEXT: S_CBRANCH_SCC0 %bb.2, implicit $scc + ; GCN-NEXT: S_BRANCH %bb.1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.1: + ; GCN-NEXT: successors: %bb.2(0x80000000) + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.2: + ; GCN-NEXT: S_ENDPGM 0 + bb.0: + successors: %bb.1(0x40000000), %bb.2(0x40000000) + liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + %0:vgpr_32 = IMPLICIT_DEF + %2:sreg_32 = COPY %0 + S_CMP_LG_U32 %2, 0, implicit-def $scc + %31:sreg_64_xexec = S_CSELECT_B64 1, 0, implicit $scc + %40:sreg_32_xm0_xexec = COPY %31.sub1:sreg_64_xexec + %41:sreg_32 = COPY %31.sub1:sreg_64_xexec + %sgpr4:sreg_32 = S_OR_B32 %40:sreg_32_xm0_xexec, %41:sreg_32, implicit-def $scc + S_CMP_LG_U32 %sgpr4, 0, implicit-def $scc + S_CBRANCH_SCC0 %bb.2, implicit $scc + S_BRANCH %bb.1 + + bb.1: + successors: %bb.2(0x80000000) + + bb.2: + S_ENDPGM 0 + +... + +--- +name: s_cselect_b64_undef_s_or_b32_s_cmp_lg_u32_0x00000000 +body: | + ; GCN-LABEL: name: s_cselect_b64_undef_s_or_b32_s_cmp_lg_u32_0x00000000 + ; GCN: bb.0: + ; GCN-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000) + ; GCN-NEXT: liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF + ; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY [[DEF]] + ; GCN-NEXT: S_CMP_LG_U32 [[COPY]], 0, implicit-def $scc + ; GCN-NEXT: [[S_CSELECT_B64_:%[0-9]+]]:sreg_64_xexec = S_CSELECT_B64 -1, 0, implicit $scc + ; GCN-NEXT: %sgpr4:sreg_32 = S_OR_B32 undef %4:sreg_32_xm0_xexec, undef %5:sreg_32_xm0_xexec, implicit-def $scc + ; GCN-NEXT: S_CBRANCH_SCC0 %bb.2, implicit $scc + ; GCN-NEXT: S_BRANCH %bb.1 + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.1: + ; GCN-NEXT: successors: %bb.2(0x80000000) + ; GCN-NEXT: {{ $}} + ; GCN-NEXT: bb.2: + ; GCN-NEXT: S_ENDPGM 0 + bb.0: + successors: %bb.1(0x40000000), %bb.2(0x40000000) + liveins: $sgpr0_sgpr1, $vgpr0_vgpr1 + %0:vgpr_32 = IMPLICIT_DEF + %2:sreg_32 = COPY %0 + S_CMP_LG_U32 %2, 0, implicit-def $scc + %31:sreg_64_xexec = S_CSELECT_B64 -1, 0, implicit $scc + %sgpr4:sreg_32 = S_OR_B32 undef %40:sreg_32_xm0_xexec, undef %41:sreg_32_xm0_xexec, implicit-def $scc + S_CMP_LG_U32 %sgpr4, 0, implicit-def $scc + S_CBRANCH_SCC0 %bb.2, implicit $scc + S_BRANCH %bb.1 + + bb.1: + successors: %bb.2(0x80000000) + + bb.2: + S_ENDPGM 0 + +... diff --git a/llvm/test/CodeGen/AMDGPU/sdiv64.ll b/llvm/test/CodeGen/AMDGPU/sdiv64.ll index 71f5a94..74a6d7f 100644 --- a/llvm/test/CodeGen/AMDGPU/sdiv64.ll +++ b/llvm/test/CodeGen/AMDGPU/sdiv64.ll @@ -8,7 +8,6 @@ define amdgpu_kernel void @s_test_sdiv(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0xd ; GCN-NEXT: s_load_dwordx4 s[4:7], s[4:5], 0x9 ; GCN-NEXT: s_mov_b32 s3, 0xf000 -; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: s_waitcnt lgkmcnt(0) ; GCN-NEXT: s_ashr_i32 s8, s1, 31 ; GCN-NEXT: s_add_u32 s0, s0, s8 @@ -17,8 +16,8 @@ define amdgpu_kernel void @s_test_sdiv(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_xor_b64 s[10:11], s[0:1], s[8:9] ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s10 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s11 -; GCN-NEXT: s_sub_u32 s12, 0, s10 -; GCN-NEXT: s_subb_u32 s13, 0, s11 +; GCN-NEXT: s_sub_u32 s0, 0, s10 +; GCN-NEXT: s_subb_u32 s1, 0, s11 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -27,128 +26,121 @@ define amdgpu_kernel void @s_test_sdiv(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GCN-NEXT: v_mul_hi_u32 v2, s12, v0 -; GCN-NEXT: v_readfirstlane_b32 s14, v1 -; GCN-NEXT: v_readfirstlane_b32 s0, v0 -; GCN-NEXT: s_mul_i32 s1, s12, s14 -; GCN-NEXT: v_readfirstlane_b32 s17, v2 -; GCN-NEXT: s_mul_i32 s15, s13, s0 -; GCN-NEXT: s_mul_i32 s16, s12, s0 -; GCN-NEXT: s_add_i32 s1, s17, s1 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s16 -; GCN-NEXT: s_add_i32 s1, s1, s15 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s1 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s16 -; GCN-NEXT: v_readfirstlane_b32 s15, v3 -; GCN-NEXT: s_mul_i32 s17, s0, s1 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s1 -; GCN-NEXT: s_add_u32 s15, s15, s17 -; GCN-NEXT: v_readfirstlane_b32 s17, v0 -; GCN-NEXT: s_addc_u32 s17, 0, s17 -; GCN-NEXT: s_mul_i32 s16, s14, s16 -; GCN-NEXT: v_readfirstlane_b32 s18, v4 -; GCN-NEXT: s_add_u32 s15, s15, s16 -; GCN-NEXT: s_addc_u32 s15, s17, s18 -; GCN-NEXT: v_readfirstlane_b32 s16, v1 -; GCN-NEXT: s_addc_u32 s16, s16, 0 -; GCN-NEXT: s_mul_i32 s1, s14, s1 -; GCN-NEXT: s_add_u32 s1, s15, s1 -; GCN-NEXT: s_addc_u32 s15, 0, s16 -; GCN-NEXT: s_add_u32 s16, s0, s1 -; GCN-NEXT: v_mov_b32_e32 v0, s16 -; GCN-NEXT: v_mul_hi_u32 v0, s12, v0 -; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_addc_u32 s14, s14, s15 -; GCN-NEXT: s_mul_i32 s0, s12, s14 -; GCN-NEXT: v_readfirstlane_b32 s1, v0 -; GCN-NEXT: s_add_i32 s0, s1, s0 -; GCN-NEXT: s_mul_i32 s13, s13, s16 -; GCN-NEXT: s_mul_i32 s1, s12, s16 -; GCN-NEXT: s_add_i32 s0, s0, s13 -; GCN-NEXT: v_mov_b32_e32 v2, s1 -; GCN-NEXT: v_mov_b32_e32 v0, s0 -; GCN-NEXT: v_mul_hi_u32 v3, s14, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s16, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s14, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s16, v0 -; GCN-NEXT: s_mul_i32 s13, s16, s0 -; GCN-NEXT: v_readfirstlane_b32 s17, v2 -; GCN-NEXT: s_add_u32 s13, s17, s13 -; GCN-NEXT: v_readfirstlane_b32 s15, v0 -; GCN-NEXT: s_mul_i32 s1, s14, s1 -; GCN-NEXT: s_addc_u32 s15, 0, s15 -; GCN-NEXT: v_readfirstlane_b32 s12, v3 -; GCN-NEXT: s_add_u32 s1, s13, s1 -; GCN-NEXT: s_addc_u32 s1, s15, s12 +; GCN-NEXT: v_mul_hi_u32 v2, s0, v0 ; GCN-NEXT: v_readfirstlane_b32 s12, v1 -; GCN-NEXT: s_addc_u32 s12, s12, 0 -; GCN-NEXT: s_mul_i32 s0, s14, s0 -; GCN-NEXT: s_add_u32 s0, s1, s0 -; GCN-NEXT: s_addc_u32 s12, 0, s12 -; GCN-NEXT: s_add_u32 s15, s16, s0 -; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_addc_u32 s14, s14, s12 +; GCN-NEXT: v_readfirstlane_b32 s2, v0 +; GCN-NEXT: s_mul_i32 s13, s0, s12 +; GCN-NEXT: v_readfirstlane_b32 s16, v2 +; GCN-NEXT: s_mul_i32 s14, s1, s2 +; GCN-NEXT: s_mul_i32 s15, s0, s2 +; GCN-NEXT: s_add_i32 s13, s16, s13 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s15 +; GCN-NEXT: s_add_i32 s13, s13, s14 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s13 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s15 +; GCN-NEXT: v_readfirstlane_b32 s14, v3 +; GCN-NEXT: s_mul_i32 s16, s2, s13 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s13 +; GCN-NEXT: s_add_u32 s14, s14, s16 +; GCN-NEXT: v_readfirstlane_b32 s16, v0 +; GCN-NEXT: s_mul_i32 s15, s12, s15 +; GCN-NEXT: s_addc_u32 s16, 0, s16 +; GCN-NEXT: v_readfirstlane_b32 s17, v4 +; GCN-NEXT: s_add_u32 s14, s14, s15 +; GCN-NEXT: s_addc_u32 s14, s16, s17 +; GCN-NEXT: v_readfirstlane_b32 s15, v1 +; GCN-NEXT: s_addc_u32 s15, s15, 0 +; GCN-NEXT: s_mul_i32 s13, s12, s13 +; GCN-NEXT: s_add_u32 s13, s14, s13 +; GCN-NEXT: s_addc_u32 s14, 0, s15 +; GCN-NEXT: s_add_u32 s13, s2, s13 +; GCN-NEXT: v_mov_b32_e32 v0, s13 +; GCN-NEXT: v_mul_hi_u32 v0, s0, v0 +; GCN-NEXT: s_addc_u32 s12, s12, s14 +; GCN-NEXT: s_mul_i32 s14, s0, s12 +; GCN-NEXT: s_mul_i32 s1, s1, s13 +; GCN-NEXT: v_readfirstlane_b32 s15, v0 +; GCN-NEXT: s_add_i32 s14, s15, s14 +; GCN-NEXT: s_mul_i32 s0, s0, s13 +; GCN-NEXT: s_add_i32 s1, s14, s1 +; GCN-NEXT: v_mov_b32_e32 v2, s0 +; GCN-NEXT: v_mov_b32_e32 v0, s1 +; GCN-NEXT: v_mul_hi_u32 v3, s12, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s13, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s12, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s13, v0 +; GCN-NEXT: s_mul_i32 s15, s13, s1 +; GCN-NEXT: v_readfirstlane_b32 s17, v2 +; GCN-NEXT: s_add_u32 s15, s17, s15 +; GCN-NEXT: v_readfirstlane_b32 s16, v0 +; GCN-NEXT: s_mul_i32 s0, s12, s0 +; GCN-NEXT: s_addc_u32 s16, 0, s16 +; GCN-NEXT: v_readfirstlane_b32 s14, v3 +; GCN-NEXT: s_add_u32 s0, s15, s0 +; GCN-NEXT: s_addc_u32 s0, s16, s14 +; GCN-NEXT: v_readfirstlane_b32 s14, v1 +; GCN-NEXT: s_addc_u32 s14, s14, 0 +; GCN-NEXT: s_mul_i32 s1, s12, s1 +; GCN-NEXT: s_add_u32 s0, s0, s1 +; GCN-NEXT: s_addc_u32 s1, 0, s14 +; GCN-NEXT: s_add_u32 s14, s13, s0 +; GCN-NEXT: s_addc_u32 s15, s12, s1 ; GCN-NEXT: s_ashr_i32 s12, s7, 31 ; GCN-NEXT: s_add_u32 s0, s6, s12 ; GCN-NEXT: s_mov_b32 s13, s12 ; GCN-NEXT: s_addc_u32 s1, s7, s12 ; GCN-NEXT: s_xor_b64 s[6:7], s[0:1], s[12:13] -; GCN-NEXT: v_mov_b32_e32 v0, s14 +; GCN-NEXT: v_mov_b32_e32 v0, s15 ; GCN-NEXT: v_mul_hi_u32 v1, s6, v0 -; GCN-NEXT: v_mov_b32_e32 v2, s15 +; GCN-NEXT: v_mov_b32_e32 v2, s14 ; GCN-NEXT: v_mul_hi_u32 v3, s6, v2 ; GCN-NEXT: s_mov_b32 s0, s4 ; GCN-NEXT: v_readfirstlane_b32 s4, v1 ; GCN-NEXT: v_mul_hi_u32 v1, s7, v2 -; GCN-NEXT: s_mul_i32 s1, s6, s14 +; GCN-NEXT: s_mul_i32 s1, s6, s15 ; GCN-NEXT: v_readfirstlane_b32 s16, v3 ; GCN-NEXT: v_mul_hi_u32 v0, s7, v0 ; GCN-NEXT: s_add_u32 s1, s16, s1 ; GCN-NEXT: s_addc_u32 s4, 0, s4 -; GCN-NEXT: s_mul_i32 s15, s7, s15 +; GCN-NEXT: s_mul_i32 s14, s7, s14 ; GCN-NEXT: v_readfirstlane_b32 s16, v1 -; GCN-NEXT: s_add_u32 s1, s1, s15 +; GCN-NEXT: s_add_u32 s1, s1, s14 ; GCN-NEXT: s_addc_u32 s1, s4, s16 ; GCN-NEXT: v_readfirstlane_b32 s4, v0 ; GCN-NEXT: s_addc_u32 s4, s4, 0 -; GCN-NEXT: s_mul_i32 s14, s7, s14 -; GCN-NEXT: s_add_u32 s16, s1, s14 -; GCN-NEXT: v_mov_b32_e32 v0, s16 +; GCN-NEXT: s_mul_i32 s14, s7, s15 +; GCN-NEXT: s_add_u32 s14, s1, s14 +; GCN-NEXT: v_mov_b32_e32 v0, s14 ; GCN-NEXT: v_mul_hi_u32 v0, s10, v0 -; GCN-NEXT: s_addc_u32 s17, 0, s4 +; GCN-NEXT: s_addc_u32 s15, 0, s4 ; GCN-NEXT: s_mov_b32 s1, s5 -; GCN-NEXT: s_mul_i32 s4, s10, s17 +; GCN-NEXT: s_mul_i32 s4, s10, s15 ; GCN-NEXT: v_readfirstlane_b32 s5, v0 ; GCN-NEXT: s_add_i32 s4, s5, s4 -; GCN-NEXT: s_mul_i32 s5, s11, s16 -; GCN-NEXT: s_add_i32 s18, s4, s5 -; GCN-NEXT: s_sub_i32 s14, s7, s18 -; GCN-NEXT: s_mul_i32 s4, s10, s16 +; GCN-NEXT: s_mul_i32 s5, s11, s14 +; GCN-NEXT: s_add_i32 s16, s4, s5 +; GCN-NEXT: s_sub_i32 s17, s7, s16 +; GCN-NEXT: s_mul_i32 s4, s10, s14 ; GCN-NEXT: s_sub_u32 s6, s6, s4 ; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s15, s4, s5 -; GCN-NEXT: s_subb_u32 s19, s14, s11 -; GCN-NEXT: s_sub_u32 s20, s6, s10 -; GCN-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GCN-NEXT: s_or_b32 s14, s14, s15 -; GCN-NEXT: s_subb_u32 s14, s19, 0 -; GCN-NEXT: s_cmp_ge_u32 s14, s11 -; GCN-NEXT: s_cselect_b32 s15, -1, 0 -; GCN-NEXT: s_cmp_ge_u32 s20, s10 +; GCN-NEXT: s_subb_u32 s17, s17, s11 +; GCN-NEXT: s_sub_u32 s18, s6, s10 +; GCN-NEXT: s_subb_u32 s17, s17, 0 +; GCN-NEXT: s_cmp_ge_u32 s17, s11 ; GCN-NEXT: s_cselect_b32 s19, -1, 0 -; GCN-NEXT: s_cmp_eq_u32 s14, s11 -; GCN-NEXT: s_cselect_b32 s14, s19, s15 -; GCN-NEXT: s_add_u32 s15, s16, 1 -; GCN-NEXT: s_addc_u32 s19, s17, 0 -; GCN-NEXT: s_add_u32 s20, s16, 2 -; GCN-NEXT: s_addc_u32 s21, s17, 0 -; GCN-NEXT: s_cmp_lg_u32 s14, 0 -; GCN-NEXT: s_cselect_b32 s14, s20, s15 -; GCN-NEXT: s_cselect_b32 s15, s21, s19 +; GCN-NEXT: s_cmp_ge_u32 s18, s10 +; GCN-NEXT: s_cselect_b32 s18, -1, 0 +; GCN-NEXT: s_cmp_eq_u32 s17, s11 +; GCN-NEXT: s_cselect_b32 s17, s18, s19 +; GCN-NEXT: s_add_u32 s18, s14, 1 +; GCN-NEXT: s_addc_u32 s19, s15, 0 +; GCN-NEXT: s_add_u32 s20, s14, 2 +; GCN-NEXT: s_addc_u32 s21, s15, 0 +; GCN-NEXT: s_cmp_lg_u32 s17, 0 +; GCN-NEXT: s_cselect_b32 s17, s20, s18 +; GCN-NEXT: s_cselect_b32 s18, s21, s19 ; GCN-NEXT: s_or_b32 s4, s4, s5 -; GCN-NEXT: s_subb_u32 s4, s7, s18 +; GCN-NEXT: s_subb_u32 s4, s7, s16 ; GCN-NEXT: s_cmp_ge_u32 s4, s11 ; GCN-NEXT: s_cselect_b32 s5, -1, 0 ; GCN-NEXT: s_cmp_ge_u32 s6, s10 @@ -156,13 +148,14 @@ define amdgpu_kernel void @s_test_sdiv(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_cmp_eq_u32 s4, s11 ; GCN-NEXT: s_cselect_b32 s4, s6, s5 ; GCN-NEXT: s_cmp_lg_u32 s4, 0 -; GCN-NEXT: s_cselect_b32 s5, s15, s17 -; GCN-NEXT: s_cselect_b32 s4, s14, s16 +; GCN-NEXT: s_cselect_b32 s5, s18, s15 +; GCN-NEXT: s_cselect_b32 s4, s17, s14 ; GCN-NEXT: s_xor_b64 s[6:7], s[12:13], s[8:9] ; GCN-NEXT: s_xor_b64 s[4:5], s[4:5], s[6:7] ; GCN-NEXT: s_sub_u32 s4, s4, s6 ; GCN-NEXT: s_subb_u32 s5, s5, s7 ; GCN-NEXT: v_mov_b32_e32 v0, s4 +; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_mov_b32_e32 v1, s5 ; GCN-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0 ; GCN-NEXT: s_endpgm @@ -202,8 +195,6 @@ define amdgpu_kernel void @s_test_sdiv(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-IR-NEXT: s_cbranch_vccz .LBB0_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s18, s16, 1 -; GCN-IR-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-IR-NEXT: s_or_b32 s10, s10, s11 ; GCN-IR-NEXT: s_addc_u32 s10, s17, 0 ; GCN-IR-NEXT: s_cselect_b64 s[10:11], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s16, 63, s16 @@ -235,8 +226,6 @@ define amdgpu_kernel void @s_test_sdiv(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-IR-NEXT: s_sub_u32 s16, s16, s20 ; GCN-IR-NEXT: s_subb_u32 s17, s17, s21 ; GCN-IR-NEXT: s_add_u32 s14, s14, 1 -; GCN-IR-NEXT: s_cselect_b64 s[20:21], -1, 0 -; GCN-IR-NEXT: s_or_b32 s20, s20, s21 ; GCN-IR-NEXT: s_addc_u32 s15, s15, 0 ; GCN-IR-NEXT: s_cselect_b64 s[20:21], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[12:13], s[8:9] @@ -1150,8 +1139,7 @@ define amdgpu_kernel void @s_test_sdiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s6 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s7 ; GCN-NEXT: s_sub_u32 s2, 0, s6 -; GCN-NEXT: s_subb_u32 s10, 0, s7 -; GCN-NEXT: s_mov_b32 s3, 0xf000 +; GCN-NEXT: s_subb_u32 s8, 0, s7 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -1161,115 +1149,109 @@ define amdgpu_kernel void @s_test_sdiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 ; GCN-NEXT: v_mul_hi_u32 v2, s2, v0 -; GCN-NEXT: v_readfirstlane_b32 s11, v1 -; GCN-NEXT: v_readfirstlane_b32 s8, v0 -; GCN-NEXT: s_mul_i32 s9, s2, s11 -; GCN-NEXT: v_readfirstlane_b32 s14, v2 -; GCN-NEXT: s_mul_i32 s12, s10, s8 -; GCN-NEXT: s_mul_i32 s13, s2, s8 -; GCN-NEXT: s_add_i32 s9, s14, s9 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s13 -; GCN-NEXT: s_add_i32 s9, s9, s12 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s9 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s13 -; GCN-NEXT: v_readfirstlane_b32 s12, v3 -; GCN-NEXT: s_mul_i32 s15, s8, s9 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s9 -; GCN-NEXT: s_add_u32 s12, s12, s15 -; GCN-NEXT: v_readfirstlane_b32 s15, v0 -; GCN-NEXT: s_mul_i32 s13, s11, s13 -; GCN-NEXT: s_addc_u32 s15, 0, s15 -; GCN-NEXT: v_readfirstlane_b32 s14, v4 -; GCN-NEXT: s_add_u32 s12, s12, s13 -; GCN-NEXT: s_addc_u32 s12, s15, s14 -; GCN-NEXT: v_readfirstlane_b32 s13, v1 -; GCN-NEXT: s_addc_u32 s13, s13, 0 -; GCN-NEXT: s_mul_i32 s9, s11, s9 -; GCN-NEXT: s_add_u32 s9, s12, s9 -; GCN-NEXT: s_addc_u32 s12, 0, s13 -; GCN-NEXT: s_add_u32 s13, s8, s9 -; GCN-NEXT: v_mov_b32_e32 v0, s13 +; GCN-NEXT: v_readfirstlane_b32 s9, v1 +; GCN-NEXT: v_readfirstlane_b32 s3, v0 +; GCN-NEXT: s_mul_i32 s10, s2, s9 +; GCN-NEXT: v_readfirstlane_b32 s13, v2 +; GCN-NEXT: s_mul_i32 s11, s8, s3 +; GCN-NEXT: s_mul_i32 s12, s2, s3 +; GCN-NEXT: s_add_i32 s10, s13, s10 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s12 +; GCN-NEXT: s_add_i32 s10, s10, s11 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s10 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s12 +; GCN-NEXT: v_readfirstlane_b32 s11, v3 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s10 +; GCN-NEXT: s_mul_i32 s14, s3, s10 +; GCN-NEXT: s_add_u32 s11, s11, s14 +; GCN-NEXT: v_readfirstlane_b32 s14, v0 +; GCN-NEXT: s_mul_i32 s12, s9, s12 +; GCN-NEXT: s_addc_u32 s14, 0, s14 +; GCN-NEXT: v_readfirstlane_b32 s13, v4 +; GCN-NEXT: s_add_u32 s11, s11, s12 +; GCN-NEXT: v_readfirstlane_b32 s15, v1 +; GCN-NEXT: s_addc_u32 s11, s14, s13 +; GCN-NEXT: s_addc_u32 s12, s15, 0 +; GCN-NEXT: s_mul_i32 s10, s9, s10 +; GCN-NEXT: s_add_u32 s10, s11, s10 +; GCN-NEXT: s_addc_u32 s11, 0, s12 +; GCN-NEXT: s_add_u32 s10, s3, s10 +; GCN-NEXT: v_mov_b32_e32 v0, s10 ; GCN-NEXT: v_mul_hi_u32 v0, s2, v0 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_addc_u32 s11, s11, s12 -; GCN-NEXT: s_mul_i32 s8, s2, s11 -; GCN-NEXT: v_readfirstlane_b32 s9, v0 -; GCN-NEXT: s_add_i32 s8, s9, s8 -; GCN-NEXT: s_mul_i32 s10, s10, s13 -; GCN-NEXT: s_mul_i32 s2, s2, s13 -; GCN-NEXT: s_add_i32 s8, s8, s10 +; GCN-NEXT: s_addc_u32 s9, s9, s11 +; GCN-NEXT: s_mul_i32 s11, s2, s9 +; GCN-NEXT: s_mul_i32 s8, s8, s10 +; GCN-NEXT: v_readfirstlane_b32 s12, v0 +; GCN-NEXT: s_add_i32 s11, s12, s11 +; GCN-NEXT: s_mul_i32 s2, s2, s10 +; GCN-NEXT: s_add_i32 s8, s11, s8 ; GCN-NEXT: v_mov_b32_e32 v2, s2 ; GCN-NEXT: v_mov_b32_e32 v0, s8 -; GCN-NEXT: v_mul_hi_u32 v3, s11, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s13, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s11, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s13, v0 -; GCN-NEXT: s_mul_i32 s10, s13, s8 +; GCN-NEXT: v_mul_hi_u32 v3, s9, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s10, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s9, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s10, v0 +; GCN-NEXT: s_mul_i32 s12, s10, s8 ; GCN-NEXT: v_readfirstlane_b32 s14, v2 -; GCN-NEXT: s_add_u32 s10, s14, s10 -; GCN-NEXT: v_readfirstlane_b32 s12, v0 -; GCN-NEXT: s_mul_i32 s2, s11, s2 -; GCN-NEXT: s_addc_u32 s12, 0, s12 -; GCN-NEXT: v_readfirstlane_b32 s9, v3 -; GCN-NEXT: s_add_u32 s2, s10, s2 -; GCN-NEXT: s_addc_u32 s2, s12, s9 -; GCN-NEXT: v_readfirstlane_b32 s9, v1 -; GCN-NEXT: s_addc_u32 s9, s9, 0 -; GCN-NEXT: s_mul_i32 s8, s11, s8 +; GCN-NEXT: s_add_u32 s12, s14, s12 +; GCN-NEXT: v_readfirstlane_b32 s13, v0 +; GCN-NEXT: s_mul_i32 s2, s9, s2 +; GCN-NEXT: s_addc_u32 s13, 0, s13 +; GCN-NEXT: v_readfirstlane_b32 s11, v3 +; GCN-NEXT: s_add_u32 s2, s12, s2 +; GCN-NEXT: s_addc_u32 s2, s13, s11 +; GCN-NEXT: v_readfirstlane_b32 s11, v1 +; GCN-NEXT: s_addc_u32 s11, s11, 0 +; GCN-NEXT: s_mul_i32 s8, s9, s8 ; GCN-NEXT: s_add_u32 s2, s2, s8 -; GCN-NEXT: s_addc_u32 s10, 0, s9 -; GCN-NEXT: s_add_u32 s2, s13, s2 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_addc_u32 s8, s11, s10 +; GCN-NEXT: s_addc_u32 s8, 0, s11 +; GCN-NEXT: s_add_u32 s2, s10, s2 +; GCN-NEXT: s_addc_u32 s8, s9, s8 ; GCN-NEXT: v_mul_hi_u32 v1, s2, 24 ; GCN-NEXT: v_mul_hi_u32 v0, s8, 24 ; GCN-NEXT: s_mul_i32 s8, s8, 24 -; GCN-NEXT: s_mov_b32 s2, -1 +; GCN-NEXT: s_mov_b32 s3, 0xf000 ; GCN-NEXT: v_readfirstlane_b32 s10, v1 ; GCN-NEXT: v_readfirstlane_b32 s9, v0 ; GCN-NEXT: s_add_u32 s8, s10, s8 -; GCN-NEXT: s_addc_u32 s12, 0, s9 -; GCN-NEXT: v_mov_b32_e32 v0, s12 +; GCN-NEXT: s_addc_u32 s10, 0, s9 +; GCN-NEXT: v_mov_b32_e32 v0, s10 ; GCN-NEXT: v_mul_hi_u32 v0, s6, v0 -; GCN-NEXT: s_mul_i32 s8, s7, s12 +; GCN-NEXT: s_mul_i32 s8, s7, s10 +; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_readfirstlane_b32 s9, v0 -; GCN-NEXT: s_add_i32 s13, s9, s8 -; GCN-NEXT: s_sub_i32 s10, 0, s13 -; GCN-NEXT: s_mul_i32 s8, s6, s12 -; GCN-NEXT: s_sub_u32 s14, 24, s8 +; GCN-NEXT: s_add_i32 s11, s9, s8 +; GCN-NEXT: s_sub_i32 s12, 0, s11 +; GCN-NEXT: s_mul_i32 s8, s6, s10 +; GCN-NEXT: s_sub_u32 s13, 24, s8 ; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s11, s8, s9 -; GCN-NEXT: s_subb_u32 s15, s10, s7 -; GCN-NEXT: s_sub_u32 s16, s14, s6 -; GCN-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-NEXT: s_or_b32 s10, s10, s11 -; GCN-NEXT: s_subb_u32 s10, s15, 0 -; GCN-NEXT: s_cmp_ge_u32 s10, s7 -; GCN-NEXT: s_cselect_b32 s11, -1, 0 -; GCN-NEXT: s_cmp_ge_u32 s16, s6 +; GCN-NEXT: s_subb_u32 s12, s12, s7 +; GCN-NEXT: s_sub_u32 s14, s13, s6 +; GCN-NEXT: s_subb_u32 s12, s12, 0 +; GCN-NEXT: s_cmp_ge_u32 s12, s7 ; GCN-NEXT: s_cselect_b32 s15, -1, 0 -; GCN-NEXT: s_cmp_eq_u32 s10, s7 -; GCN-NEXT: s_cselect_b32 s10, s15, s11 -; GCN-NEXT: s_add_u32 s11, s12, 1 +; GCN-NEXT: s_cmp_ge_u32 s14, s6 +; GCN-NEXT: s_cselect_b32 s14, -1, 0 +; GCN-NEXT: s_cmp_eq_u32 s12, s7 +; GCN-NEXT: s_cselect_b32 s12, s14, s15 +; GCN-NEXT: s_add_u32 s14, s10, 1 ; GCN-NEXT: s_addc_u32 s15, 0, 0 -; GCN-NEXT: s_add_u32 s16, s12, 2 +; GCN-NEXT: s_add_u32 s16, s10, 2 ; GCN-NEXT: s_addc_u32 s17, 0, 0 -; GCN-NEXT: s_cmp_lg_u32 s10, 0 -; GCN-NEXT: s_cselect_b32 s10, s16, s11 -; GCN-NEXT: s_cselect_b32 s11, s17, s15 +; GCN-NEXT: s_cmp_lg_u32 s12, 0 +; GCN-NEXT: s_cselect_b32 s12, s16, s14 +; GCN-NEXT: s_cselect_b32 s14, s17, s15 ; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_subb_u32 s8, 0, s13 +; GCN-NEXT: s_subb_u32 s8, 0, s11 ; GCN-NEXT: s_cmp_ge_u32 s8, s7 ; GCN-NEXT: s_cselect_b32 s9, -1, 0 -; GCN-NEXT: s_cmp_ge_u32 s14, s6 +; GCN-NEXT: s_cmp_ge_u32 s13, s6 ; GCN-NEXT: s_cselect_b32 s6, -1, 0 ; GCN-NEXT: s_cmp_eq_u32 s8, s7 ; GCN-NEXT: s_cselect_b32 s6, s6, s9 ; GCN-NEXT: s_cmp_lg_u32 s6, 0 -; GCN-NEXT: s_cselect_b32 s7, s11, 0 -; GCN-NEXT: s_cselect_b32 s6, s10, s12 +; GCN-NEXT: s_cselect_b32 s7, s14, 0 +; GCN-NEXT: s_cselect_b32 s6, s12, s10 ; GCN-NEXT: s_xor_b64 s[6:7], s[6:7], s[4:5] ; GCN-NEXT: s_sub_u32 s6, s6, s4 ; GCN-NEXT: s_subb_u32 s7, s7, s4 @@ -1303,8 +1285,6 @@ define amdgpu_kernel void @s_test_sdiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_cbranch_vccz .LBB10_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s12, s10, 1 -; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-IR-NEXT: s_or_b32 s8, s8, s9 ; GCN-IR-NEXT: s_addc_u32 s8, s11, 0 ; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s10, 63, s10 @@ -1335,8 +1315,6 @@ define amdgpu_kernel void @s_test_sdiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_sub_u32 s12, s12, s18 ; GCN-IR-NEXT: s_subb_u32 s13, s13, s19 ; GCN-IR-NEXT: s_add_u32 s16, s16, 1 -; GCN-IR-NEXT: s_cselect_b64 s[18:19], -1, 0 -; GCN-IR-NEXT: s_or_b32 s18, s18, s19 ; GCN-IR-NEXT: s_addc_u32 s17, s17, 0 ; GCN-IR-NEXT: s_cselect_b64 s[18:19], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[10:11], s[6:7] diff --git a/llvm/test/CodeGen/AMDGPU/GlobalISel/shlN_add.ll b/llvm/test/CodeGen/AMDGPU/shlN_add.ll index 9f4a6f2..3e507a0 100644 --- a/llvm/test/CodeGen/AMDGPU/GlobalISel/shlN_add.ll +++ b/llvm/test/CodeGen/AMDGPU/shlN_add.ll @@ -1,4 +1,9 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9-SDAG %s +; RUN: llc -mtriple=amdgcn -mcpu=fiji < %s | FileCheck -check-prefixes=GCN,GFX8-SDAG %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GCN,GFX10-SDAG %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -amdgpu-enable-delay-alu=0 < %s | FileCheck -check-prefixes=GCN,GFX10-SDAG %s + ; RUN: llc -global-isel -new-reg-bank-select -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9 %s ; RUN: llc -global-isel -new-reg-bank-select -mtriple=amdgcn -mcpu=fiji < %s | FileCheck -check-prefixes=GCN,GFX8 %s ; RUN: llc -global-isel -new-reg-bank-select -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GCN,GFX10 %s @@ -7,6 +12,24 @@ ; Test gfx9+ s_shl[1-4]_add_u32 pattern matching define amdgpu_ps i32 @s_shl1_add_u32(i32 inreg %src0, i32 inreg %src1) { +; GFX9-SDAG-LABEL: s_shl1_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl1_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl1_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl1_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl1_add_u32 s0, s0, s1 @@ -28,6 +51,24 @@ define amdgpu_ps i32 @s_shl1_add_u32(i32 inreg %src0, i32 inreg %src1) { } define amdgpu_ps i32 @s_shl2_add_u32(i32 inreg %src0, i32 inreg %src1) { +; GFX9-SDAG-LABEL: s_shl2_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl2_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl2_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl2_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl2_add_u32 s0, s0, s1 @@ -49,6 +90,24 @@ define amdgpu_ps i32 @s_shl2_add_u32(i32 inreg %src0, i32 inreg %src1) { } define amdgpu_ps i32 @s_shl3_add_u32(i32 inreg %src0, i32 inreg %src1) { +; GFX9-SDAG-LABEL: s_shl3_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl3_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl3_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl3_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl3_add_u32 s0, s0, s1 @@ -70,6 +129,24 @@ define amdgpu_ps i32 @s_shl3_add_u32(i32 inreg %src0, i32 inreg %src1) { } define amdgpu_ps i32 @s_shl4_add_u32(i32 inreg %src0, i32 inreg %src1) { +; GFX9-SDAG-LABEL: s_shl4_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl4_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl4_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s1 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl4_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl4_add_u32 s0, s0, s1 @@ -102,6 +179,25 @@ define amdgpu_ps i32 @s_shl5_add_u32(i32 inreg %src0, i32 inreg %src1) { } define i32 @v_shl1_add_u32(i32 %src0, i32 %src1) { +; GFX9-SDAG-LABEL: v_shl1_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, v0, 1, v1 +; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX8-SDAG-LABEL: v_shl1_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX8-SDAG-NEXT: v_lshlrev_b32_e32 v0, 1, v0 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, v0, v1 +; GFX8-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX10-SDAG-LABEL: v_shl1_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, v0, 1, v1 +; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31] +; ; GFX9-LABEL: v_shl1_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -126,6 +222,25 @@ define i32 @v_shl1_add_u32(i32 %src0, i32 %src1) { } define i32 @v_shl2_add_u32(i32 %src0, i32 %src1) { +; GFX9-SDAG-LABEL: v_shl2_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, v0, 2, v1 +; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX8-SDAG-LABEL: v_shl2_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX8-SDAG-NEXT: v_lshlrev_b32_e32 v0, 2, v0 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, v0, v1 +; GFX8-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX10-SDAG-LABEL: v_shl2_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, v0, 2, v1 +; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31] +; ; GFX9-LABEL: v_shl2_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -150,6 +265,25 @@ define i32 @v_shl2_add_u32(i32 %src0, i32 %src1) { } define i32 @v_shl3_add_u32(i32 %src0, i32 %src1) { +; GFX9-SDAG-LABEL: v_shl3_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, v0, 3, v1 +; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX8-SDAG-LABEL: v_shl3_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX8-SDAG-NEXT: v_lshlrev_b32_e32 v0, 3, v0 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, v0, v1 +; GFX8-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX10-SDAG-LABEL: v_shl3_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, v0, 3, v1 +; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31] +; ; GFX9-LABEL: v_shl3_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -174,6 +308,25 @@ define i32 @v_shl3_add_u32(i32 %src0, i32 %src1) { } define i32 @v_shl4_add_u32(i32 %src0, i32 %src1) { +; GFX9-SDAG-LABEL: v_shl4_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, v0, 4, v1 +; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX8-SDAG-LABEL: v_shl4_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX8-SDAG-NEXT: v_lshlrev_b32_e32 v0, 4, v0 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, v0, v1 +; GFX8-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX10-SDAG-LABEL: v_shl4_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, v0, 4, v1 +; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31] +; ; GFX9-LABEL: v_shl4_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -198,6 +351,25 @@ define i32 @v_shl4_add_u32(i32 %src0, i32 %src1) { } define