diff options
245 files changed, 10934 insertions, 1310 deletions
diff --git a/.github/workflows/containers/github-action-ci-windows/Dockerfile b/.github/workflows/containers/github-action-ci-windows/Dockerfile index 640d34d..9ddf501 100644 --- a/.github/workflows/containers/github-action-ci-windows/Dockerfile +++ b/.github/workflows/containers/github-action-ci-windows/Dockerfile @@ -90,7 +90,7 @@ RUN powershell -Command \ RUN git config --system core.longpaths true & \ git config --global core.autocrlf false -ARG RUNNER_VERSION=2.328.0 +ARG RUNNER_VERSION=2.329.0 ENV RUNNER_VERSION=$RUNNER_VERSION RUN powershell -Command \ diff --git a/.github/workflows/containers/github-action-ci/Dockerfile b/.github/workflows/containers/github-action-ci/Dockerfile index 4540d64..1b376dd 100644 --- a/.github/workflows/containers/github-action-ci/Dockerfile +++ b/.github/workflows/containers/github-action-ci/Dockerfile @@ -99,7 +99,7 @@ WORKDIR /home/gha FROM ci-container AS ci-container-agent -ENV GITHUB_RUNNER_VERSION=2.328.0 +ENV GITHUB_RUNNER_VERSION=2.329.0 RUN mkdir actions-runner && \ cd actions-runner && \ diff --git a/.github/workflows/pr-code-format.yml b/.github/workflows/pr-code-format.yml index 1e0dc70..2b85d8b 100644 --- a/.github/workflows/pr-code-format.yml +++ b/.github/workflows/pr-code-format.yml @@ -12,6 +12,8 @@ on: jobs: code_formatter: runs-on: ubuntu-24.04 + container: + image: 'ghcr.io/llvm/ci-ubuntu-24.04-format' timeout-minutes: 30 concurrency: group: ${{ github.workflow }}-${{ github.event.pull_request.number }} @@ -23,6 +25,14 @@ jobs: with: fetch-depth: 2 + # We need to set the repo checkout as safe, otherwise tj-actions/changed-files + # will fail due to the changed ownership inside the container. + # TODO(boomanaiden154): We should probably fix this by having the default user + # in the container have the same ID as the GHA user on the host. + - name: Set Safe Directory + run: | + chown -R root $(pwd) + - name: Get changed files id: changed-files uses: tj-actions/changed-files@ed68ef82c095e0d48ec87eccea555d944a631a4c # v46.0.5 @@ -39,24 +49,6 @@ jobs: echo "Formatting files:" echo "$CHANGED_FILES" - # The clang format version should always be upgraded to the first version - # of a release cycle (x.1.0) or the last version of a release cycle, or - # if there have been relevant clang-format backports. - - name: Install clang-format - uses: aminya/setup-cpp@a276e6e3d1db9160db5edc458e99a30d3b109949 # v1.7.1 - with: - clangformat: 21.1.0 - - - name: Setup Python env - uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0 - with: - python-version: '3.13' - cache: 'pip' - cache-dependency-path: 'llvm/utils/git/requirements_formatting.txt' - - - name: Install python dependencies - run: pip install -r llvm/utils/git/requirements_formatting.txt - - name: Run code formatter env: GITHUB_PR_NUMBER: ${{ github.event.pull_request.number }} diff --git a/bolt/docs/BinaryAnalysis.md b/bolt/docs/BinaryAnalysis.md index b13410c..07f096e 100644 --- a/bolt/docs/BinaryAnalysis.md +++ b/bolt/docs/BinaryAnalysis.md @@ -1,7 +1,7 @@ # BOLT-based binary analysis As part of post-link-time optimizing, BOLT needs to perform a range of analyses -on binaries such as recontructing control flow graphs, and more. +on binaries such as reconstructing control flow graphs, and more. The `llvm-bolt-binary-analysis` tool enables running requested binary analyses on binaries, and generating reports. It does this by building on top of the diff --git a/bolt/docs/CommandLineArgumentReference.md b/bolt/docs/CommandLineArgumentReference.md index 151399d..43cecee 100644 --- a/bolt/docs/CommandLineArgumentReference.md +++ b/bolt/docs/CommandLineArgumentReference.md @@ -375,7 +375,7 @@ - `--use-old-text` - Re-use space in old .text if possible (relocation mode) + Reuse space in old .text if possible (relocation mode) - `-v <uint>` diff --git a/bolt/docs/RuntimeLibrary.md b/bolt/docs/RuntimeLibrary.md index 58d9497..b969ebd 100644 --- a/bolt/docs/RuntimeLibrary.md +++ b/bolt/docs/RuntimeLibrary.md @@ -15,7 +15,7 @@ However, this approach quickly becomes awkward if we want to insert a lot of cod Currently, our runtime library is written in C++ and contains code that helps us instrument a binary. ### Limitations -Our library is not written with regular C++ code as it is not linked against any other libraries (this means we cannnot rely on anything defined on libstdc++, glibc, libgcc etc), but is self sufficient. In runtime/CMakeLists.txt, we can see it is built with -ffreestanding, which requires the compiler to avoid using a runtime library by itself. +Our library is not written with regular C++ code as it is not linked against any other libraries (this means we cannot rely on anything defined on libstdc++, glibc, libgcc etc), but is self sufficient. In runtime/CMakeLists.txt, we can see it is built with -ffreestanding, which requires the compiler to avoid using a runtime library by itself. While this requires us to make our own syscalls, it does simplify our linker a lot, which is very limited and can only do basic function name resolving. However, this is a big improvement in comparison with programmatically generating the code in assembly language using MCInsts. diff --git a/bolt/docs/doxygen.cfg.in b/bolt/docs/doxygen.cfg.in index 538285f..de8b1f7b 100644 --- a/bolt/docs/doxygen.cfg.in +++ b/bolt/docs/doxygen.cfg.in @@ -1070,7 +1070,7 @@ HTML_STYLESHEET = # defined cascading style sheet that is included after the standard style sheets # created by doxygen. Using this option one can overrule certain style aspects. # This is preferred over using HTML_STYLESHEET since it does not replace the -# standard style sheet and is therefor more robust against future updates. +# standard style sheet and is therefore more robust against future updates. # Doxygen will copy the style sheet file to the output directory. For an example # see the documentation. # This tag requires that the tag GENERATE_HTML is set to YES. diff --git a/bolt/include/bolt/Core/BinaryFunction.h b/bolt/include/bolt/Core/BinaryFunction.h index f5e9887..7b10b2d 100644 --- a/bolt/include/bolt/Core/BinaryFunction.h +++ b/bolt/include/bolt/Core/BinaryFunction.h @@ -1336,7 +1336,7 @@ public: ColdCodeSectionName = Name.str(); } - /// Return true iif the function will halt execution on entry. + /// Return true if the function will halt execution on entry. bool trapsOnEntry() const { return TrapsOnEntry; } /// Make the function always trap on entry. Other than the trap instruction, diff --git a/bolt/include/bolt/Core/DIEBuilder.h b/bolt/include/bolt/Core/DIEBuilder.h index 4c3c277..95e958f 100644 --- a/bolt/include/bolt/Core/DIEBuilder.h +++ b/bolt/include/bolt/Core/DIEBuilder.h @@ -60,7 +60,7 @@ public: uint32_t UnitLength = 0; bool IsConstructed = false; // A map of DIE offsets in original DWARF section to DIE ID. - // Whih is used to access DieInfoVector. + // Which is used to access DieInfoVector. std::unordered_map<uint64_t, uint32_t> DIEIDMap; // Some STL implementations don't have a noexcept move constructor for diff --git a/bolt/include/bolt/Core/DebugData.h b/bolt/include/bolt/Core/DebugData.h index 8149789..7c8ea12ee 100644 --- a/bolt/include/bolt/Core/DebugData.h +++ b/bolt/include/bolt/Core/DebugData.h @@ -326,8 +326,8 @@ public: /// Write out entries in to .debug_addr section for CUs. virtual std::optional<uint64_t> finalize(const size_t BufferSize); - /// Return buffer