i32 @v_shl5_add_u32(i32 %src0, i32 %src1) { +; GFX9-SDAG-LABEL: v_shl5_add_u32: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, v0, 5, v1 +; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX8-SDAG-LABEL: v_shl5_add_u32: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX8-SDAG-NEXT: v_lshlrev_b32_e32 v0, 5, v0 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, v0, v1 +; GFX8-SDAG-NEXT: s_setpc_b64 s[30:31] +; +; GFX10-SDAG-LABEL: v_shl5_add_u32: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, v0, 5, v1 +; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31] +; ; GFX9-LABEL: v_shl5_add_u32: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) @@ -224,6 +396,22 @@ define i32 @v_shl5_add_u32(i32 %src0, i32 %src1) { ; FIXME: Use v_lshl_add_u32 ; shift is scalar, but add is vector. define amdgpu_ps float @shl1_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { +; GFX9-SDAG-LABEL: shl1_add_u32_vgpr1: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, s0, 1, v0 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: shl1_add_u32_vgpr1: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, s0, v0 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: shl1_add_u32_vgpr1: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, s0, 1, v0 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: shl1_add_u32_vgpr1: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl_b32 s0, s0, 1 @@ -248,6 +436,22 @@ define amdgpu_ps float @shl1_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { } define amdgpu_ps float @shl2_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { +; GFX9-SDAG-LABEL: shl2_add_u32_vgpr1: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, s0, 2, v0 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: shl2_add_u32_vgpr1: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, s0, v0 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: shl2_add_u32_vgpr1: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, s0, 2, v0 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: shl2_add_u32_vgpr1: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl_b32 s0, s0, 2 @@ -272,6 +476,22 @@ define amdgpu_ps float @shl2_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { } define amdgpu_ps float @shl3_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { +; GFX9-SDAG-LABEL: shl3_add_u32_vgpr1: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, s0, 3, v0 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: shl3_add_u32_vgpr1: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, s0, v0 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: shl3_add_u32_vgpr1: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, s0, 3, v0 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: shl3_add_u32_vgpr1: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl_b32 s0, s0, 3 @@ -296,6 +516,22 @@ define amdgpu_ps float @shl3_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { } define amdgpu_ps float @shl4_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { +; GFX9-SDAG-LABEL: shl4_add_u32_vgpr1: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, s0, 4, v0 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: shl4_add_u32_vgpr1: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, s0, v0 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: shl4_add_u32_vgpr1: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, s0, 4, v0 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: shl4_add_u32_vgpr1: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl_b32 s0, s0, 4 @@ -320,6 +556,22 @@ define amdgpu_ps float @shl4_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { } define amdgpu_ps float @shl5_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { +; GFX9-SDAG-LABEL: shl5_add_u32_vgpr1: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: v_lshl_add_u32 v0, s0, 5, v0 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: shl5_add_u32_vgpr1: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 5 +; GFX8-SDAG-NEXT: v_add_u32_e32 v0, vcc, s0, v0 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: shl5_add_u32_vgpr1: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: v_lshl_add_u32 v0, s0, 5, v0 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: shl5_add_u32_vgpr1: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl_b32 s0, s0, 5 @@ -344,6 +596,30 @@ define amdgpu_ps float @shl5_add_u32_vgpr1(i32 inreg %src0, i32 %src1) { } define amdgpu_ps <2 x i32> @s_shl1_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> inreg %src1) { +; GFX9-SDAG-LABEL: s_shl1_add_u32_v2: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX9-SDAG-NEXT: s_lshl_b32 s1, s1, 1 +; GFX9-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl1_add_u32_v2: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX8-SDAG-NEXT: s_lshl_b32 s1, s1, 1 +; GFX8-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl1_add_u32_v2: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 1 +; GFX10-SDAG-NEXT: s_lshl_b32 s1, s1, 1 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX10-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl1_add_u32_v2: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl1_add_u32 s0, s0, s2 @@ -369,6 +645,30 @@ define amdgpu_ps <2 x i32> @s_shl1_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> i } define amdgpu_ps <2 x i32> @s_shl2_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> inreg %src1) { +; GFX9-SDAG-LABEL: s_shl2_add_u32_v2: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX9-SDAG-NEXT: s_lshl_b32 s1, s1, 2 +; GFX9-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl2_add_u32_v2: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX8-SDAG-NEXT: s_lshl_b32 s1, s1, 2 +; GFX8-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl2_add_u32_v2: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX10-SDAG-NEXT: s_lshl_b32 s1, s1, 2 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX10-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl2_add_u32_v2: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl2_add_u32 s0, s0, s2 @@ -394,6 +694,30 @@ define amdgpu_ps <2 x i32> @s_shl2_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> i } define amdgpu_ps <2 x i32> @s_shl3_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> inreg %src1) { +; GFX9-SDAG-LABEL: s_shl3_add_u32_v2: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX9-SDAG-NEXT: s_lshl_b32 s1, s1, 3 +; GFX9-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl3_add_u32_v2: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX8-SDAG-NEXT: s_lshl_b32 s1, s1, 3 +; GFX8-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl3_add_u32_v2: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 3 +; GFX10-SDAG-NEXT: s_lshl_b32 s1, s1, 3 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX10-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl3_add_u32_v2: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl3_add_u32 s0, s0, s2 @@ -419,6 +743,30 @@ define amdgpu_ps <2 x i32> @s_shl3_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> i } define amdgpu_ps <2 x i32> @s_shl4_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> inreg %src1) { +; GFX9-SDAG-LABEL: s_shl4_add_u32_v2: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX9-SDAG-NEXT: s_lshl_b32 s1, s1, 4 +; GFX9-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl4_add_u32_v2: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX8-SDAG-NEXT: s_lshl_b32 s1, s1, 4 +; GFX8-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl4_add_u32_v2: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 4 +; GFX10-SDAG-NEXT: s_lshl_b32 s1, s1, 4 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX10-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl4_add_u32_v2: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl4_add_u32 s0, s0, s2 @@ -444,6 +792,30 @@ define amdgpu_ps <2 x i32> @s_shl4_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> i } define amdgpu_ps <2 x i32> @s_shl_2_4_add_u32_v2(<2 x i32> inreg %src0, <2 x i32> inreg %src1) { +; GFX9-SDAG-LABEL: s_shl_2_4_add_u32_v2: +; GFX9-SDAG: ; %bb.0: +; GFX9-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX9-SDAG-NEXT: s_lshl_b32 s1, s1, 4 +; GFX9-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX9-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX9-SDAG-NEXT: ; return to shader part epilog +; +; GFX8-SDAG-LABEL: s_shl_2_4_add_u32_v2: +; GFX8-SDAG: ; %bb.0: +; GFX8-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX8-SDAG-NEXT: s_lshl_b32 s1, s1, 4 +; GFX8-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX8-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX8-SDAG-NEXT: ; return to shader part epilog +; +; GFX10-SDAG-LABEL: s_shl_2_4_add_u32_v2: +; GFX10-SDAG: ; %bb.0: +; GFX10-SDAG-NEXT: s_lshl_b32 s0, s0, 2 +; GFX10-SDAG-NEXT: s_lshl_b32 s1, s1, 4 +; GFX10-SDAG-NEXT: s_add_i32 s0, s0, s2 +; GFX10-SDAG-NEXT: s_add_i32 s1, s1, s3 +; GFX10-SDAG-NEXT: ; return to shader part epilog +; ; GFX9-LABEL: s_shl_2_4_add_u32_v2: ; GFX9: ; %bb.0: ; GFX9-NEXT: s_lshl2_add_u32 s0, s0, s2 diff --git a/llvm/test/CodeGen/AMDGPU/spill-vgpr-to-agpr-update-regscavenger.ll b/llvm/test/CodeGen/AMDGPU/spill-vgpr-to-agpr-update-regscavenger.ll index ef96944..586579f 100644 --- a/llvm/test/CodeGen/AMDGPU/spill-vgpr-to-agpr-update-regscavenger.ll +++ b/llvm/test/CodeGen/AMDGPU/spill-vgpr-to-agpr-update-regscavenger.ll @@ -20,33 +20,38 @@ define void @test() { ; CHECK-NEXT: ; in Loop: Header=BB0_1 Depth=1 ; CHECK-NEXT: .LBB0_3: ; %bb.3 ; CHECK-NEXT: ; in Loop: Header=BB0_1 Depth=1 +; CHECK-NEXT: ; implicit-def: $sgpr4 +; CHECK-NEXT: v_mov_b32_e32 v0, s4 +; CHECK-NEXT: v_readfirstlane_b32 s6, v0 ; CHECK-NEXT: s_mov_b64 s[4:5], -1 +; CHECK-NEXT: s_mov_b32 s7, 0 +; CHECK-NEXT: s_cmp_eq_u32 s6, s7 ; CHECK-NEXT: ; implicit-def: $vgpr1 : SGPR spill to VGPR lane ; CHECK-NEXT: v_writelane_b32 v1, s4, 0 ; CHECK-NEXT: v_writelane_b32 v1, s5, 1 -; CHECK-NEXT: s_or_saveexec_b64 s[8:9], -1 -; CHECK-NEXT: s_nop 0 +; CHECK-NEXT: s_mov_b64 s[10:11], exec +; CHECK-NEXT: s_mov_b64 exec, -1 ; CHECK-NEXT: v_accvgpr_write_b32 a0, v1 ; Reload Reuse -; CHECK-NEXT: s_mov_b64 exec, s[8:9] +; CHECK-NEXT: s_mov_b64 exec, s[10:11] ; CHECK-NEXT: s_cbranch_scc1 .LBB0_5 ; CHECK-NEXT: ; %bb.4: ; %bb.4 ; CHECK-NEXT: ; in Loop: Header=BB0_1 Depth=1 -; CHECK-NEXT: s_or_saveexec_b64 s[8:9], -1 +; CHECK-NEXT: s_or_saveexec_b64 s[10:11], -1 ; CHECK-NEXT: v_accvgpr_read_b32 v1, a0 ; Reload Reuse -; CHECK-NEXT: s_mov_b64 exec, s[8:9] +; CHECK-NEXT: s_mov_b64 exec, s[10:11] ; CHECK-NEXT: s_mov_b64 s[4:5], 0 ; CHECK-NEXT: v_writelane_b32 v1, s4, 0 ; CHECK-NEXT: v_writelane_b32 v1, s5, 1 -; CHECK-NEXT: s_or_saveexec_b64 s[8:9], -1 +; CHECK-NEXT: s_or_saveexec_b64 s[10:11], -1 ; CHECK-NEXT: s_nop 0 ; CHECK-NEXT: v_accvgpr_write_b32 a0, v1 ; Reload Reuse -; CHECK-NEXT: s_mov_b64 exec, s[8:9] +; CHECK-NEXT: s_mov_b64 exec, s[10:11] ; CHECK-NEXT: .LBB0_5: ; %Flow ; CHECK-NEXT: ; in Loop: Header=BB0_1 Depth=1 -; CHECK-NEXT: s_or_saveexec_b64 s[8:9], -1 +; CHECK-NEXT: s_or_saveexec_b64 s[10:11], -1 ; CHECK-NEXT: s_nop 0 ; CHECK-NEXT: v_accvgpr_read_b32 v1, a0 ; Reload Reuse -; CHECK-NEXT: s_mov_b64 exec, s[8:9] +; CHECK-NEXT: s_mov_b64 exec, s[10:11] ; CHECK-NEXT: v_readlane_b32 s4, v1, 0 ; CHECK-NEXT: v_readlane_b32 s5, v1, 1 ; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, 1, s[4:5] diff --git a/llvm/test/CodeGen/AMDGPU/splitkit-getsubrangeformask.ll b/llvm/test/CodeGen/AMDGPU/splitkit-getsubrangeformask.ll index 364598f..5aafb0f 100644 --- a/llvm/test/CodeGen/AMDGPU/splitkit-getsubrangeformask.ll +++ b/llvm/test/CodeGen/AMDGPU/splitkit-getsubrangeformask.ll @@ -31,8 +31,8 @@ define amdgpu_gs void @_amdgpu_gs_main(i32 inreg %primShaderTableAddrLow, <31 x ; CHECK-NEXT: [[COPY13:%[0-9]+]]:sgpr_32 = COPY $sgpr10 ; CHECK-NEXT: [[COPY14:%[0-9]+]]:sgpr_32 = COPY $sgpr8 ; CHECK-NEXT: undef [[S_LOAD_DWORDX2_IMM:%[0-9]+]].sub0_sub1:sgpr_128 = S_LOAD_DWORDX2_IMM [[COPY]], 232, 0 :: (invariant load (s64) from %ir.39, addrspace 4) - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM undef %117:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: KILL undef %117:sgpr_128 + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM undef %125:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: KILL undef %125:sgpr_128 ; CHECK-NEXT: [[S_LSHL_B32_:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY5]], 4, implicit-def dead $scc ; CHECK-NEXT: [[S_LSHL_B32_1:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY4]], 4, implicit-def dead $scc ; CHECK-NEXT: [[S_LSHL_B32_2:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY3]], 4, implicit-def dead $scc @@ -44,85 +44,87 @@ define amdgpu_gs void @_amdgpu_gs_main(i32 inreg %primShaderTableAddrLow, <31 x ; CHECK-NEXT: [[S_SUB_I32_1:%[0-9]+]]:sreg_32 = S_SUB_I32 [[S_BUFFER_LOAD_DWORD_IMM]], 30, implicit-def dead $scc ; CHECK-NEXT: undef [[S_ADD_U32_:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY6]], [[S_LSHL_B32_2]], implicit-def $scc ; CHECK-NEXT: [[S_ADD_U32_:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %54:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_]], 16, 0 :: (invariant load (s128) from %ir.71, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_]], 16, 0 :: (invariant load (s128) from %ir.81, addrspace 4) ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM1:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM undef %74:sreg_64, 0, 0 :: (invariant load (s128) from `ptr addrspace(4) poison`, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM2:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_]], 64, 0 :: (invariant load (s128) from %ir.88, addrspace 4) ; CHECK-NEXT: KILL undef %74:sreg_64 ; CHECK-NEXT: KILL [[S_ADD_U32_]].sub0, [[S_ADD_U32_]].sub1 ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM1:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM [[S_LOAD_DWORDX4_IMM]], 0, 0 :: (dereferenceable invariant load (s32)) ; CHECK-NEXT: [[V_MOV_B32_e32_:%[0-9]+]]:vgpr_32 = V_MOV_B32_e32 0, implicit $exec ; CHECK-NEXT: undef [[S_MOV_B32_:%[0-9]+]].sub1:sgpr_128 = S_MOV_B32 0 - ; CHECK-NEXT: [[BUFFER_LOAD_DWORD_OFFSET:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET undef %112:sgpr_128, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], undef %87:sgpr_128, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) + ; CHECK-NEXT: [[BUFFER_LOAD_DWORD_OFFSET:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET undef %118:sgpr_128, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) + ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], undef %89:sgpr_128, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN1:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM1]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: KILL undef %112:sgpr_128 - ; CHECK-NEXT: KILL undef %87:sgpr_128 + ; CHECK-NEXT: KILL undef %89:sgpr_128 + ; CHECK-NEXT: KILL undef %118:sgpr_128 ; CHECK-NEXT: [[S_SUB_I32_2:%[0-9]+]]:sreg_32 = S_SUB_I32 [[S_BUFFER_LOAD_DWORD_IMM1]], 31, implicit-def dead $scc ; CHECK-NEXT: undef [[S_ADD_U32_1:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY6]], [[S_LSHL_B32_]], implicit-def $scc ; CHECK-NEXT: [[S_ADD_U32_1:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %54:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc ; CHECK-NEXT: undef [[S_ADD_U32_2:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY6]], [[S_LSHL_B32_1]], implicit-def $scc ; CHECK-NEXT: [[S_ADD_U32_2:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %54:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_ASHR_I32_3:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 undef %148:sreg_32, 31, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_3:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY6]], undef %148:sreg_32, implicit-def $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM3:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_1]], 64, 0 :: (invariant load (s128) from %ir.77, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM4:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_2]], 64, 0 :: (invariant load (s128) from %ir.83, addrspace 4) - ; CHECK-NEXT: KILL [[S_ADD_U32_2]].sub0, [[S_ADD_U32_2]].sub1 + ; CHECK-NEXT: undef [[S_ADD_U32_3:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY6]], [[S_LSHL_B32_2]], implicit-def $scc + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM2:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_1]], 64, 0 :: (invariant load (s128) from %ir.87, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM3:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_2]], 64, 0 :: (invariant load (s128) from %ir.93, addrspace 4) ; CHECK-NEXT: KILL [[S_ADD_U32_1]].sub0, [[S_ADD_U32_1]].sub1 - ; CHECK-NEXT: [[S_ADD_U32_3:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %54:sreg_32, [[S_ASHR_I32_3]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_4:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, [[S_LSHL_B32_]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_4:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_5:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, [[S_LSHL_B32_1]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_5:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_6:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, undef %148:sreg_32, implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_6:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_3]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_7:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, [[S_LSHL_B32_2]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_7:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_8:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY8]], [[S_LSHL_B32_]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_8:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %48:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_9:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY9]], [[S_LSHL_B32_1]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_9:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %45:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_10:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY9]], [[S_LSHL_B32_2]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_10:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %45:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: KILL [[S_ADD_U32_2]].sub0, [[S_ADD_U32_2]].sub1 + ; CHECK-NEXT: [[S_ADD_U32_3:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %54:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_ASHR_I32_3:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 undef %169:sreg_32, 31, implicit-def dead $scc + ; CHECK-NEXT: undef [[S_ADD_U32_4:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY6]], undef %169:sreg_32, implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_4:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %54:sreg_32, [[S_ASHR_I32_3]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_5:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, [[S_LSHL_B32_]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_5:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_6:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, [[S_LSHL_B32_1]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_6:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_7:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, undef %169:sreg_32, implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_7:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_3]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_8:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY7]].sub0, [[S_LSHL_B32_2]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_8:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %51:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_9:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY8]], [[S_LSHL_B32_]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_9:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %48:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_10:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY9]], [[S_LSHL_B32_1]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_10:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %45:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_11:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY9]], [[S_LSHL_B32_2]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_11:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %45:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc ; CHECK-NEXT: [[S_ADD_I32_:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_LSHL_B32_]], 16, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_1:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_LSHL_B32_2]], 16, implicit-def dead $scc ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM [[S_MOV_B32_]], [[S_ADD_I32_]], 0, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM1:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM [[S_MOV_B32_]], undef %279:sreg_32, 0, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM1:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM [[S_MOV_B32_]], undef %302:sreg_32, 0, 0 :: (dereferenceable invariant load (s32)) ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM2:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM [[S_MOV_B32_]], [[S_ADD_I32_1]], 0, 0 :: (dereferenceable invariant load (s32)) ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM2:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM [[S_MOV_B32_]], 16, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM3:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %334:sgpr_128, undef %335:sreg_32, 0, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM3:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM undef %345:sgpr_128, 16, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM5:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_3]], 64, 0 :: (invariant load (s128) from %ir.95, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM6:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_4]], 0, 0 :: (invariant load (s128) from %ir.100, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM7:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_5]], 0, 0 :: (invariant load (s128) from %ir.105, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM8:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_6]], 0, 0 :: (invariant load (s128) from %ir.112, addrspace 4) - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM4:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %329:sgpr_128, [[S_ADD_I32_]], 0, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM5:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %340:sgpr_128, [[S_ADD_I32_1]], 0, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN2:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM3]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN3:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM4]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM3:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %357:sgpr_128, undef %358:sreg_32, 0, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM3:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM undef %368:sgpr_128, 16, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM4:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_3]], 64, 0 :: (invariant load (s128) from %ir.99, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM5:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_4]], 64, 0 :: (invariant load (s128) from %ir.107, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM6:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_5]], 0, 0 :: (invariant load (s128) from %ir.112, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM7:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_6]], 0, 0 :: (invariant load (s128) from %ir.117, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM8:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_7]], 0, 0 :: (invariant load (s128) from %ir.124, addrspace 4) + ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN2:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM2]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM4:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %352:sgpr_128, [[S_ADD_I32_]], 0, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM5:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %363:sgpr_128, [[S_ADD_I32_1]], 0, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN3:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM3]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ADD_I32_2:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM]], -98, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_3:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM1]], -114, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_4:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM2]], -130, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_5:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_IMM2]], -178, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_11:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY10]], [[S_LSHL_B32_]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_11:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %42:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_12:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY11]], [[S_LSHL_B32_]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_12:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %39:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_13:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY11]], [[S_LSHL_B32_1]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_13:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %39:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: undef [[S_ADD_U32_14:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY11]], [[S_LSHL_B32_2]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_14:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %39:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_12:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY10]], [[S_LSHL_B32_]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_12:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %42:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_13:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY11]], [[S_LSHL_B32_]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_13:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %39:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_14:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY11]], [[S_LSHL_B32_1]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_14:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %39:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_15:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY11]], [[S_LSHL_B32_2]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_15:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %39:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc ; CHECK-NEXT: [[S_LSHL_B32_3:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY12]], 4, implicit-def dead $scc - ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN4:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM2]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) + ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN4:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM4]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ADD_I32_6:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_LSHL_B32_3]], 16, implicit-def dead $scc - ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM6:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %361:sgpr_128, [[S_ADD_I32_6]], 0, 0 :: (dereferenceable invariant load (s32)) + ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_SGPR_IMM6:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM undef %384:sgpr_128, [[S_ADD_I32_6]], 0, 0 :: (dereferenceable invariant load (s32)) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN5:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM5]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM9:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_4]], 224, 0 :: (invariant load (s128) from %ir.117, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM10:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[COPY7]], 224, 0 :: (invariant load (s128) from %ir.133, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM11:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_4]], 576, 0 :: (invariant load (s128) from %ir.138, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM9:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_5]], 224, 0 :: (invariant load (s128) from %ir.129, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM10:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[COPY7]], 224, 0 :: (invariant load (s128) from %ir.145, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM11:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_5]], 576, 0 :: (invariant load (s128) from %ir.150, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN6:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM6]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM12:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_5]], 224, 0 :: (invariant load (s128) from %ir.122, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM13:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_6]], 576, 0 :: (invariant load (s128) from %ir.150, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM14:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_7]], 224, 0 :: (invariant load (s128) from %ir.128, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM12:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_6]], 224, 0 :: (invariant load (s128) from %ir.134, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM13:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_7]], 576, 0 :: (invariant load (s128) from %ir.162, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM14:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_8]], 224, 0 :: (invariant load (s128) from %ir.140, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN7:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM7]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN8:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM8]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ADD_I32_7:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM4]], -217, implicit-def dead $scc @@ -133,49 +135,49 @@ define amdgpu_gs void @_amdgpu_gs_main(i32 inreg %primShaderTableAddrLow, <31 x ; CHECK-NEXT: [[S_ADD_I32_12:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM3]], -329, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_13:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM3]], -345, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_14:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM6]], -441, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_15:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY2]], [[S_LSHL_B32_2]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_15:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %36:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_16:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY2]], [[S_LSHL_B32_2]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_16:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %36:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc ; CHECK-NEXT: [[S_LSHL_B32_4:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY13]], 4, implicit-def dead $scc ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN9:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM9]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ASHR_I32_4:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 [[S_LSHL_B32_4]], 31, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_16:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY2]], [[S_LSHL_B32_4]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_16:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %36:sreg_32, [[S_ASHR_I32_4]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: undef [[S_ADD_U32_17:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY2]], [[S_LSHL_B32_4]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_17:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %36:sreg_32, [[S_ASHR_I32_4]], implicit-def dead $scc, implicit $scc ; CHECK-NEXT: [[S_LSHL_B32_5:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY5]], 3, implicit-def dead $scc ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN10:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM12]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ASHR_I32_5:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 [[S_LSHL_B32_5]], 31, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_17:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_5]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_17:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_5]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORD_IMM:%[0-9]+]]:sreg_32_xm0_xexec = S_LOAD_DWORD_IMM [[S_ADD_U32_17]], 168, 0 :: (invariant load (s32) from %ir.260, align 8, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM15:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_7]], 576, 0 :: (invariant load (s128) from %ir.145, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_18:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_5]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_18:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_5]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORD_IMM:%[0-9]+]]:sreg_32_xm0_xexec = S_LOAD_DWORD_IMM [[S_ADD_U32_18]], 168, 0 :: (invariant load (s32) from %ir.273, align 8, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM15:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_8]], 576, 0 :: (invariant load (s128) from %ir.157, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN11:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM14]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN12:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM10]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN13:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM11]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_LOAD_DWORDX2_IMM:%[0-9]+]].sub3:sgpr_128 = S_MOV_B32 553734060 ; CHECK-NEXT: [[S_LOAD_DWORDX2_IMM:%[0-9]+]].sub2:sgpr_128 = S_MOV_B32 -1 ; CHECK-NEXT: [[COPY15:%[0-9]+]]:sgpr_128 = COPY [[S_LOAD_DWORDX2_IMM]] - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM16:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_8]], 0, 0 :: (invariant load (s128) from %ir.158, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM16:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_9]], 0, 0 :: (invariant load (s128) from %ir.170, addrspace 4) ; CHECK-NEXT: [[COPY15:%[0-9]+]].sub1:sgpr_128 = COPY [[S_MOV_B32_]].sub1 ; CHECK-NEXT: [[COPY15:%[0-9]+]].sub0:sgpr_128 = COPY [[S_LOAD_DWORD_IMM]] ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM4:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM [[COPY15]], 0, 0 :: (dereferenceable invariant load (s32)) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN14:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM15]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN15:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM13]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM17:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_9]], 0, 0 :: (invariant load (s128) from %ir.166, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM18:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_10]], 0, 0 :: (invariant load (s128) from %ir.171, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM17:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_10]], 0, 0 :: (invariant load (s128) from %ir.178, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM18:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_11]], 0, 0 :: (invariant load (s128) from %ir.183, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN16:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM16]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_LSHL_B32_6:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY4]], 3, implicit-def dead $scc ; CHECK-NEXT: [[BUFFER_LOAD_DWORD_OFFSET1:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET [[S_LOAD_DWORDX4_IMM1]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ASHR_I32_6:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 [[S_LSHL_B32_6]], 31, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_15:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_IMM4]], -467, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_18:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_6]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_18:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_6]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX2_IMM1:%[0-9]+]]:sreg_64_xexec = S_LOAD_DWORDX2_IMM [[S_ADD_U32_18]], 168, 0 :: (invariant load (s64) from %ir.269, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_19:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_6]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_19:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_6]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORDX2_IMM1:%[0-9]+]]:sreg_64_xexec = S_LOAD_DWORDX2_IMM [[S_ADD_U32_19]], 168, 0 :: (invariant load (s64) from %ir.282, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_DWORD_OFFSET2:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET [[S_LOAD_DWORDX4_IMM17]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_DWORD_OFFSET3:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET [[S_LOAD_DWORDX4_IMM18]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM19:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_11]], 0, 0 :: (invariant load (s128) from %ir.193, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM20:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_12]], 0, 0 :: (invariant load (s128) from %ir.199, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM19:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_12]], 0, 0 :: (invariant load (s128) from %ir.205, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM20:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_13]], 0, 0 :: (invariant load (s128) from %ir.211, addrspace 4) ; CHECK-NEXT: [[COPY16:%[0-9]+]]:sgpr_128 = COPY [[S_LOAD_DWORDX2_IMM]] - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM21:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_13]], 0, 0 :: (invariant load (s128) from %ir.204, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM22:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_14]], 0, 0 :: (invariant load (s128) from %ir.209, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM21:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_14]], 0, 0 :: (invariant load (s128) from %ir.216, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM22:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_15]], 0, 0 :: (invariant load (s128) from %ir.221, addrspace 4) ; CHECK-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32 = S_AND_B32 [[S_LOAD_DWORDX2_IMM1]].sub1, 65535, implicit-def dead $scc ; CHECK-NEXT: [[COPY16:%[0-9]+]].sub0:sgpr_128 = COPY [[S_LOAD_DWORDX2_IMM1]].sub0 ; CHECK-NEXT: [[COPY16:%[0-9]+]].sub1:sgpr_128 = COPY [[S_AND_B32_]] @@ -187,30 +189,30 @@ define amdgpu_gs void @_amdgpu_gs_main(i32 inreg %primShaderTableAddrLow, <31 x ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN20:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM22]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[S_ASHR_I32_7:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 [[S_LSHL_B32_7]], 31, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_16:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_IMM5]], -468, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_19:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_7]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_19:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_7]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX2_IMM2:%[0-9]+]]:sreg_64_xexec = S_LOAD_DWORDX2_IMM [[S_ADD_U32_19]], 168, 0 :: (invariant load (s64) from %ir.280, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_20:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_7]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_20:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_7]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORDX2_IMM2:%[0-9]+]]:sreg_64_xexec = S_LOAD_DWORDX2_IMM [[S_ADD_U32_20]], 168, 0 :: (invariant load (s64) from %ir.293, addrspace 4) ; CHECK-NEXT: [[COPY17:%[0-9]+]]:sgpr_128 = COPY [[S_LOAD_DWORDX2_IMM]] ; CHECK-NEXT: [[S_AND_B32_1:%[0-9]+]]:sreg_32 = S_AND_B32 [[S_LOAD_DWORDX2_IMM2]].sub1, 65535, implicit-def dead $scc ; CHECK-NEXT: [[COPY17:%[0-9]+]].sub0:sgpr_128 = COPY [[S_LOAD_DWORDX2_IMM2]].sub0 ; CHECK-NEXT: [[COPY17:%[0-9]+]].sub1:sgpr_128 = COPY [[S_AND_B32_1]] ; CHECK-NEXT: [[S_BUFFER_LOAD_DWORD_IMM6:%[0-9]+]]:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM [[COPY17]], 0, 0 :: (dereferenceable invariant load (s32)) - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM23:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_15]], 160, 0 :: (invariant load (s128) from %ir.244, addrspace 4) - ; CHECK-NEXT: [[S_LOAD_DWORD_IMM1:%[0-9]+]]:sreg_32_xm0_xexec = S_LOAD_DWORD_IMM undef %443:sreg_64, 0, 0 :: (invariant load (s32) from `ptr addrspace(4) poison`, addrspace 4) - ; CHECK-NEXT: KILL [[S_ADD_U32_15]].sub0, [[S_ADD_U32_15]].sub1 + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM23:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_16]], 160, 0 :: (invariant load (s128) from %ir.256, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORD_IMM1:%[0-9]+]]:sreg_32_xm0_xexec = S_LOAD_DWORD_IMM undef %470:sreg_64, 0, 0 :: (invariant load (s32) from `ptr addrspace(4) poison`, addrspace 4) + ; CHECK-NEXT: KILL [[S_ADD_U32_16]].sub0, [[S_ADD_U32_16]].sub1 + ; CHECK-NEXT: KILL undef %470:sreg_64 ; CHECK-NEXT: KILL [[COPY17]].sub0_sub1_sub2, [[COPY17]].sub3 - ; CHECK-NEXT: KILL undef %443:sreg_64 ; CHECK-NEXT: [[S_LSHL_B32_8:%[0-9]+]]:sreg_32 = S_LSHL_B32 [[COPY14]], 3, implicit-def dead $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM24:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_16]], 160, 0 :: (invariant load (s128) from %ir.252, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM24:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_17]], 160, 0 :: (invariant load (s128) from %ir.265, addrspace 4) ; CHECK-NEXT: [[S_ASHR_I32_8:%[0-9]+]]:sreg_32_xm0 = S_ASHR_I32 [[S_LSHL_B32_8]], 31, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_17:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_IMM6]], -469, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_20:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_8]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_20:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_8]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORD_IMM2:%[0-9]+]]:sreg_32_xm0_xexec = S_LOAD_DWORD_IMM [[S_ADD_U32_20]], 168, 0 :: (invariant load (s32) from %ir.291, align 8, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_21:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY]].sub0, [[S_LSHL_B32_8]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_21:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %57:sreg_32, [[S_ASHR_I32_8]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORD_IMM2:%[0-9]+]]:sreg_32_xm0_xexec = S_LOAD_DWORD_IMM [[S_ADD_U32_21]], 168, 0 :: (invariant load (s32) from %ir.305, align 8, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN21:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM23]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN22:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM24]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) - ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM23]] ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM24]] + ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM23]] ; CHECK-NEXT: [[S_AND_B32_2:%[0-9]+]]:sreg_32 = S_AND_B32 [[S_LOAD_DWORD_IMM1]], 65535, implicit-def dead $scc ; CHECK-NEXT: [[COPY18:%[0-9]+]]:sgpr_128 = COPY [[S_LOAD_DWORDX2_IMM]] ; CHECK-NEXT: [[COPY18:%[0-9]+]].sub1:sgpr_128 = COPY [[S_AND_B32_2]] @@ -222,22 +224,22 @@ define amdgpu_gs void @_amdgpu_gs_main(i32 inreg %primShaderTableAddrLow, <31 x ; CHECK-NEXT: [[S_ADD_I32_21:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM3]], -507, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_22:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_SGPR_IMM3]], -539, implicit-def dead $scc ; CHECK-NEXT: [[S_ADD_I32_23:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_IMM7]], -473, implicit-def dead $scc - ; CHECK-NEXT: undef [[S_ADD_U32_21:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY1]], [[S_LSHL_B32_]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_21:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %33:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM25:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_21]], 96, 0 :: (invariant load (s128) from %ir.309, addrspace 4) - ; CHECK-NEXT: undef [[S_ADD_U32_22:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY1]], [[S_LSHL_B32_1]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_22:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %33:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM26:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_22]], 96, 0 :: (invariant load (s128) from %ir.315, addrspace 4) - ; CHECK-NEXT: undef [[S_ADD_U32_23:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY1]], [[S_LSHL_B32_2]], implicit-def $scc - ; CHECK-NEXT: [[S_ADD_U32_23:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %33:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc - ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM27:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_23]], 96, 0 :: (invariant load (s128) from %ir.321, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_22:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY1]], [[S_LSHL_B32_]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_22:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %33:sreg_32, [[S_ASHR_I32_]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM25:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_22]], 96, 0 :: (invariant load (s128) from %ir.323, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_23:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY1]], [[S_LSHL_B32_1]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_23:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %33:sreg_32, [[S_ASHR_I32_1]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM26:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_23]], 96, 0 :: (invariant load (s128) from %ir.329, addrspace 4) + ; CHECK-NEXT: undef [[S_ADD_U32_24:%[0-9]+]].sub0:sreg_64 = S_ADD_U32 [[COPY1]], [[S_LSHL_B32_2]], implicit-def $scc + ; CHECK-NEXT: [[S_ADD_U32_24:%[0-9]+]].sub1:sreg_64 = S_ADDC_U32 undef %33:sreg_32, [[S_ASHR_I32_2]], implicit-def dead $scc, implicit $scc + ; CHECK-NEXT: [[S_LOAD_DWORDX4_IMM27:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM [[S_ADD_U32_24]], 96, 0 :: (invariant load (s128) from %ir.335, addrspace 4) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN23:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM25]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN24:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM26]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) ; CHECK-NEXT: [[BUFFER_LOAD_FORMAT_X_IDXEN25:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM27]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s32), align 1, addrspace 8) + ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM27]] ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM25]] - ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM26]] ; CHECK-NEXT: KILL [[V_MOV_B32_e32_]] - ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM27]] + ; CHECK-NEXT: KILL [[S_LOAD_DWORDX4_IMM26]] ; CHECK-NEXT: [[V_ADD_U32_e64_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e64 -2, [[BUFFER_LOAD_FORMAT_X_IDXEN]], 0, implicit $exec ; CHECK-NEXT: [[V_ADD_U32_e64_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e64 -1, [[BUFFER_LOAD_FORMAT_X_IDXEN1]], 0, implicit $exec ; CHECK-NEXT: [[V_ADD_U32_e64_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e64 -3, [[BUFFER_LOAD_FORMAT_X_IDXEN]], 0, implicit $exec @@ -349,13 +351,13 @@ define amdgpu_gs void @_amdgpu_gs_main(i32 inreg %primShaderTableAddrLow, <31 x ; CHECK-NEXT: [[V_OR_B32_e64_64:%[0-9]+]]:vgpr_32 = V_OR_B32_e64 [[V_OR_B32_e64_63]], [[V_ADD_U32_e64_28]], implicit $exec ; CHECK-NEXT: [[V_ADD_U32_e64_30:%[0-9]+]]:vgpr_32 = V_ADD_U32_e64 -593, [[BUFFER_LOAD_FORMAT_X_IDXEN]], 0, implicit $exec ; CHECK-NEXT: [[V_OR_B32_e64_65:%[0-9]+]]:vgpr_32 = V_OR_B32_e64 [[V_OR_B32_e64_64]], [[V_ADD_U32_e64_29]], implicit $exec - ; CHECK-NEXT: [[S_LOAD_DWORDX8_IMM:%[0-9]+]]:sgpr_256 = S_LOAD_DWORDX8_IMM undef %516:sreg_64, 0, 0 :: (invariant load (s256) from `ptr addrspace(4) poison`, addrspace 4) + ; CHECK-NEXT: [[S_LOAD_DWORDX8_IMM:%[0-9]+]]:sgpr_256 = S_LOAD_DWORDX8_IMM undef %543:sreg_64, 0, 0 :: (invariant load (s256) from `ptr addrspace(4) poison`, addrspace 4) ; CHECK-NEXT: [[V_OR_B32_e64_66:%[0-9]+]]:vgpr_32 = V_OR_B32_e64 [[V_OR_B32_e64_65]], [[V_ADD_U32_e64_30]], implicit $exec ; CHECK-NEXT: [[S_ADD_I32_24:%[0-9]+]]:sreg_32 = S_ADD_I32 [[S_BUFFER_LOAD_DWORD_IMM8]], -594, implicit-def dead $scc ; CHECK-NEXT: [[V_OR_B32_e64_67:%[0-9]+]]:vgpr_32 = V_OR_B32_e64 [[S_ADD_I32_24]], [[V_OR_B32_e64_66]], implicit $exec ; CHECK-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_32_xm0_xexec = V_CMP_EQ_U32_e64 0, [[V_OR_B32_e64_67]], implicit $exec ; CHECK-NEXT: undef [[V_CNDMASK_B32_e64_:%[0-9]+]].sub3:vreg_128 = V_CNDMASK_B32_e64 0, 0, 0, 1, [[V_CMP_EQ_U32_e64_]], implicit $exec - ; CHECK-NEXT: IMAGE_STORE_V4_V2_nsa_gfx10 [[V_CNDMASK_B32_e64_]], undef %530:vgpr_32, undef %532:vgpr_32, [[S_LOAD_DWORDX8_IMM]], 15, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), addrspace 8) + ; CHECK-NEXT: IMAGE_STORE_V4_V2_nsa_gfx10 [[V_CNDMASK_B32_e64_]], undef %557:vgpr_32, undef %559:vgpr_32, [[S_LOAD_DWORDX8_IMM]], 15, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), addrspace 8) ; CHECK-NEXT: S_ENDPGM 0 .expVert: %0 = extractelement <31 x i32> %userData, i64 2 diff --git a/llvm/test/CodeGen/AMDGPU/srem64.ll b/llvm/test/CodeGen/AMDGPU/srem64.ll index ea9bb04..862e2dd 100644 --- a/llvm/test/CodeGen/AMDGPU/srem64.ll +++ b/llvm/test/CodeGen/AMDGPU/srem64.ll @@ -8,12 +8,11 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_load_dwordx2 s[8:9], s[4:5], 0xd ; GCN-NEXT: s_load_dwordx4 s[4:7], s[4:5], 0x9 ; GCN-NEXT: s_mov_b32 s3, 0xf000 -; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: s_waitcnt lgkmcnt(0) ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s8 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s9 -; GCN-NEXT: s_sub_u32 s10, 0, s8 -; GCN-NEXT: s_subb_u32 s11, 0, s9 +; GCN-NEXT: s_sub_u32 s0, 0, s8 +; GCN-NEXT: s_subb_u32 s1, 0, s9 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -22,69 +21,65 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GCN-NEXT: v_mul_hi_u32 v2, s10, v0 -; GCN-NEXT: v_readfirstlane_b32 s12, v1 -; GCN-NEXT: v_readfirstlane_b32 s0, v0 -; GCN-NEXT: s_mul_i32 s1, s10, s12 -; GCN-NEXT: v_readfirstlane_b32 s15, v2 -; GCN-NEXT: s_mul_i32 s13, s11, s0 -; GCN-NEXT: s_mul_i32 s14, s10, s0 -; GCN-NEXT: s_add_i32 s1, s15, s1 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s14 -; GCN-NEXT: s_add_i32 s1, s1, s13 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s1 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s14 -; GCN-NEXT: v_readfirstlane_b32 s13, v3 -; GCN-NEXT: s_mul_i32 s15, s0, s1 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s1 -; GCN-NEXT: s_add_u32 s13, s13, s15 +; GCN-NEXT: v_mul_hi_u32 v2, s0, v0 +; GCN-NEXT: v_readfirstlane_b32 s10, v1 +; GCN-NEXT: v_readfirstlane_b32 s2, v0 +; GCN-NEXT: s_mul_i32 s11, s0, s10 +; GCN-NEXT: v_readfirstlane_b32 s14, v2 +; GCN-NEXT: s_mul_i32 s12, s1, s2 +; GCN-NEXT: s_mul_i32 s13, s0, s2 +; GCN-NEXT: s_add_i32 s11, s14, s11 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s13 +; GCN-NEXT: s_add_i32 s11, s11, s12 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s11 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s13 +; GCN-NEXT: v_readfirstlane_b32 s12, v3 +; GCN-NEXT: s_mul_i32 s15, s2, s11 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s11 +; GCN-NEXT: s_add_u32 s12, s12, s15 ; GCN-NEXT: v_readfirstlane_b32 s15, v0 -; GCN-NEXT: s_mul_i32 s14, s12, s14 +; GCN-NEXT: s_mul_i32 s13, s10, s13 ; GCN-NEXT: s_addc_u32 s15, 0, s15 -; GCN-NEXT: v_readfirstlane_b32 s16, v4 -; GCN-NEXT: s_add_u32 s13, s13, s14 -; GCN-NEXT: s_addc_u32 s13, s15, s16 -; GCN-NEXT: v_readfirstlane_b32 s14, v1 -; GCN-NEXT: s_addc_u32 s14, s14, 0 -; GCN-NEXT: s_mul_i32 s1, s12, s1 -; GCN-NEXT: s_add_u32 s1, s13, s1 -; GCN-NEXT: s_addc_u32 s13, 0, s14 -; GCN-NEXT: s_add_u32 s14, s0, s1 -; GCN-NEXT: v_mov_b32_e32 v0, s14 -; GCN-NEXT: v_mul_hi_u32 v0, s10, v0 -; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_addc_u32 s12, s12, s13 -; GCN-NEXT: s_mul_i32 s0, s10, s12 -; GCN-NEXT: v_readfirstlane_b32 s1, v0 -; GCN-NEXT: s_add_i32 s0, s1, s0 -; GCN-NEXT: s_mul_i32 s11, s11, s14 -; GCN-NEXT: s_mul_i32 s1, s10, s14 -; GCN-NEXT: s_add_i32 s0, s0, s11 -; GCN-NEXT: v_mov_b32_e32 v2, s1 -; GCN-NEXT: v_mov_b32_e32 v0, s0 -; GCN-NEXT: v_mul_hi_u32 v3, s12, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s14, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s12, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s14, v0 -; GCN-NEXT: s_mul_i32 s11, s14, s0 -; GCN-NEXT: v_readfirstlane_b32 s15, v2 -; GCN-NEXT: s_add_u32 s11, s15, s11 +; GCN-NEXT: v_readfirstlane_b32 s14, v4 +; GCN-NEXT: s_add_u32 s12, s12, s13 +; GCN-NEXT: s_addc_u32 s12, s15, s14 +; GCN-NEXT: v_readfirstlane_b32 s13, v1 +; GCN-NEXT: s_addc_u32 s13, s13, 0 +; GCN-NEXT: s_mul_i32 