with all the entries in .debug_addr already writen out using - /// update(...). + /// Return buffer with all the entries in .debug_addr already written out + /// using update(...). virtual std::unique_ptr<AddressSectionBuffer> releaseBuffer() { return std::move(Buffer); } @@ -409,7 +409,7 @@ protected: std::mutex WriterMutex; std::unique_ptr<AddressSectionBuffer> Buffer; std::unique_ptr<raw_svector_ostream> AddressStream; - /// Used to track sections that were not modified so that they can be re-used. + /// Used to track sections that were not modified so that they can be reused. static DenseMap<uint64_t, uint64_t> UnmodifiedAddressOffsets; }; diff --git a/bolt/include/bolt/Core/DebugNames.h b/bolt/include/bolt/Core/DebugNames.h index cc4e13a..4ec49ca 100644 --- a/bolt/include/bolt/Core/DebugNames.h +++ b/bolt/include/bolt/Core/DebugNames.h @@ -65,7 +65,7 @@ public: void setCurrentUnit(DWARFUnit &Unit, const uint64_t UnitStartOffset); /// Emit Accelerator table. void emitAccelTable(); - /// Returns true if the table was crated. + /// Returns true if the table was created. bool isCreated() const { return NeedToCreate; } /// Returns buffer containing the accelerator table. std::unique_ptr<DebugBufferVector> releaseBuffer() { @@ -91,7 +91,7 @@ private: uint64_t CurrentUnitOffset = 0; const DWARFUnit *CurrentUnit = nullptr; std::unordered_map<uint32_t, uint32_t> AbbrevTagToIndexMap; - /// Contains a map of TU hashes to a Foreign TU indecies. + /// Contains a map of TU hashes to a Foreign TU indices. /// This is used to reduce the size of Foreign TU list since there could be /// multiple TUs with the same hash. DenseMap<uint64_t, uint32_t> TUHashToIndexMap; diff --git a/bolt/include/bolt/Core/MCPlusBuilder.h b/bolt/include/bolt/Core/MCPlusBuilder.h index 2772de7..d666c10 100644 --- a/bolt/include/bolt/Core/MCPlusBuilder.h +++ b/bolt/include/bolt/Core/MCPlusBuilder.h @@ -432,7 +432,7 @@ public: return Analysis->isConditionalBranch(Inst); } - /// Returns true if Inst is a condtional move instruction + /// Returns true if Inst is a conditional move instruction virtual bool isConditionalMove(const MCInst &Inst) const { llvm_unreachable("not implemented"); return false; @@ -1564,7 +1564,7 @@ public: } /// Get the default def_in and live_out registers for the function - /// Currently only used for the Stoke optimzation + /// Currently only used for the Stoke optimization virtual void getDefaultDefIn(BitVector &Regs) const { llvm_unreachable("not implemented"); } diff --git a/bolt/include/bolt/Passes/FrameAnalysis.h b/bolt/include/bolt/Passes/FrameAnalysis.h index d71c338..5ce85be 100644 --- a/bolt/include/bolt/Passes/FrameAnalysis.h +++ b/bolt/include/bolt/Passes/FrameAnalysis.h @@ -37,7 +37,7 @@ struct FrameIndexEntry { int64_t StackOffset; uint8_t Size; - /// If this is false, we will never atempt to remove or optimize this + /// If this is false, we will never attempt to remove or optimize this /// instruction. We just use it to keep track of stores we don't fully /// understand but we know it may write to a frame position. bool IsSimple; diff --git a/bolt/include/bolt/Passes/LongJmp.h b/bolt/include/bolt/Passes/LongJmp.h index df3ea96..84da453 100644 --- a/bolt/include/bolt/Passes/LongJmp.h +++ b/bolt/include/bolt/Passes/LongJmp.h @@ -30,7 +30,7 @@ namespace bolt { /// 64-bit range, we guarantee it can reach any code location. /// class LongJmpPass : public BinaryFunctionPass { - /// Used to implement stub grouping (re-using a stub from one function into + /// Used to implement stub grouping (reusing a stub from one function into /// another) using StubTy = std::pair<uint64_t, BinaryBasicBlock *>; using StubGroupTy = SmallVector<StubTy, 4>; diff --git a/bolt/include/bolt/Passes/ProfileQualityStats.h b/bolt/include/bolt/Passes/ProfileQualityStats.h index 86fc88c..ee74b12 100644 --- a/bolt/include/bolt/Passes/ProfileQualityStats.h +++ b/bolt/include/bolt/Passes/ProfileQualityStats.h @@ -49,7 +49,7 @@ // aggregates the block gaps into 2 values for the function: "weighted" is the // weighted average of the block conservation gaps, where the weights depend on // each block's execution count and instruction count; "worst" is the worst -// (biggest) block gap acorss all basic blocks in the function with an execution +// (biggest) block gap across all basic blocks in the function with an execution // count of > 500. The pass then reports the 95th percentile of the weighted and // worst values of the 1000 functions in a single BOLT-INFO line. The smaller // the reported values are, the better the BOLT profile satisfies the function diff --git a/bolt/include/bolt/Passes/ReorderAlgorithm.h b/bolt/include/bolt/Passes/ReorderAlgorithm.h index 95d9e83..42bb333 100644 --- a/bolt/include/bolt/Passes/ReorderAlgorithm.h +++ b/bolt/include/bolt/Passes/ReorderAlgorithm.h @@ -26,7 +26,7 @@ namespace bolt { /// Objects of this class implement various basic block clustering algorithms. /// Basic block clusters are chains of basic blocks that should be laid out -/// in this order to maximize performace. These algorithms group basic blocks +/// in this order to maximize performance. These algorithms group basic blocks /// into clusters using execution profile data and various heuristics. class ClusterAlgorithm { public: diff --git a/bolt/include/bolt/Rewrite/DWARFRewriter.h b/bolt/include/bolt/Rewrite/DWARFRewriter.h index 6242456..cab346b 100644 --- a/bolt/include/bolt/Rewrite/DWARFRewriter.h +++ b/bolt/include/bolt/Rewrite/DWARFRewriter.h @@ -128,7 +128,7 @@ private: CUOffsetMap finalizeTypeSections(DIEBuilder &DIEBlder, DIEStreamer &Streamer, GDBIndex &GDBIndexSection); - /// Process and write out CUs that are passsed in. + /// Process and write out CUs that are passed in. void finalizeCompileUnits(DIEBuilder &DIEBlder, DIEStreamer &Streamer, CUOffsetMap &CUMap, const std::list<DWARFUnit *> &CUs, diff --git a/bolt/lib/Core/BinaryContext.cpp b/bolt/lib/Core/BinaryContext.cpp index 206d8eef..46b3372 100644 --- a/bolt/lib/Core/BinaryContext.cpp +++ b/bolt/lib/Core/BinaryContext.cpp @@ -844,7 +844,7 @@ BinaryContext::getOrCreateJumpTable(BinaryFunction &Function, uint64_t Address, auto isSibling = std::bind(&BinaryContext::areRelatedFragments, this, &Function, std::placeholders::_1); assert(llvm::all_of(JT->Parents, isSibling) && - "cannot re-use jump table of a different function"); + "cannot reuse jump table of a different function"); (void)isSibling; if (opts::Verbosity > 2) { this->outs() << "BOLT-INFO: multiple fragments access the same jump table" @@ -860,7 +860,7 @@ BinaryContext::getOrCreateJumpTable(BinaryFunction &Function, uint64_t Address, return JT->getFirstLabel(); } - // Re-use the existing symbol if possible. + // Reuse the existing symbol if possible. MCSymbol *JTLabel = nullptr; if (BinaryData *Object = getBinaryDataAtAddress(Address)) { if (!isInternalSymbolName(Object->getSymbol()->getName())) diff --git a/bolt/lib/Core/BinaryFunction.cpp b/bolt/lib/Core/BinaryFunction.cpp index 9687892..4dfd4ba 100644 --- a/bolt/lib/Core/BinaryFunction.cpp +++ b/bolt/lib/Core/BinaryFunction.cpp @@ -3875,7 +3875,7 @@ uint64_t BinaryFunction::getEntryIDForSymbol(const MCSymbol *Symbol) const { if (FunctionSymbol == Symbol) return 0; - // Check all secondary entries available as either basic blocks or lables. + // Check all secondary entries available as either basic blocks or labels. uint64_t NumEntries = 1; for (const BinaryBasicBlock *BB : BasicBlocks) { MCSymbol *EntrySymbol = getSecondaryEntryPointSymbol(*BB); diff --git a/bolt/lib/Core/BinaryFunctionCallGraph.cpp b/bolt/lib/Core/BinaryFunctionCallGraph.cpp index f0c46a8..af22419 100644 --- a/bolt/lib/Core/BinaryFunctionCallGraph.cpp +++ b/bolt/lib/Core/BinaryFunctionCallGraph.cpp @@ -122,7 +122,7 @@ buildCallGraph(BinaryContext &BC, CgFilterFunction Filter, bool CgFromPerfData, // create a node for a function unless it was the target of a call from // a hot block. The alternative would be to set the count to one or // accumulate the number of calls from the callsite into the function - // samples. Results from perfomance testing seem to favor the zero + // samples. Results from performance testing seem to favor the zero // count though, so I'm leaving it this way for now. return Cg.addNode(Function, Size, Function->getKnownExecutionCount()); } diff --git a/bolt/lib/Core/DIEBuilder.cpp b/bolt/lib/Core/DIEBuilder.cpp index 7ce55f9..5b628f6 100644 --- a/bolt/lib/Core/DIEBuilder.cpp +++ b/bolt/lib/Core/DIEBuilder.cpp @@ -137,7 +137,7 @@ void DIEBuilder::updateReferences() { DIEInteger(NewAddr)); } - // Handling referenes in location expressions. + // Handling references in location expressions. for (LocWithReference &LocExpr : getState().LocWithReferencesToProcess) { SmallVector<uint8_t, 32> Buffer; DataExtractor Data(StringRef((const char *)LocExpr.BlockData.data(), @@ -336,7 +336,7 @@ void DIEBuilder::buildCompileUnits(const bool Init) { registerUnit(*DU, false); } - // Using DULIst since it can be modified by cross CU refrence resolution. + // Using DULIst since it can be modified by cross CU reference resolution. for (DWARFUnit *DU : getState().DUList) { if (DU->isTypeUnit()) continue; @@ -508,7 +508,7 @@ void DIEBuilder::finish() { UnitStartOffset += CurUnitInfo.UnitLength; }; // Computing offsets for .debug_types section. - // It's processed first when CU is registered so will be at the begginnig of + // It's processed first when CU is registered so will be at the beginning of // the vector. uint64_t TypeUnitStartOffset = 0; for (DWARFUnit *CU : getState().DUList) { diff --git a/bolt/lib/Core/DebugData.cpp b/bolt/lib/Core/DebugData.cpp index e05f28f..24a4c85 100644 --- a/bolt/lib/Core/DebugData.cpp +++ b/bolt/lib/Core/DebugData.cpp @@ -876,7 +876,7 @@ void DebugStrOffsetsWriter::finalizeSection(DWARFUnit &Unit, DIEValue StrListBaseAttrInfo = Die.findAttribute(dwarf::DW_AT_str_offsets_base); auto RetVal = ProcessedBaseOffsets.find(*Val); - // Handling re-use of str-offsets section. + // Handling reuse of str-offsets section. if (RetVal == ProcessedBaseOffsets.end() || StrOffsetSectionWasModified) { initialize(Unit); // Update String Offsets that were modified. @@ -1167,7 +1167,7 @@ void DwarfLineTable::emitCU(MCStreamer *MCOS, MCDwarfLineTableParams Params, // For functions that we do not modify we output them as raw data. // Re-constructing .debug_line_str so that offsets are correct for those // debug line tables. -// Bonus is that when we output a final binary we can re-use .debug_line_str +// Bonus is that when we output a final binary we can reuse .debug_line_str // section. So we don't have to do the SHF_ALLOC trick we did with // .debug_line. static void parseAndPopulateDebugLineStr(BinarySection &LineStrSection, diff --git a/bolt/lib/Core/DebugNames.cpp b/bolt/lib/Core/DebugNames.cpp index a9d98a6..6be2c5a 100644 --- a/bolt/lib/Core/DebugNames.cpp +++ b/bolt/lib/Core/DebugNames.cpp @@ -55,7 +55,7 @@ DWARF5AcceleratorTable::DWARF5AcceleratorTable( llvm::hash_value(llvm::StringRef(CStr)), StrOffset); if (!R.second) BC.errs() - << "BOLT-WARNING: [internal-dwarf-error]: collision occured on " + << "BOLT-WARNING: [internal-dwarf-error]: collision occurred on " << CStr << " at offset : 0x" << Twine::utohexstr(StrOffset) << ". Previous string offset is: 0x" << Twine::utohexstr(R.first->second) << ".\n"; @@ -86,7 +86,7 @@ void DWARF5AcceleratorTable::addUnit(DWARFUnit &Unit, if (Unit.isTypeUnit()) { if (DWOID) { // We adding an entry for a DWO TU. The DWO CU might not have any entries, - // so need to add it to the list pre-emptively. + // so need to add it to the list preemptively. auto Iter = CUOffsetsToPatch.insert({*DWOID, CUList.size()}); if (Iter.second) CUList.push_back(BADCUOFFSET); diff --git a/bolt/lib/Passes/Aligner.cpp b/bolt/lib/Passes/Aligner.cpp index c3ddeda..5d21bdb 100644 --- a/bolt/lib/Passes/Aligner.cpp +++ b/bolt/lib/Passes/Aligner.cpp @@ -60,7 +60,7 @@ namespace llvm { namespace bolt { // Align function to the specified byte-boundary (typically, 64) offsetting -// the fuction by not more than the corresponding value +// the function by not more than the corresponding value static void alignMaxBytes(BinaryFunction &Function) { Function.setAlignment(opts::AlignFunctions); Function.setMaxAlignmentBytes(opts::AlignFunctionsMaxBytes); @@ -68,7 +68,7 @@ static void alignMaxBytes(BinaryFunction &Function) { } // Align function to the specified byte-boundary (typically, 64) offsetting -// the fuction by not more than the minimum over +// the function by not more than the minimum over // -- the size of the function // -- the specified number of bytes static void alignCompact(BinaryFunction &Function, diff --git a/bolt/lib/Passes/RegReAssign.cpp b/bolt/lib/Passes/RegReAssign.cpp index 60349f1..0859cd2 100644 --- a/bolt/lib/Passes/RegReAssign.cpp +++ b/bolt/lib/Passes/RegReAssign.cpp @@ -145,7 +145,7 @@ void RegReAssign::rankRegisters(BinaryFunction &Function) { const bool CannotUseREX = BC.MIB->cannotUseREX(Inst); const MCInstrDesc &Desc = BC.MII->get(Inst.getOpcode()); - // Disallow substituitions involving regs in implicit uses lists + // Disallow substitutions involving regs in implicit uses lists for (MCPhysReg ImplicitUse : Desc.implicit_uses()) { const size_t RegEC = BC.MIB->getAliases(ImplicitUse, false).find_first(); @@ -153,7 +153,7 @@ void RegReAssign::rankRegisters(BinaryFunction &Function) { std::numeric_limits<decltype(RegScore)::value_type>::min(); } - // Disallow substituitions involving regs in implicit defs lists + // Disallow substitutions involving regs in implicit defs lists for (MCPhysReg ImplicitDef : Desc.implicit_defs()) { const size_t RegEC = BC.MIB->getAliases(ImplicitDef, false).find_first(); @@ -174,7 +174,7 @@ void RegReAssign::rankRegisters(BinaryFunction &Function) { if (RegEC == 0) continue; - // Disallow substituitions involving regs in instrs that cannot use REX + // Disallow substitutions involving regs in instrs that cannot use REX // The relationship of X86 registers is shown in the diagram. BL and BH // do not have a direct alias relationship. However, if the BH register // cannot be swapped, then the BX/EBX/RBX registers cannot be swapped as diff --git a/bolt/lib/Passes/SplitFunctions.cpp b/bolt/lib/Passes/SplitFunctions.cpp index eab669b..66a373a 100644 --- a/bolt/lib/Passes/SplitFunctions.cpp +++ b/bolt/lib/Passes/SplitFunctions.cpp @@ -386,7 +386,7 @@ private: } /// Compute sum of scores over jumps within \p BlockOrder given \p SplitIndex. - /// Increament Score.LocalScore in place by the sum. + /// Increment Score.LocalScore in place by the sum. void computeJumpScore(const BasicBlockOrder &BlockOrder, const size_t SplitIndex, SplitScore &Score) { @@ -413,7 +413,7 @@ private: } /// Compute sum of scores over calls originated in the current function - /// given \p SplitIndex. Increament Score.LocalScore in place by the