s11, s10, s11 +; GCN-NEXT: s_add_u32 s11, s12, s11 +; GCN-NEXT: s_addc_u32 s12, 0, s13 +; GCN-NEXT: s_add_u32 s11, s2, s11 +; GCN-NEXT: v_mov_b32_e32 v0, s11 +; GCN-NEXT: v_mul_hi_u32 v0, s0, v0 +; GCN-NEXT: s_addc_u32 s10, s10, s12 +; GCN-NEXT: s_mul_i32 s12, s0, s10 +; GCN-NEXT: s_mul_i32 s1, s1, s11 ; GCN-NEXT: v_readfirstlane_b32 s13, v0 -; GCN-NEXT: s_mul_i32 s1, s12, s1 -; GCN-NEXT: s_addc_u32 s13, 0, s13 -; GCN-NEXT: v_readfirstlane_b32 s10, v3 -; GCN-NEXT: s_add_u32 s1, s11, s1 -; GCN-NEXT: s_addc_u32 s1, s13, s10 -; GCN-NEXT: v_readfirstlane_b32 s10, v1 -; GCN-NEXT: s_addc_u32 s10, s10, 0 -; GCN-NEXT: s_mul_i32 s0, s12, s0 -; GCN-NEXT: s_add_u32 s0, s1, s0 -; GCN-NEXT: s_addc_u32 s10, 0, s10 -; GCN-NEXT: s_add_u32 s11, s14, s0 -; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_addc_u32 s1, s12, s10 +; GCN-NEXT: s_add_i32 s12, s13, s12 +; GCN-NEXT: s_mul_i32 s0, s0, s11 +; GCN-NEXT: s_add_i32 s1, s12, s1 +; GCN-NEXT: v_mov_b32_e32 v2, s0 +; GCN-NEXT: v_mov_b32_e32 v0, s1 +; GCN-NEXT: v_mul_hi_u32 v3, s10, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s11, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s10, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s11, v0 +; GCN-NEXT: s_mul_i32 s13, s11, s1 +; GCN-NEXT: v_readfirstlane_b32 s15, v2 +; GCN-NEXT: s_add_u32 s13, s15, s13 +; GCN-NEXT: v_readfirstlane_b32 s14, v0 +; GCN-NEXT: s_mul_i32 s0, s10, s0 +; GCN-NEXT: s_addc_u32 s14, 0, s14 +; GCN-NEXT: v_readfirstlane_b32 s12, v3 +; GCN-NEXT: s_add_u32 s0, s13, s0 +; GCN-NEXT: s_addc_u32 s0, s14, s12 +; GCN-NEXT: v_readfirstlane_b32 s12, v1 +; GCN-NEXT: s_addc_u32 s12, s12, 0 +; GCN-NEXT: s_mul_i32 s1, s10, s1 +; GCN-NEXT: s_add_u32 s0, s0, s1 +; GCN-NEXT: s_addc_u32 s1, 0, s12 +; GCN-NEXT: s_add_u32 s11, s11, s0 +; GCN-NEXT: s_addc_u32 s1, s10, s1 ; GCN-NEXT: v_mov_b32_e32 v0, s1 ; GCN-NEXT: v_mul_hi_u32 v1, s6, v0 ; GCN-NEXT: v_mov_b32_e32 v2, s11 @@ -118,11 +113,9 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_mul_i32 s4, s8, s4 ; GCN-NEXT: s_sub_u32 s6, s6, s4 ; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s11, s4, s5 ; GCN-NEXT: s_subb_u32 s13, s10, s9 ; GCN-NEXT: s_sub_u32 s14, s6, s8 ; GCN-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-NEXT: s_or_b32 s15, s10, s11 ; GCN-NEXT: s_subb_u32 s15, s13, 0 ; GCN-NEXT: s_cmp_ge_u32 s15, s9 ; GCN-NEXT: s_cselect_b32 s16, -1, 0 @@ -131,13 +124,11 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_cmp_eq_u32 s15, s9 ; GCN-NEXT: s_cselect_b32 s16, s17, s16 ; GCN-NEXT: s_or_b32 s10, s10, s11 -; GCN-NEXT: s_subb_u32 s13, s13, s9 -; GCN-NEXT: s_sub_u32 s17, s14, s8 -; GCN-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-NEXT: s_or_b32 s10, s10, s11 -; GCN-NEXT: s_subb_u32 s10, s13, 0 +; GCN-NEXT: s_subb_u32 s10, s13, s9 +; GCN-NEXT: s_sub_u32 s11, s14, s8 +; GCN-NEXT: s_subb_u32 s10, s10, 0 ; GCN-NEXT: s_cmp_lg_u32 s16, 0 -; GCN-NEXT: s_cselect_b32 s11, s17, s14 +; GCN-NEXT: s_cselect_b32 s11, s11, s14 ; GCN-NEXT: s_cselect_b32 s10, s10, s15 ; GCN-NEXT: s_or_b32 s4, s4, s5 ; GCN-NEXT: s_subb_u32 s4, s7, s12 @@ -150,6 +141,7 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-NEXT: s_cmp_lg_u32 s5, 0 ; GCN-NEXT: s_cselect_b32 s4, s10, s4 ; GCN-NEXT: s_cselect_b32 s5, s11, s6 +; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_mov_b32_e32 v0, s5 ; GCN-NEXT: v_mov_b32_e32 v1, s4 ; GCN-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0 @@ -180,8 +172,6 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-IR-NEXT: s_cbranch_vccz .LBB0_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s14, s12, 1 -; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-IR-NEXT: s_or_b32 s8, s8, s9 ; GCN-IR-NEXT: s_addc_u32 s8, s13, 0 ; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s12, 63, s12 @@ -213,8 +203,6 @@ define amdgpu_kernel void @s_test_srem(ptr addrspace(1) %out, i64 %x, i64 %y) { ; GCN-IR-NEXT: s_sub_u32 s12, s12, s18 ; GCN-IR-NEXT: s_subb_u32 s13, s13, s19 ; GCN-IR-NEXT: s_add_u32 s16, s16, 1 -; GCN-IR-NEXT: s_cselect_b64 s[18:19], -1, 0 -; GCN-IR-NEXT: s_or_b32 s18, s18, s19 ; GCN-IR-NEXT: s_addc_u32 s17, s17, 0 ; GCN-IR-NEXT: s_cselect_b64 s[18:19], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[10:11], s[4:5] @@ -968,81 +956,76 @@ define amdgpu_kernel void @s_test_srem33_64(ptr addrspace(1) %out, i64 %x, i64 % ; GCN-NEXT: s_xor_b64 s[4:5], s[2:3], s[4:5] ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s4 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s5 -; GCN-NEXT: s_sub_u32 s10, 0, s4 -; GCN-NEXT: s_subb_u32 s11, 0, s5 +; GCN-NEXT: s_sub_u32 s8, 0, s4 +; GCN-NEXT: s_subb_u32 s9, 0, s5 ; GCN-NEXT: s_mov_b32 s3, 0xf000 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 -; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 ; GCN-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0 ; GCN-NEXT: v_trunc_f32_e32 v1, v1 ; GCN-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GCN-NEXT: v_mul_hi_u32 v2, s10, v0 -; GCN-NEXT: v_readfirstlane_b32 s12, v1 -; GCN-NEXT: v_readfirstlane_b32 s8, v0 -; GCN-NEXT: s_mul_i32 s9, s10, s12 -; GCN-NEXT: v_readfirstlane_b32 s15, v2 -; GCN-NEXT: s_mul_i32 s13, s11, s8 -; GCN-NEXT: s_mul_i32 s14, s10, s8 -; GCN-NEXT: s_add_i32 s9, s15, s9 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s14 -; GCN-NEXT: s_add_i32 s9, s9, s13 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s9 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s14 -; GCN-NEXT: v_readfirstlane_b32 s13, v3 -; GCN-NEXT: s_mul_i32 s15, s8, s9 -; GCN-NEXT: s_add_u32 s13, s13, s15 -; GCN-NEXT: v_readfirstlane_b32 s15, v0 -; GCN-NEXT: v_mul_hi_u32 v0, v1, s9 -; GCN-NEXT: s_addc_u32 s15, 0, s15 -; GCN-NEXT: s_mul_i32 s14, s12, s14 -; GCN-NEXT: v_readfirstlane_b32 s16, v4 -; GCN-NEXT: s_add_u32 s13, s13, s14 -; GCN-NEXT: s_addc_u32 s13, s15, s16 +; GCN-NEXT: v_mul_hi_u32 v2, s8, v0 +; GCN-NEXT: v_readfirstlane_b32 s10, v1 +; GCN-NEXT: v_readfirstlane_b32 s2, v0 +; GCN-NEXT: s_mul_i32 s11, s8, s10 +; GCN-NEXT: v_readfirstlane_b32 s14, v2 +; GCN-NEXT: s_mul_i32 s12, s9, s2 +; GCN-NEXT: s_mul_i32 s13, s8, s2 +; GCN-NEXT: s_add_i32 s11, s14, s11 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s13 +; GCN-NEXT: s_add_i32 s11, s11, s12 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s11 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s13 +; GCN-NEXT: v_readfirstlane_b32 s12, v3 +; GCN-NEXT: s_mul_i32 s14, s2, s11 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s11 +; GCN-NEXT: s_add_u32 s12, s12, s14 ; GCN-NEXT: v_readfirstlane_b32 s14, v0 -; GCN-NEXT: s_addc_u32 s14, s14, 0 -; GCN-NEXT: s_mul_i32 s9, s12, s9 -; GCN-NEXT: s_add_u32 s9, s13, s9 -; GCN-NEXT: s_addc_u32 s13, 0, s14 -; GCN-NEXT: s_add_u32 s14, s8, s9 -; GCN-NEXT: v_mov_b32_e32 v0, s14 -; GCN-NEXT: v_mul_hi_u32 v0, s10, v0 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_addc_u32 s12, s12, s13 -; GCN-NEXT: s_mul_i32 s8, s10, s12 -; GCN-NEXT: v_readfirstlane_b32 s9, v0 -; GCN-NEXT: s_add_i32 s8, s9, s8 -; GCN-NEXT: s_mul_i32 s11, s11, s14 -; GCN-NEXT: s_mul_i32 s9, s10, s14 -; GCN-NEXT: s_add_i32 s8, s8, s11 -; GCN-NEXT: v_mov_b32_e32 v2, s9 -; GCN-NEXT: v_mov_b32_e32 v0, s8 -; GCN-NEXT: v_mul_hi_u32 v3, s12, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s14, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s12, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s14, v0 -; GCN-NEXT: s_mul_i32 s11, s14, s8 -; GCN-NEXT: v_readfirstlane_b32 s15, v2 -; GCN-NEXT: s_add_u32 s11, s15, s11 +; GCN-NEXT: s_addc_u32 s14, 0, s14 +; GCN-NEXT: s_mul_i32 s13, s10, s13 +; GCN-NEXT: v_readfirstlane_b32 s15, v4 +; GCN-NEXT: s_add_u32 s12, s12, s13 +; GCN-NEXT: s_addc_u32 s12, s14, s15 +; GCN-NEXT: v_readfirstlane_b32 s13, v1 +; GCN-NEXT: s_addc_u32 s13, s13, 0 +; GCN-NEXT: s_mul_i32 s11, s10, s11 +; GCN-NEXT: s_add_u32 s11, s12, s11 +; GCN-NEXT: s_addc_u32 s12, 0, s13 +; GCN-NEXT: s_add_u32 s11, s2, s11 +; GCN-NEXT: v_mov_b32_e32 v0, s11 +; GCN-NEXT: v_mul_hi_u32 v0, s8, v0 +; GCN-NEXT: s_addc_u32 s10, s10, s12 +; GCN-NEXT: s_mul_i32 s12, s8, s10 +; GCN-NEXT: s_mul_i32 s9, s9, s11 ; GCN-NEXT: v_readfirstlane_b32 s13, v0 -; GCN-NEXT: s_mul_i32 s9, s12, s9 -; GCN-NEXT: s_addc_u32 s13, 0, s13 -; GCN-NEXT: v_readfirstlane_b32 s10, v3 -; GCN-NEXT: s_add_u32 s9, s11, s9 -; GCN-NEXT: s_addc_u32 s9, s13, s10 -; GCN-NEXT: v_readfirstlane_b32 s10, v1 -; GCN-NEXT: s_addc_u32 s10, s10, 0 -; GCN-NEXT: s_mul_i32 s8, s12, s8 -; GCN-NEXT: s_add_u32 s8, s9, s8 -; GCN-NEXT: s_addc_u32 s10, 0, s10 -; GCN-NEXT: s_add_u32 s11, s14, s8 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_addc_u32 s10, s12, s10 +; GCN-NEXT: s_add_i32 s12, s13, s12 +; GCN-NEXT: s_mul_i32 s8, s8, s11 +; GCN-NEXT: s_add_i32 s9, s12, s9 +; GCN-NEXT: v_mov_b32_e32 v2, s8 +; GCN-NEXT: v_mov_b32_e32 v0, s9 +; GCN-NEXT: v_mul_hi_u32 v3, s10, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s11, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s10, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s11, v0 +; GCN-NEXT: s_mul_i32 s13, s11, s9 +; GCN-NEXT: v_readfirstlane_b32 s15, v2 +; GCN-NEXT: s_add_u32 s13, s15, s13 +; GCN-NEXT: v_readfirstlane_b32 s14, v0 +; GCN-NEXT: s_mul_i32 s8, s10, s8 +; GCN-NEXT: s_addc_u32 s14, 0, s14 +; GCN-NEXT: v_readfirstlane_b32 s12, v3 +; GCN-NEXT: s_add_u32 s8, s13, s8 +; GCN-NEXT: s_addc_u32 s8, s14, s12 +; GCN-NEXT: v_readfirstlane_b32 s12, v1 +; GCN-NEXT: s_addc_u32 s12, s12, 0 +; GCN-NEXT: s_mul_i32 s9, s10, s9 +; GCN-NEXT: s_add_u32 s8, s8, s9 +; GCN-NEXT: s_addc_u32 s9, 0, s12 +; GCN-NEXT: s_add_u32 s11, s11, s8 +; GCN-NEXT: s_addc_u32 s10, s10, s9 ; GCN-NEXT: s_ashr_i32 s8, s7, 31 ; GCN-NEXT: s_add_u32 s6, s6, s8 ; GCN-NEXT: s_mov_b32 s9, s8 @@ -1071,6 +1054,7 @@ define amdgpu_kernel void @s_test_srem33_64(ptr addrspace(1) %out, i64 %x, i64 % ; GCN-NEXT: v_mul_hi_u32 v0, s4, v0 ; GCN-NEXT: s_addc_u32 s11, 0, s12 ; GCN-NEXT: s_mul_i32 s11, s4, s11 +; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_readfirstlane_b32 s12, v0 ; GCN-NEXT: s_add_i32 s11, s12, s11 ; GCN-NEXT: s_mul_i32 s12, s5, s10 @@ -1079,11 +1063,9 @@ define amdgpu_kernel void @s_test_srem33_64(ptr addrspace(1) %out, i64 %x, i64 % ; GCN-NEXT: s_mul_i32 s10, s4, s10 ; GCN-NEXT: s_sub_u32 s6, s6, s10 ; GCN-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-NEXT: s_or_b32 s13, s10, s11 ; GCN-NEXT: s_subb_u32 s15, s12, s5 ; GCN-NEXT: s_sub_u32 s16, s6, s4 ; GCN-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GCN-NEXT: s_or_b32 s17, s12, s13 ; GCN-NEXT: s_subb_u32 s17, s15, 0 ; GCN-NEXT: s_cmp_ge_u32 s17, s5 ; GCN-NEXT: s_cselect_b32 s18, -1, 0 @@ -1092,13 +1074,11 @@ define amdgpu_kernel void @s_test_srem33_64(ptr addrspace(1) %out, i64 %x, i64 % ; GCN-NEXT: s_cmp_eq_u32 s17, s5 ; GCN-NEXT: s_cselect_b32 s18, s19, s18 ; GCN-NEXT: s_or_b32 s12, s12, s13 -; GCN-NEXT: s_subb_u32 s15, s15, s5 -; GCN-NEXT: s_sub_u32 s19, s16, s4 -; GCN-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GCN-NEXT: s_or_b32 s12, s12, s13 -; GCN-NEXT: s_subb_u32 s12, s15, 0 +; GCN-NEXT: s_subb_u32 s12, s15, s5 +; GCN-NEXT: s_sub_u32 s13, s16, s4 +; GCN-NEXT: s_subb_u32 s12, s12, 0 ; GCN-NEXT: s_cmp_lg_u32 s18, 0 -; GCN-NEXT: s_cselect_b32 s13, s19, s16 +; GCN-NEXT: s_cselect_b32 s13, s13, s16 ; GCN-NEXT: s_cselect_b32 s12, s12, s17 ; GCN-NEXT: s_or_b32 s10, s10, s11 ; GCN-NEXT: s_subb_u32 s7, s7, s14 @@ -1156,8 +1136,6 @@ define amdgpu_kernel void @s_test_srem33_64(ptr addrspace(1) %out, i64 %x, i64 % ; GCN-IR-NEXT: s_cbranch_vccz .LBB8_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s16, s14, 1 -; GCN-IR-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-IR-NEXT: s_or_b32 s10, s10, s11 ; GCN-IR-NEXT: s_addc_u32 s10, s15, 0 ; GCN-IR-NEXT: s_cselect_b64 s[10:11], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s14, 63, s14 @@ -1189,8 +1167,6 @@ define amdgpu_kernel void @s_test_srem33_64(ptr addrspace(1) %out, i64 %x, i64 % ; GCN-IR-NEXT: s_sub_u32 s14, s14, s20 ; GCN-IR-NEXT: s_subb_u32 s15, s15, s21 ; GCN-IR-NEXT: s_add_u32 s18, s18, 1 -; GCN-IR-NEXT: s_cselect_b64 s[20:21], -1, 0 -; GCN-IR-NEXT: s_or_b32 s20, s20, s21 ; GCN-IR-NEXT: s_addc_u32 s19, s19, 0 ; GCN-IR-NEXT: s_cselect_b64 s[20:21], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[12:13], s[2:3] @@ -1316,8 +1292,7 @@ define amdgpu_kernel void @s_test_srem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s4 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s5 ; GCN-NEXT: s_sub_u32 s2, 0, s4 -; GCN-NEXT: s_subb_u32 s8, 0, s5 -; GCN-NEXT: s_mov_b32 s3, 0xf000 +; GCN-NEXT: s_subb_u32 s6, 0, s5 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -1327,72 +1302,68 @@ define amdgpu_kernel void @s_test_srem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 ; GCN-NEXT: v_mul_hi_u32 v2, s2, v0 -; GCN-NEXT: v_readfirstlane_b32 s9, v1 -; GCN-NEXT: v_readfirstlane_b32 s6, v0 -; GCN-NEXT: s_mul_i32 s7, s2, s9 -; GCN-NEXT: v_readfirstlane_b32 s12, v2 -; GCN-NEXT: s_mul_i32 s10, s8, s6 -; GCN-NEXT: s_mul_i32 s11, s2, s6 -; GCN-NEXT: s_add_i32 s7, s12, s7 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s11 -; GCN-NEXT: s_add_i32 s7, s7, s10 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s7 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s11 -; GCN-NEXT: v_readfirstlane_b32 s10, v3 -; GCN-NEXT: s_mul_i32 s13, s6, s7 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s7 -; GCN-NEXT: s_add_u32 s10, s10, s13 -; GCN-NEXT: v_readfirstlane_b32 s13, v0 -; GCN-NEXT: s_mul_i32 s11, s9, s11 -; GCN-NEXT: s_addc_u32 s13, 0, s13 -; GCN-NEXT: v_readfirstlane_b32 s12, v4 -; GCN-NEXT: s_add_u32 s10, s10, s11 -; GCN-NEXT: s_addc_u32 s10, s13, s12 -; GCN-NEXT: v_readfirstlane_b32 s11, v1 -; GCN-NEXT: s_addc_u32 s11, s11, 0 -; GCN-NEXT: s_mul_i32 s7, s9, s7 -; GCN-NEXT: s_add_u32 s7, s10, s7 -; GCN-NEXT: s_addc_u32 s10, 0, s11 -; GCN-NEXT: s_add_u32 s11, s6, s7 -; GCN-NEXT: v_mov_b32_e32 v0, s11 +; GCN-NEXT: v_readfirstlane_b32 s7, v1 +; GCN-NEXT: v_readfirstlane_b32 s3, v0 +; GCN-NEXT: s_mul_i32 s8, s2, s7 +; GCN-NEXT: v_readfirstlane_b32 s11, v2 +; GCN-NEXT: s_mul_i32 s9, s6, s3 +; GCN-NEXT: s_mul_i32 s10, s2, s3 +; GCN-NEXT: s_add_i32 s8, s11, s8 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s10 +; GCN-NEXT: s_add_i32 s8, s8, s9 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s8 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s10 +; GCN-NEXT: v_readfirstlane_b32 s9, v3 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s8 +; GCN-NEXT: s_mul_i32 s12, s3, s8 +; GCN-NEXT: s_add_u32 s9, s9, s12 +; GCN-NEXT: v_readfirstlane_b32 s12, v0 +; GCN-NEXT: s_mul_i32 s10, s7, s10 +; GCN-NEXT: s_addc_u32 s12, 0, s12 +; GCN-NEXT: v_readfirstlane_b32 s11, v4 +; GCN-NEXT: s_add_u32 s9, s9, s10 +; GCN-NEXT: v_readfirstlane_b32 s13, v1 +; GCN-NEXT: s_addc_u32 s9, s12, s11 +; GCN-NEXT: s_addc_u32 s10, s13, 0 +; GCN-NEXT: s_mul_i32 s8, s7, s8 +; GCN-NEXT: s_add_u32 s8, s9, s8 +; GCN-NEXT: s_addc_u32 s9, 0, s10 +; GCN-NEXT: s_add_u32 s8, s3, s8 +; GCN-NEXT: v_mov_b32_e32 v0, s8 ; GCN-NEXT: v_mul_hi_u32 v0, s2, v0 -; GCN-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-NEXT: s_or_b32 s6, s6, s7 -; GCN-NEXT: s_addc_u32 s9, s9, s10 -; GCN-NEXT: s_mul_i32 s6, s2, s9 -; GCN-NEXT: v_readfirstlane_b32 s7, v0 -; GCN-NEXT: s_add_i32 s6, s7, s6 -; GCN-NEXT: s_mul_i32 s8, s8, s11 -; GCN-NEXT: s_mul_i32 s2, s2, s11 -; GCN-NEXT: s_add_i32 s6, s6, s8 +; GCN-NEXT: s_addc_u32 s7, s7, s9 +; GCN-NEXT: s_mul_i32 s9, s2, s7 +; GCN-NEXT: s_mul_i32 s6, s6, s8 +; GCN-NEXT: v_readfirstlane_b32 s10, v0 +; GCN-NEXT: s_add_i32 s9, s10, s9 +; GCN-NEXT: s_mul_i32 s2, s2, s8 +; GCN-NEXT: s_add_i32 s6, s9, s6 ; GCN-NEXT: v_mov_b32_e32 v2, s2 ; GCN-NEXT: v_mov_b32_e32 v0, s6 -; GCN-NEXT: v_mul_hi_u32 v3, s9, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s11, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s9, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s11, v0 -; GCN-NEXT: s_mul_i32 s8, s11, s6 +; GCN-NEXT: v_mul_hi_u32 v3, s7, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s8, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s7, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s8, v0 +; GCN-NEXT: s_mul_i32 s10, s8, s6 ; GCN-NEXT: v_readfirstlane_b32 s12, v2 -; GCN-NEXT: s_add_u32 s8, s12, s8 -; GCN-NEXT: v_readfirstlane_b32 s10, v0 -; GCN-NEXT: s_mul_i32 s2, s9, s2 -; GCN-NEXT: s_addc_u32 s10, 0, s10 -; GCN-NEXT: v_readfirstlane_b32 s7, v3 -; GCN-NEXT: s_add_u32 s2, s8, s2 -; GCN-NEXT: s_addc_u32 s2, s10, s7 -; GCN-NEXT: v_readfirstlane_b32 s7, v1 -; GCN-NEXT: s_addc_u32 s7, s7, 0 -; GCN-NEXT: s_mul_i32 s6, s9, s6 +; GCN-NEXT: s_add_u32 s10, s12, s10 +; GCN-NEXT: v_readfirstlane_b32 s11, v0 +; GCN-NEXT: s_mul_i32 s2, s7, s2 +; GCN-NEXT: s_addc_u32 s11, 0, s11 +; GCN-NEXT: v_readfirstlane_b32 s9, v3 +; GCN-NEXT: s_add_u32 s2, s10, s2 +; GCN-NEXT: s_addc_u32 s2, s11, s9 +; GCN-NEXT: v_readfirstlane_b32 s9, v1 +; GCN-NEXT: s_addc_u32 s9, s9, 0 +; GCN-NEXT: s_mul_i32 s6, s7, s6 ; GCN-NEXT: s_add_u32 s2, s2, s6 -; GCN-NEXT: s_addc_u32 s8, 0, s7 -; GCN-NEXT: s_add_u32 s2, s11, s2 -; GCN-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-NEXT: s_or_b32 s6, s6, s7 -; GCN-NEXT: s_addc_u32 s6, s9, s8 +; GCN-NEXT: s_addc_u32 s6, 0, s9 +; GCN-NEXT: s_add_u32 s2, s8, s2 +; GCN-NEXT: s_addc_u32 s6, s7, s6 ; GCN-NEXT: v_mul_hi_u32 v1, s2, 24 ; GCN-NEXT: v_mul_hi_u32 v0, s6, 24 ; GCN-NEXT: s_mul_i32 s6, s6, 24 -; GCN-NEXT: s_mov_b32 s2, -1 +; GCN-NEXT: s_mov_b32 s3, 0xf000 ; GCN-NEXT: v_readfirstlane_b32 s8, v1 ; GCN-NEXT: v_readfirstlane_b32 s7, v0 ; GCN-NEXT: s_add_u32 s6, s8, s6 @@ -1401,16 +1372,15 @@ define amdgpu_kernel void @s_test_srem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_mul_hi_u32 v0, s4, v0 ; GCN-NEXT: s_mul_i32 s7, s5, s6 ; GCN-NEXT: s_mul_i32 s6, s4, s6 +; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_readfirstlane_b32 s8, v0 ; GCN-NEXT: s_add_i32 s10, s8, s7 ; GCN-NEXT: s_sub_i32 s8, 0, s10 ; GCN-NEXT: s_sub_u32 s11, 24, s6 ; GCN-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-NEXT: s_or_b32 s9, s6, s7 ; GCN-NEXT: s_subb_u32 s12, s8, s5 ; GCN-NEXT: s_sub_u32 s13, s11, s4 ; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s14, s8, s9 ; GCN-NEXT: s_subb_u32 s14, s12, 0 ; GCN-NEXT: s_cmp_ge_u32 s14, s5 ; GCN-NEXT: s_cselect_b32 s15, -1, 0 @@ -1419,13 +1389,11 @@ define amdgpu_kernel void @s_test_srem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: s_cmp_eq_u32 s14, s5 ; GCN-NEXT: s_cselect_b32 s15, s16, s15 ; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_subb_u32 s12, s12, s5 -; GCN-NEXT: s_sub_u32 s16, s13, s4 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_subb_u32 s8, s12, 0 +; GCN-NEXT: s_subb_u32 s8, s12, s5 +; GCN-NEXT: s_sub_u32 s9, s13, s4 +; GCN-NEXT: s_subb_u32 s8, s8, 0 ; GCN-NEXT: s_cmp_lg_u32 s15, 0 -; GCN-NEXT: s_cselect_b32 s9, s16, s13 +; GCN-NEXT: s_cselect_b32 s9, s9, s13 ; GCN-NEXT: s_cselect_b32 s8, s8, s14 ; GCN-NEXT: s_or_b32 s6, s6, s7 ; GCN-NEXT: s_subb_u32 s6, 0, s10 @@ -1468,8 +1436,6 @@ define amdgpu_kernel void @s_test_srem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_cbranch_vccz .LBB10_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s8, s2, 1 -; GCN-IR-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-IR-NEXT: s_or_b32 s9, s10, s11 ; GCN-IR-NEXT: s_addc_u32 s3, s3, 0 ; GCN-IR-NEXT: s_cselect_b64 s[10:11], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s2, 63, s2 @@ -1500,8 +1466,6 @@ define amdgpu_kernel void @s_test_srem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_sub_u32 s10, s10, s16 ; GCN-IR-NEXT: s_subb_u32 s11, s11, s17 ; GCN-IR-NEXT: s_add_u32 s14, s14, 1 -; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 -; GCN-IR-NEXT: s_or_b32 s16, s16, s17 ; GCN-IR-NEXT: s_addc_u32 s15, s15, 0 ; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[8:9], s[6:7] diff --git a/llvm/test/CodeGen/AMDGPU/uaddo.ll b/llvm/test/CodeGen/AMDGPU/uaddo.ll index bdd22f25..b000fae 100644 --- a/llvm/test/CodeGen/AMDGPU/uaddo.ll +++ b/llvm/test/CodeGen/AMDGPU/uaddo.ll @@ -15,10 +15,8 @@ define amdgpu_kernel void @s_uaddo_i64_zext(ptr addrspace(1) %out, i64 %a, i64 % ; SI-NEXT: s_waitcnt lgkmcnt(0) ; SI-NEXT: s_mov_b32 s4, s0 ; SI-NEXT: s_add_u32 s2, s2, s8 -; SI-NEXT: s_mov_b32 s5, s1 -; SI-NEXT: s_cselect_b64 s[0:1], -1, 0 -; SI-NEXT: s_or_b32 s0, s0, s1 ; SI-NEXT: s_addc_u32 s3, s3, s9 +; SI-NEXT: s_mov_b32 s5, s1 ; SI-NEXT: s_cselect_b64 s[0:1], -1, 0 ; SI-NEXT: v_cndmask_b32_e64 v0, 0, 1, s[0:1] ; SI-NEXT: v_mov_b32_e32 v1, s3 @@ -433,8 +431,6 @@ define amdgpu_kernel void @s_uaddo_i64(ptr addrspace(1) %out, ptr addrspace(1) % ; SI-NEXT: s_mov_b32 s10, -1 ; SI-NEXT: s_waitcnt lgkmcnt(0) ; SI-NEXT: s_add_u32 s4, s4, s6 -; SI-NEXT: s_cselect_b64 s[12:13], -1, 0 -; SI-NEXT: s_or_b32 s6, s12, s13 ; SI-NEXT: s_addc_u32 s5, s5, s7 ; SI-NEXT: s_mov_b32 s8, s0 ; SI-NEXT: s_mov_b32 s9, s1 diff --git a/llvm/test/CodeGen/AMDGPU/udiv64.ll b/llvm/test/CodeGen/AMDGPU/udiv64.ll index fd461ac..775483c 100644 --- a/llvm/test/CodeGen/AMDGPU/udiv64.ll +++ b/llvm/test/CodeGen/AMDGPU/udiv64.ll @@ -146,8 +146,6 @@ define amdgpu_kernel void @s_test_udiv_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-IR-NEXT: s_cbranch_vccz .LBB0_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s14, s12, 1 -; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-IR-NEXT: s_or_b32 s8, s8, s9 ; GCN-IR-NEXT: s_addc_u32 s8, s13, 0 ; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s12, 63, s12 @@ -179,8 +177,6 @@ define amdgpu_kernel void @s_test_udiv_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-IR-NEXT: s_sub_u32 s12, s12, s16 ; GCN-IR-NEXT: s_subb_u32 s13, s13, s17 ; GCN-IR-NEXT: s_add_u32 s10, s10, 1 -; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 -; GCN-IR-NEXT: s_or_b32 s16, s16, s17 ; GCN-IR-NEXT: s_addc_u32 s11, s11, 0 ; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[2:3], s[4:5] @@ -786,12 +782,11 @@ define amdgpu_kernel void @s_test_udiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-LABEL: s_test_udiv_k_num_i64: ; GCN: ; %bb.0: ; GCN-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x9 -; GCN-NEXT: s_mov_b32 s7, 0xf000 ; GCN-NEXT: s_waitcnt lgkmcnt(0) ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s2 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s3 -; GCN-NEXT: s_sub_u32 s6, 0, s2 -; GCN-NEXT: s_subb_u32 s8, 0, s3 +; GCN-NEXT: s_sub_u32 s4, 0, s2 +; GCN-NEXT: s_subb_u32 s5, 0, s3 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -800,118 +795,112 @@ define amdgpu_kernel void @s_test_udiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GCN-NEXT: v_mul_hi_u32 v2, s6, v0 +; GCN-NEXT: v_mul_hi_u32 v2, s4, v0 +; GCN-NEXT: v_readfirstlane_b32 s6, v1 +; GCN-NEXT: v_readfirstlane_b32 s7, v0 +; GCN-NEXT: s_mul_i32 s8, s4, s6 +; GCN-NEXT: v_readfirstlane_b32 s11, v2 +; GCN-NEXT: s_mul_i32 s9, s5, s7 +; GCN-NEXT: s_mul_i32 s10, s4, s7 +; GCN-NEXT: s_add_i32 s8, s11, s8 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s10 +; GCN-NEXT: s_add_i32 s8, s8, s9 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s8 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s10 +; GCN-NEXT: v_readfirstlane_b32 s9, v3 +; GCN-NEXT: s_mul_i32 s12, s7, s8 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s8 +; GCN-NEXT: s_add_u32 s9, s9, s12 +; GCN-NEXT: v_readfirstlane_b32 s12, v0 +; GCN-NEXT: s_mul_i32 s10, s6, s10 +; GCN-NEXT: s_addc_u32 s12, 0, s12 +; GCN-NEXT: v_readfirstlane_b32 s11, v4 +; GCN-NEXT: s_add_u32 s9, s9, s10 +; GCN-NEXT: v_readfirstlane_b32 s13, v1 +; GCN-NEXT: s_addc_u32 s9, s12, s11 +; GCN-NEXT: s_mul_i32 s8, s6, s8 +; GCN-NEXT: s_addc_u32 s10, s13, 0 +; GCN-NEXT: s_add_u32 s8, s9, s8 +; GCN-NEXT: s_addc_u32 s9, 0, s10 +; GCN-NEXT: s_add_u32 s8, s7, s8 +; GCN-NEXT: v_mov_b32_e32 v0, s8 +; GCN-NEXT: v_mul_hi_u32 v0, s4, v0 +; GCN-NEXT: s_addc_u32 s6, s6, s9 +; GCN-NEXT: s_mul_i32 s9, s4, s6 +; GCN-NEXT: s_mul_i32 s5, s5, s8 +; GCN-NEXT: v_readfirstlane_b32 s10, v0 +; GCN-NEXT: s_add_i32 s9, s10, s9 +; GCN-NEXT: s_mul_i32 s4, s4, s8 +; GCN-NEXT: s_add_i32 s5, s9, s5 +; GCN-NEXT: v_mov_b32_e32 v2, s4 +; GCN-NEXT: v_mov_b32_e32 v0, s5 +; GCN-NEXT: v_mul_hi_u32 v3, s6, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s8, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s6, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s8, v0 +; GCN-NEXT: s_mul_i32 s10, s8, s5 +; GCN-NEXT: v_readfirstlane_b32 s12, v2 +; GCN-NEXT: s_add_u32 s10, s12, s10 +; GCN-NEXT: v_readfirstlane_b32 s11, v0 +; GCN-NEXT: s_mul_i32 s4, s6, s4 +; GCN-NEXT: s_addc_u32 s11, 0, s11 +; GCN-NEXT: v_readfirstlane_b32 s9, v3 +; GCN-NEXT: s_add_u32 s4, s10, s4 +; GCN-NEXT: s_addc_u32 s4, s11, s9 ; GCN-NEXT: v_readfirstlane_b32 s9, v1 +; GCN-NEXT: s_addc_u32 s9, s9, 0 +; GCN-NEXT: s_mul_i32 s5, s6, s5 +; GCN-NEXT: s_add_u32 s4, s4, s5 +; GCN-NEXT: s_addc_u32 s5, 0, s9 +; GCN-NEXT: s_add_u32 s4, s8, s4 +; GCN-NEXT: s_addc_u32 s5, s6, s5 +; GCN-NEXT: v_mul_hi_u32 v1, s4, 24 +; GCN-NEXT: v_mul_hi_u32 v0, s5, 24 +; GCN-NEXT: s_mul_i32 s5, s5, 24 +; GCN-NEXT: s_mov_b32 s7, 0xf000 +; GCN-NEXT: v_readfirstlane_b32 s8, v1 ; GCN-NEXT: v_readfirstlane_b32 s4, v0 -; GCN-NEXT: s_mul_i32 s5, s6, s9 -; GCN-NEXT: v_readfirstlane_b32 s12, v2 -; GCN-NEXT: s_mul_i32 s10, s8, s4 -; GCN-NEXT: s_mul_i32 s11, s6, s4 -; GCN-NEXT: s_add_i32 s5, s12, s5 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s11 -; GCN-NEXT: s_add_i32 s5, s5, s10 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s5 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s11 -; GCN-NEXT: v_readfirstlane_b32 s10, v3 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s5 -; GCN-NEXT: s_mul_i32 s13, s4, s5 -; GCN-NEXT: s_add_u32 s10, s10, s13 -; GCN-NEXT: v_readfirstlane_b32 s13, v0 -; GCN-NEXT: s_mul_i32 s11, s9, s11 -; GCN-NEXT: s_addc_u32 s13, 0, s13 -; GCN-NEXT: v_readfirstlane_b32 s12, v4 -; GCN-NEXT: s_add_u32 s10, s10, s11 -; GCN-NEXT: v_readfirstlane_b32 s14, v1 -; GCN-NEXT: s_addc_u32 s10, s13, s12 -; GCN-NEXT: s_addc_u32 s11, s14, 0 -; GCN-NEXT: s_mul_i32 s5, s9, s5 -; GCN-NEXT: s_add_u32 s5, s10, s5 -; GCN-NEXT: s_addc_u32 s10, 0, s11 -; GCN-NEXT: s_add_u32 s11, s4, s5 -; GCN-NEXT: v_mov_b32_e32 v0, s11 -; GCN-NEXT: v_mul_hi_u32 v0, s6, v0 -; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s4, s4, s5 -; GCN-NEXT: s_addc_u32 s9, s9, s10 -; GCN-NEXT: s_mul_i32 s4, s6, s9 -; GCN-NEXT: v_readfirstlane_b32 s5, v0 -; GCN-NEXT: s_add_i32 s4, s5, s4 -; GCN-NEXT: s_mul_i32 s8, s8, s11 -; GCN-NEXT: s_mul_i32 s5, s6, s11 -; GCN-NEXT: s_add_i32 s4, s4, s8 -; GCN-NEXT: v_mov_b32_e32 v2, s5 -; GCN-NEXT: v_mov_b32_e32 v0, s4 -; GCN-NEXT: v_mul_hi_u32 v3, s9, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s11, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s9, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s11, v0 -; GCN-NEXT: s_mul_i32 s8, s11, s4 -; GCN-NEXT: v_readfirstlane_b32 s12, v2 -; GCN-NEXT: s_add_u32 s8, s12, s8 -; GCN-NEXT: v_readfirstlane_b32 s10, v0 -; GCN-NEXT: s_mul_i32 s5, s9, s5 -; GCN-NEXT: s_addc_u32 s10, 0, s10 -; GCN-NEXT: v_readfirstlane_b32 s6, v3 ; GCN-NEXT: s_add_u32 s5, s8, s5 -; GCN-NEXT: s_addc_u32 s5, s10, s6 -; GCN-NEXT: v_readfirstlane_b32 s6, v1 -; GCN-NEXT: s_addc_u32 s6, s6, 0 -; GCN-NEXT: s_mul_i32 s4, s9, s4 -; GCN-NEXT: s_add_u32 s4, s5, s4 -; GCN-NEXT: s_addc_u32 s6, 0, s6 -; GCN-NEXT: s_add_u32 s8, s11, s4 -; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s4, s4, s5 -; GCN-NEXT: s_addc_u32 s4, s9, s6 -; GCN-NEXT: v_mul_hi_u32 v1, s8, 24 -; GCN-NEXT: v_mul_hi_u32 v0, s4, 24 -; GCN-NEXT: s_mul_i32 s4, s4, 24 -; GCN-NEXT: s_mov_b32 s6, -1 -; GCN-NEXT: v_readfirstlane_b32 s8, v1 -; GCN-NEXT: v_readfirstlane_b32 s5, v0 -; GCN-NEXT: s_add_u32 s4, s8, s4 -; GCN-NEXT: s_addc_u32 s10, 0, s5 -; GCN-NEXT: v_mov_b32_e32 v0, s10 +; GCN-NEXT: s_addc_u32 s8, 0, s4 +; GCN-NEXT: v_mov_b32_e32 v0, s8 ; GCN-NEXT: v_mul_hi_u32 v0, s2, v0 ; GCN-NEXT: s_mov_b32 s4, s0 ; GCN-NEXT: s_mov_b32 s5, s1 -; GCN-NEXT: s_mul_i32 s0, s3, s10 +; GCN-NEXT: s_mul_i32 s0, s3, s8 ; GCN-NEXT: v_readfirstlane_b32 s1, v0 -; GCN-NEXT: s_add_i32 s11, s1, s0 -; GCN-NEXT: s_sub_i32 s8, 0, s11 -; GCN-NEXT: s_mul_i32 s0, s2, s10 -; GCN-NEXT: s_sub_u32 s12, 24, s0 +; GCN-NEXT: s_add_i32 s9, s1, s0 +; GCN-NEXT: s_sub_i32 s10, 0, s9 +; GCN-NEXT: s_mul_i32 s0, s2, s8 +; GCN-NEXT: s_sub_u32 s11, 24, s0 ; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s9, s0, s1 -; GCN-NEXT: s_subb_u32 s13, s8, s3 -; GCN-NEXT: s_sub_u32 s14, s12, s2 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_subb_u32 s8, s13, 0 -; GCN-NEXT: s_cmp_ge_u32 s8, s3 -; GCN-NEXT: s_cselect_b32 s9, -1, 0 -; GCN-NEXT: s_cmp_ge_u32 s14, s2 +; GCN-NEXT: s_subb_u32 s10, s10, s3 +; GCN-NEXT: s_sub_u32 s12, s11, s2 +; GCN-NEXT: s_subb_u32 s10, s10, 0 +; GCN-NEXT: s_cmp_ge_u32 s10, s3 ; GCN-NEXT: s_cselect_b32 s13, -1, 0 -; GCN-NEXT: s_cmp_eq_u32 s8, s3 -; GCN-NEXT: s_cselect_b32 s8, s13, s9 -; GCN-NEXT: s_add_u32 s9, s10, 1 +; GCN-NEXT: s_cmp_ge_u32 s12, s2 +; GCN-NEXT: s_cselect_b32 s12, -1, 0 +; GCN-NEXT: s_cmp_eq_u32 s10, s3 +; GCN-NEXT: s_cselect_b32 s10, s12, s13 +; GCN-NEXT: s_add_u32 s12, s8, 1 ; GCN-NEXT: s_addc_u32 s13, 0, 0 -; GCN-NEXT: s_add_u32 s14, s10, 2 +; GCN-NEXT: s_add_u32 s14, s8, 2 ; GCN-NEXT: s_addc_u32 s15, 0, 0 -; GCN-NEXT: s_cmp_lg_u32 s8, 0 -; GCN-NEXT: s_cselect_b32 s8, s14, s9 -; GCN-NEXT: s_cselect_b32 s9, s15, s13 +; GCN-NEXT: s_cmp_lg_u32 s10, 0 +; GCN-NEXT: s_cselect_b32 s10, s14, s12 +; GCN-NEXT: s_cselect_b32 s12, s15, s13 ; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_subb_u32 s0, 0, s11 +; GCN-NEXT: s_subb_u32 s0, 0, s9 ; GCN-NEXT: s_cmp_ge_u32 s0, s3 ; GCN-NEXT: s_cselect_b32 s1, -1, 0 -; GCN-NEXT: s_cmp_ge_u32 s12, s2 +; GCN-NEXT: s_cmp_ge_u32 s11, s2 ; GCN-NEXT: s_cselect_b32 s2, -1, 0 ; GCN-NEXT: s_cmp_eq_u32 s0, s3 ; GCN-NEXT: s_cselect_b32 s0, s2, s1 ; GCN-NEXT: s_cmp_lg_u32 s0, 0 -; GCN-NEXT: s_cselect_b32 s0, s9, 0 -; GCN-NEXT: s_cselect_b32 s1, s8, s10 +; GCN-NEXT: s_cselect_b32 s0, s12, 0 +; GCN-NEXT: s_cselect_b32 s1, s10, s8 +; GCN-NEXT: s_mov_b32 s6, -1 ; GCN-NEXT: v_mov_b32_e32 v0, s1 ; GCN-NEXT: v_mov_b32_e32 v1, s0 ; GCN-NEXT: buffer_store_dwordx2 v[0:1], off, s[4:7], 0 @@ -937,8 +926,6 @@ define amdgpu_kernel void @s_test_udiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_cbranch_vccz .LBB8_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s10, s8, 1 -; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-IR-NEXT: s_or_b32 s6, s6, s7 ; GCN-IR-NEXT: s_addc_u32 s6, s9, 0 ; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s8, 63, s8 @@ -969,8 +956,6 @@ define amdgpu_kernel void @s_test_udiv_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_sub_u32 s10, s10, s16 ; GCN-IR-NEXT: s_subb_u32 s11, s11, s17 ; GCN-IR-NEXT: s_add_u32 s14, s14, 1 -; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 -; GCN-IR-NEXT: s_or_b32 s16, s16, s17 ; GCN-IR-NEXT: s_addc_u32 s15, s15, 0 ; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[8:9], s[4:5] @@ -1307,8 +1292,6 @@ define amdgpu_kernel void @s_test_udiv_k_den_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_cbranch_vccz .LBB11_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s11, s8, 1 -; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-IR-NEXT: s_or_b32 s6, s6, s7 ; GCN-IR-NEXT: s_addc_u32 s6, s9, 0 ; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s8, 63, s8 @@ -1336,8 +1319,6 @@ define amdgpu_kernel void @s_test_udiv_k_den_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_sub_u32 s2, s2, s8 ; GCN-IR-NEXT: s_subb_u32 s3, s3, 0 ; GCN-IR-NEXT: s_add_u32 s10, s10, 1 -; GCN-IR-NEXT: s_cselect_b64 s[12:13], -1, 0 -; GCN-IR-NEXT: s_or_b32 s12, s12, s13 ; GCN-IR-NEXT: s_addc_u32 s11, s11, 0 ; GCN-IR-NEXT: s_cselect_b64 s[12:13], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[8:9], s[4:5] diff --git a/llvm/test/CodeGen/AMDGPU/urem64.ll b/llvm/test/CodeGen/AMDGPU/urem64.ll index 137dc1f..28e6627 100644 --- a/llvm/test/CodeGen/AMDGPU/urem64.ll +++ b/llvm/test/CodeGen/AMDGPU/urem64.ll @@ -8,12 +8,11 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-NEXT: s_load_dwordx2 s[8:9], s[4:5], 0xd ; GCN-NEXT: s_load_dwordx4 s[4:7], s[4:5], 0x9 ; GCN-NEXT: s_mov_b32 s3, 0xf000 -; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: s_waitcnt lgkmcnt(0) ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s8 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s9 -; GCN-NEXT: s_sub_u32 s10, 0, s8 -; GCN-NEXT: s_subb_u32 s11, 0, s9 +; GCN-NEXT: s_sub_u32 s0, 0, s8 +; GCN-NEXT: s_subb_u32 s1, 0, s9 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -22,69 +21,65 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GCN-NEXT: v_mul_hi_u32 v2, s10, v0 -; GCN-NEXT: v_readfirstlane_b32 s12, v1 -; GCN-NEXT: v_readfirstlane_b32 s0, v0 -; GCN-NEXT: s_mul_i32 s1, s10, s12 -; GCN-NEXT: v_readfirstlane_b32 s15, v2 -; GCN-NEXT: s_mul_i32 s13, s11, s0 -; GCN-NEXT: s_mul_i32 s14, s10, s0 -; GCN-NEXT: s_add_i32 s1, s15, s1 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s14 -; GCN-NEXT: s_add_i32 s1, s1, s13 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s1 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s14 -; GCN-NEXT: v_readfirstlane_b32 s13, v3 -; GCN-NEXT: s_mul_i32 s15, s0, s1 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s1 -; GCN-NEXT: s_add_u32 s13, s13, s15 +; GCN-NEXT: v_mul_hi_u32 v2, s0, v0 +; GCN-NEXT: v_readfirstlane_b32 s10, v1 +; GCN-NEXT: v_readfirstlane_b32 s2, v0 +; GCN-NEXT: s_mul_i32 s11, s0, s10 +; GCN-NEXT: v_readfirstlane_b32 s14, v2 +; GCN-NEXT: s_mul_i32 s12, s1, s2 +; GCN-NEXT: s_mul_i32 s13, s0, s2 +; GCN-NEXT: s_add_i32 s11, s14, s11 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s13 +; GCN-NEXT: s_add_i32 s11, s11, s12 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s11 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s13 +; GCN-NEXT: v_readfirstlane_b32 s12, v3 +; GCN-NEXT: s_mul_i32 s15, s2, s11 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s11 +; GCN-NEXT: s_add_u32 s12, s12, s15 ; GCN-NEXT: v_readfirstlane_b32 s15, v0 -; GCN-NEXT: s_mul_i32 s14, s12, s14 +; GCN-NEXT: s_mul_i32 s13, s10, s13 ; GCN-NEXT: s_addc_u32 s15, 0, s15 -; GCN-NEXT: v_readfirstlane_b32 s16, v4 -; GCN-NEXT: s_add_u32 s13, s13, s14 -; GCN-NEXT: s_addc_u32 s13, s15, s16 -; GCN-NEXT: v_readfirstlane_b32 s14, v1 -; GCN-NEXT: s_addc_u32 s14, s14, 0 -; GCN-NEXT: s_mul_i32 s1, s12, s1 -; GCN-NEXT: s_add_u32 s1, s13, s1 -; GCN-NEXT: s_addc_u32 s13, 0, s14 -; GCN-NEXT: s_add_u32 s14, s0, s1 -; GCN-NEXT: v_mov_b32_e32 v0, s14 -; GCN-NEXT: v_mul_hi_u32 v0, s10, v0 -; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_addc_u32 s12, s12, s13 -; GCN-NEXT: s_mul_i32 s0, s10, s12 -; GCN-NEXT: v_readfirstlane_b32 s1, v0 -; GCN-NEXT: s_add_i32 s0, s1, s0 -; GCN-NEXT: s_mul_i32 s11, s11, s14 -; GCN-NEXT: s_mul_i32 s1, s10, s14 -; GCN-NEXT: s_add_i32 s0, s0, s11 -; GCN-NEXT: v_mov_b32_e32 v2, s1 -; GCN-NEXT: v_mov_b32_e32 v0, s0 -; GCN-NEXT: v_mul_hi_u32 v3, s12, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s14, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s12, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s14, v0 -; GCN-NEXT: s_mul_i32 s11, s14, s0 -; GCN-NEXT: v_readfirstlane_b32 s15, v2 -; GCN-NEXT: s_add_u32 s11, s15, s11 +; GCN-NEXT: v_readfirstlane_b32 s14, v4 +; GCN-NEXT: s_add_u32 s12, s12, s13 +; GCN-NEXT: s_addc_u32 s12, s15, s14 +; GCN-NEXT: v_readfirstlane_b32 s13, v1 +; GCN-NEXT: s_addc_u32 s13, s13, 0 +; GCN-NEXT: s_mul_i32 s11, s10, s11 +; GCN-NEXT: s_add_u32 s11, s12, s11 +; GCN-NEXT: s_addc_u32 s12, 0, s13 +; GCN-NEXT: s_add_u32 s11, s2, s11 +; GCN-NEXT: v_mov_b32_e32 v0, s11 +; GCN-NEXT: v_mul_hi_u32 v0, s0, v0 +; GCN-NEXT: s_addc_u32 s10, s10, s12 +; GCN-NEXT: s_mul_i32 s12, s0, s10 +; GCN-NEXT: s_mul_i32 s1, s1, s11 ; GCN-NEXT: v_readfirstlane_b32 s13, v0 -; GCN-NEXT: s_mul_i32 s1, s12, s1 -; GCN-NEXT: s_addc_u32 s13, 0, s13 -; GCN-NEXT: v_readfirstlane_b32 s10, v3 -; GCN-NEXT: s_add_u32 s1, s11, s1 -; GCN-NEXT: s_addc_u32 s1, s13, s10 -; GCN-NEXT: v_readfirstlane_b32 s10, v1 -; GCN-NEXT: s_addc_u32 s10, s10, 0 -; GCN-NEXT: s_mul_i32 s0, s12, s0 -; GCN-NEXT: s_add_u32 s0, s1, s0 -; GCN-NEXT: s_addc_u32 s10, 0, s10 -; GCN-NEXT: s_add_u32 s11, s14, s0 -; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s0, s0, s1 -; GCN-NEXT: s_addc_u32 s1, s12, s10 +; GCN-NEXT: s_add_i32 s12, s13, s12 +; GCN-NEXT: s_mul_i32 s0, s0, s11 +; GCN-NEXT: s_add_i32 s1, s12, s1 +; GCN-NEXT: v_mov_b32_e32 v2, s0 +; GCN-NEXT: v_mov_b32_e32 v0, s1 +; GCN-NEXT: v_mul_hi_u32 v3, s10, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s11, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s10, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s11, v0 +; GCN-NEXT: s_mul_i32 s13, s11, s1 +; GCN-NEXT: v_readfirstlane_b32 s15, v2 +; GCN-NEXT: s_add_u32 s13, s15, s13 +; GCN-NEXT: v_readfirstlane_b32 s14, v0 +; GCN-NEXT: s_mul_i32 s0, s10, s0 +; GCN-NEXT: s_addc_u32 s14, 0, s14 +; GCN-NEXT: v_readfirstlane_b32 s12, v3 +; GCN-NEXT: s_add_u32 s0, s13, s0 +; GCN-NEXT: s_addc_u32 s0, s14, s12 +; GCN-NEXT: v_readfirstlane_b32 s12, v1 +; GCN-NEXT: s_addc_u32 s12, s12, 0 +; GCN-NEXT: s_mul_i32 s1, s10, s1 +; GCN-NEXT: s_add_u32 s0, s0, s1 +; GCN-NEXT: s_addc_u32 s1, 0, s12 +; GCN-NEXT: s_add_u32 s11, s11, s0 +; GCN-NEXT: s_addc_u32 s1, s10, s1 ; GCN-NEXT: v_mov_b32_e32 v0, s1 ; GCN-NEXT: v_mul_hi_u32 v1, s6, v0 ; GCN-NEXT: v_mov_b32_e32 v2, s11 @@ -118,11 +113,9 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-NEXT: s_mul_i32 s4, s8, s4 ; GCN-NEXT: s_sub_u32 s6, s6, s4 ; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s11, s4, s5 ; GCN-NEXT: s_subb_u32 s13, s10, s9 ; GCN-NEXT: s_sub_u32 s14, s6, s8 ; GCN-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-NEXT: s_or_b32 s15, s10, s11 ; GCN-NEXT: s_subb_u32 s15, s13, 0 ; GCN-NEXT: s_cmp_ge_u32 s15, s9 ; GCN-NEXT: s_cselect_b32 s16, -1, 0 @@ -131,13 +124,11 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-NEXT: s_cmp_eq_u32 s15, s9 ; GCN-NEXT: s_cselect_b32 s16, s17, s16 ; GCN-NEXT: s_or_b32 s10, s10, s11 -; GCN-NEXT: s_subb_u32 s13, s13, s9 -; GCN-NEXT: s_sub_u32 s17, s14, s8 -; GCN-NEXT: s_cselect_b64 s[10:11], -1, 0 -; GCN-NEXT: s_or_b32 s10, s10, s11 -; GCN-NEXT: s_subb_u32 s10, s13, 0 +; GCN-NEXT: s_subb_u32 s10, s13, s9 +; GCN-NEXT: s_sub_u32 s11, s14, s8 +; GCN-NEXT: s_subb_u32 s10, s10, 0 ; GCN-NEXT: s_cmp_lg_u32 s16, 0 -; GCN-NEXT: s_cselect_b32 s11, s17, s14 +; GCN-NEXT: s_cselect_b32 s11, s11, s14 ; GCN-NEXT: s_cselect_b32 s10, s10, s15 ; GCN-NEXT: s_or_b32 s4, s4, s5 ; GCN-NEXT: s_subb_u32 s4, s7, s12 @@ -150,6 +141,7 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-NEXT: s_cmp_lg_u32 s5, 0 ; GCN-NEXT: s_cselect_b32 s4, s10, s4 ; GCN-NEXT: s_cselect_b32 s5, s11, s6 +; GCN-NEXT: s_mov_b32 s2, -1 ; GCN-NEXT: v_mov_b32_e32 v0, s5 ; GCN-NEXT: v_mov_b32_e32 v1, s4 ; GCN-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0 @@ -180,8 +172,6 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-IR-NEXT: s_cbranch_vccz .LBB0_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s14, s12, 1 -; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-IR-NEXT: s_or_b32 s8, s8, s9 ; GCN-IR-NEXT: s_addc_u32 s8, s13, 0 ; GCN-IR-NEXT: s_cselect_b64 s[8:9], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s12, 63, s12 @@ -213,8 +203,6 @@ define amdgpu_kernel void @s_test_urem_i64(ptr addrspace(1) %out, i64 %x, i64 %y ; GCN-IR-NEXT: s_sub_u32 s12, s12, s18 ; GCN-IR-NEXT: s_subb_u32 s13, s13, s19 ; GCN-IR-NEXT: s_add_u32 s16, s16, 1 -; GCN-IR-NEXT: s_cselect_b64 s[18:19], -1, 0 -; GCN-IR-NEXT: s_or_b32 s18, s18, s19 ; GCN-IR-NEXT: s_addc_u32 s17, s17, 0 ; GCN-IR-NEXT: s_cselect_b64 s[18:19], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[10:11], s[4:5] @@ -803,12 +791,11 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-LABEL: s_test_urem_k_num_i64: ; GCN: ; %bb.0: ; GCN-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x9 -; GCN-NEXT: s_mov_b32 s7, 0xf000 ; GCN-NEXT: s_waitcnt lgkmcnt(0) ; GCN-NEXT: v_cvt_f32_u32_e32 v0, s2 ; GCN-NEXT: v_cvt_f32_u32_e32 v1, s3 -; GCN-NEXT: s_sub_u32 s6, 0, s2 -; GCN-NEXT: s_subb_u32 s8, 0, s3 +; GCN-NEXT: s_sub_u32 s4, 0, s2 +; GCN-NEXT: s_subb_u32 s5, 0, s3 ; GCN-NEXT: v_madmk_f32 v0, v1, 0x4f800000, v0 ; GCN-NEXT: v_rcp_f32_e32 v0, v0 ; GCN-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0 @@ -817,77 +804,73 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: v_madmk_f32 v0, v1, 0xcf800000, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v0, v0 ; GCN-NEXT: v_cvt_u32_f32_e32 v1, v1 -; GCN-NEXT: v_mul_hi_u32 v2, s6, v0 +; GCN-NEXT: v_mul_hi_u32 v2, s4, v0 +; GCN-NEXT: v_readfirstlane_b32 s6, v1 +; GCN-NEXT: v_readfirstlane_b32 s7, v0 +; GCN-NEXT: s_mul_i32 s8, s4, s6 +; GCN-NEXT: v_readfirstlane_b32 s11, v2 +; GCN-NEXT: s_mul_i32 s9, s5, s7 +; GCN-NEXT: s_mul_i32 s10, s4, s7 +; GCN-NEXT: s_add_i32 s8, s11, s8 +; GCN-NEXT: v_mul_hi_u32 v3, v0, s10 +; GCN-NEXT: s_add_i32 s8, s8, s9 +; GCN-NEXT: v_mul_hi_u32 v0, v0, s8 +; GCN-NEXT: v_mul_hi_u32 v4, v1, s10 +; GCN-NEXT: v_readfirstlane_b32 s9, v3 +; GCN-NEXT: s_mul_i32 s12, s7, s8 +; GCN-NEXT: v_mul_hi_u32 v1, v1, s8 +; GCN-NEXT: s_add_u32 s9, s9, s12 +; GCN-NEXT: v_readfirstlane_b32 s12, v0 +; GCN-NEXT: s_mul_i32 s10, s6, s10 +; GCN-NEXT: s_addc_u32 s12, 0, s12 +; GCN-NEXT: v_readfirstlane_b32 s11, v4 +; GCN-NEXT: s_add_u32 s9, s9, s10 +; GCN-NEXT: v_readfirstlane_b32 s13, v1 +; GCN-NEXT: s_addc_u32 s9, s12, s11 +; GCN-NEXT: s_mul_i32 s8, s6, s8 +; GCN-NEXT: s_addc_u32 s10, s13, 0 +; GCN-NEXT: s_add_u32 s8, s9, s8 +; GCN-NEXT: s_addc_u32 s9, 0, s10 +; GCN-NEXT: s_add_u32 s8, s7, s8 +; GCN-NEXT: v_mov_b32_e32 v0, s8 +; GCN-NEXT: v_mul_hi_u32 v0, s4, v0 +; GCN-NEXT: s_addc_u32 s6, s6, s9 +; GCN-NEXT: s_mul_i32 s9, s4, s6 +; GCN-NEXT: s_mul_i32 s5, s5, s8 +; GCN-NEXT: v_readfirstlane_b32 s10, v0 +; GCN-NEXT: s_add_i32 s9, s10, s9 +; GCN-NEXT: s_mul_i32 s4, s4, s8 +; GCN-NEXT: s_add_i32 s5, s9, s5 +; GCN-NEXT: v_mov_b32_e32 v2, s4 +; GCN-NEXT: v_mov_b32_e32 v0, s5 +; GCN-NEXT: v_mul_hi_u32 v3, s6, v2 +; GCN-NEXT: v_mul_hi_u32 v2, s8, v2 +; GCN-NEXT: v_mul_hi_u32 v1, s6, v0 +; GCN-NEXT: v_mul_hi_u32 v0, s8, v0 +; GCN-NEXT: s_mul_i32 s10, s8, s5 +; GCN-NEXT: v_readfirstlane_b32 s12, v2 +; GCN-NEXT: s_add_u32 s10, s12, s10 +; GCN-NEXT: v_readfirstlane_b32 s11, v0 +; GCN-NEXT: s_mul_i32 s4, s6, s4 +; GCN-NEXT: s_addc_u32 s11, 0, s11 +; GCN-NEXT: v_readfirstlane_b32 s9, v3 +; GCN-NEXT: s_add_u32 s4, s10, s4 +; GCN-NEXT: s_addc_u32 s4, s11, s9 ; GCN-NEXT: v_readfirstlane_b32 s9, v1 +; GCN-NEXT: s_addc_u32 s9, s9, 0 +; GCN-NEXT: s_mul_i32 s5, s6, s5 +; GCN-NEXT: s_add_u32 s4, s4, s5 +; GCN-NEXT: s_addc_u32 s5, 0, s9 +; GCN-NEXT: s_add_u32 s4, s8, s4 +; GCN-NEXT: s_addc_u32 s5, s6, s5 +; GCN-NEXT: v_mul_hi_u32 v1, s4, 24 +; GCN-NEXT: v_mul_hi_u32 v0, s5, 24 +; GCN-NEXT: s_mul_i32 s5, s5, 24 +; GCN-NEXT: s_mov_b32 s7, 0xf000 +; GCN-NEXT: v_readfirstlane_b32 s8, v1 ; GCN-NEXT: v_readfirstlane_b32 s4, v0 -; GCN-NEXT: s_mul_i32 s5, s6, s9 -; GCN-NEXT: v_readfirstlane_b32 s12, v2 -; GCN-NEXT: s_mul_i32 s10, s8, s4 -; GCN-NEXT: s_mul_i32 s11, s6, s4 -; GCN-NEXT: s_add_i32 s5, s12, s5 -; GCN-NEXT: v_mul_hi_u32 v3, v0, s11 -; GCN-NEXT: s_add_i32 s5, s5, s10 -; GCN-NEXT: v_mul_hi_u32 v0, v0, s5 -; GCN-NEXT: v_mul_hi_u32 v4, v1, s11 -; GCN-NEXT: v_readfirstlane_b32 s10, v3 -; GCN-NEXT: v_mul_hi_u32 v1, v1, s5 -; GCN-NEXT: s_mul_i32 s13, s4, s5 -; GCN-NEXT: s_add_u32 s10, s10, s13 -; GCN-NEXT: v_readfirstlane_b32 s13, v0 -; GCN-NEXT: s_mul_i32 s11, s9, s11 -; GCN-NEXT: s_addc_u32 s13, 0, s13 -; GCN-NEXT: v_readfirstlane_b32 s12, v4 -; GCN-NEXT: s_add_u32 s10, s10, s11 -; GCN-NEXT: v_readfirstlane_b32 s14, v1 -; GCN-NEXT: s_addc_u32 s10, s13, s12 -; GCN-NEXT: s_addc_u32 s11, s14, 0 -; GCN-NEXT: s_mul_i32 s5, s9, s5 -; GCN-NEXT: s_add_u32 s5, s10, s5 -; GCN-NEXT: s_addc_u32 s10, 0, s11 -; GCN-NEXT: s_add_u32 s11, s4, s5 -; GCN-NEXT: v_mov_b32_e32 v0, s11 -; GCN-NEXT: v_mul_hi_u32 v0, s6, v0 -; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s4, s4, s5 -; GCN-NEXT: s_addc_u32 s9, s9, s10 -; GCN-NEXT: s_mul_i32 s4, s6, s9 -; GCN-NEXT: v_readfirstlane_b32 s5, v0 -; GCN-NEXT: s_add_i32 s4, s5, s4 -; GCN-NEXT: s_mul_i32 s8, s8, s11 -; GCN-NEXT: s_mul_i32 s5, s6, s11 -; GCN-NEXT: s_add_i32 s4, s4, s8 -; GCN-NEXT: v_mov_b32_e32 v2, s5 -; GCN-NEXT: v_mov_b32_e32 v0, s4 -; GCN-NEXT: v_mul_hi_u32 v3, s9, v2 -; GCN-NEXT: v_mul_hi_u32 v2, s11, v2 -; GCN-NEXT: v_mul_hi_u32 v1, s9, v0 -; GCN-NEXT: v_mul_hi_u32 v0, s11, v0 -; GCN-NEXT: s_mul_i32 s8, s11, s4 -; GCN-NEXT: v_readfirstlane_b32 s12, v2 -; GCN-NEXT: s_add_u32 s8, s12, s8 -; GCN-NEXT: v_readfirstlane_b32 s10, v0 -; GCN-NEXT: s_mul_i32 s5, s9, s5 -; GCN-NEXT: s_addc_u32 s10, 0, s10 -; GCN-NEXT: v_readfirstlane_b32 s6, v3 ; GCN-NEXT: s_add_u32 s5, s8, s5 -; GCN-NEXT: s_addc_u32 s5, s10, s6 -; GCN-NEXT: v_readfirstlane_b32 s6, v1 -; GCN-NEXT: s_addc_u32 s6, s6, 0 -; GCN-NEXT: s_mul_i32 s4, s9, s4 -; GCN-NEXT: s_add_u32 s4, s5, s4 -; GCN-NEXT: s_addc_u32 s6, 0, s6 -; GCN-NEXT: s_add_u32 s8, s11, s4 -; GCN-NEXT: s_cselect_b64 s[4:5], -1, 0 -; GCN-NEXT: s_or_b32 s4, s4, s5 -; GCN-NEXT: s_addc_u32 s4, s9, s6 -; GCN-NEXT: v_mul_hi_u32 v1, s8, 24 -; GCN-NEXT: v_mul_hi_u32 v0, s4, 24 -; GCN-NEXT: s_mul_i32 s4, s4, 24 -; GCN-NEXT: s_mov_b32 s6, -1 -; GCN-NEXT: v_readfirstlane_b32 s8, v1 -; GCN-NEXT: v_readfirstlane_b32 s5, v0 -; GCN-NEXT: s_add_u32 s4, s8, s4 -; GCN-NEXT: s_addc_u32 s8, 0, s5 +; GCN-NEXT: s_addc_u32 s8, 0, s4 ; GCN-NEXT: v_mov_b32_e32 v0, s8 ; GCN-NEXT: v_mul_hi_u32 v0, s2, v0 ; GCN-NEXT: s_mov_b32 s4, s0 @@ -899,11 +882,9 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: s_mul_i32 s0, s2, s8 ; GCN-NEXT: s_sub_u32 s11, 24, s0 ; GCN-NEXT: s_cselect_b64 s[0:1], -1, 0 -; GCN-NEXT: s_or_b32 s8, s0, s1 ; GCN-NEXT: s_subb_u32 s12, s9, s3 ; GCN-NEXT: s_sub_u32 s13, s11, s2 ; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s14, s8, s9 ; GCN-NEXT: s_subb_u32 s14, s12, 0 ; GCN-NEXT: s_cmp_ge_u32 s14, s3 ; GCN-NEXT: s_cselect_b32 s15, -1, 0 @@ -912,13 +893,11 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: s_cmp_eq_u32 s14, s3 ; GCN-NEXT: s_cselect_b32 s15, s16, s15 ; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_subb_u32 s12, s12, s3 -; GCN-NEXT: s_sub_u32 s16, s13, s2 -; GCN-NEXT: s_cselect_b64 s[8:9], -1, 0 -; GCN-NEXT: s_or_b32 s8, s8, s9 -; GCN-NEXT: s_subb_u32 s8, s12, 0 +; GCN-NEXT: s_subb_u32 s8, s12, s3 +; GCN-NEXT: s_sub_u32 s9, s13, s2 +; GCN-NEXT: s_subb_u32 s8, s8, 0 ; GCN-NEXT: s_cmp_lg_u32 s15, 0 -; GCN-NEXT: s_cselect_b32 s9, s16, s13 +; GCN-NEXT: s_cselect_b32 s9, s9, s13 ; GCN-NEXT: s_cselect_b32 s8, s8, s14 ; GCN-NEXT: s_or_b32 s0, s0, s1 ; GCN-NEXT: s_subb_u32 s0, 0, s10 @@ -931,6 +910,7 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-NEXT: s_cmp_lg_u32 s1, 0 ; GCN-NEXT: s_cselect_b32 s0, s8, s0 ; GCN-NEXT: s_cselect_b32 s1, s9, s11 +; GCN-NEXT: s_mov_b32 s6, -1 ; GCN-NEXT: v_mov_b32_e32 v0, s1 ; GCN-NEXT: v_mov_b32_e32 v1, s0 ; GCN-NEXT: buffer_store_dwordx2 v[0:1], off, s[4:7], 0 @@ -956,8 +936,6 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_cbranch_vccz .LBB6_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s10, s8, 1 -; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-IR-NEXT: s_or_b32 s6, s6, s7 ; GCN-IR-NEXT: s_addc_u32 s6, s9, 0 ; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s8, 63, s8 @@ -988,8 +966,6 @@ define amdgpu_kernel void @s_test_urem_k_num_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_sub_u32 s10, s10, s16 ; GCN-IR-NEXT: s_subb_u32 s11, s11, s17 ; GCN-IR-NEXT: s_add_u32 s14, s14, 1 -; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 -; GCN-IR-NEXT: s_or_b32 s16, s16, s17 ; GCN-IR-NEXT: s_addc_u32 s15, s15, 0 ; GCN-IR-NEXT: s_cselect_b64 s[16:17], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[8:9], s[4:5] @@ -1077,8 +1053,6 @@ define amdgpu_kernel void @s_test_urem_k_den_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_cbranch_vccz .LBB7_5 ; GCN-IR-NEXT: ; %bb.1: ; %udiv-bb1 ; GCN-IR-NEXT: s_add_u32 s11, s8, 1 -; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 -; GCN-IR-NEXT: s_or_b32 s6, s6, s7 ; GCN-IR-NEXT: s_addc_u32 s6, s9, 0 ; GCN-IR-NEXT: s_cselect_b64 s[6:7], -1, 0 ; GCN-IR-NEXT: s_sub_i32 s8, 63, s8 @@ -1106,8 +1080,6 @@ define amdgpu_kernel void @s_test_urem_k_den_i64(ptr addrspace(1) %out, i64 %x) ; GCN-IR-NEXT: s_sub_u32 s8, s8, s10 ; GCN-IR-NEXT: s_subb_u32 s9, s9, 0 ; GCN-IR-NEXT: s_add_u32 s12, s12, 1 -; GCN-IR-NEXT: s_cselect_b64 s[14:15], -1, 0 -; GCN-IR-NEXT: s_or_b32 s14, s14, s15 ; GCN-IR-NEXT: s_addc_u32 s13, s13, 0 ; GCN-IR-NEXT: s_cselect_b64 s[14:15], -1, 0 ; GCN-IR-NEXT: s_mov_b64 s[10:11], s[4:5] diff --git a/llvm/test/CodeGen/AMDGPU/usubo.ll b/llvm/test/CodeGen/AMDGPU/usubo.ll index e8db647..8a54ad3 100644 --- a/llvm/test/CodeGen/AMDGPU/usubo.ll +++ b/llvm/test/CodeGen/AMDGPU/usubo.ll @@ -15,10 +15,8 @@ define amdgpu_kernel void @s_usubo_i64_zext(ptr addrspace(1) %out, i64 %a, i64 % ; SI-NEXT: s_waitcnt lgkmcnt(0) ; SI-NEXT: s_mov_b32 s4, s0 ; SI-NEXT: s_sub_u32 s2, s2, s8 -; SI-NEXT: s_mov_b32 s5, s1 -; SI-NEXT: s_cselect_b64 s[0:1], -1, 0 -; SI-NEXT: s_or_b32 s0, s0, s1 ; SI-NEXT: s_subb_u32 s3, s3, s9 +; SI-NEXT: s_mov_b32 s5, s1 ; SI-NEXT: s_cselect_b64 s[0:1], -1, 0 ; SI-NEXT: v_cndmask_b32_e64 v0, 0, 1, s[0:1] ; SI-NEXT: v_mov_b32_e32 v1, s3 @@ -432,8 +430,6 @@ define amdgpu_kernel void @s_usubo_i64(ptr addrspace(1) %out, ptr addrspace(1) % ; SI-NEXT: s_mov_b32 s10, -1 ; SI-NEXT: s_waitcnt lgkmcnt(0) ; SI-NEXT: s_sub_u32 s4, s4, s6 -; SI-NEXT: s_cselect_b64 s[12:13], -1, 0 -; SI-NEXT: s_or_b32 s6, s12, s13 ; SI-NEXT: s_subb_u32 s5, s5, s7 ; SI-NEXT: s_mov_b32 s8, s0 ; SI-NEXT: s_mov_b32 s9, s1 diff --git a/llvm/test/CodeGen/PowerPC/vp-ld-st.ll b/llvm/test/CodeGen/PowerPC/vp-ld-st.ll new file mode 100644 index 0000000..f0f9943 --- /dev/null +++ b/llvm/test/CodeGen/PowerPC/vp-ld-st.ll @@ -0,0 +1,160 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6 +; RUN: llc -verify-machineinstrs -mcpu=pwr10 \ +; RUN: -mtriple=powerpc64le-unknown-unknown < %s | FileCheck %s +; RUN: llc -verify-machineinstrs -mcpu=future \ +; RUN: -mtriple=powerpc64le-unknown-unknown < %s | FileCheck -check-prefix=FUTURE %s + +; RUN: llc -verify-machineinstrs -mcpu=pwr10 \ +; RUN: -mtriple=powerpc64-unknown-unknown < %s | FileCheck %s +; RUN: llc -verify-machineinstrs -mcpu=future \ +; RUN: -mtriple=powerpc64-unknown-unknown < %s | FileCheck --check-prefix=FUTURE %s + +; Function Attrs: nounwind readnone +define void @stxvl1(<16 x i8> %a, ptr %b, i64 %c) { +; CHECK-LABEL: stxvl1: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 3, 6, 56 +; CHECK-NEXT: stxvl 34, 5, 3 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: stxvl1: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: stxvrl 34, 5, 6 +; FUTURE-NEXT: blr +entry: + %cconv = trunc i64 %c to i32 + tail call void @llvm.vp.store.v16i8.p0(<16 x i8> %a, ptr %b, <16 x i1> splat (i1 true), i32 %cconv) + ret void +} + +; Function Attrs: nounwind readnone +define void @stxvl2(<8 x i16> %a, ptr %b, i64 %c) { +; CHECK-LABEL: stxvl2: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 3, 6, 57 +; CHECK-NEXT: stxvl 34, 5, 3 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: stxvl2: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: sldi 3, 6, 1 +; FUTURE-NEXT: stxvrl 34, 5, 3 +; FUTURE-NEXT: blr +entry: + %cconv = trunc i64 %c to i32 + tail call void @llvm.vp.store.v8i16.p0(<8 x i16> %a, ptr %b, <8 x i1> splat (i1 true), i32 %cconv) + ret void +} + +; Function Attrs: nounwind readnone +define void @stxvl4(<4 x i32> %a, ptr %b, i64 %c) { +; CHECK-LABEL: stxvl4: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 3, 6, 58 +; CHECK-NEXT: stxvl 34, 5, 3 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: stxvl4: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: sldi 3, 6, 2 +; FUTURE-NEXT: stxvrl 34, 5, 3 +; FUTURE-NEXT: blr +entry: + %cconv = trunc i64 %c to i32 + tail call void @llvm.vp.store.v4i32.p0(<4 x i32> %a, ptr %b, <4 x i1> splat (i1 true), i32 %cconv) + ret void +} + +; Function Attrs: nounwind readnone +define void @stxvl8(<2 x i64> %a, ptr %b, i64 %c) { +; CHECK-LABEL: stxvl8: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 3, 6, 59 +; CHECK-NEXT: stxvl 34, 5, 3 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: stxvl8: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: sldi 3, 6, 3 +; FUTURE-NEXT: stxvrl 34, 5, 3 +; FUTURE-NEXT: blr +entry: + %cconv = trunc i64 %c to i32 + tail call void @llvm.vp.store.v2i64.p0(<2 x i64> %a, ptr %b, <2 x i1> splat (i1 true), i32 %cconv) + ret void +} + +; Function Attrs: nounwind readnone +define <16 x i8> @lxvl1(ptr %a, i64 %b) { +; CHECK-LABEL: lxvl1: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 4, 4, 56 +; CHECK-NEXT: lxvl 34, 3, 4 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: lxvl1: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: lxvrl 34, 3, 4 +; FUTURE-NEXT: blr +entry: + %bconv = trunc i64 %b to i32 + %0 = tail call <16 x i8> @llvm.vp.load.v16i8.p0(ptr %a, <16 x i1> splat (i1 true), i32 %bconv) + ret <16 x i8> %0 +} + +; Function Attrs: nounwind readnone +define <8 x i16> @lxvl2(ptr %a, i64 %b) { +; CHECK-LABEL: lxvl2: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 4, 4, 57 +; CHECK-NEXT: lxvl 34, 3, 4 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: lxvl2: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: sldi 4, 4, 1 +; FUTURE-NEXT: lxvrl 34, 3, 4 +; FUTURE-NEXT: blr +entry: + %bconv = trunc i64 %b to i32 + %0 = tail call <8 x i16> @llvm.vp.load.v8i16.p0(ptr %a, <8 x i1> splat (i1 true), i32 %bconv) + ret <8 x i16> %0 +} + +; Function Attrs: nounwind readnone +define <4 x i32> @lxvl4(ptr %a, i64 %b) { +; CHECK-LABEL: lxvl4: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 4, 4, 58 +; CHECK-NEXT: lxvl 34, 3, 4 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: lxvl4: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: sldi 4, 4, 2 +; FUTURE-NEXT: lxvrl 34, 3, 4 +; FUTURE-NEXT: blr +entry: + %bconv = trunc i64 %b to i32 + %0 = tail call <4 x i32> @llvm.vp.load.v4i32.p0(ptr %a, <4 x i1> splat (i1 true), i32 %bconv) + ret <4 x i32> %0 +} + +; Function Attrs: nounwind readnone +define <2 x i64> @lxvl8(ptr %a, i64 %b) { +; CHECK-LABEL: lxvl8: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: sldi 4, 4, 59 +; CHECK-NEXT: lxvl 34, 3, 4 +; CHECK-NEXT: blr +; +; FUTURE-LABEL: lxvl8: +; FUTURE: # %bb.0: # %entry +; FUTURE-NEXT: sldi 4, 4, 3 +; FUTURE-NEXT: lxvrl 34, 3, 4 +; FUTURE-NEXT: blr +entry: + %bconv = trunc i64 %b to i32 + %0 = tail call <2 x i64> @llvm.vp.load.v2i64.p0(ptr %a, <2 x i1> splat (i1 true), i32 %bconv) + ret <2 x i64> %0 +} diff --git a/llvm/test/DebugInfo/extradata-node-reference.ll b/llvm/test/DebugInfo/extradata-node-reference.ll index 0ec9312..1881756 100644 --- a/llvm/test/DebugInfo/extradata-node-reference.ll +++ b/llvm/test/DebugInfo/extradata-node-reference.ll @@ -29,7 +29,7 @@ !1 = !DIFile(filename: "test.cpp", directory: ".") !2 = !{i32 2, !"Debug Info Version", i32 3} !3 = !{i32 1, !"wchar_size", i32 4} -!4 = !{i32 2, !"Dwarf Version", i32 5} +!4 = !{i32 2, !"Dwarf Version", i32 4} !8 = !{!9, !16, !24, !35} ; extraData node definitions diff --git a/llvm/test/Transforms/InstCombine/sink-dereferenceable-assume.ll b/llvm/test/Transforms/InstCombine/sink-dereferenceable-assume.ll new file mode 100644 index 0000000..9531323 --- /dev/null +++ b/llvm/test/Transforms/InstCombine/sink-dereferenceable-assume.ll @@ -0,0 +1,31 @@ +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 6 +; RUN: opt -p instcombine -S %s | FileCheck %s + +define i64 @test_sink_with_dereferenceable_assume(ptr %p, ptr %q, i1 %cond) { +; CHECK-LABEL: define i64 @test_sink_with_dereferenceable_assume( +; CHECK-SAME: ptr [[P:%.*]], ptr [[Q:%.*]], i1 [[COND:%.*]]) { +; CHECK-NEXT: [[ENTRY:.*:]] +; CHECK-NEXT: br i1 [[COND]], label %[[THEN:.*]], label %[[ELSE:.*]] +; CHECK: [[THEN]]: +; CHECK-NEXT: [[Q_INT:%.*]] = ptrtoint ptr [[Q]] to i64 +; CHECK-NEXT: [[P_INT:%.*]] = ptrtoint ptr [[P]] to i64 +; CHECK-NEXT: [[DIFF:%.*]] = sub i64 [[Q_INT]], [[P_INT]] +; CHECK-NEXT: ret i64 [[DIFF]] +; CHECK: [[ELSE]]: +; CHECK-NEXT: ret i64 0 +; +entry: + %p_int = ptrtoint ptr %p to i64 + %q_int = ptrtoint ptr %q to i64 + %diff = sub i64 %q_int, %p_int + call void @llvm.assume(i1 true) [ "dereferenceable"(ptr %p, i64 %diff) ] + br i1 %cond, label %then, label %else + +then: + ret i64 %diff + +else: + ret i64 0 +} + +declare void @llvm.assume(i1 noundef) diff --git a/llvm/test/Transforms/PhaseOrdering/AArch64/std-find.ll b/llvm/test/Transforms/PhaseOrdering/AArch64/std-find.ll index 338d925..33e3e83 100644 --- a/llvm/test/Transforms/PhaseOrdering/AArch64/std-find.ll +++ b/llvm/test/Transforms/PhaseOrdering/AArch64/std-find.ll @@ -49,10 +49,10 @@ entry: call void @llvm.assume(i1 true) [ "dereferenceable"(ptr %0, i64 256) ] %start.ptr = load ptr, ptr %first, align 8 %1 = load i64, ptr %first, align 8 - %coerce.val.pi.i = add i64 %1, 256 - %coerce.val.ip = inttoptr i64 %coerce.val.pi.i to ptr - %cmp.not6.i.i = icmp eq ptr %start.ptr, %coerce.val.ip - br i1 %cmp.not6.i.i, label %return, label %loop.ph + %coerce.val.p = add i64 %1, 256 + %coerce.val.ip = inttoptr i64 %coerce.val.p to ptr + %ec6. = icmp eq ptr %start.ptr, %coerce.val.ip + br i1 %ec6., label %return, label %loop.ph loop.ph: %2 = load i16, ptr %s.addr, align 2 @@ -61,13 +61,13 @@ loop.ph: loop.header: %ptr.iv = phi ptr [ %start.ptr, %loop.ph ], [ %ptr.iv.next, %loop.latch ] %3 = load i16, ptr %ptr.iv, align 2 - %cmp2.i.i = icmp eq i16 %3, %2 - br i1 %cmp2.i.i, label %return, label %loop.latch + %cmp2. = icmp eq i16 %3, %2 + br i1 %cmp2., label %return, label %loop.latch loop.latch: %ptr.iv.next = getelementptr inbounds nuw i8, ptr %ptr.iv, i64 2 - %cmp.not.i.i = icmp eq ptr %ptr.iv.next, %coerce.val.ip - br i1 %cmp.not.i.i, label %return, label %loop.header + %ec. = icmp eq ptr %ptr.iv.next, %coerce.val.ip + br i1 %ec., label %return, label %loop.header return: %merge = phi ptr [ %start.ptr, %entry ], [ %coerce.val.ip, %loop.latch ], [ %ptr.iv, %loop.header ] @@ -103,10 +103,10 @@ entry: %0 = load ptr, ptr %first, align 8 %start.ptr = load ptr, ptr %first, align 8 %1 = load i64, ptr %first, align 8 - %coerce.val.pi.i = add i64 %1, 256 - %coerce.val.ip = inttoptr i64 %coerce.val.pi.i to ptr - %cmp.not6.i.i = icmp eq ptr %start.ptr, %coerce.val.ip - br i1 %cmp.not6.i.i, label %return, label %loop.ph + %coerce.val.p = add i64 %1, 256 + %coerce.val.ip = inttoptr i64 %coerce.val.p to ptr + %ec6. = icmp eq ptr %start.ptr, %coerce.val.ip + br i1 %ec6., label %return, label %loop.ph loop.ph: %2 = load i16, ptr %s.addr, align 2 @@ -115,13 +115,13 @@ loop.ph: loop.header: %ptr.iv = phi ptr [ %start.ptr, %loop.ph ], [ %ptr.iv.next, %loop.latch ] %3 = load i16, ptr %ptr.iv, align 2 - %cmp2.i.i = icmp eq i16 %3, %2 - br i1 %cmp2.i.i, label %return, label %loop.latch + %cmp2. = icmp eq i16 %3, %2 + br i1 %cmp2., label %return, label %loop.latch loop.latch: %ptr.iv.next = getelementptr inbounds nuw i8, ptr %ptr.iv, i64 2 - %cmp.not.i.i = icmp eq ptr %ptr.iv.next, %coerce.val.ip - br i1 %cmp.not.i.i, label %return, label %loop.header + %ec. = icmp eq ptr %ptr.iv.next, %coerce.val.ip + br i1 %ec., label %return, label %loop.header return: %merge = phi ptr [ %start.ptr, %entry ], [ %coerce.val.ip, %loop.latch ], [ %ptr.iv, %loop.header ] @@ -129,9 +129,118 @@ return: ret i64 %res } +define ptr @std_find_caller(ptr noundef %first, ptr noundef %last) { +; CHECK-LABEL: define noundef ptr @std_find_caller( +; CHECK-SAME: ptr noundef [[FIRST:%.*]], ptr noundef [[LAST:%.*]]) local_unnamed_addr #[[ATTR0]] { +; CHECK-NEXT: [[ENTRY:.*]]: +; CHECK-NEXT: call void @llvm.assume(i1 true) [ "align"(ptr [[FIRST]], i64 2) ] +; CHECK-NEXT: call void @llvm.assume(i1 true) [ "align"(ptr [[LAST]], i64 2) ] +; CHECK-NEXT: [[PRE_I:%.*]] = icmp eq ptr [[FIRST]], [[LAST]] +; CHECK-NEXT: br i1 [[PRE_I]], label %[[STD_FIND_GENERIC_IMPL_EXIT:.*]], label %[[LOOP_HEADER_I_PREHEADER:.*]] +; CHECK: [[LOOP_HEADER_I_PREHEADER]]: +; CHECK-NEXT: [[LAST2:%.*]] = ptrtoint ptr [[LAST]] to i64 +; CHECK-NEXT: [[FIRST3:%.*]] = ptrtoint ptr [[FIRST]] to i64 +; CHECK-NEXT: [[LAST_I64:%.*]] = ptrtoint ptr [[LAST]] to i64 +; CHECK-NEXT: [[FIRST1:%.*]] = ptrtoint ptr [[FIRST]] to i64 +; CHECK-NEXT: [[PTR_SUB:%.*]] = sub i64 [[LAST_I64]], [[FIRST1]] +; CHECK-NEXT: [[SCEVGEP:%.*]] = getelementptr i8, ptr [[FIRST]], i64 [[PTR_SUB]] +; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[LAST2]], -2 +; CHECK-NEXT: [[TMP1:%.*]] = sub i64 [[TMP0]], [[FIRST3]] +; CHECK-NEXT: [[TMP2:%.*]] = lshr exact i64 [[TMP1]], 1 +; CHECK-NEXT: [[TMP3:%.*]] = add nuw i64 [[TMP2]], 1 +; CHECK-NEXT: [[XTRAITER:%.*]] = and i64 [[TMP3]], 3 +; CHECK-NEXT: [[TMP4:%.*]] = and i64 [[TMP1]], 6 +; CHECK-NEXT: [[LCMP_MOD_NOT:%.*]] = icmp eq i64 [[TMP4]], 6 +; CHECK-NEXT: br i1 [[LCMP_MOD_NOT]], label %[[LOOP_HEADER_I_PROL_LOOPEXIT:.*]], label %[[LOOP_HEADER_I_PROL:.*]] +; CHECK: [[LOOP_HEADER_I_PROL]]: +; CHECK-NEXT: [[PTR_IV_I_PROL:%.*]] = phi ptr [ [[PTR_IV_NEXT_I_PROL:%.*]], %[[LOOP_LATCH_I_PROL:.*]] ], [ [[FIRST]], %[[LOOP_HEADER_I_PREHEADER]] ] +; CHECK-NEXT: [[PROL_ITER:%.*]] = phi i64 [ [[PROL_ITER_NEXT:%.*]], %[[LOOP_LATCH_I_PROL]] ], [ 0, %[[LOOP_HEADER_I_PREHEADER]] ] +; CHECK-NEXT: [[L_I_PROL:%.*]] = load i16, ptr [[PTR_IV_I_PROL]], align 2 +; CHECK-NEXT: [[C_1_I_PROL:%.*]] = icmp eq i16 [[L_I_PROL]], 1 +; CHECK-NEXT: br i1 [[C_1_I_PROL]], label %[[STD_FIND_GENERIC_IMPL_EXIT]], label %[[LOOP_LATCH_I_PROL]] +; CHECK: [[LOOP_LATCH_I_PROL]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_PROL]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I_PROL]], i64 2 +; CHECK-NEXT: [[PROL_ITER_NEXT]] = add i64 [[PROL_ITER]], 1 +; CHECK-NEXT: [[PROL_ITER_CMP_NOT:%.*]] = icmp eq i64 [[PROL_ITER_NEXT]], [[XTRAITER]] +; CHECK-NEXT: br i1 [[PROL_ITER_CMP_NOT]], label %[[LOOP_HEADER_I_PROL_LOOPEXIT]], label %[[LOOP_HEADER_I_PROL]], !llvm.loop [[LOOP3:![0-9]+]] +; CHECK: [[LOOP_HEADER_I_PROL_LOOPEXIT]]: +; CHECK-NEXT: [[PTR_IV_I_UNR:%.*]] = phi ptr [ [[FIRST]], %[[LOOP_HEADER_I_PREHEADER]] ], [ [[PTR_IV_NEXT_I_PROL]], %[[LOOP_LATCH_I_PROL]] ] +; CHECK-NEXT: [[TMP5:%.*]] = icmp ult i64 [[TMP1]], 6 +; CHECK-NEXT: br i1 [[TMP5]], label %[[STD_FIND_GENERIC_IMPL_EXIT]], label %[[LOOP_HEADER_I:.*]] +; CHECK: [[LOOP_HEADER_I]]: +; CHECK-NEXT: [[PTR_IV_I:%.*]] = phi ptr [ [[PTR_IV_NEXT_I_3:%.*]], %[[LOOP_LATCH_I_3:.*]] ], [ [[PTR_IV_I_UNR]], %[[LOOP_HEADER_I_PROL_LOOPEXIT]] ] +; CHECK-NEXT: [[L_I:%.*]] = load i16, ptr [[PTR_IV_I]], align 2 +; CHECK-NEXT: [[C_1_I:%.*]] = icmp eq i16 [[L_I]], 1 +; CHECK-NEXT: br i1 [[C_1_I]], label %[[STD_FIND_GENERIC_IMPL_EXIT]], label %[[LOOP_LATCH_I:.*]] +; CHECK: [[LOOP_LATCH_I]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I:%.*]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 2 +; CHECK-NEXT: [[L_I_1:%.*]] = load i16, ptr [[PTR_IV_NEXT_I]], align 2 +; CHECK-NEXT: [[C_1_I_1:%.*]] = icmp eq i16 [[L_I_1]], 1 +; CHECK-NEXT: br i1 [[C_1_I_1]], label %[[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT11:.*]], label %[[LOOP_LATCH_I_1:.*]] +; CHECK: [[LOOP_LATCH_I_1]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_1:%.*]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 4 +; CHECK-NEXT: [[L_I_2:%.*]] = load i16, ptr [[PTR_IV_NEXT_I_1]], align 2 +; CHECK-NEXT: [[C_1_I_2:%.*]] = icmp eq i16 [[L_I_2]], 1 +; CHECK-NEXT: br i1 [[C_1_I_2]], label %[[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT9:.*]], label %[[LOOP_LATCH_I_2:.*]] +; CHECK: [[LOOP_LATCH_I_2]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_2:%.*]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 6 +; CHECK-NEXT: [[L_I_3:%.*]] = load i16, ptr [[PTR_IV_NEXT_I_2]], align 2 +; CHECK-NEXT: [[C_1_I_3:%.*]] = icmp eq i16 [[L_I_3]], 1 +; CHECK-NEXT: br i1 [[C_1_I_3]], label %[[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT7:.*]], label %[[LOOP_LATCH_I_3]] +; CHECK: [[LOOP_LATCH_I_3]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_3]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 8 +; CHECK-NEXT: [[EC_I_3:%.*]] = icmp eq ptr [[PTR_IV_NEXT_I_3]], [[LAST]] +; CHECK-NEXT: br i1 [[EC_I_3]], label %[[STD_FIND_GENERIC_IMPL_EXIT]], label %[[LOOP_HEADER_I]] +; CHECK: [[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT7]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_2_LE:%.*]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 6 +; CHECK-NEXT: br label %[[STD_FIND_GENERIC_IMPL_EXIT]] +; CHECK: [[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT9]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_1_LE:%.*]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 4 +; CHECK-NEXT: br label %[[STD_FIND_GENERIC_IMPL_EXIT]] +; CHECK: [[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT11]]: +; CHECK-NEXT: [[PTR_IV_NEXT_I_LE:%.*]] = getelementptr inbounds nuw i8, ptr [[PTR_IV_I]], i64 2 +; CHECK-NEXT: br label %[[STD_FIND_GENERIC_IMPL_EXIT]] +; CHECK: [[STD_FIND_GENERIC_IMPL_EXIT]]: +; CHECK-NEXT: [[RES_I:%.*]] = phi ptr [ [[FIRST]], %[[ENTRY]] ], [ [[SCEVGEP]], %[[LOOP_HEADER_I_PROL_LOOPEXIT]] ], [ [[PTR_IV_NEXT_I_2_LE]], %[[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT7]] ], [ [[PTR_IV_NEXT_I_1_LE]], %[[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT9]] ], [ [[PTR_IV_NEXT_I_LE]], %[[STD_FIND_GENERIC_IMPL_EXIT_LOOPEXIT_UNR_LCSSA_LOOPEXIT_SPLIT_LOOP_EXIT11]] ], [ [[SCEVGEP]], %[[LOOP_LATCH_I_3]] ], [ [[PTR_IV_I]], %[[LOOP_HEADER_I]] ], [ [[PTR_IV_I_PROL]], %[[LOOP_HEADER_I_PROL]] ] +; CHECK-NEXT: ret ptr [[RES_I]] +; +entry: + %last.i64 = ptrtoint ptr %last to i64 + %first.i64 = ptrtoint ptr %first to i64 + %ptr.sub = sub i64 %last.i64, %first.i64 + call void @llvm.assume(i1 true) [ "align"(ptr %first, i64 2) ] + call void @llvm.assume(i1 true) [ "align"(ptr %last, i64 2) ] + call void @llvm.assume(i1 true) [ "dereferenceable"(ptr %first, i64 %ptr.sub) ] + %call = call noundef ptr @std_find_generic_impl(ptr noundef nonnull %first, ptr noundef %last, i16 noundef signext 1) + ret ptr %call +} + +define linkonce_odr noundef ptr @std_find_generic_impl(ptr noundef %first, ptr noundef %last, i16 noundef %value) { +entry: + %pre = icmp eq ptr %first, %last + br i1 %pre, label %exit, label %loop.header + +loop.header: + %ptr.iv = phi ptr [ %ptr.iv.next, %loop.latch ], [ %first, %entry ] + %l = load i16, ptr %ptr.iv, align 2 + %c.1 = icmp eq i16 %l, %value + br i1 %c.1, label %exit, label %loop.latch + +loop.latch: + %ptr.iv.next = getelementptr inbounds nuw i8, ptr %ptr.iv, i64 2 + %ec = icmp eq ptr %ptr.iv.next, %last + br i1 %ec, label %exit, label %loop.header + +exit: + %res = phi ptr [ %first, %entry ], [ %ptr.iv, %loop.header ], [ %ptr.iv.next, %loop.latch ] + ret ptr %res +} + declare void @llvm.assume(i1 noundef) ;. ; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]} ; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1} ; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"} +; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META4:![0-9]+]]} +; CHECK: [[META4]] = !{!"llvm.loop.unroll.disable"} ;. diff --git a/llvm/test/Transforms/PhaseOrdering/X86/addsub-inseltpoison.ll b/llvm/test/Transforms/PhaseOrdering/X86/addsub-inseltpoison.ll index 2c1d73e..9f3244d 100644 --- a/llvm/test/Transforms/PhaseOrdering/X86/addsub-inseltpoison.ll +++ b/llvm/test/Transforms/PhaseOrdering/X86/addsub-inseltpoison.ll @@ -498,11 +498,9 @@ define void @add_aggregate_store(<2 x float> %a0, <2 x float> %a1, <2 x float> % ; PR58139 define <2 x double> @_mm_complexmult_pd_naive(<2 x double> %a, <2 x double> %b) { ; SSE-LABEL: @_mm_complexmult_pd_naive( -; SSE-NEXT: [[B1:%.*]] = extractelement <2 x double> [[B:%.*]], i64 1 -; SSE-NEXT: [[TMP1:%.*]] = fneg double [[B1]] ; SSE-NEXT: [[TMP2:%.*]] = shufflevector <2 x double> [[A:%.*]], <2 x double> poison, <2 x i32> <i32 1, i32 1> -; SSE-NEXT: [[TMP3:%.*]] = shufflevector <2 x double> [[B]], <2 x double> poison, <2 x i32> <i32 poison, i32 0> -; SSE-NEXT: [[TMP4:%.*]] = insertelement <2 x double> [[TMP3]], double [[TMP1]], i64 0 +; SSE-NEXT: [[TMP3:%.*]] = fneg <2 x double> [[B:%.*]] +; SSE-NEXT: [[TMP4:%.*]] = shufflevector <2 x double> [[TMP3]], <2 x double> [[B]], <2 x i32> <i32 1, i32 2> ; SSE-NEXT: [[TMP5:%.*]] = fmul <2 x double> [[TMP2]], [[TMP4]] ; SSE-NEXT: [[TMP6:%.*]] = shufflevector <2 x double> [[A]], <2 x double> poison, <2 x i32> zeroinitializer ; SSE-NEXT: [[TMP7:%.*]] = tail call <2 x double> @llvm.fmuladd.v2f64(<2 x double> [[TMP6]], <2 x double> [[B]], <2 x double> [[TMP5]]) diff --git a/llvm/test/Transforms/PhaseOrdering/X86/addsub.ll b/llvm/test/Transforms/PhaseOrdering/X86/addsub.ll index fa6403f..de64bf2 100644 --- a/llvm/test/Transforms/PhaseOrdering/X86/addsub.ll +++ b/llvm/test/Transforms/PhaseOrdering/X86/addsub.ll @@ -502,11 +502,9 @@ define void @add_aggregate_store(<2 x float> %a0, <2 x float> %a1, <2 x float> % ; PR58139 define <2 x double> @_mm_complexmult_pd_naive(<2 x double> %a, <2 x double> %b) { ; SSE-LABEL: @_mm_complexmult_pd_naive( -; SSE-NEXT: [[B1:%.*]] = extractelement <2 x double> [[B:%.*]], i64 1 -; SSE-NEXT: [[TMP1:%.*]] = fneg double [[B1]] ; SSE-NEXT: [[TMP2:%.*]] = shufflevector <2 x double> [[A:%.*]], <2 x double> poison, <2 x i32> <i32 1, i32 1> -; SSE-NEXT: [[TMP3:%.*]] = shufflevector <2 x double> [[B]], <2 x double> poison, <2 x i32> <i32 poison, i32 0> -; SSE-NEXT: [[TMP4:%.*]] = insertelement <2 x double> [[TMP3]], double [[TMP1]], i64 0 +; SSE-NEXT: [[TMP3:%.*]] = fneg <2 x double> [[B:%.*]] +; SSE-NEXT: [[TMP4:%.*]] = shufflevector <2 x double> [[TMP3]], <2 x double> [[B]], <2 x i32> <i32 1, i32 2> ; SSE-NEXT: [[TMP5:%.*]] = fmul <2 x double> [[TMP2]], [[TMP4]] ; SSE-NEXT: [[TMP6:%.*]] = shufflevector <2 x double> [[A]], <2 x double> poison, <2 x i32> zeroinitializer ; SSE-NEXT: [[TMP7:%.*]] = tail call <2 x double> @llvm.fmuladd.v2f64(<2 x double> [[TMP6]], <2 x double> [[B]], <2 x double> [[TMP5]]) diff --git a/llvm/test/Transforms/VectorCombine/X86/extract-fneg-insert.ll b/llvm/test/Transforms/VectorCombine/X86/extract-fneg-insert.ll index 5358e04..88fcf35 100644 --- a/llvm/test/Transforms/VectorCombine/X86/extract-fneg-insert.ll +++ b/llvm/test/Transforms/VectorCombine/X86/extract-fneg-insert.ll @@ -58,6 +58,19 @@ define <4 x float> @ext2_v2f32v4f32(<2 x float> %x, <4 x float> %y) { ret <4 x float> %r } +define <2 x float> @ext2_v4f32v2f32(<4 x float> %x, <2 x float> %y) { +; CHECK-LABEL: @ext2_v4f32v2f32( +; CHECK-NEXT: [[TMP1:%.*]] = fneg <4 x float> [[X:%.*]] +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x float> [[TMP1]], <4 x float> poison, <2 x i32> <i32 poison, i32 3> +; CHECK-NEXT: [[R:%.*]] = shufflevector <2 x float> [[Y:%.*]], <2 x float> [[TMP2]], <2 x i32> <i32 0, i32 3> +; CHECK-NEXT: ret <2 x float> [[R]] +; + %e = extractelement <4 x float> %x, i32 3 + %n = fneg float %e + %r = insertelement <2 x float> %y, float %n, i32 1 + ret <2 x float> %r +} + ; Eliminating extract/insert is still profitable. Flags propagate. define <2 x double> @ext1_v2f64(<2 x double> %x, <2 x double> %y) { @@ -73,17 +86,11 @@ define <2 x double> @ext1_v2f64(<2 x double> %x, <2 x double> %y) { } define <4 x double> @ext1_v2f64v4f64(<2 x double> %x, <4 x double> %y) { -; SSE-LABEL: @ext1_v2f64v4f64( -; SSE-NEXT: [[E:%.*]] = extractelement <2 x double> [[X:%.*]], i32 1 -; SSE-NEXT: [[N:%.*]] = fneg nsz double [[E]] -; SSE-NEXT: [[R:%.*]] = insertelement <4 x double> [[Y:%.*]], double [[N]], i32 1 -; SSE-NEXT: ret <4 x double> [[R]] -; -; AVX-LABEL: @ext1_v2f64v4f64( -; AVX-NEXT: [[TMP1:%.*]] = fneg nsz <2 x double> [[X:%.*]] -; AVX-NEXT: [[TMP2:%.*]] = shufflevector <2 x double> [[TMP1]], <2 x double> poison, <4 x i32> <i32 poison, i32 1, i32 poison, i32 poison> -; AVX-NEXT: [[R:%.*]] = shufflevector <4 x double> [[Y:%.*]], <4 x double> [[TMP2]], <4 x i32> <i32 0, i32 5, i32 2, i32 3> -; AVX-NEXT: ret <4 x double> [[R]] +; CHECK-LABEL: @ext1_v2f64v4f64( +; CHECK-NEXT: [[TMP1:%.*]] = fneg nsz <2 x double> [[X:%.*]] +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <2 x double> [[TMP1]], <2 x double> poison, <4 x i32> <i32 poison, i32 1, i32 poison, i32 poison> +; CHECK-NEXT: [[R:%.*]] = shufflevector <4 x double> [[Y:%.*]], <4 x double> [[TMP2]], <4 x i32> <i32 0, i32 5, i32 2, i32 3> +; CHECK-NEXT: ret <4 x double> [[R]] ; %e = extractelement <2 x double> %x, i32 1 %n = fneg nsz double %e @@ -91,6 +98,19 @@ define <4 x double> @ext1_v2f64v4f64(<2 x double> %x, <4 x double> %y) { ret <4 x double> %r } +define <2 x double> @ext1_v4f64v2f64(<4 x double> %x, <2 x double> %y) { +; CHECK-LABEL: @ext1_v4f64v2f64( +; CHECK-NEXT: [[TMP1:%.*]] = fneg nsz <4 x double> [[X:%.*]] +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x double> [[TMP1]], <4 x double> poison, <2 x i32> <i32 poison, i32 3> +; CHECK-NEXT: [[R:%.*]] = shufflevector <2 x double> [[Y:%.*]], <2 x double> [[TMP2]], <2 x i32> <i32 0, i32 3> +; CHECK-NEXT: ret <2 x double> [[R]] +; + %e = extractelement <4 x double> %x, i32 3 + %n = fneg nsz double %e + %r = insertelement <2 x double> %y, double %n, i32 1 + ret <2 x double> %r +} + define <8 x float> @ext7_v8f32(<8 x float> %x, <8 x float> %y) { ; CHECK-LABEL: @ext7_v8f32( ; CHECK-NEXT: [[TMP1:%.*]] = fneg <8 x float> [[X:%.*]] @@ -105,9 +125,9 @@ define <8 x float> @ext7_v8f32(<8 x float> %x, <8 x float> %y) { define <8 x float> @ext7_v4f32v8f32(<4 x float> %x, <8 x float> %y) { ; CHECK-LABEL: @ext7_v4f32v8f32( -; CHECK-NEXT: [[E:%.*]] = extractelement <4 x float> [[X:%.*]], i32 3 -; CHECK-NEXT: [[N:%.*]] = fneg float [[E]] -; CHECK-NEXT: [[R:%.*]] = insertelement <8 x float> [[Y:%.*]], float [[N]], i32 7 +; CHECK-NEXT: [[TMP1:%.*]] = fneg <4 x float> [[X:%.*]] +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x float> [[TMP1]], <4 x float> poison, <8 x i32> <i32 poison, i32 poison, i32 poison, i32 3, i32 poison, i32 poison, i32 poison, i32 poison> +; CHECK-NEXT: [[R:%.*]] = shufflevector <8 x float> [[Y:%.*]], <8 x float> [[TMP2]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 11> ; CHECK-NEXT: ret <8 x float> [[R]] ; %e = extractelement <4 x float> %x, i32 3 @@ -116,6 +136,19 @@ define <8 x float> @ext7_v4f32v8f32(<4 x float> %x, <8 x float> %y) { ret <8 x float> %r } +define <4 x float> @ext7_v8f32v4f32(<8 x float> %x, <4 x float> %y) { +; CHECK-LABEL: @ext7_v8f32v4f32( +; CHECK-NEXT: [[TMP1:%.*]] = fneg <8 x float> [[X:%.*]] +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <8 x float> [[TMP1]], <8 x float> poison, <4 x i32> <i32 poison, i32 poison, i32 poison, i32 7> +; CHECK-NEXT: [[R:%.*]] = shufflevector <4 x float> [[Y:%.*]], <4 x float> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 2, i32 7> +; CHECK-NEXT: ret <4 x float> [[R]] +; + %e = extractelement <8 x float> %x, i32 7 + %n = fneg float %e + %r = insertelement <4 x float> %y, float %n, i32 3 + ret <4 x float> %r +} + ; Same as above with an extra use of the extracted element. define <8 x float> @ext7_v8f32_use1(<8 x float> %x, <8 x float> %y) { @@ -141,12 +174,20 @@ define <8 x float> @ext7_v8f32_use1(<8 x float> %x, <8 x float> %y) { } define <8 x float> @ext7_v4f32v8f32_use1(<4 x float> %x, <8 x float> %y) { -; CHECK-LABEL: @ext7_v4f32v8f32_use1( -; CHECK-NEXT: [[E:%.*]] = extractelement <4 x float> [[X:%.*]], i32 3 -; CHECK-NEXT: call void @use(float [[E]]) -; CHECK-NEXT: [[N:%.*]] = fneg float [[E]] -; CHECK-NEXT: [[R:%.*]] = insertelement <8 x float> [[Y:%.*]], float [[N]], i32 3 -; CHECK-NEXT: ret <8 x float> [[R]] +; SSE-LABEL: @ext7_v4f32v8f32_use1( +; SSE-NEXT: [[E:%.*]] = extractelement <4 x float> [[X:%.*]], i32 3 +; SSE-NEXT: call void @use(float [[E]]) +; SSE-NEXT: [[TMP1:%.*]] = fneg <4 x float> [[X]] +; SSE-NEXT: [[TMP2:%.*]] = shufflevector <4 x float> [[TMP1]], <4 x float> poison, <8 x i32> <i32 poison, i32 poison, i32 poison, i32 3, i32 poison, i32 poison, i32 poison, i32 poison> +; SSE-NEXT: [[R:%.*]] = shufflevector <8 x float> [[Y:%.*]], <8 x float> [[TMP2]], <8 x i32> <i32 0, i32 1, i32 2, i32 11, i32 4, i32 5, i32 6, i32 7> +; SSE-NEXT: ret <8 x float> [[R]] +; +; AVX-LABEL: @ext7_v4f32v8f32_use1( +; AVX-NEXT: [[E:%.*]] = extractelement <4 x float> [[X:%.*]], i32 3 +; AVX-NEXT: call void @use(float [[E]]) +; AVX-NEXT: [[N:%.*]] = fneg float [[E]] +; AVX-NEXT: [[R:%.*]] = insertelement <8 x float> [[Y:%.*]], float [[N]], i32 3 +; AVX-NEXT: ret <8 x float> [[R]] ; %e = extractelement <4 x float> %x, i32 3 call void @use(float %e) @@ -155,6 +196,29 @@ define <8 x float> @ext7_v4f32v8f32_use1(<4 x float> %x, <8 x float> %y) { ret <8 x float> %r } +define <4 x float> @ext7_v8f32v4f32_use1(<8 x float> %x, <4 x float> %y) { +; SSE-LABEL: @ext7_v8f32v4f32_use1( +; SSE-NEXT: [[E:%.*]] = extractelement <8 x float> [[X:%.*]], i32 7 +; SSE-NEXT: call void @use(float [[E]]) +; SSE-NEXT: [[TMP1:%.*]] = fneg <8 x float> [[X]] +; SSE-NEXT: [[TMP2:%.*]] = shufflevector <8 x float> [[TMP1]], <8 x float> poison, <4 x i32> <i32 poison, i32 poison, i32 poison, i32 7> +; SSE-NEXT: [[R:%.*]] = shufflevector <4 x float> [[Y:%.*]], <4 x float> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 2, i32 7> +; SSE-NEXT: ret <4 x float> [[R]] +; +; AVX-LABEL: @ext7_v8f32v4f32_use1( +; AVX-NEXT: [[E:%.*]] = extractelement <8 x float> [[X:%.*]], i32 7 +; AVX-NEXT: call void @use(float [[E]]) +; AVX-NEXT: [[N:%.*]] = fneg float [[E]] +; AVX-NEXT: [[R:%.*]] = insertelement <4 x float> [[Y:%.*]], float [[N]], i32 3 +; AVX-NEXT: ret <4 x float> [[R]] +; + %e = extractelement <8 x float> %x, i32 7 + call void @use(float %e) + %n = fneg float %e + %r = insertelement <4 x float> %y, float %n, i32 3 + ret <4 x float> %r +} + ; Negative test - the transform is likely not profitable if the fneg has another use. define <8 x float> @ext7_v8f32_use2(<8 x float> %x, <8 x float> %y) { @@ -187,6 +251,21 @@ define <8 x float> @ext7_v4f32v8f32_use2(<4 x float> %x, <8 x float> %y) { ret <8 x float> %r } +define <4 x float> @ext7_v8f32v4f32_use2(<8 x float> %x, <4 x float> %y) { +; CHECK-LABEL: @ext7_v8f32v4f32_use2( +; CHECK-NEXT: [[E:%.*]] = extractelement <8 x float> [[X:%.*]], i32 7 +; CHECK-NEXT: [[N:%.*]] = fneg float [[E]] +; CHECK-NEXT: call void @use(float [[N]]) +; CHECK-NEXT: [[R:%.*]] = insertelement <4 x float> [[Y:%.*]], float [[N]], i32 3 +; CHECK-NEXT: ret <4 x float> [[R]] +; + %e = extractelement <8 x float> %x, i32 7 + %n = fneg float %e + call void @use(float %n) + %r = insertelement <4 x float> %y, float %n, i32 3 + ret <4 x float> %r +} + ; Negative test - can't convert variable index to a shuffle. define <2 x double> @ext_index_var_v2f64(<2 x double> %x, <2 x double> %y, i32 %index) { @@ -215,14 +294,10 @@ define <4 x double> @ext_index_var_v2f64v4f64(<2 x double> %x, <4 x double> %y, ret <4 x double> %r } -; Negative test - require same extract/insert index for simple shuffle. -; TODO: We could handle this by adjusting the cost calculation. - define <2 x double> @ext1_v2f64_ins0(<2 x double> %x, <2 x double> %y) { ; CHECK-LABEL: @ext1_v2f64_ins0( -; CHECK-NEXT: [[E:%.*]] = extractelement <2 x double> [[X:%.*]], i32 1 -; CHECK-NEXT: [[N:%.*]] = fneg nsz double [[E]] -; CHECK-NEXT: [[R:%.*]] = insertelement <2 x double> [[Y:%.*]], double [[N]], i32 0 +; CHECK-NEXT: [[TMP1:%.*]] = fneg nsz <2 x double> [[X:%.*]] +; CHECK-NEXT: [[R:%.*]] = shufflevector <2 x double> [[Y:%.*]], <2 x double> [[TMP1]], <2 x i32> <i32 3, i32 1> ; CHECK-NEXT: ret <2 x double> [[R]] ; %e = extractelement <2 x double> %x, i32 1 @@ -231,12 +306,11 @@ define <2 x double> @ext1_v2f64_ins0(<2 x double> %x, <2 x double> %y) { ret <2 x double> %r } -; Negative test - extract from an index greater than the vector width of the destination define <2 x double> @ext3_v4f64v2f64(<4 x double> %x, <2 x double> %y) { ; CHECK-LABEL: @ext3_v4f64v2f64( -; CHECK-NEXT: [[E:%.*]] = extractelement <4 x double> [[X:%.*]], i32 3 -; CHECK-NEXT: [[N:%.*]] = fneg nsz double [[E]] -; CHECK-NEXT: [[R:%.*]] = insertelement <2 x double> [[Y:%.*]], double [[N]], i32 1 +; CHECK-NEXT: [[TMP1:%.*]] = fneg nsz <4 x double> [[X:%.*]] +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x double> [[TMP1]], <4 x double> poison, <2 x i32> <i32 poison, i32 3> +; CHECK-NEXT: [[R:%.*]] = shufflevector <2 x double> [[Y:%.*]], <2 x double> [[TMP2]], <2 x i32> <i32 0, i32 3> ; CHECK-NEXT: ret <2 x double> [[R]] ; %e = extractelement <4 x double> %x, i32 3 @@ -246,11 +320,17 @@ define <2 x double> @ext3_v4f64v2f64(<4 x double> %x, <2 x double> %y) { } define <4 x double> @ext1_v2f64v4f64_ins0(<2 x double> %x, <4 x double> %y) { -; CHECK-LABEL: @ext1_v2f64v4f64_ins0( -; CHECK-NEXT: [[E:%.*]] = extractelement <2 x double> [[X:%.*]], i32 1 -; CHECK-NEXT: [[N:%.*]] = fneg nsz double [[E]] -; CHECK-NEXT: [[R:%.*]] = insertelement <4 x double> [[Y:%.*]], double [[N]], i32 0 -; CHECK-NEXT: ret <4 x double> [[R]] +; SSE-LABEL: @ext1_v2f64v4f64_ins0( +; SSE-NEXT: [[TMP1:%.*]] = fneg nsz <2 x double> [[X:%.*]] +; SSE-NEXT: [[TMP2:%.*]] = shufflevector <2 x double> [[TMP1]], <2 x double> poison, <4 x i32> <i32 poison, i32 1, i32 poison, i32 poison> +; SSE-NEXT: [[R:%.*]] = shufflevector <4 x double> [[Y:%.*]], <4 x double> [[TMP2]], <4 x i32> <i32 5, i32 1, i32 2, i32 3> +; SSE-NEXT: ret <4 x double> [[R]] +; +; AVX-LABEL: @ext1_v2f64v4f64_ins0( +; AVX-NEXT: [[E:%.*]] = extractelement <2 x double> [[X:%.*]], i32 1 +; AVX-NEXT: [[N:%.*]] = fneg nsz double [[E]] +; AVX-NEXT: [[R:%.*]] = insertelement <4 x double> [[Y:%.