sum. + /// given \p SplitIndex. Increment Score.LocalScore in place by the sum. void computeLocalCallScore(const BasicBlockOrder &BlockOrder, const size_t SplitIndex, SplitScore &Score) { if (opts::CallScale == 0) @@ -455,7 +455,7 @@ private: } /// Compute sum of splitting scores for cover calls of the input function. - /// Increament Score.CoverCallScore in place by the sum. + /// Increment Score.CoverCallScore in place by the sum. void computeCoverCallScore(const BasicBlockOrder &BlockOrder, const size_t SplitIndex, const std::vector<CallInfo> &CoverCalls, @@ -467,7 +467,7 @@ private: assert(CI.Length >= Score.HotSizeReduction && "Length of cover calls must exceed reduced size of hot fragment."); // Compute the new length of the call, which is shorter than the original - // one by the size of the splitted fragment minus the total size increase. + // one by the size of the split fragment minus the total size increase. const size_t NewCallLength = CI.Length - Score.HotSizeReduction; Score.CoverCallScore += computeCallScore(CI.Count, NewCallLength); } @@ -502,12 +502,12 @@ private: // First part of LocalScore is the sum over call edges originated in the // input function. These edges can get shorter or longer depending on - // SplitIndex. Score.LocalScore is increamented in place. + // SplitIndex. Score.LocalScore is incremented in place. computeLocalCallScore(BlockOrder, SplitIndex, Score); // Second part of LocalScore is the sum over jump edges with src basic block // and dst basic block in the current function. Score.LocalScore is - // increamented in place. + // incremented in place. computeJumpScore(BlockOrder, SplitIndex, Score); // Compute CoverCallScore and store in Score in place. diff --git a/bolt/lib/Profile/DataReader.cpp b/bolt/lib/Profile/DataReader.cpp index afe2421..277d4bb 100644 --- a/bolt/lib/Profile/DataReader.cpp +++ b/bolt/lib/Profile/DataReader.cpp @@ -907,7 +907,7 @@ ErrorOr<uint64_t> DataReader::parseHexField(char EndChar, bool EndNl) { StringRef NumStr = NumStrRes.get(); uint64_t Num; if (NumStr.getAsInteger(16, Num)) { - reportError("expected hexidecimal number"); + reportError("expected hexadecimal number"); Diag << "Found: " << NumStr << "\n"; return make_error_code(llvm::errc::io_error); } diff --git a/bolt/lib/Rewrite/BuildIDRewriter.cpp b/bolt/lib/Rewrite/BuildIDRewriter.cpp index d50416f..706a3d0 100644 --- a/bolt/lib/Rewrite/BuildIDRewriter.cpp +++ b/bolt/lib/Rewrite/BuildIDRewriter.cpp @@ -48,7 +48,7 @@ public: }; Error BuildIDRewriter::sectionInitializer() { - // Typically, build ID will reside in .note.gnu.build-id section. Howerver, + // Typically, build ID will reside in .note.gnu.build-id section. However, // a linker script can change the section name and such is the case with // the Linux kernel. Hence, we iterate over all note sections. for (BinarySection &NoteSection : BC.sections()) { diff --git a/bolt/lib/Rewrite/DWARFRewriter.cpp b/bolt/lib/Rewrite/DWARFRewriter.cpp index 7366d2a..a7a4b66 100644 --- a/bolt/lib/Rewrite/DWARFRewriter.cpp +++ b/bolt/lib/Rewrite/DWARFRewriter.cpp @@ -1723,7 +1723,7 @@ StringRef getSectionName(const SectionRef &Section) { return Name; } -// Exctracts an appropriate slice if input is DWP. +// Extracts an appropriate slice if input is DWP. // Applies patches or overwrites the section. std::optional<StringRef> updateDebugData( DWARFContext &DWCtx, StringRef SectionName, StringRef SectionContents, @@ -1759,7 +1759,7 @@ std::optional<StringRef> updateDebugData( auto Iter = OverridenSections.find(Kind); if (Iter == OverridenSections.end()) { errs() - << "BOLT-WARNING: [internal-dwarf-error]: Could not find overriden " + << "BOLT-WARNING: [internal-dwarf-error]: Could not find overridden " "section for: " << Twine::utohexstr(DWOId) << ".\n"; return std::nullopt; @@ -1991,7 +1991,7 @@ void DWARFRewriter::convertToRangesPatchDebugInfo( } } - // HighPC was conveted into DW_AT_ranges. + // HighPC was converted into DW_AT_ranges. // For DWARF5 we only access ranges through index. DIEBldr.replaceValue(&Die, HighPCAttrInfo.getAttribute(), dwarf::DW_AT_ranges, diff --git a/bolt/lib/Rewrite/PseudoProbeRewriter.cpp b/bolt/lib/Rewrite/PseudoProbeRewriter.cpp index ee021fe..947d899 100644 --- a/bolt/lib/Rewrite/PseudoProbeRewriter.cpp +++ b/bolt/lib/Rewrite/PseudoProbeRewriter.cpp @@ -308,7 +308,7 @@ void PseudoProbeRewriter::encodePseudoProbes() { Contents.append(OSE.str().begin(), OSE.str().end()); }; - // Emit indiviual pseudo probes in a inline tree node + // Emit individual pseudo probes in a inline tree node // Probe index, type, attribute, address type and address are encoded // Address of the first probe is absolute. // Other probes' address are represented by delta diff --git a/bolt/lib/Rewrite/RewriteInstance.cpp b/bolt/lib/Rewrite/RewriteInstance.cpp index a0e7995..9580163 100644 --- a/bolt/lib/Rewrite/RewriteInstance.cpp +++ b/bolt/lib/Rewrite/RewriteInstance.cpp @@ -1087,7 +1087,7 @@ void RewriteInstance::discoverFileObjects() { if (SymbolAddress == Section->getAddress() + Section->getSize()) { assert(SymbolSize == 0 && - "unexpect non-zero sized symbol at end of section"); + "unexpected non-zero sized symbol at end of section"); LLVM_DEBUG( dbgs() << "BOLT-DEBUG: rejecting as symbol points to end of its section\n"); @@ -2440,7 +2440,7 @@ void RewriteInstance::processDynamicRelocations() { } // The rest of dynamic relocations - DT_RELA. - // The static executable might have .rela.dyn secion and not have PT_DYNAMIC + // The static executable might have .rela.dyn section and not have PT_DYNAMIC if (!DynamicRelocationsSize && BC->IsStaticExecutable) { ErrorOr<BinarySection &> DynamicRelSectionOrErr = BC->getUniqueSectionByName(getRelaDynSectionName()); @@ -5017,7 +5017,7 @@ void RewriteInstance::updateELFSymbolTable( if (!Section) return false; - // Remove the section symbol iif the corresponding section was stripped. + // Remove the section symbol if the corresponding section was stripped. if (Symbol.getType() == ELF::STT_SECTION) { if (!getNewSectionIndex(Symbol.st_shndx)) return true; diff --git a/bolt/lib/Target/AArch64/AArch64MCPlusBuilder.cpp b/bolt/lib/Target/AArch64/AArch64MCPlusBuilder.cpp index df4f421..6954cb2 100644 --- a/bolt/lib/Target/AArch64/AArch64MCPlusBuilder.cpp +++ b/bolt/lib/Target/AArch64/AArch64MCPlusBuilder.cpp @@ -1296,7 +1296,7 @@ public: AArch64_AM::ShiftExtendType ExtendType = AArch64_AM::getArithExtendType(OperandExtension); if (ShiftVal != 2) { - // TODO: Handle the patten where ShiftVal != 2. + // TODO: Handle the pattern where ShiftVal != 2. // The following code sequence below has no shift amount, // the range could be 0 to 4. // The pattern comes from libc, it occurs when the binary is static. @@ -1626,7 +1626,7 @@ public: int getUncondBranchEncodingSize() const override { return 28; } // This helper function creates the snippet of code that compares a register - // RegNo with an immedaite Imm, and jumps to Target if they are equal. + // RegNo with an immediate Imm, and jumps to Target if they are equal. // cmp RegNo, #Imm // b.eq Target // where cmp is an alias for subs, which results in the code below: @@ -1648,7 +1648,7 @@ public: } // This helper function creates the snippet of code that compares a register - // RegNo with an immedaite Imm, and jumps to Target if they are not equal. + // RegNo with an immediate Imm, and jumps to Target if they are not equal. // cmp RegNo, #Imm // b.ne Target // where cmp is an alias for subs, which results in the code below: diff --git a/bolt/lib/Target/X86/X86MCPlusBuilder.cpp b/bolt/lib/Target/X86/X86MCPlusBuilder.cpp index 9026a9d..5fca5e8 100644 --- a/bolt/lib/Target/X86/X86MCPlusBuilder.cpp +++ b/bolt/lib/Target/X86/X86MCPlusBuilder.cpp @@ -2715,7 +2715,7 @@ public: bool FoundOne = false; - // Iterate only through src operands that arent also dest operands + // Iterate only through src operands that aren't also dest operands for (unsigned Index = InstDesc.getNumDefs() + (HasLHS ? 