*]], double [[N]], i32 0 +; AVX-NEXT: ret <4 x double> [[R]] ; %e = extractelement <2 x double> %x, i32 1 %n = fneg nsz double %e diff --git a/llvm/test/lit.cfg.py b/llvm/test/lit.cfg.py index 35ea8b8..725ddb87 100644 --- a/llvm/test/lit.cfg.py +++ b/llvm/test/lit.cfg.py @@ -233,6 +233,7 @@ tools.extend( "llvm-addr2line", "llvm-bcanalyzer", "llvm-bitcode-strip", + "llvm-cas", "llvm-cgdata", "llvm-config", "llvm-cov", @@ -796,10 +797,19 @@ if config.have_opt_viewer_modules: if config.expensive_checks: config.available_features.add("expensive_checks") +if config.have_ondisk_cas: + config.available_features.add("ondisk_cas") + if "MemoryWithOrigins" in config.llvm_use_sanitizer: config.available_features.add("use_msan_with_origins") +# Restrict the size of the on-disk CAS for tests. This allows testing in +# constrained environments (e.g. small TMPDIR). It also prevents leaving +# behind large files on file systems that do not support sparse files if a test +# crashes before resizing the file. +config.environment["LLVM_CAS_MAX_MAPPING_SIZE"] = "%d" % (100 * 1024 * 1024) + # Some tools support an environment variable "OBJECT_MODE" on AIX OS, which # controls the kind of objects they will support. If there is no "OBJECT_MODE" # environment variable specified, the default behaviour is to support 32-bit diff --git a/llvm/test/lit.site.cfg.py.in b/llvm/test/lit.site.cfg.py.in index 973e0ec9..c5cb716 100644 --- a/llvm/test/lit.site.cfg.py.in +++ b/llvm/test/lit.site.cfg.py.in @@ -66,6 +66,7 @@ config.spirv_tools_tests = @LLVM_INCLUDE_SPIRV_TOOLS_TESTS@ config.have_vc_rev = @LLVM_APPEND_VC_REV@ config.force_vc_rev = "@LLVM_FORCE_VC_REVISION@" config.has_logf128 = @LLVM_HAS_LOGF128@ +config.have_ondisk_cas = @LLVM_ENABLE_ONDISK_CAS@ import lit.llvm lit.llvm.initialize(lit_config, config) diff --git a/llvm/test/tools/dsymutil/ARM/typedefs-with-same-name.test b/llvm/test/tools/dsymutil/ARM/typedefs-with-same-name.test new file mode 100644 index 0000000..cef40b4 --- /dev/null +++ b/llvm/test/tools/dsymutil/ARM/typedefs-with-same-name.test @@ -0,0 +1,41 @@ +#RUN: dsymutil --linker=parallel -f -oso-prepend-path=%p/../Inputs/ -y %s -o %t.dwarf +#RUN: llvm-dwarfdump %t.dwarf | FileCheck %s + +# There should be two typedef DIE named "BarInt" in the resultant .dwarf file. +# The second should refer to the first, which refer to "Foo<int>". +# CHECK: 0x[[FIRST_BARINT_ADDR:[0-9a-f]*]]: DW_TAG_typedef +# CHECK-NEXT: DW_AT_type (0x{{([[:xdigit:]]*)}} "Foo<int>") +# CHECK-NEXT: DW_AT_name ("BarInt") +# CHECK: 0x{{([[:xdigit:]]*)}}: DW_TAG_typedef +# CHECK-NEXT: DW_AT_type (0x[[FIRST_BARINT_ADDR]] "BarInt") +# CHECK-NEXT: DW_AT_name ("BarInt") + +# Source: +# +# template <typename T> struct Foo; +# typedef Foo<int> BarInt; +# template <typename T> +# struct [[clang::preferred_name(BarInt)]] Foo{}; +# int main() { +# BarInt barInt; +# return 0; +# } +# +# Compile with: +# +# $ clang++ -g -O0 -c typedefs-with-same-name.cpp -o typedefs-with-same-name.o +# +# To generate the debug map: +# +# $ clang++ typedefs-with-same-name.o -o typedefs-with-same-name +# $ dsymutil -dump-debug-map typedefs-with-same-name + +--- +triple: 'arm64-apple-darwin' +objects: + - filename: '/typedefs-with-same-name.o' + timestamp: 1762438746 + type: 102 + symbols: + - { sym: _main, objAddr: 0x0, binAddr: 0x100000360, size: 0x14 } +... diff --git a/llvm/test/tools/dsymutil/Inputs/typedefs-with-same-name.o b/llvm/test/tools/dsymutil/Inputs/typedefs-with-same-name.o Binary files differnew file mode 100644 index 0000000..6cc47c1 --- /dev/null +++ b/llvm/test/tools/dsymutil/Inputs/typedefs-with-same-name.o diff --git a/llvm/test/tools/dsymutil/X86/DWARFLinkerParallel/odr-fwd-declaration.test b/llvm/test/tools/dsymutil/X86/DWARFLinkerParallel/odr-fwd-declaration.test index d028194..fd15ce3 100644 --- a/llvm/test/tools/dsymutil/X86/DWARFLinkerParallel/odr-fwd-declaration.test +++ b/llvm/test/tools/dsymutil/X86/DWARFLinkerParallel/odr-fwd-declaration.test @@ -35,14 +35,14 @@ void foo() { Sptrptr ptr1 = 0; } // CHECK: DW_TAG_member // CHECK-NEXT: DW_AT_name{{.*}}"field" -// CHECK: 0x[[TYPEDEF_PTR_S]]: DW_TAG_typedef -// CHECK-NEXT: DW_AT_type{{.*}}{0x[[PTR_S]]} "S *" -// CHECK-NEXT: DW_AT_name{{.*}}"Sptr" - // CHECK: 0x[[TYPEDEF_PTR_PTR_S:[a-f0-9]*]]: DW_TAG_typedef // CHECK-NEXT: DW_AT_type{{.*}}{0x[[PTR_PTR_S]]} "Sptr *" // CHECK-NEXT: DW_AT_name{{.*}}"Sptrptr" +// CHECK: 0x[[TYPEDEF_PTR_S]]: DW_TAG_typedef +// CHECK-NEXT: DW_AT_type{{.*}}{0x[[PTR_S]]} "S *" +// CHECK-NEXT: DW_AT_name{{.*}}"Sptr" + // First we confirm that first compile unit properly references type. // // CHECK: DW_TAG_compile_unit diff --git a/llvm/test/tools/llvm-cas/Inputs/oneline b/llvm/test/tools/llvm-cas/Inputs/oneline new file mode 100644 index 0000000..d95f3ad --- /dev/null +++ b/llvm/test/tools/llvm-cas/Inputs/oneline @@ -0,0 +1 @@ +content diff --git a/llvm/test/tools/llvm-cas/Inputs/oneline-nonewline b/llvm/test/tools/llvm-cas/Inputs/oneline-nonewline new file mode 100644 index 0000000..6b584e8 --- /dev/null +++ b/llvm/test/tools/llvm-cas/Inputs/oneline-nonewline @@ -0,0 +1 @@ +content
\ No newline at end of file diff --git a/llvm/test/tools/llvm-cas/action-cache.test b/llvm/test/tools/llvm-cas/action-cache.test new file mode 100644 index 0000000..fcb212c --- /dev/null +++ b/llvm/test/tools/llvm-cas/action-cache.test @@ -0,0 +1,14 @@ +RUN: rm -rf %t %t.cas +RUN: mkdir %t + +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data %S/Inputs/oneline >%t/oneline.casid +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data %S/Inputs/oneline-nonewline >%t/oneline-nonewline.casid + +RUN: llvm-cas --cas %t.cas --put-cache-key @%t/oneline.casid @%t/oneline-nonewline.casid +RUN: llvm-cas --cas %t.cas --get-cache-result @%t/oneline.casid > %t/result.casid +RUN: diff %t/oneline-nonewline.casid %t/result.casid + +RUN: not llvm-cas --cas %t.cas --get-cache-result @%t/oneline-nonewline.casid 2>&1 | FileCheck %s +CHECK: result not found diff --git a/llvm/test/tools/llvm-cas/cache.test b/llvm/test/tools/llvm-cas/cache.test new file mode 100644 index 0000000..f0ce691 --- /dev/null +++ b/llvm/test/tools/llvm-cas/cache.test @@ -0,0 +1,14 @@ +RUN: rm -rf %t %t.cas +RUN: mkdir %t + +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data /dev/null > %t/empty.casid +RUN: echo "abc" | \ +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data - >%t/abc.casid + +RUN: llvm-cas --cas %t/cas --put-cache-key @%t/abc.casid @%t/empty.casid +RUN: llvm-cas --cas %t/cas --get-cache-result @%t/abc.casid > %t/empty2.casid +RUN: diff %t/empty.casid %t/empty2.casid + +RUN: not llvm-cas --cas %t/cas --get-cache-result @%t/empty.casid diff --git a/llvm/test/tools/llvm-cas/dump.test b/llvm/test/tools/llvm-cas/dump.test new file mode 100644 index 0000000..f23bac6c --- /dev/null +++ b/llvm/test/tools/llvm-cas/dump.test @@ -0,0 +1,27 @@ +RUN: rm -rf %t +RUN: mkdir %t + +RUN: llvm-cas --cas %t/cas --make-blob \ +RUN: --data - </dev/null + +RUN: llvm-cas --cas %t/cas --make-blob \ +RUN: --data %s + +RUN: llvm-cas --cas %t/cas --dump | FileCheck %s + +// check the dump format. +CHECK: index: +CHECK-NEXT: hash-num-bits= +CHECK-NEXT: root addr= +// it should has at least one index +CHECK-NEXT: - index= + +// two records +CHECK: record +CHECK-NEXT: - addr= +CHECK-NEXT: - addr= + +// both should be small enough to be in data pool +CHECK: pool: +CHECK-NEXT: - addr= +CHECK-NEXT: - addr= diff --git a/llvm/test/tools/llvm-cas/lit.local.cfg b/llvm/test/tools/llvm-cas/lit.local.cfg new file mode 100644 index 0000000..379945b --- /dev/null +++ b/llvm/test/tools/llvm-cas/lit.local.cfg @@ -0,0 +1,2 @@ +if not config.have_ondisk_cas: + config.unsupported = True diff --git a/llvm/test/tools/llvm-cas/make-blob.test b/llvm/test/tools/llvm-cas/make-blob.test new file mode 100644 index 0000000..532a3a3 --- /dev/null +++ b/llvm/test/tools/llvm-cas/make-blob.test @@ -0,0 +1,41 @@ +RUN: rm -rf %t %t.cas +RUN: mkdir %t + +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data - </dev/null >%t/empty.casid +RUN: sed -e 's,^.,CHECK: ,' <%t/empty.casid >%t/empty.check +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data /dev/null | FileCheck %t/empty.check +RUN: echo "abc" | \ +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data - >%t/abc.casid +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data %S/Inputs/oneline >%t/oneline.casid +RUN: llvm-cas --cas %t.cas --make-blob \ +RUN: --data %S/Inputs/oneline-nonewline >%t/oneline-nonewline.casid + +RUN: llvm-cas --cas %t.cas --cat-node-data @%t/empty.casid |\ +RUN: FileCheck %s -check-prefix CHECK-EMPTY -allow-empty +CHECK-EMPTY-NOT: {{.}} + +RUN: llvm-cas --cas %t.cas --cat-node-data @%t/abc.casid |\ +RUN: FileCheck %s -check-prefix CHECK-ABC +CHECK-ABC: abc + +RUN: llvm-cas --cas %t.cas --cat-node-data @%t/oneline-nonewline.casid |\ +RUN: FileCheck %s -check-prefix CHECK-ONELINE +RUN: llvm-cas --cas %t.cas --cat-node-data @%t/oneline.casid |\ +RUN: FileCheck %s -check-prefix CHECK-ONELINE +CHECK-ONELINE: content + +# Double-check newlines. +RUN: llvm-cas --cas %t.cas --cat-node-data @%t/oneline-nonewline.casid \ +RUN: >%t/oneline-nonewline +RUN: diff %S/Inputs/oneline-nonewline %t/oneline-nonewline +RUN: llvm-cas --cas %t.cas --cat-node-data @%t/oneline.casid \ +RUN: >%t/oneline +RUN: diff %S/Inputs/oneline %t/oneline + +# Validate +RUN: llvm-cas --cas %t.cas --validate-object @%t/oneline-nonewline.casid +RUN: llvm-cas --cas %t.cas --validate-object @%t/oneline.casid diff --git a/llvm/test/tools/llvm-cas/make-node.test b/llvm/test/tools/llvm-cas/make-node.test new file mode 100644 index 0000000..de548af --- /dev/null +++ b/llvm/test/tools/llvm-cas/make-node.test @@ -0,0 +1,37 @@ +RUN: rm -rf %t +RUN: mkdir %t + +# Make some empty objects. +RUN: llvm-cas --cas %t/cas --make-node \ +RUN: --data - </dev/null >%t/empty.casid + +RUN: llvm-cas --cas %t/cas --cat-node-data @%t/empty.casid |\ +RUN: FileCheck %s -check-prefix CHECK-EMPTY -allow-empty +RUN: llvm-cas --cas %t/cas --ls-node-refs @%t/empty.casid |\ +RUN: FileCheck %s -check-prefix CHECK-EMPTY -allow-empty +CHECK-EMPTY-NOT: {{.}} + +# Make a complex object, which references existing ones. Reference a blob and +# other objects, and reference one of them twice to be sure they don't get +# deduped. +RUN: llvm-cas --cas %t/cas --make-blob --data /dev/null \ +RUN: >%t/empty-blob.casid +RUN: cat %t/empty.casid %t/empty.casid %t/empty-blob.casid \ +RUN: >%t/complex.refs +RUN: cat %t/complex.refs | sed -e 's,^.,CHECK: ,' > %t/complex.check +RUN: llvm-cas --cas %t/cas --make-node \ +RUN: --data %S/Inputs/oneline @%t/complex.refs \ +RUN: >%t/complex.casid +RUN: llvm-cas --cas %t/cas --cat-node-data \ +RUN: @%t/complex.casid | FileCheck %s -check-prefix COMPLEX-DATA +RUN: llvm-cas --cas %t/cas --ls-node-refs @%t/complex.casid |\ +RUN: FileCheck %t/complex.check +COMPLEX-DATA: content + +RUN: llvm-cas --cas %t/cas --validate-object @%t/complex.casid + +# Import from a new CAS. +RUN: llvm-cas --cas %t/new-cas --upstream-cas %t/cas --import @%t/complex.casid +RUN: llvm-cas --cas %t/new-cas --cat-node-data \ +RUN: @%t/complex.casid | FileCheck %s -check-prefix COMPLEX-DATA +RUN: llvm-cas --cas %t/new-cas --validate diff --git a/llvm/test/tools/llvm-cas/print-id.test b/llvm/test/tools/llvm-cas/print-id.test new file mode 100644 index 0000000..5a2efd5 --- /dev/null +++ b/llvm/test/tools/llvm-cas/print-id.test @@ -0,0 +1,13 @@ +RUN: rm -rf %t +RUN: mkdir %t + +RUN: llvm-cas --cas %t/cas --make-blob --data %S/Inputs/oneline > %t/id + +# Confirm that the ID has the right prefix, is well-formed, and that there's +# nothing else on the line. +RUN: FileCheck %s --match-full-lines --strict-whitespace <%t/id +CHECK:llvmcas://{{[a-z0-9]+}} + +# Confirm that there's a newline after. +RUN: wc -l <%t/id | FileCheck %s -check-prefix=NEWLINE +NEWLINE: 1 diff --git a/llvm/test/tools/llvm-cas/validation.test b/llvm/test/tools/llvm-cas/validation.test new file mode 100644 index 0000000..13f24f0 --- /dev/null +++ b/llvm/test/tools/llvm-cas/validation.test @@ -0,0 +1,31 @@ +RUN: rm -rf %t +RUN: mkdir %t + +# Ingest a blob which just fits inside the CAS data pool to make sure the validate passes. +RUN: truncate -s 7 %t/file +RUN: cat %t/file | \ +RUN: llvm-cas --cas %t/cas --make-blob \ +RUN: --data - +RUN: llvm-cas --cas %t/cas --validate --check-hash + +RUN: llvm-cas --cas %t/cas --validate +RUN: llvm-cas --cas %t/cas --validate --check-hash + +RUN: rm %t/cas/v1.1/data.v1 +RUN: not llvm-cas --cas %t/cas --validate +RUN: not llvm-cas --cas %t/cas --validate --check-hash + +RUN: mkdir %t/ac + +RUN: llvm-cas --cas %t/ac --make-blob \ +RUN: --data /dev/null > %t/empty.casid +RUN: echo "abc" | \ +RUN: llvm-cas --cas %t/ac --make-blob \ +RUN: --data - >%t/abc.casid + +RUN: llvm-cas --cas %t/ac --put-cache-key @%t/abc.casid @%t/empty.casid +RUN: llvm-cas --cas %t/ac --validate +# Note: records are 40 bytes (32 hash bytes + 8 byte value), so trim the last +# allocated record, leaving it invalid. +RUN: truncate -s -40 %t/ac/v1.1/actions.v1 +RUN: not llvm-cas --cas %t/ac --validate diff --git a/llvm/tools/llvm-cas/CMakeLists.txt b/llvm/tools/llvm-cas/CMakeLists.txt new file mode 100644 index 0000000..e9d40cb --- /dev/null +++ b/llvm/tools/llvm-cas/CMakeLists.txt @@ -0,0 +1,17 @@ +set(LLVM_TARGET_DEFINITIONS Options.td) +tablegen(LLVM Options.inc -gen-opt-parser-defs) +add_public_tablegen_target(LLVMCASToolTableGen) + +set(LLVM_LINK_COMPONENTS + Support + CAS + Option + ) + +add_llvm_tool(llvm-cas + llvm-cas.cpp + + DEPENDS + ${tablegen_deps} + LLVMCASToolTableGen + ) diff --git a/llvm/tools/llvm-cas/Options.td b/llvm/tools/llvm-cas/Options.td new file mode 100644 index 0000000..5ae64c1 --- /dev/null +++ b/llvm/tools/llvm-cas/Options.td @@ -0,0 +1,63 @@ +include "llvm/Option/OptParser.td" + +class F<string name> : Flag<["--", "-"], name>; + +def grp_action : OptionGroup<"Actions">, HelpText<"llvm-cas actions">; + +def help : F<"help">, HelpText<"Prints this help output">; +def : Flag<["-"], "h">, Alias<help>, HelpText<"Alias for --help">; + +// Tool actions + +def cas_dump : F<"dump">, HelpText<"Dump internal contents">, Group<grp_action>; +def cat_node_data : F<"cat-node-data">, + HelpText<"Cat node data">, + Group<grp_action>; +def make_blob : F<"make-blob">, HelpText<"Make blob">, Group<grp_action>; +def make_node : F<"make-node">, HelpText<"Make node">, Group<grp_action>; +def ls_node_refs : F<"ls-node-refs">, + HelpText<"List node refs">, + Group<grp_action>; +def import : F<"import">, + HelpText<"Import objects from another CAS">, + Group<grp_action>; +def put_cache_key : F<"put-cache-key">, + HelpText<"Set a value for a cache key">, + Group<grp_action>; +def get_cache_result : F<"get-cache-result">, + HelpText<"Get the result value from a cache key">, + Group<grp_action>; +def validate : F<"validate">, + HelpText<"Validate ObjectStore">, + Group<grp_action>; +def validate_object : F<"validate-object">, + HelpText<"Validate the object for CASID">, + Group<grp_action>; +def validate_if_needed : F<"validate-if-needed">, + HelpText<"Validate cas contents if needed">, + Group<grp_action>; +def prune : F<"prune">, HelpText<"Prune local cas storage">, Group<grp_action>; + +// Tool options + +def cas_path : Separate<["-", "--"], "cas">, + MetaVarName<"<path>">, + HelpText<"Path to CAS on disk">; + +def upstream_cas : Separate<["-", "--"], "upstream-cas">, + MetaVarName<"<path>">, + HelpText<"Path to another upstream CAS">; + +def data : Separate<["-", "--"], "data">, + MetaVarName<"<path>">, + HelpText<"Path to data or '-' for stdin">; + +def check_hash : F<"check-hash">, + HelpText<"Check all hashes during validation">; + +def allow_recovery : F<"allow-recovery">, + HelpText<"Allow recovery of CAS data">; + +def force : F<"force">, HelpText<"Force validation even if unnecessary">; + +def in_process : F<"in-process">, HelpText<"Validation in-process">; diff --git a/llvm/tools/llvm-cas/llvm-cas.cpp b/llvm/tools/llvm-cas/llvm-cas.cpp new file mode 100644 index 0000000..e72ee47 --- /dev/null +++ b/llvm/tools/llvm-cas/llvm-cas.cpp @@ -0,0 +1,405 @@ +//===----------------------------------------------------------------------===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// +/// +/// \file A utility for operating on LLVM CAS. +/// +//===----------------------------------------------------------------------===// + +#include "llvm/CAS/ActionCache.h" +#include "llvm/CAS/BuiltinUnifiedCASDatabases.h" +#include "llvm/CAS/ObjectStore.h" +#include "llvm/Option/Arg.h" +#include "llvm/Option/ArgList.h" +#include "llvm/Option/Option.h" +#include "llvm/Support/CommandLine.h" +#include "llvm/Support/Error.h" +#include "llvm/Support/InitLLVM.h" +#include "llvm/Support/MemoryBuffer.h" +#include "llvm/Support/raw_ostream.h" + +using namespace llvm; +using namespace llvm::cas; + +namespace { +enum ID { + OPT_INVALID = 0, // This is not an option ID. +#define OPTION(...) LLVM_MAKE_OPT_ID(__VA_ARGS__), +#include "Options.inc" +#undef OPTION +}; + +#define OPTTABLE_STR_TABLE_CODE +#include "Options.inc" +#undef OPTTABLE_STR_TABLE_CODE + +#define OPTTABLE_PREFIXES_TABLE_CODE +#include "Options.inc" +#undef OPTTABLE_PREFIXES_TABLE_CODE + +using namespace llvm::opt; +static constexpr opt::OptTable::Info InfoTable[] = { +#define OPTION(...) LLVM_CONSTRUCT_OPT_INFO(__VA_ARGS__), +#include "Options.inc" +#undef OPTION +}; + +class LLVMCASOptTable : public opt::GenericOptTable { +public: + LLVMCASOptTable() + : opt::GenericOptTable(OptionStrTable, OptionPrefixesTable, InfoTable) {} +}; + +enum class CommandKind { + Invalid, + Dump, + CatNodeData, + MakeBlob, + MakeNode, + ListObjectReferences, + Import, + PutCacheKey, + GetCacheResult, + Validate, + ValidateObject, + ValidateIfNeeded, + Prune, +}; + +struct CommandOptions { + CommandKind Command = CommandKind::Invalid; + std::vector<std::string> Inputs; + std::string CASPath; + std::string UpstreamCASPath; + std::string DataPath; + bool CheckHash; + bool AllowRecovery; + bool Force; + bool InProcess; + + static CommandKind getCommandKind(opt::Arg &A) { + switch (A.getOption().getID()) { + case OPT_cas_dump: + return CommandKind::Dump; + case OPT_cat_node_data: + return CommandKind::CatNodeData; + case OPT_make_blob: + return CommandKind::MakeBlob; + case OPT_make_node: + return CommandKind::MakeNode; + case OPT_ls_node_refs: + return CommandKind::ListObjectReferences; + case OPT_import: + return CommandKind::Import; + case OPT_put_cache_key: + return CommandKind::PutCacheKey; + case OPT_get_cache_result: + return CommandKind::GetCacheResult; + case OPT_validate: + return CommandKind::Validate; + case OPT_validate_object: + return CommandKind::ValidateObject; + case OPT_validate_if_needed: + return CommandKind::ValidateIfNeeded; + case OPT_prune: + return CommandKind::Prune; + } + return CommandKind::Invalid; + } + + // Command requires input. + static bool requiresInput(CommandKind Kind) { + return Kind != CommandKind::ValidateIfNeeded && + Kind != CommandKind::Validate && Kind != CommandKind::MakeBlob && + Kind != CommandKind::MakeNode && Kind != CommandKind::Dump && + Kind != CommandKind::Prune; + } +}; +} // namespace + +static int dump(ObjectStore &CAS); +static int listObjectReferences(ObjectStore &CAS, const CASID &ID); +static int catNodeData(ObjectStore &CAS, const CASID &ID); +static int makeBlob(ObjectStore &CAS, StringRef DataPath); +static int makeNode(ObjectStore &CAS, ArrayRef<std::string> References, + StringRef DataPath); +static int import(ObjectStore &FromCAS, ObjectStore &ToCAS, + ArrayRef<std::string> Objects); +static int putCacheKey(ObjectStore &CAS, ActionCache &AC, + ArrayRef<std::string> Objects); +static int getCacheResult(ObjectStore &CAS, ActionCache &AC, const CASID &ID); +static int validateObject(ObjectStore &CAS, const CASID &ID); +static int validate(ObjectStore &CAS, ActionCache &AC, bool CheckHash); +static int validateIfNeeded(StringRef Path, bool CheckHash, bool Force, + bool AllowRecovery, bool InProcess, + const char *Argv0); +static int prune(cas::ObjectStore &CAS); + +static Expected<CommandOptions> parseOptions(int Argc, char **Argv) { + BumpPtrAllocator Alloc; + StringSaver Saver(Alloc); + SmallVector<const char *> ExpanedArgs; + if (!cl::expandResponseFiles(Argc, Argv, nullptr, Saver, ExpanedArgs)) + return createStringError("cannot expand response file"); + + LLVMCASOptTable T; + unsigned MI, MC; + opt::InputArgList Args = T.ParseArgs(ExpanedArgs, MI, MC); + + for (auto *Arg : Args.filtered(OPT_UNKNOWN)) { + llvm::errs() << "ignoring unknown option: " << Arg->getSpelling() << '\n'; + } + + if (Args.hasArg(OPT_help)) { + T.printHelp( + outs(), + (std::string(Argv[0]) + " [action] [options] <input files>").c_str(), + "llvm-cas tool that performs CAS actions.", false); + exit(0); + } + + CommandOptions Opts; + for (auto *A : Args.filtered(OPT_grp_action)) + Opts.Command = CommandOptions::getCommandKind(*A); + + if (Opts.Command == CommandKind::Invalid) + return createStringError("no command action is specified"); + + for (auto *File : Args.filtered(OPT_INPUT)) + Opts.Inputs.push_back(File->getValue()); + Opts.CASPath = Args.getLastArgValue(OPT_cas_path); + Opts.UpstreamCASPath = Args.getLastArgValue(OPT_upstream_cas); + Opts.DataPath = Args.getLastArgValue(OPT_data); + Opts.CheckHash = Args.hasArg(OPT_check_hash); + Opts.AllowRecovery = Args.hasArg(OPT_allow_recovery); + Opts.Force = Args.hasArg(OPT_force); + Opts.InProcess = Args.hasArg(OPT_in_process); + + // Validate options. + if (Opts.CASPath.empty()) + return createStringError("missing --cas <path>"); + + if (Opts.Inputs.empty() && CommandOptions::requiresInput(Opts.Command)) + return createStringError("missing <input> to operate on"); + + return Opts; +} + +int main(int Argc, char **Argv) { + InitLLVM X(Argc, Argv); + + ExitOnError ExitOnErr; + auto Opts = ExitOnErr(parseOptions(Argc, Argv)); + + if (Opts.Command == CommandKind::ValidateIfNeeded) + return validateIfNeeded(Opts.CASPath, Opts.CheckHash, Opts.Force, + Opts.AllowRecovery, Opts.InProcess, Argv[0]); + + auto [CAS, AC] = ExitOnErr(createOnDiskUnifiedCASDatabases(Opts.CASPath)); + assert(CAS); + + if (Opts.Command == CommandKind::Dump) + return dump(*CAS); + + if (Opts.Command == CommandKind::Validate) + return validate(*CAS, *AC, Opts.CheckHash); + + if (Opts.Command == CommandKind::MakeBlob) + return makeBlob(*CAS, Opts.DataPath); + + if (Opts.Command == CommandKind::MakeNode) + return makeNode(*CAS, Opts.Inputs, Opts.DataPath); + + if (Opts.Command == CommandKind::Prune) + return prune(*CAS); + + if (Opts.Command == CommandKind::Import) { + if (Opts.UpstreamCASPath.empty()) + ExitOnErr(createStringError("missing '-upstream-cas'")); + + auto [UpstreamCAS, _] = + ExitOnErr(createOnDiskUnifiedCASDatabases(Opts.UpstreamCASPath)); + return import(*UpstreamCAS, *CAS, Opts.Inputs); + } + + if (Opts.Command == CommandKind::PutCacheKey || + Opts.Command == CommandKind::GetCacheResult) { + if (!AC) + ExitOnErr(createStringError("no action-cache available")); + } + + if (Opts.Command == CommandKind::PutCacheKey) + return putCacheKey(*CAS, *AC, Opts.Inputs); + + // Remaining commands need exactly one CAS object. + if (Opts.Inputs.size() > 1) + ExitOnErr(createStringError("too many <object>s, expected 1")); + CASID ID = ExitOnErr(CAS->parseID(Opts.Inputs.front())); + + if (Opts.Command == CommandKind::GetCacheResult) + return getCacheResult(*CAS, *AC, ID); + + if (Opts.Command == CommandKind::ListObjectReferences) + return listObjectReferences(*CAS, ID); + + if (Opts.Command == CommandKind::CatNodeData) + return catNodeData(*CAS, ID); + + assert(Opts.Command == CommandKind::ValidateObject); + return validateObject(*CAS, ID); +} + +static Expected<std::unique_ptr<MemoryBuffer>> openBuffer(StringRef DataPath) { + if (DataPath.empty()) + return createStringError("--data missing"); + return errorOrToExpected(DataPath == "-" + ? llvm::MemoryBuffer::getSTDIN() + : llvm::MemoryBuffer::getFile(DataPath)); +} + +int dump(ObjectStore &CAS) { + ExitOnError ExitOnErr("llvm-cas: dump: "); + CAS.print(llvm::outs()); + return 0; +} + +int makeBlob(ObjectStore &CAS, StringRef DataPath) { + ExitOnError ExitOnErr("llvm-cas: make-blob: "); + std::unique_ptr<MemoryBuffer> Buffer = ExitOnErr(openBuffer(DataPath)); + + ObjectProxy Blob = ExitOnErr(CAS.createProxy({}, Buffer->getBuffer())); + llvm::outs() << Blob.getID() << "\n"; + return 0; +} + +int catNodeData(ObjectStore &CAS, const CASID &ID) { + ExitOnError ExitOnErr("llvm-cas: cat-node-data: "); + llvm::outs() << ExitOnErr(CAS.getProxy(ID)).getData(); + return 0; +} + +int listObjectReferences(ObjectStore &CAS, const CASID &ID) { + ExitOnError ExitOnErr("llvm-cas: ls-node-refs: "); + + ObjectProxy Object = ExitOnErr(CAS.getProxy(ID)); + ExitOnErr(Object.forEachReference([&](ObjectRef Ref) -> Error { + llvm::outs() << CAS.getID(Ref) << "\n"; + return Error::success(); + })); + + return 0; +} + +static int makeNode(ObjectStore &CAS, ArrayRef<std::string> Objects, + StringRef DataPath) { + std::unique_ptr<MemoryBuffer> Data = + ExitOnError("llvm-cas: make-node: data: ")(openBuffer(DataPath)); + + SmallVector<ObjectRef> IDs; + for (StringRef Object : Objects) { + ExitOnError ObjectErr("llvm-cas: make-node: ref: "); + std::optional<ObjectRef> ID = + CAS.getReference(ObjectErr(CAS.parseID(Object))); + if (!ID) + ObjectErr(createStringError("unknown object '" + Object + "'")); + IDs.push_back(*ID); + } + + ExitOnError ExitOnErr("llvm-cas: make-node: "); + ObjectProxy Object = ExitOnErr(CAS.createProxy(IDs, Data->getBuffer())); + llvm::outs() << Object.getID() << "\n"; + return 0; +} + +static int import(ObjectStore &FromCAS, ObjectStore &ToCAS, + ArrayRef<std::string> Objects) { + ExitOnError ExitOnErr("llvm-cas: import: "); + + for (StringRef Object : Objects) { + CASID ID = ExitOnErr(FromCAS.parseID(Object)); + auto Ref = FromCAS.getReference(ID); + if (!Ref) + ExitOnErr(createStringError("input not found: " + ID.toString())); + + auto Imported = ExitOnErr(ToCAS.importObject(FromCAS, *Ref)); + llvm::outs() << ToCAS.getID(Imported).toString() << "\n"; + } + return 0; +} + +static int putCacheKey(ObjectStore &CAS, ActionCache &AC, + ArrayRef<std::string> Objects) { + ExitOnError ExitOnErr("llvm-cas: put-cache-key: "); + + if (Objects.size() % 2 != 0) + ExitOnErr(createStringError("expected pairs of inputs")); + while (!Objects.empty()) { + CASID Key = ExitOnErr(CAS.parseID(Objects[0])); + CASID Result = ExitOnErr(CAS.parseID(Objects[1])); + Objects = Objects.drop_front(2); + ExitOnErr(AC.put(Key, Result)); + } + return 0; +} + +static int getCacheResult(ObjectStore &CAS, ActionCache &AC, const CASID &ID) { + ExitOnError ExitOnErr("llvm-cas: get-cache-result: "); + + auto Result = ExitOnErr(AC.get(ID)); + if (!Result) { + outs() << "result not found\n"; + return 1; + } + outs() << *Result << "\n"; + return 0; +} + +int validateObject(ObjectStore &CAS, const CASID &ID) { + ExitOnError ExitOnErr("llvm-cas: validate-object: "); + ExitOnErr(CAS.validateObject(ID)); + outs() << ID << ": validated successfully\n"; + return 0; +} + +int validate(ObjectStore &CAS, ActionCache &AC, bool CheckHash) { + ExitOnError ExitOnErr("llvm-cas: validate: "); + ExitOnErr(CAS.validate(CheckHash)); + ExitOnErr(AC.validate()); + outs() << "validated successfully\n"; + return 0; +} + +int validateIfNeeded(StringRef Path, bool CheckHash, bool Force, + bool AllowRecovery, bool InProcess, const char *Argv0) { + ExitOnError ExitOnErr("llvm-cas: validate-if-needed: "); + std::string ExecStorage; + std::optional<StringRef> Exec; + if (!InProcess) { + ExecStorage = sys::fs::getMainExecutable(Argv0, (void *)validateIfNeeded); + Exec = ExecStorage; + } + ValidationResult Result = ExitOnErr(validateOnDiskUnifiedCASDatabasesIfNeeded( + Path, CheckHash, AllowRecovery, Force, Exec)); + switch (Result) { + case ValidationResult::Valid: + outs() << "validated successfully\n"; + break; + case ValidationResult::Recovered: + outs() << "recovered from invalid data\n"; + break; + case ValidationResult::Skipped: + outs() << "validation skipped\n"; + break; + } + return 0; +} + +static int prune(cas::ObjectStore &CAS) { + ExitOnError ExitOnErr("llvm-cas: prune: "); + ExitOnErr(CAS.pruneStorageData()); + return 0; +} diff --git a/llvm/unittests/Frontend/OpenMPDecompositionTest.cpp b/llvm/unittests/Frontend/OpenMPDecompositionTest.cpp index a8706ce..e0c6b39 100644 --- a/llvm/unittests/Frontend/OpenMPDecompositionTest.cpp +++ b/llvm/unittests/Frontend/OpenMPDecompositionTest.cpp @@ -88,6 +88,7 @@ using DistSchedule = tomp::clause::DistScheduleT<TypeTy, IdTy, ExprTy>; using Doacross = tomp::clause::DoacrossT<TypeTy, IdTy, ExprTy>; using DynamicAllocators = tomp::clause::DynamicAllocatorsT<TypeTy, IdTy, ExprTy>; +using DynGroupprivate = tomp::clause::DynGroupprivateT<TypeTy, IdTy, ExprTy>; using Enter = tomp::clause::EnterT<TypeTy, IdTy, ExprTy>; using Exclusive = tomp::clause::ExclusiveT<TypeTy, IdTy, ExprTy>; using Fail = tomp::clause::FailT<TypeTy, IdTy, ExprTy>; diff --git a/llvm/utils/prepare-code-coverage-artifact.py b/llvm/utils/prepare-code-coverage-artifact.py index b865211..5d4f549 100755 --- a/llvm/utils/prepare-code-coverage-artifact.py +++ b/llvm/utils/prepare-code-coverage-artifact.py @@ -1,6 +1,4 @@ -#!/usr/bin/env python - -from __future__ import print_function +#!/usr/bin/env python3 """Prepare a code coverage artifact. diff --git a/mlir/include/mlir/Conversion/LLVMCommon/VectorPattern.h b/mlir/include/mlir/Conversion/LLVMCommon/VectorPattern.h index cad6cec..e7ab63a 100644 --- a/mlir/include/mlir/Conversion/LLVMCommon/VectorPattern.h +++ b/mlir/include/mlir/Conversion/LLVMCommon/VectorPattern.h @@ -86,7 +86,8 @@ private: /// ArrayRef<NamedAttribute>. template <typename SourceOp, typename TargetOp, template <typename, typename> typename AttrConvert = - AttrConvertPassThrough> + AttrConvertPassThrough, + bool FailOnUnsupportedFP = false> class VectorConvertToLLVMPattern : public ConvertOpToLLVMPattern<SourceOp> { public: using ConvertOpToLLVMPattern<SourceOp>::ConvertOpToLLVMPattern; @@ -123,11 +124,13 @@ public: "unsupported floating point type"); return success(); }; - for (Value operand : op->getOperands()) - if (failed(checkType(operand))) + if (FailOnUnsupportedFP) { + for (Value operand : op->getOperands()) + if (failed(checkType(operand))) + return failure(); + if (failed(checkType(op->getResult(0)))) return failure(); - if (failed(checkType(op->getResult(0)))) - return failure(); + } // Determine attributes for the target op AttrConvert<SourceOp, TargetOp> attrConvert(op); diff --git a/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td b/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td index 80bc0e5..1cc5b74 100644 --- a/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td +++ b/mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td @@ -1236,6 +1236,23 @@ def NVVM_FenceProxyAcquireOp : NVVM_Op<"fence.proxy.acquire">, let hasVerifier = 1; } +def NVVM_MembarOp : NVVM_Op<"memory.barrier">, + Arguments<(ins MemScopeKindAttr:$scope)> { + let summary = "Memory barrier operation"; + let description = [{ + `membar` operation guarantees that prior memory accesses requested by this + thread are performed at the specified `scope`, before later memory + operations requested by this thread following the membar instruction. + + [For more information, see PTX ISA](https://docs.nvidia.com/cuda/parallel-thread-execution/#parallel-synchronization-and-communication-instructions-membar) + }]; + + let assemblyFormat = "$scope attr-dict"; + let llvmBuilder = [{ + createIntrinsicCall(builder, getMembarIntrinsicID($scope), {}); + }]; +} + def NVVM_FenceProxyReleaseOp : NVVM_Op<"fence.proxy.release">, Arguments<(ins MemScopeKindAttr:$scope, DefaultValuedAttr<ProxyKindAttr, diff --git a/mlir/include/mlir/Dialect/LLVMIR/XeVMOps.td b/mlir/include/mlir/Dialect/LLVMIR/XeVMOps.td index 2dd6121..388efaa 100644 --- a/mlir/include/mlir/Dialect/LLVMIR/XeVMOps.td +++ b/mlir/include/mlir/Dialect/LLVMIR/XeVMOps.td @@ -463,10 +463,9 @@ def XeVM_PrefetchOp def XeVM_BlockPrefetch2dOp : XeVM_Op<"blockprefetch2d">, - Arguments<(ins Arg<LLVM_AnyPointer, "", [MemRead]>:$ptr, I32:$base_width, - I32:$base_height, I32:$base_pitch, I32:$x, I32:$y, - I32Attr:$elem_size_in_bits, I32Attr:$tile_width, I32Attr:$tile_height, - I32Attr:$v_blocks, + Arguments<(ins LLVM_AnyPointer:$ptr, I32:$base_width, I32:$base_height, + I32:$base_pitch, I32:$x, I32:$y, I32Attr:$elem_size_in_bits, + I32Attr:$tile_width, I32Attr:$tile_height, I32Attr:$v_blocks, OptionalAttr<XeVM_LoadCacheControlAttr>:$cache_control)> { let summary = "2D block prefetch"; diff --git a/mlir/lib/Conversion/ArithToLLVM/ArithToLLVM.cpp b/mlir/lib/Conversion/ArithToLLVM/ArithToLLVM.cpp index 03ed4d5..b609990 100644 --- a/mlir/lib/Conversion/ArithToLLVM/ArithToLLVM.cpp +++ b/mlir/lib/Conversion/ArithToLLVM/ArithToLLVM.cpp @@ -36,20 +36,23 @@ namespace { /// attribute. template <typename SourceOp, typename TargetOp, bool Constrained, template <typename, typename> typename AttrConvert = - AttrConvertPassThrough> + AttrConvertPassThrough, + bool FailOnUnsupportedFP = false> struct ConstrainedVectorConvertToLLVMPattern - : public VectorConvertToLLVMPattern<SourceOp, TargetOp, AttrConvert> { - using VectorConvertToLLVMPattern<SourceOp, TargetOp, - AttrConvert>::VectorConvertToLLVMPattern; + : public VectorConvertToLLVMPattern<SourceOp, TargetOp, AttrConvert, + FailOnUnsupportedFP> { + using VectorConvertToLLVMPattern< + SourceOp, TargetOp, AttrConvert, + FailOnUnsupportedFP>::VectorConvertToLLVMPattern; LogicalResult matchAndRewrite(SourceOp op, typename SourceOp::Adaptor adaptor, ConversionPatternRewriter &rewriter) const override { if (Constrained != static_cast<bool>(op.getRoundingModeAttr())) return failure(); - return VectorConvertToLLVMPattern<SourceOp, TargetOp, - AttrConvert>::matchAndRewrite(op, adaptor, - rewriter); + return VectorConvertToLLVMPattern< + SourceOp, TargetOp, AttrConvert, + FailOnUnsupportedFP>::matchAndRewrite(op, adaptor, rewriter); } }; @@ -78,7 +81,8 @@ struct IdentityBitcastLowering final using AddFOpLowering = VectorConvertToLLVMPattern<arith::AddFOp, LLVM::FAddOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using AddIOpLowering = VectorConvertToLLVMPattern<arith::AddIOp, LLVM::AddOp, arith::AttrConvertOverflowToLLVM>; @@ -87,53 +91,67 @@ using BitcastOpLowering = VectorConvertToLLVMPattern<arith::BitcastOp, LLVM::BitcastOp>; using DivFOpLowering = VectorConvertToLLVMPattern<arith::DivFOp, LLVM::FDivOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using DivSIOpLowering = VectorConvertToLLVMPattern<arith::DivSIOp, LLVM::SDivOp>; using DivUIOpLowering = VectorConvertToLLVMPattern<arith::DivUIOp, LLVM::UDivOp>; -using ExtFOpLowering = VectorConvertToLLVMPattern<arith::ExtFOp, LLVM::FPExtOp>; +using ExtFOpLowering = VectorConvertToLLVMPattern<arith::ExtFOp, LLVM::FPExtOp, + AttrConvertPassThrough, + /*FailOnUnsupportedFP=*/true>; using ExtSIOpLowering = VectorConvertToLLVMPattern<arith::ExtSIOp, LLVM::SExtOp>; using ExtUIOpLowering = VectorConvertToLLVMPattern<arith::ExtUIOp, LLVM::ZExtOp>; using FPToSIOpLowering = - VectorConvertToLLVMPattern<arith::FPToSIOp, LLVM::FPToSIOp>; + VectorConvertToLLVMPattern<arith::FPToSIOp, LLVM::FPToSIOp, + AttrConvertPassThrough, + /*FailOnUnsupportedFP=*/true>; using FPToUIOpLowering = - VectorConvertToLLVMPattern<arith::FPToUIOp, LLVM::FPToUIOp>; + VectorConvertToLLVMPattern<arith::FPToUIOp, LLVM::FPToUIOp, + AttrConvertPassThrough, + /*FailOnUnsupportedFP=*/true>; using MaximumFOpLowering = VectorConvertToLLVMPattern<arith::MaximumFOp, LLVM::MaximumOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using MaxNumFOpLowering = VectorConvertToLLVMPattern<arith::MaxNumFOp, LLVM::MaxNumOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using MaxSIOpLowering = VectorConvertToLLVMPattern<arith::MaxSIOp, LLVM::SMaxOp>; using MaxUIOpLowering = VectorConvertToLLVMPattern<arith::MaxUIOp, LLVM::UMaxOp>; using MinimumFOpLowering = VectorConvertToLLVMPattern<arith::MinimumFOp, LLVM::MinimumOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using MinNumFOpLowering = VectorConvertToLLVMPattern<arith::MinNumFOp, LLVM::MinNumOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using MinSIOpLowering = VectorConvertToLLVMPattern<arith::MinSIOp, LLVM::SMinOp>; using MinUIOpLowering = VectorConvertToLLVMPattern<arith::MinUIOp, LLVM::UMinOp>; using MulFOpLowering = VectorConvertToLLVMPattern<arith::MulFOp, LLVM::FMulOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using MulIOpLowering = VectorConvertToLLVMPattern<arith::MulIOp, LLVM::MulOp, arith::AttrConvertOverflowToLLVM>; using NegFOpLowering = VectorConvertToLLVMPattern<arith::NegFOp, LLVM::FNegOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using OrIOpLowering = VectorConvertToLLVMPattern<arith::OrIOp, LLVM::OrOp>; using RemFOpLowering = VectorConvertToLLVMPattern<arith::RemFOp, LLVM::FRemOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using RemSIOpLowering = VectorConvertToLLVMPattern<arith::RemSIOp, LLVM::SRemOp>; using RemUIOpLowering = @@ -151,21 +169,25 @@ using SIToFPOpLowering = VectorConvertToLLVMPattern<arith::SIToFPOp, LLVM::SIToFPOp>; using SubFOpLowering = VectorConvertToLLVMPattern<arith::SubFOp, LLVM::FSubOp, - arith::AttrConvertFastMathToLLVM>; + arith::AttrConvertFastMathToLLVM, + /*FailOnUnsupportedFP=*/true>; using SubIOpLowering = VectorConvertToLLVMPattern<arith::SubIOp, LLVM::SubOp, arith::AttrConvertOverflowToLLVM>; using TruncFOpLowering = ConstrainedVectorConvertToLLVMPattern<arith::TruncFOp, LLVM::FPTruncOp, - false>; + false, AttrConvertPassThrough, + /*FailOnUnsupportedFP=*/true>; using ConstrainedTruncFOpLowering = ConstrainedVectorConvertToLLVMPattern< arith::TruncFOp, LLVM::ConstrainedFPTruncIntr, true, - arith::AttrConverterConstrainedFPToLLVM>; + arith::AttrConverterConstrainedFPToLLVM, /*FailOnUnsupportedFP=*/true>; using TruncIOpLowering = VectorConvertToLLVMPattern<arith::TruncIOp, LLVM::TruncOp, arith::AttrConvertOverflowToLLVM>; using UIToFPOpLowering = - VectorConvertToLLVMPattern<arith::UIToFPOp, LLVM::UIToFPOp>; + VectorConvertToLLVMPattern<arith::UIToFPOp, LLVM::UIToFPOp, + AttrConvertPassThrough, + /*FailOnUnsupportedFP=*/true>; using XOrIOpLowering = VectorConvertToLLVMPattern<arith::XOrIOp, LLVM::XOrOp>; //===----------------------------------------------------------------------===// diff --git a/mlir/lib/Dialect/Bufferization/IR/BufferizationDialect.cpp b/mlir/lib/Dialect/Bufferization/IR/BufferizationDialect.cpp index 6c08cdf..d6c3cd6 100644 --- a/mlir/lib/Dialect/Bufferization/IR/BufferizationDialect.cpp +++ b/mlir/lib/Dialect/Bufferization/IR/BufferizationDialect.cpp @@ -21,25 +21,6 @@ using namespace mlir::bufferization; #include "mlir/Dialect/Bufferization/IR/BufferizationOpsDialect.cpp.inc" -/// Attribute name used to mark function arguments who's buffers can be written -/// to during One-Shot Module Bufferize. -constexpr const ::llvm::StringLiteral BufferizationDialect::kWritableAttrName; - -/// Attribute name used to mark the bufferization layout for region arguments -/// during One-Shot Module Bufferize. -constexpr const ::llvm::StringLiteral - BufferizationDialect::kBufferLayoutAttrName; - -/// An attribute that can be attached to ops with an allocation and/or -/// deallocation side effect. It indicates that the op is under a "manual -/// deallocation" scheme. In the case of an allocation op, the returned -/// value is *not* an automatically managed allocation and assigned an -/// ownership of "false". Furthermore, only deallocation ops that are -/// guaranteed to deallocate a buffer under "manual deallocation" are -/// allowed to have this attribute. (Deallocation ops without this -/// attribute are rejected by the ownership-based buffer deallocation pass.) -constexpr const ::llvm::StringLiteral BufferizationDialect::kManualDeallocation; - //===----------------------------------------------------------------------===// // Bufferization Dialect Interfaces //===----------------------------------------------------------------------===// diff --git a/mlir/lib/Target/LLVMIR/Dialect/NVVM/NVVMToLLVMIRTranslation.cpp b/mlir/lib/Target/LLVMIR/Dialect/NVVM/NVVMToLLVMIRTranslation.cpp index 0964e1b..cecff51 100644 --- a/mlir/lib/Target/LLVMIR/Dialect/NVVM/NVVMToLLVMIRTranslation.cpp +++ b/mlir/lib/Target/LLVMIR/Dialect/NVVM/NVVMToLLVMIRTranslation.cpp @@ -291,6 +291,20 @@ static unsigned getUnidirectionalFenceProxyID(NVVM::ProxyKind fromProxy, llvm_unreachable("Unsupported proxy kinds"); } +static unsigned getMembarIntrinsicID(NVVM::MemScopeKind scope) { + switch (scope) { + case NVVM::MemScopeKind::CTA: + return llvm::Intrinsic::nvvm_membar_cta; + case NVVM::MemScopeKind::CLUSTER: + return llvm::Intrinsic::nvvm_fence_sc_cluster; + case NVVM::MemScopeKind::GPU: + return llvm::Intrinsic::nvvm_membar_gl; + case NVVM::MemScopeKind::SYS: + return llvm::Intrinsic::nvvm_membar_sys; + } + llvm_unreachable("Unknown scope for memory barrier"); +} + #define TCGEN05LD(SHAPE, NUM) llvm::Intrinsic::nvvm_tcgen05_ld_##SHAPE##_##NUM static llvm::Intrinsic::ID diff --git a/mlir/test/Conversion/ArithToLLVM/arith-to-llvm.mlir b/mlir/test/Conversion/ArithToLLVM/arith-to-llvm.mlir index b5dcb01..5f1ec66 100644 --- a/mlir/test/Conversion/ArithToLLVM/arith-to-llvm.mlir +++ b/mlir/test/Conversion/ArithToLLVM/arith-to-llvm.mlir @@ -754,11 +754,13 @@ func.func @memref_bitcast(%1: memref<?xi16>) -> memref<?xbf16> { // CHECK: arith.addf {{.*}} : f4E2M1FN // CHECK: arith.addf {{.*}} : vector<4xf4E2M1FN> // CHECK: arith.addf {{.*}} : vector<8x4xf4E2M1FN> -func.func @unsupported_fp_type(%arg0: f4E2M1FN, %arg1: vector<4xf4E2M1FN>, %arg2: vector<8x4xf4E2M1FN>) -> (f4E2M1FN, vector<4xf4E2M1FN>, vector<8x4xf4E2M1FN>) { +// CHECK: llvm.select {{.*}} : i1, i4 +func.func @unsupported_fp_type(%arg0: f4E2M1FN, %arg1: vector<4xf4E2M1FN>, %arg2: vector<8x4xf4E2M1FN>, %arg3: f4E2M1FN, %arg4: i1) { %0 = arith.addf %arg0, %arg0 : f4E2M1FN %1 = arith.addf %arg1, %arg1 : vector<4xf4E2M1FN> %2 = arith.addf %arg2, %arg2 : vector<8x4xf4E2M1FN> - return %0, %1, %2 : f4E2M1FN, vector<4xf4E2M1FN>, vector<8x4xf4E2M1FN> + %3 = arith.select %arg4, %arg0, %arg3 : f4E2M1FN + return } // ----- diff --git a/mlir/test/Conversion/XeGPUToXeVM/loadstoreprefetch.mlir b/mlir/test/Conversion/XeGPUToXeVM/loadstoreprefetch.mlir index 9c552d8..d606cf5 100644 --- a/mlir/test/Conversion/XeGPUToXeVM/loadstoreprefetch.mlir +++ b/mlir/test/Conversion/XeGPUToXeVM/loadstoreprefetch.mlir @@ -1,15 +1,16 @@ -// RUN: mlir-opt %s --split-input-file -convert-xegpu-to-xevm | FileCheck %s +// RUN: mlir-opt %s --split-input-file -convert-xegpu-to-xevm -canonicalize | FileCheck %s gpu.module @test { // CHECK-LABEL: @load_gather_i64_src_value_offset -// CHECK-SAME: %[[ARG0:.*]]: i64, %[[ARG1:.*]]: vector<1xindex> -gpu.func @load_gather_i64_src_value_offset(%src: i64, %offset: vector<1xindex>) { +// CHECK-SAME: %[[ARG0:.*]]: i64, %[[ARG1:.*]]: vector<1xindex>, %[[ARG2:.*]]: memref<1xf16> +// CHECK-SAME: %[[ARG3:.*]]: vector<1xi1> +gpu.func @load_gather_i64_src_value_offset(%src: i64, %offset: vector<1xindex>, %dst: memref<1xf16>, %mask: vector<1xi1>) { + // CHECK: %[[C0:.*]] = arith.constant 0 : index + // CHECK: %[[CST_0:.*]] = arith.constant 0.000000e+00 : f16 + // CHECK: %[[C2_I64:.*]] = arith.constant 2 : i64 + // CHECK: %[[VAR2:.*]] = vector.extract %[[ARG3]][0] : i1 from vector<1xi1> // CHECK: %[[VAR0:.*]] = vector.extract %[[ARG1]][0] : index from vector<1xindex> // CHECK: %[[VAR1:.*]] = arith.index_castui %[[VAR0]] : index to i64 - // CHECK: %[[CST:.*]] = arith.constant dense<true> : vector<1xi1> - // CHECK: %[[VAR2:.*]] = vector.extract %[[CST]][0] : i1 from vector<1xi1> - %1 = arith.constant dense<1>: vector<1xi1> - // CHECK: %[[C2_I64:.*]] = arith.constant 2 : i64 // CHECK: %[[VAR3:.*]] = arith.muli %[[VAR1]], %[[C2_I64]] : i64 // CHECK: %[[VAR4:.*]] = arith.addi %[[ARG0]], %[[VAR3]] : i64 // CHECK: %[[VAR5:.*]] = llvm.inttoptr %[[VAR4]] : i64 to !llvm.ptr<1> @@ -17,11 +18,12 @@ gpu.func @load_gather_i64_src_value_offset(%src: i64, %offset: vector<1xindex>) // CHECK: %[[VAR7:.*]] = llvm.load %[[VAR5]] {cache_control = #xevm.load_cache_control<L1c_L2uc_L3uc>} : !llvm.ptr<1> -> f16 // CHECK: scf.yield %[[VAR7]] : f16 // CHECK: } else { - // CHECK: %[[CST_0:.*]] = arith.constant 0.000000e+00 : f16 // CHECK: scf.yield %[[CST_0]] : f16 // CHECK: } - %3 = xegpu.load %src[%offset], %1 <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}> + %0 = xegpu.load %src[%offset], %mask <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}> : i64, vector<1xindex>, vector<1xi1> -> vector<1xf16> + %c0 = arith.constant 0 : index + vector.store %0, %dst[%c0] : memref<1xf16>, vector<1xf16> gpu.return } } @@ -30,16 +32,16 @@ gpu.func @load_gather_i64_src_value_offset(%src: i64, %offset: vector<1xindex>) gpu.module @test { // CHECK-LABEL: @source_materialize_single_elem_vec // CHECK-SAME: %[[ARG0:.*]]: i64, %[[ARG1:.*]]: vector<1xindex>, %[[ARG2:.*]]: memref<1xf16> -gpu.func @source_materialize_single_elem_vec(%src: i64, %offset: vector<1xindex>, %dst: memref<1xf16>) { - %1 = arith.constant dense<1>: vector<1xi1> - %3 = xegpu.load %src[%offset], %1 <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}> +// CHECK-SAME: %[[ARG3:.*]]: vector<1xi1> +gpu.func @source_materialize_single_elem_vec(%src: i64, %offset: vector<1xindex>, %dst: memref<1xf16>, %mask: vector<1xi1>) { + %0 = xegpu.load %src[%offset], %mask <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}> : i64, vector<1xindex>, vector<1xi1> -> vector<1xf16> + // CHECK: %[[C0:.*]] = arith.constant 0 : index // CHECK: %[[VAR_IF:.*]] = scf.if // CHECK: %[[VAR_RET:.*]] = vector.broadcast %[[VAR_IF]] : f16 to vector<1xf16> - // CHECK: %[[C0:.*]] = arith.constant 0 : index // CHECK: vector.store %[[VAR_RET]], %[[ARG2]][%[[C0]]] : memref<1xf16>, vector<1xf16> %c0 = arith.constant 0 : index - vector.store %3, %dst[%c0] : memref<1xf16>, vector<1xf16> + vector.store %0, %dst[%c0] : memref<1xf16>, vector<1xf16> gpu.return } } @@ -48,24 +50,21 @@ gpu.func @source_materialize_single_elem_vec(%src: i64, %offset: vector<1xindex> gpu.module @test { // CHECK-LABEL: @store_scatter_i64_src_value_offset -// CHECK-SAME: %[[ARG0:.*]]: i64, %[[ARG1:.*]]: vector<1xindex> -gpu.func @store_scatter_i64_src_value_offset(%src: i64, %offset: vector<1xindex>) { +// CHECK-SAME: %[[ARG0:.*]]: i64, %[[ARG1:.*]]: vector<1xindex>, %[[ARG2:.*]]: vector<1xi1> +gpu.func @store_scatter_i64_src_value_offset(%src: i64, %offset: vector<1xindex>, %mask: vector<1xi1>) { + // CHECK: %[[CST_0:.*]] = arith.constant 2.900000e+00 : f32 + // CHECK: %[[C4_I64:.*]] = arith.constant 4 : i64 + // CHECK: %[[VAR2:.*]] = vector.extract %[[ARG2]][0] : i1 from vector<1xi1> // CHECK: %[[VAR0:.*]] = vector.extract %[[ARG1]][0] : index from vector<1xindex> // CHECK: %[[VAR1:.*]] = arith.index_castui %[[VAR0]] : index to i64 - // CHECK: %[[CST:.*]] = arith.constant dense<true> : vector<1xi1> - // CHECK: %[[VAR2:.*]] = vector.extract %[[CST]][0] : i1 from vector<1xi1> - %1 = arith.constant dense<1>: vector<1xi1> - // CHECK: %[[CST_0:.*]] = arith.constant dense<2.900000e+00> : vector<1xf32> - // CHECK: %[[VAR3:.*]] = vector.extract %[[CST_0]][0] : f32 from vector<1xf32> - %2 = arith.constant dense<2.9>: vector<1xf32> - // CHECK: %[[C4_I64:.*]] = arith.constant 4 : i64 + %0 = arith.constant dense<2.9>: vector<1xf32> // CHECK: %[[VAR4:.*]] = arith.muli %[[VAR1]], %[[C4_I64]] : i64 // CHECK: %[[VAR5:.*]] = arith.addi %[[ARG0]], %[[VAR4]] : i64 // CHECK: %[[VAR6:.*]] = llvm.inttoptr %[[VAR5]] : i64 to !llvm.ptr<1> // CHECK: scf.if %[[VAR2]] { - // CHECK: llvm.store %[[VAR3]], %[[VAR6]] {cache_control = #xevm.store_cache_control<L1wb_L2uc_L3uc>} : f32, !llvm.ptr<1> + // CHECK: llvm.store %[[CST_0]], %[[VAR6]] {cache_control = #xevm.store_cache_control<L1wb_L2uc_L3uc>} : f32, !llvm.ptr<1> // CHECK: } - xegpu.store %2, %src[%offset], %1 <{l1_hint = #xegpu.cache_hint<write_back>, l2_hint = #xegpu.cache_hint<uncached>}> + xegpu.store %0, %src[%offset], %mask <{l1_hint = #xegpu.cache_hint<write_back>, l2_hint = #xegpu.cache_hint<uncached>}> : vector<1xf32>, i64, vector<1xindex>, vector<1xi1> gpu.return } @@ -76,9 +75,9 @@ gpu.module @test { // CHECK-LABEL: @prefetch_i64_src_value_offset // CHECK-SAME: %[[ARG0:.*]]: i64, %[[ARG1:.*]]: vector<1xindex> gpu.func @prefetch_i64_src_value_offset(%src: i64, %offset: vector<1xindex>) { + // CHECK: %[[C4_I64:.*]] = arith.constant 4 : i64 // CHECK: %[[VAR0:.*]] = vector.extract %[[ARG1]][0] : index from vector<1xindex> // CHECK: %[[VAR1:.*]] = arith.index_castui %[[VAR0]] : index to i64 - // CHECK: %[[C4_I64:.*]] = arith.constant 4 : i64 // CHECK: %[[VAR2:.*]] = arith.muli %[[VAR1]], %[[C4_I64]] : i64 // CHECK: %[[VAR3:.*]] = arith.addi %[[ARG0]], %[[VAR2]] : i64 // CHECK: %[[VAR4:.*]] = llvm.inttoptr %[[VAR3]] : i64 to !llvm.ptr<1> @@ -94,11 +93,11 @@ gpu.module @test { // CHECK-LABEL: @prefetch_memref_src_value_offset // CHECK-SAME: %[[ARG0:.*]]: memref<256xf32>, %[[ARG1:.*]]: vector<1xindex> gpu.func @prefetch_memref_src_value_offset(%src: memref<256xf32>, %offset: vector<1xindex>) { + // CHECK: %[[C4_I64:.*]] = arith.constant 4 : i64 // CHECK: %[[VAR0:.*]] = vector.extract %[[ARG1]][0] : index from vector<1xindex> // CHECK: %[[VAR1:.*]] = arith.index_castui %[[VAR0]] : index to i64 // CHECK: %[[INTPTR:.*]] = memref.extract_aligned_pointer_as_index %[[ARG0]] : memref<256xf32> -> index // CHECK: %[[VAR2:.*]] = arith.index_castui %[[INTPTR]] : index to i64 - // CHECK: %[[C4_I64:.*]] = arith.constant 4 : i64 // CHECK: %[[VAR3:.*]] = arith.muli %[[VAR1]], %[[C4_I64]] : i64 // CHECK: %[[VAR4:.*]] = arith.addi %[[VAR2]], %[[VAR3]] : i64 // CHECK: %[[VAR5:.*]] = llvm.inttoptr %[[VAR4]] : i64 to !llvm.ptr<1> diff --git a/mlir/test/Conversion/XeGPUToXeVM/prefetch_nd.mlir b/mlir/test/Conversion/XeGPUToXeVM/prefetch_nd.mlir index 873478a..e4b3030 100644 --- a/mlir/test/Conversion/XeGPUToXeVM/prefetch_nd.mlir +++ b/mlir/test/Conversion/XeGPUToXeVM/prefetch_nd.mlir @@ -1,34 +1,29 @@ -// RUN: mlir-opt -convert-xegpu-to-xevm -split-input-file %s | FileCheck %s +// RUN: mlir-opt -convert-xegpu-to-xevm -canonicalize %s | FileCheck %s -gpu.module @fence_check { - gpu.func @fence(%src: memref<8x16xf32, 1>, %dst: memref<8x16xf32, 1>) kernel { +gpu.module @prefetch_nd_check { + // CHECK-LABEL: gpu.func @prefetch_nd + gpu.func @prefetch_nd(%src: memref<8x16xf32, 1>, %dst: memref<8x16xf32, 1>) kernel { + // CHECK: %[[PREF_BASE_ROW_IN_BYTES:.*]] = arith.constant 64 : i32 + // CHECK: %[[LD_CREATE_DESC_I64:.*]] = arith.constant dense<0> : vector<4xi64> + // CHECK: %[[PREF_BASE_H:.*]] = arith.constant 8 : i32 + // CHECK: %[[PREF_BASE_W:.*]] = arith.constant 16 : i32 + // CHECK: %[[OFFSET_ZERO:.*]] = arith.constant 0 : i32 %srcce = memref.memory_space_cast %src : memref<8x16xf32, 1> to memref<8x16xf32> - %dstte = memref.memory_space_cast %dst : memref<8x16xf32, 1> to memref<8x16xf32> - // CHECK: %[[LD_PTR_AS_I64:.*]] = arith.index_castui {{.*}} : index to i64 - // CHECK: %[[LD_CREATE_DESC_I64:.*]] = vector.bitcast {{.*}} : vector<8xi32> to vector<4xi64> // CHECK: %[[LD_DESC_0:.*]] = vector.insert %[[LD_PTR_AS_I64]], %[[LD_CREATE_DESC_I64]] [0] : i64 into vector<4xi64> // CHECK: %[[LD_DESC_1:.*]] = vector.bitcast %[[LD_DESC_0]] : vector<4xi64> to vector<8xi32> - // CHECK: %[[LD_DESC_2:.*]] = vector.insert {{.*}}, %[[LD_DESC_1]] [2] : i32 into vector<8xi32> - // CHECK: %[[LD_DESC_3:.*]] = vector.insert {{.*}}, %[[LD_DESC_2]] [3] : i32 into vector<8xi32> - // CHECK: %[[LD_DESC_4:.*]] = vector.insert {{.*}}, %[[LD_DESC_3]] [4] : i32 into vector<8xi32> - // CHECK: %[[LD_DESC:.*]] = vector.insert {{.*}}, %[[LD_DESC_4]] [5] : i32 into vector<8xi32> + // CHECK: %[[LD_DESC_2:.*]] = vector.insert %[[PREF_BASE_W]], %[[LD_DESC_1]] [2] : i32 into vector<8xi32> + // CHECK: %[[LD_DESC_3:.*]] = vector.insert %[[PREF_BASE_H]], %[[LD_DESC_2]] [3] : i32 into vector<8xi32> + // CHECK: %[[LD_DESC_4:.*]] = vector.insert %[[OFFSET_ZERO]], %[[LD_DESC_3]] [4] : i32 into vector<8xi32> + // CHECK: %[[LD_DESC:.*]] = vector.insert %[[OFFSET_ZERO]], %[[LD_DESC_4]] [5] : i32 into vector<8xi32> %src_tdesc = xegpu.create_nd_tdesc %srcce : memref<8x16xf32> -> !xegpu.tensor_desc<8x16xf32, #xegpu.block_tdesc_attr<memory_space = global>, #xegpu.layout<lane_layout = [1, 16], lane_data = [1, 1]>> //CHECK: %[[LD_DESC_I64:.*]] = vector.bitcast %[[LD_DESC]] : vector<8xi32> to vector<4xi64> //CHECK: %[[PREF_INTPTR:.*]] = vector.extract %[[LD_DESC_I64]][0] : i64 from vector<4xi64> - //CHECK: %[[PREF_BASE_W:.*]] = vector.extract %[[LD_DESC]][2] : i32 from vector<8xi32> - //CHECK: %[[PREF_BASE_H:.*]] = vector.extract %[[LD_DESC]][3] : i32 from vector<8xi32> - //CHECK: %[[PREF_TILE_W64:.*]] = arith.constant 0 : i64 - //CHECK: %[[PREF_TILE_W:.*]] = arith.trunci %[[PREF_TILE_W64]] : i64 to i32 - //CHECK: %[[PREF_TILE_H64:.*]] = arith.constant 0 : i64 - //CHECK: %[[PREF_TILE_H:.*]] = arith.trunci %[[PREF_TILE_H64]] : i64 to i32 //CHECK: %[[PREF_LLVMPTR:.*]] = llvm.inttoptr %[[PREF_INTPTR]] : i64 to !llvm.ptr<1> - //CHECK: %[[PREF_SIZEOF_F32:.*]] = arith.constant 4 : i32 - //CHECK: %[[PREF_BASE_ROW_IN_BYTES:.*]] = arith.muli %[[PREF_BASE_W]], %[[PREF_SIZEOF_F32]] : i32 //CHECK: xevm.blockprefetch2d %[[PREF_LLVMPTR]], %[[PREF_BASE_ROW_IN_BYTES]], %[[PREF_BASE_H]], - //CHECK-SAME: %[[PREF_BASE_ROW_IN_BYTES]], %[[PREF_TILE_W]], %[[PREF_TILE_H]] + //CHECK-SAME: %[[PREF_BASE_ROW_IN_BYTES]], %[[OFFSET_ZERO]], %[[OFFSET_ZERO]] //CHECK-SAME: <{cache_control = #xevm.load_cache_control<L1c_L2uc_L3uc>, elem_size_in_bits = 32 : i32, //CHECK-SAME: tile_height = 8 : i32, tile_width = 16 : i32, v_blocks = 1 : i32}> //CHECK-SAME: : (!llvm.ptr<1>, i32, i32, i32, i32, i32) diff --git a/mlir/test/Target/LLVMIR/nvvm/membar.mlir b/mlir/test/Target/LLVMIR/nvvm/membar.mlir new file mode 100644 index 0000000..1b794f6 --- /dev/null +++ b/mlir/test/Target/LLVMIR/nvvm/membar.mlir @@ -0,0 +1,14 @@ +// RUN: mlir-translate -mlir-to-llvmir %s -split-input-file --verify-diagnostics | FileCheck %s + +// CHECK-lABEL: @memorybarrier() +llvm.func @memorybarrier() { + // CHECK: call void @llvm.nvvm.membar.cta() + nvvm.memory.barrier #nvvm.mem_scope<cta> + // CHECK: call void @llvm.nvvm.fence.sc.cluster() + nvvm.memory.barrier #nvvm.mem_scope<cluster> + // CHECK: call void @llvm.nvvm.membar.gl() + nvvm.memory.barrier #nvvm.mem_scope<gpu> + // CHECK: call void @llvm.nvvm.membar.sys() + nvvm.memory.barrier #nvvm.mem_scope<sys> + llvm.return +} diff --git a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel index 248e1e6..a4ea627 100644 --- a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel +++ b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel @@ -8,6 +8,7 @@ load("@bazel_skylib//rules:common_settings.bzl", "bool_flag") load("@bazel_skylib//rules:expand_template.bzl", "expand_template") load("@rules_cc//cc:defs.bzl", "cc_binary", "cc_library") +load("@rules_python//python:defs.bzl", "py_binary") load("//llvm:targets.bzl", "llvm_targets") load( ":build_defs.bzl", @@ -14428,3 +14429,8 @@ cc_library( "//llvm:Support", ], ) + +py_binary( + name = "generate-test-checks", + srcs = ["utils/generate-test-checks.py"], +) |