1 : 0), E = InstDesc.getNumOperands(); Index != E; ++Index) { diff --git a/bolt/lib/Utils/CommandLineOpts.cpp b/bolt/lib/Utils/CommandLineOpts.cpp index 095612a..5be04d2 100644 --- a/bolt/lib/Utils/CommandLineOpts.cpp +++ b/bolt/lib/Utils/CommandLineOpts.cpp @@ -285,7 +285,7 @@ cl::opt<bool> TimeRewrite("time-rewrite", cl::opt<bool> UseOldText( "use-old-text", - cl::desc("re-use space in old .text if possible (relocation mode)"), + cl::desc("reuse space in old .text if possible (relocation mode)"), cl::cat(BoltCategory)); cl::opt<bool> UpdateDebugSections( diff --git a/bolt/runtime/hugify.cpp b/bolt/runtime/hugify.cpp index 672b042..de89630 100644 --- a/bolt/runtime/hugify.cpp +++ b/bolt/runtime/hugify.cpp @@ -24,7 +24,7 @@ {} #endif -// Function constains trampoline to _start, +// Function constrains trampoline to _start, // so we can resume regular execution of the function that we hooked. extern void __bolt_hugify_start_program(); diff --git a/bolt/runtime/instr.cpp b/bolt/runtime/instr.cpp index 1f54a50..f586db2 100644 --- a/bolt/runtime/instr.cpp +++ b/bolt/runtime/instr.cpp @@ -214,7 +214,7 @@ private: /// __bolt_instr_setup, our initialization routine. BumpPtrAllocator *GlobalAlloc; -// Base address which we substract from recorded PC values when searching for +// Base address which we subtract from recorded PC values when searching for // indirect call description entries. Needed because indCall descriptions are // mapped read-only and contain static addresses. Initialized in // __bolt_instr_setup. @@ -261,7 +261,7 @@ struct SimpleHashTableEntryBase { // Currently we have to do it the ugly way because // we want every message to be printed atomically via a single call to // __write. If we use reportNumber() and others nultiple times, we'll get - // garbage in mulithreaded environment + // garbage in multithreaded environment char Buf[BufSize]; char *Ptr = Buf; Ptr = intToStr(Ptr, __getpid(), 10); @@ -1585,7 +1585,7 @@ __bolt_instr_data_dump(int FD, const char *LibPath = nullptr, /// at user-specified intervals void watchProcess() { timespec ts, rem; - uint64_t Ellapsed = 0ull; + uint64_t Elapsed = 0ull; int FD = openProfile(); uint64_t ppid; if (__bolt_instr_wait_forks) { @@ -1615,10 +1615,10 @@ void watchProcess() { break; } - if (++Ellapsed < __bolt_instr_sleep_time) + if (++Elapsed < __bolt_instr_sleep_time) continue; - Ellapsed = 0; + Elapsed = 0; __bolt_instr_data_dump(FD); if (__bolt_instr_no_counters_clear == false) __bolt_instr_clear_counters(); diff --git a/bolt/runtime/sys_aarch64.h b/bolt/runtime/sys_aarch64.h index 77c9cfc..b1d04f9 100644 --- a/bolt/runtime/sys_aarch64.h +++ b/bolt/runtime/sys_aarch64.h @@ -41,7 +41,7 @@ // Anonymous namespace covering everything but our library entry point namespace { -// Get the difference between runtime addrress of .text section and +// Get the difference between runtime address of .text section and // static address in section header table. Can be extracted from arbitrary // pc value recorded at runtime to get the corresponding static address, which // in turn can be used to search for indirect call description. Needed because diff --git a/bolt/runtime/sys_riscv64.h b/bolt/runtime/sys_riscv64.h index 00a21e4..442fa2e0 100644 --- a/bolt/runtime/sys_riscv64.h +++ b/bolt/runtime/sys_riscv64.h @@ -105,7 +105,7 @@ // Anonymous namespace covering everything but our library entry point namespace { -// Get the difference between runtime addrress of .text section and +// Get the difference between runtime address of .text section and // static address in section header table. Can be extracted from arbitrary // pc value recorded at runtime to get the corresponding static address, which // in turn can be used to search for indirect call description. Needed because diff --git a/bolt/runtime/sys_x86_64.h b/bolt/runtime/sys_x86_64.h index ca2c693..933e939 100644 --- a/bolt/runtime/sys_x86_64.h +++ b/bolt/runtime/sys_x86_64.h @@ -40,7 +40,7 @@ namespace { -// Get the difference between runtime addrress of .text section and +// Get the difference between runtime address of .text section and // static address in section header table. Can be extracted from arbitrary // pc value recorded at runtime to get the corresponding static address, which // in turn can be used to search for indirect call description. Needed because @@ -171,8 +171,9 @@ uint64_t __exit(uint64_t code) { #if !defined(__APPLE__) // We use a stack-allocated buffer for string manipulation in many pieces of // this code, including the code that prints each line of the fdata file. This -// buffer needs to accomodate large function names, but shouldn't be arbitrarily -// large (dynamically allocated) for simplicity of our memory space usage. +// buffer needs to accommodate large function names, but shouldn't be +// arbitrarily large (dynamically allocated) for simplicity of our memory space +// usage. // Declare some syscall wrappers we use throughout this code to avoid linking // against system libc. diff --git a/bolt/test/AArch64/constant-island-alignment.s b/bolt/test/AArch64/constant-island-alignment.s index 957c470..99fe733 100644 --- a/bolt/test/AArch64/constant-island-alignment.s +++ b/bolt/test/AArch64/constant-island-alignment.s @@ -3,7 +3,7 @@ # RUN: split-file %s %t // For the first test case, in case the nop before .Lci will be removed -// the pointer to exit function won't be alinged and the test will fail. +// the pointer to exit function won't be aligned and the test will fail. # RUN: llvm-mc -filetype=obj -triple aarch64-unknown-unknown \ # RUN: %t/xword_align.s -o %t_xa.o diff --git a/bolt/test/AArch64/ifunc.test b/bolt/test/AArch64/ifunc.test index 3da42c6..15ecc35 100644 --- a/bolt/test/AArch64/ifunc.test +++ b/bolt/test/AArch64/ifunc.test @@ -9,7 +9,7 @@ // RUN: FileCheck --check-prefix=REL_CHECK %s // Non-pie static executable doesn't generate PT_DYNAMIC, check relocation -// is readed successfully and IPLT trampoline has been identified by bolt. +// is read successfully and IPLT trampoline has been identified by bolt. // RUN: %clang %cflags -nostdlib -O3 %p/../Inputs/ifunc.c -fuse-ld=lld -no-pie \ // RUN: -o %t.O3_nopie.exe -Wl,-q // RUN: llvm-readelf -l %t.O3_nopie.exe | \ diff --git a/bolt/test/X86/cdsplit-call-scale.s b/bolt/test/X86/cdsplit-call-scale.s index 66f3003..caa11b6 100644 --- a/bolt/test/X86/cdsplit-call-scale.s +++ b/bolt/test/X86/cdsplit-call-scale.s @@ -1,8 +1,8 @@ ## Test the control of aggressiveness of 3-way splitting by -call-scale. -## When -call-scale=0.0, the tested function is 2-way splitted. -## When -call-scale=1.0, the tested function is 3-way splitted with 5 blocks +## When -call-scale=0.0, the tested function is 2-way split. +## When -call-scale=1.0, the tested function is 3-way split with 5 blocks ## in warm because of the increased benefit of shortening the call edges. -## When -call-scale=1000.0, the tested function is still 3-way splitted with +## When -call-scale=1000.0, the tested function is still 3-way split with ## 5 blocks in warm because cdsplit does not allow hot-warm splitting to break ## a fall through branch from a basic block to its most likely successor. diff --git a/bolt/test/X86/dwarf5-two-cu-str-offset-table.test b/bolt/test/X86/dwarf5-two-cu-str-offset-table.test index e59664e..488635b 100644 --- a/bolt/test/X86/dwarf5-two-cu-str-offset-table.test +++ b/bolt/test/X86/dwarf5-two-cu-str-offset-table.test @@ -8,7 +8,7 @@ # RUN: llvm-dwarfdump --show-form --verbose --debug-str-offsets %t.bolt >> %t.txt # RUN: cat %t.txt | FileCheck --check-prefix=CHECK %s -## This test checks we correclty re-renerate .debug_str_offsets. +## This test checks we correctly re-renerate .debug_str_offsets. # CHECK: .debug_str_offsets contents # CHECK-NEXT: 0x00000000: Contribution size = 52, Format = DWARF32, Version = 5 diff --git a/bolt/test/X86/dwarf5-type-unit-no-cu-str-offset-table.test b/bolt/test/X86/dwarf5-type-unit-no-cu-str-offset-table.test index dc6255f..0cb62ed 100644 --- a/bolt/test/X86/dwarf5-type-unit-no-cu-str-offset-table.test +++ b/bolt/test/X86/dwarf5-type-unit-no-cu-str-offset-table.test @@ -7,7 +7,7 @@ # RUN: llvm-dwarfdump --show-form --verbose --debug-str-offsets %t.exe | FileCheck -check-prefix=PRE-BOLT %s # RUN: llvm-dwarfdump --show-form --verbose --debug-str-offsets %t.bolt | FileCheck -check-prefix=POST-BOLT %s -## This test checks we correclty re-generate .debug_str_offsets when there are type units that have an offset not shared with CU. +## This test checks we correctly re-generate .debug_str_offsets when there are type units that have an offset not shared with CU. # PRE-BOLT: .debug_str_offsets contents # PRE-BOLT-NEXT: Contribution size = 24, Format = DWARF32, Version = 5 diff --git a/bolt/test/X86/jt-symbol-disambiguation-3.s b/bolt/test/X86/jt-symbol-disambiguation-3.s index 22b34ce..c06fd3b 100644 --- a/bolt/test/X86/jt-symbol-disambiguation-3.s +++ b/bolt/test/X86/jt-symbol-disambiguation-3.s @@ -1,6 +1,6 @@ ## In this test case, we reproduce the behavior seen in gcc where the ## base address of a jump table is decremented by some number and ends up -## at the exact addess of a jump table from another function. After +## at the exact address of a jump table from another function. After ## linking, the instruction references another jump table and that ## confuses BOLT. ## We repro here the following issue: @@ -28,7 +28,7 @@ # ---- # Func foo contains a jump table whose start is colocated with a # jump table reference in another function. However, the other function -# does not use the first entries of it and is merely doing arithmetics +# does not use the first entries of it and is merely doing arithmetic # to save the creation of unused first entries. # ---- .globl foo diff --git a/bolt/test/X86/split-landing-pad.s b/bolt/test/X86/split-landing-pad.s index 681f14f..149193d 100644 --- a/bolt/test/X86/split-landing-pad.s +++ b/bolt/test/X86/split-landing-pad.s @@ -1,5 +1,5 @@ ## This test reproduces the case where C++ exception handling is used and split -## function optimization is enabled. In particular, function foo is splitted +## function optimization is enabled. In particular, function foo is split ## to two fragments: ## foo: contains 2 try blocks, which invokes bar to throw exception ## foo.cold.1: contains 2 corresponding catch blocks (landing pad) diff --git a/bolt/test/runtime/X86/asm-dump.c b/bolt/test/runtime/X86/asm-dump.c index 7656fda..fa0de9b 100644 --- a/bolt/test/runtime/X86/asm-dump.c +++ b/bolt/test/runtime/X86/asm-dump.c @@ -30,7 +30,7 @@ * Reconstruct fdata * RUN: link_fdata %t/main.s %t.o %t.fdata.reconst * - * XXX: reenable once dumping data is supported + * XXX: re-enable once dumping data is supported * Check if reoptimized file produces the same results * dontrun: %t.exe.reopt > %t.result.reopt * dontrun: cmp %t.result %t.result.reopt diff --git a/bolt/test/runtime/wait_file.sh b/bolt/test/runtime/wait_file.sh index 42d4c5b..7346476 100644 --- a/bolt/test/runtime/wait_file.sh +++ b/bolt/test/runtime/wait_file.sh @@ -12,7 +12,7 @@ check_file() { fuser -s "$file" local ret=$? - if [ $ret -eq 1 ]; then # noone has file open + if [ $ret -eq 1 ]; then # no one has file open return 0 fi if [ $ret -eq 0 ]; then # file open by some processes diff --git a/bolt/utils/bughunter.sh b/bolt/utils/bughunter.sh index c5dddc4..d5ce059 100755 --- a/bolt/utils/bughunter.sh +++ b/bolt/utils/bughunter.sh @@ -28,7 +28,7 @@ # # TIMEOUT_OR_CMD - optional timeout or command on optimized binary command # if the value is a number with an optional trailing letter -# [smhd] it is considered a paramter to "timeout", +# [smhd] it is considered a parameter to "timeout", # otherwise it's a shell command that wraps the optimized # binary command. # diff --git a/clang-tools-extra/clangd/FindTarget.cpp b/clang-tools-extra/clangd/FindTarget.cpp index 799c64b..5d5388e 100644 --- a/clang-tools-extra/clangd/FindTarget.cpp +++ b/clang-tools-extra/clangd/FindTarget.cpp @@ -1040,8 +1040,8 @@ private: if (auto *S = N.get<Stmt>()) return refInStmt(S, Resolver); if (auto *NNSL = N.get<NestedNameSpecifierLoc>()) { - if (auto TL = NNSL->getAsTypeLoc()) - return refInTypeLoc(NNSL->getAsTypeLoc(), Resolver); + if (TypeLoc TL = NNSL->getAsTypeLoc()) + return refInTypeLoc(TL, Resolver); // (!) 'DeclRelation::Alias' ensures we do not lose namespace aliases. NestedNameSpecifierLoc Qualifier = NNSL->getAsNamespaceAndPrefix().Prefix; SourceLocation NameLoc = NNSL->getLocalBeginLoc(); diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst index db2b0f6..edb872c 100644 --- a/clang/docs/ReleaseNotes.rst +++ b/clang/docs/ReleaseNotes.rst @@ -435,6 +435,7 @@ Bug Fixes to C++ Support - Suppress ``-Wdeprecated-declarations`` in implicitly generated functions. (#GH147293) - Fix a crash when deleting a pointer to an incomplete array (#GH150359). - Fixed a mismatched lambda scope bug when propagating up ``consteval`` within nested lambdas. (#GH145776) +- Disallow immediate escalation in destructors. (#GH109096) - Fix an assertion failure when expression in assumption attribute (``[[assume(expr)]]``) creates temporary objects. - Fix the dynamic_cast to final class optimization to correctly handle diff --git a/clang/include/clang/AST/OpenACCClause.h b/clang/include/clang/AST/OpenACCClause.h index 1e351f3..83f2b18 100644 --- a/clang/include/clang/AST/OpenACCClause.h +++ b/clang/include/clang/AST/OpenACCClause.h @@ -1301,46 +1301,25 @@ struct OpenACCReductionRecipe { // AST), or in a separate collection when being semantically analyzed. llvm::ArrayRef<CombinerRecipe> CombinerRecipes; + bool isSet() const { return AllocaDecl; } + +private: + friend class OpenACCReductionClause; OpenACCReductionRecipe(VarDecl *A, llvm::ArrayRef<CombinerRecipe> Combiners) : AllocaDecl(A), CombinerRecipes(Combiners) {} - - bool isSet() const { return AllocaDecl; } }; // A version of the above that is used for semantic analysis, at a time before // the OpenACCReductionClause node has been created. This one has storage for // the CombinerRecipe, since Trailing storage for it doesn't exist yet. -struct OpenACCReductionRecipeWithStorage : OpenACCReductionRecipe { -private: - llvm::SmallVector<CombinerRecipe, 1> CombinerRecipeStorage; - -public: - OpenACCReductionRecipeWithStorage(VarDecl *A, - llvm::ArrayRef<CombinerRecipe> Combiners) - : OpenACCReductionRecipe(A, {}), CombinerRecipeStorage(Combiners) { - CombinerRecipes = CombinerRecipeStorage; - } +struct OpenACCReductionRecipeWithStorage { + VarDecl *AllocaDecl; + llvm::SmallVector<OpenACCReductionRecipe::CombinerRecipe, 1> CombinerRecipes; OpenACCReductionRecipeWithStorage( - const OpenACCReductionRecipeWithStorage &Other) - : OpenACCReductionRecipe(Other), - CombinerRecipeStorage(Other.CombinerRecipeStorage) { - CombinerRecipes = CombinerRecipeStorage; - } - - OpenACCReductionRecipeWithStorage(OpenACCReductionRecipeWithStorage &&Other) - : OpenACCReductionRecipe(std::move(Other)), - CombinerRecipeStorage(std::move(Other.CombinerRecipeStorage)) { - CombinerRecipes = CombinerRecipeStorage; - } - - // There is no real problem implementing these, we just have to make sure the - // array-ref this inherits from stays in sync. But as we don't need it at the - // moment, make sure we don't accidentially call these. - OpenACCReductionRecipeWithStorage & - operator=(OpenACCReductionRecipeWithStorage &&) = delete; - OpenACCReductionRecipeWithStorage & - operator=(const OpenACCReductionRecipeWithStorage &) = delete; + VarDecl *A, + llvm::ArrayRef<OpenACCReductionRecipe::CombinerRecipe> Combiners) + : AllocaDecl(A), CombinerRecipes(Combiners) {} static OpenACCReductionRecipeWithStorage Empty() { return OpenACCReductionRecipeWithStorage(/*AllocaDecl=*/nullptr, {}); diff --git a/clang/include/clang/Basic/BuiltinsX86.td b/clang/include/clang/Basic/BuiltinsX86.td index a55dff2..006a453 100644 --- a/clang/include/clang/Basic/BuiltinsX86.td +++ b/clang/include/clang/Basic/BuiltinsX86.td @@ -2137,24 +2137,18 @@ let Features = "avx512vl", Attributes = [NoThrow, RequiredVectorWidth<256>] in { def movdqa64store256_mask : X86Builtin<"void(_Vector<4, long long int *>, _Vector<4, long long int>, unsigned char)">; } -let Features = "avx512ifma", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in { +let Features = "avx512ifma", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in { def vpmadd52huq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Vector<8, long long int>, _Vector<8, long long int>)">; def vpmadd52luq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Vector<8, long long int>, _Vector<8, long long int>)">; } -let Features = "avx512ifma,avx512vl|avxifma", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in { +let Features = "avx512ifma,avx512vl|avxifma", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in { def vpmadd52huq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>, _Vector<2, long long int>)">; -} - -let Features = "avx512ifma,avx512vl|avxifma", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in { - def vpmadd52huq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<4, long long int>, _Vector<4, long long int>)">; -} - -let Features = "avx512ifma,avx512vl|avxifma", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in { def vpmadd52luq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>, _Vector<2, long long int>)">; } -let Features = "avx512ifma,avx512vl|avxifma", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in { +let Features = "avx512ifma,avx512vl|avxifma", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<256>] in { + def vpmadd52huq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<4, long long int>, _Vector<4, long long int>)">; def vpmadd52luq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<4, long long int>, _Vector<4, long long int>)">; } diff --git a/clang/include/clang/CIR/Dialect/IR/CIROps.td b/clang/include/clang/CIR/Dialect/IR/CIROps.td index 4c15d9e..baab156 100644 --- a/clang/include/clang/CIR/Dialect/IR/CIROps.td +++ b/clang/include/clang/CIR/Dialect/IR/CIROps.td @@ -4329,7 +4329,7 @@ def CIR_AllocExceptionOp : CIR_Op<"alloc.exception"> { // Atomic operations //===----------------------------------------------------------------------===// -def CIR_AtomicXchg : CIR_Op<"atomic.xchg", [ +def CIR_AtomicXchgOp : CIR_Op<"atomic.xchg", [ AllTypesMatch<["result", "val"]>, TypesMatchWith<"type of 'val' must match the pointee type of 'ptr'", "ptr", "val", "mlir::cast<cir::PointerType>($_self).getPointee()"> @@ -4347,9 +4347,7 @@ def CIR_AtomicXchg : CIR_Op<"atomic.xchg", [ Example: ```mlir - %res = cir.atomic.xchg(%ptr : !cir.ptr<!u64i>, - %val : !u64i, - seq_cst) : !u64i + %res = cir.atomic.xchg seq_cst %ptr, %val : !cir.ptr<!u64i> -> !u64i ``` }]; @@ -4364,12 +4362,16 @@ def CIR_AtomicXchg : CIR_Op<"atomic.xchg", [ let assemblyFormat = [{ $mem_order (`volatile` $is_volatile^)? $ptr `,` $val - `:` qualified(type($ptr)) `->` type($result) attr-dict + `:` functional-type(operands, results) attr-dict }]; } -def CIR_AtomicCmpXchg : CIR_Op<"atomic.cmpxchg", [ - AllTypesMatch<["old", "expected", "desired"]> +def CIR_AtomicCmpXchgOp : CIR_Op<"atomic.cmpxchg", [ + AllTypesMatch<["old", "expected", "desired"]>, + TypesMatchWith<"type of 'expected' must match the pointee type of 'ptr'", + "ptr", "expected", "mlir::cast<cir::PointerType>($_self).getPointee()">, + TypesMatchWith<"type of 'desired' must match the pointee type of 'ptr'", + "ptr", "desired", "mlir::cast<cir::PointerType>($_self).getPointee()"> ]> { let summary = "Atomic compare and exchange"; let description = [{ @@ -4402,12 +4404,9 @@ def CIR_AtomicCmpXchg : CIR_Op<"atomic.cmpxchg", [ Example: ```mlir - %old, %success = cir.atomic.cmpxchg(%ptr : !cir.ptr<!u64i>, - %expected : !u64i, - %desired : !u64i, - success = seq_cst, - failure = seq_cst) weak - : (!u64i, !cir.bool) + %old, %success = cir.atomic.cmpxchg weak success(seq_cst) failure(acquire) + %ptr, %expected, %desired + : (!cir.ptr<!u64i>, !u64i, !u64i) -> (!u64i, !cir.bool) ``` }]; let results = (outs CIR_AnyType:$old, CIR_BoolType:$success); @@ -4421,20 +4420,13 @@ def CIR_AtomicCmpXchg : CIR_Op<"atomic.cmpxchg", [ UnitAttr:$is_volatile); let assemblyFormat = [{ - `(` - $ptr `:` qualified(type($ptr)) `,` - $expected `:` type($expected) `,` - $desired `:` type($desired) `,` - `success` `=` $succ_order `,` - `failure` `=` $fail_order - `)` - (`align` `(` $alignment^ `)`)? (`weak` $weak^)? + `success` `(` $succ_order `)` `failure` `(` $fail_order `)` + $ptr `,` $expected `,` $desired + (`align` `(` $alignment^ `)`)? (`volatile` $is_volatile^)? - `:` `(` type($old) `,` type($success) `)` attr-dict + `:` functional-type(operands, results) attr-dict }]; - - let hasVerifier = 1; } #endif // CLANG_CIR_DIALECT_IR_CIROPS_TD diff --git a/clang/include/clang/Sema/Template.h b/clang/include/clang/Sema/Template.h index 60c7d27..e963439b 100644 --- a/clang/include/clang/Sema/Template.h +++ b/clang/include/clang/Sema/Template.h @@ -205,8 +205,8 @@ enum class TemplateSubstitutionKind : char { /// Add a new outmost level to the multi-level template argument /// list. - /// A 'Final' substitution means that Subst* nodes won't be built - /// for the replacements. + /// A 'Final' substitution means that these Args don't need to be + /// resugared later. void addOuterTemplateArguments(Decl *AssociatedDecl, ArgList Args, bool Final) { assert(!NumRetainedOuterLevels && diff --git a/clang/lib/AST/ByteCode/InterpBuiltin.cpp b/clang/lib/AST/ByteCode/InterpBuiltin.cpp index 5f0a77c..2d3cb6a 100644 --- a/clang/lib/AST/ByteCode/InterpBuiltin.cpp +++ b/clang/lib/AST/ByteCode/InterpBuiltin.cpp @@ -3818,6 +3818,21 @@ bool InterpretBuiltin(InterpState &S, CodePtr OpPC, const CallExpr *Call, return F; }); + case X86::BI__builtin_ia32_vpmadd52luq128: + case X86::BI__builtin_ia32_vpmadd52luq256: + case X86::BI__builtin_ia32_vpmadd52luq512: + return interp__builtin_elementwise_triop( + S, OpPC, Call, [](const APSInt &A, const APSInt &B, const APSInt &C) { + return A + (B.trunc(52) * C.trunc(52)).zext(64); + }); + case X86::BI__builtin_ia32_vpmadd52huq128: + case X86::BI__builtin_ia32_vpmadd52huq256: + case X86::BI__builtin_ia32_vpmadd52huq512: + return interp__builtin_elementwise_triop( + S, OpPC, Call, [](const APSInt &A, const APSInt &B, const APSInt &C) { + return A + llvm::APIntOps::mulhu(B.trunc(52), C.trunc(52)).zext(64); + }); + case X86::BI__builtin_ia32_vpshldd128: case X86::BI__builtin_ia32_vpshldd256: case X86::BI__builtin_ia32_vpshldd512: diff --git a/clang/lib/AST/Decl.cpp b/clang/lib/AST/Decl.cpp index c734155..69cbf6e 100644 --- a/clang/lib/AST/Decl.cpp +++ b/clang/lib/AST/Decl.cpp @@ -3316,6 +3316,10 @@ bool FunctionDecl::isImmediateEscalating() const { CD && CD->isInheritingConstructor()) return CD->getInheritedConstructor().getConstructor(); + // Destructors are not immediate escalating. + if (isa<CXXDestructorDecl>(this)) + return false; + // - a function that results from the instantiation of a templated entity // defined with the constexpr specifier. TemplatedKind TK = getTemplatedKind(); diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp index 59b4f4f..51c0382 100644 --- a/clang/lib/AST/ExprConstant.cpp +++ b/clang/lib/AST/ExprConstant.cpp @@ -11974,6 +11974,54 @@ bool VectorExprEvaluator::VisitCallExpr(const CallExpr *E) { return Success(APValue(ResultElements.data(), ResultElements.size()), E); } + + case X86::BI__builtin_ia32_vpmadd52luq128: + case X86::BI__builtin_ia32_vpmadd52luq256: + case X86::BI__builtin_ia32_vpmadd52luq512: { + APValue A, B, C; + if (!EvaluateAsRValue(Info, E->getArg(0), A) || + !EvaluateAsRValue(Info, E->getArg(1), B) || + !EvaluateAsRValue(Info, E->getArg(2), C)) + return false; + + unsigned ALen = A.getVectorLength(); + SmallVector<APValue, 4> ResultElements; + ResultElements.reserve(ALen); + + for (unsigned EltNum = 0; EltNum < ALen; EltNum += 1) { + APInt AElt = A.getVectorElt(EltNum).getInt(); + APInt BElt = B.getVectorElt(EltNum).getInt().trunc(52); + APInt CElt = C.getVectorElt(EltNum).getInt().trunc(52); + APSInt ResElt(AElt + (BElt * CElt).zext(64), false); + ResultElements.push_back(APValue(ResElt)); + } + + return Success(APValue(ResultElements.data(), ResultElements.size()), E); + } + case X86::BI__builtin_ia32_vpmadd52huq128: + case X86::BI__builtin_ia32_vpmadd52huq256: + case X86::BI__builtin_ia32_vpmadd52huq512: { + APValue A, B, C; + if (!EvaluateAsRValue(Info, E->getArg(0), A) || + !EvaluateAsRValue(Info, E->getArg(1), B) || + !EvaluateAsRValue(Info, E->getArg(2), C)) + return false; + + unsigned ALen = A.getVectorLength(); + SmallVector<APValue, 4> ResultElements; + ResultElements.reserve(ALen); + + for (unsigned EltNum = 0; EltNum < ALen; EltNum += 1) { + APInt AElt = A.getVectorElt(EltNum).getInt(); + APInt BElt = B.getVectorElt(EltNum).getInt().trunc(52); + APInt CElt = C.getVectorElt(EltNum).getInt().trunc(52); + APSInt ResElt(AElt + llvm::APIntOps::mulhu(BElt, CElt).zext(64), false); + ResultElements.push_back(APValue(ResElt)); + } + + return Success(APValue(ResultElements.data(), ResultElements.size()), E); + } + case clang::X86::BI__builtin_ia32_vprotbi: case clang::X86::BI__builtin_ia32_vprotdi: case clang::X86::BI__builtin_ia32_vprotqi: diff --git a/clang/lib/AST/OpenACCClause.cpp b/clang/lib/AST/OpenACCClause.cpp index 17c6bec..142c932 100644 --- a/clang/lib/AST/OpenACCClause.cpp +++ b/clang/lib/AST/OpenACCClause.cpp @@ -509,7 +509,7 @@ OpenACCReductionClause *OpenACCReductionClause::Create( ArrayRef<OpenACCReductionRecipeWithStorage> Recipes, SourceLocation EndLoc) { size_t NumCombiners = llvm::accumulate( - Recipes, 0, [](size_t Num, const OpenACCReductionRecipe &R) { + Recipes, 0, [](size_t Num, const OpenACCReductionRecipeWithStorage &R) { return Num + R.CombinerRecipes.size(); }); diff --git a/clang/lib/Analysis/LifetimeSafety/FactsGenerator.cpp b/clang/lib/Analysis/LifetimeSafety/FactsGenerator.cpp index 485308f..9b68de1 100644 --- a/clang/lib/Analysis/LifetimeSafety/FactsGenerator.cpp +++ b/clang/lib/Analysis/LifetimeSafety/FactsGenerator.cpp @@ -9,9 +9,11 @@ #include "clang/Analysis/Analyses/LifetimeSafety/FactsGenerator.h" #include "clang/Analysis/Analyses/LifetimeSafety/LifetimeAnnotations.h" #include "clang/Analysis/Analyses/PostOrderCFGView.h" +#include "llvm/Support/Casting.h" #include "llvm/Support/TimeProfiler.h" namespace clang::lifetimes::internal { +using llvm::isa_and_present; static bool isGslPointerType(QualType QT) { if (const auto *RD = QT->getAsCXXRecordDecl()) { @@ -108,7 +110,7 @@ void FactsGenerator::VisitCXXMemberCallExpr(const CXXMemberCallExpr *MCE) { // Specifically for conversion operators, // like `std::string_view p = std::string{};` if (isGslPointerType(MCE->getType()) && - isa<CXXConversionDecl>(MCE->getCalleeDecl())) { + isa_and_present<CXXConversionDecl>(MCE->getCalleeDecl())) { // The argument is the implicit object itself. handleFunctionCall(MCE, MCE->getMethodDecl(), {MCE->getImplicitObjectArgument()}, diff --git a/clang/lib/CIR/CodeGen/CIRGenAtomic.cpp b/clang/lib/CIR/CodeGen/CIRGenAtomic.cpp index 0f4d6d2..a9983f8 100644 --- a/clang/lib/CIR/CodeGen/CIRGenAtomic.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenAtomic.cpp @@ -255,7 +255,7 @@ static void emitAtomicCmpXchg(CIRGenFunction &cgf, AtomicExpr *e, bool isWeak, mlir::Value expected = builder.createLoad(loc, val1); mlir::Value desired = builder.createLoad(loc, val2); - auto cmpxchg = cir::AtomicCmpXchg::create( + auto cmpxchg = cir::AtomicCmpXchgOp::create( builder, loc, expected.getType(), builder.getBoolTy(), ptr.getPointer(), expected, desired, cir::MemOrderAttr::get(&cgf.getMLIRContext(), successOrder), @@ -404,7 +404,7 @@ static void emitAtomicOp(CIRGenFunction &cgf, AtomicExpr *expr, Address dest, case AtomicExpr::AO__c11_atomic_exchange: case AtomicExpr::AO__atomic_exchange_n: case AtomicExpr::AO__atomic_exchange: - opName = cir::AtomicXchg::getOperationName(); + opName = cir::AtomicXchgOp::getOperationName(); break; case AtomicExpr::AO__opencl_atomic_init: diff --git a/clang/lib/CIR/CodeGen/CIRGenOpenACCClause.cpp b/clang/lib/CIR/CodeGen/CIRGenOpenACCClause.cpp index 3d86f71..ce4ae7e 100644 --- a/clang/lib/CIR/CodeGen/CIRGenOpenACCClause.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenOpenACCClause.cpp @@ -1005,7 +1005,7 @@ public: /*temporary=*/nullptr, OpenACCReductionOperator::Invalid, Decl::castToDeclContext(cgf.curFuncDecl), opInfo.origType, opInfo.bounds.size(), opInfo.boundTypes, opInfo.baseType, - privateOp); + privateOp, /*reductionCombinerRecipes=*/{}); // TODO: OpenACC: The dialect is going to change in the near future to // have these be on a different operation, so when that changes, we // probably need to change these here. @@ -1046,7 +1046,7 @@ public: OpenACCReductionOperator::Invalid, Decl::castToDeclContext(cgf.curFuncDecl), opInfo.origType, opInfo.bounds.size(), opInfo.boundTypes, opInfo.baseType, - firstPrivateOp); + firstPrivateOp, /*reductionCombinerRecipe=*/{}); // TODO: OpenACC: The dialect is going to change in the near future to // have these be on a different operation, so when that changes, we @@ -1088,7 +1088,7 @@ public: /*temporary=*/nullptr, clause.getReductionOp(), Decl::castToDeclContext(cgf.curFuncDecl), opInfo.origType, opInfo.bounds.size(), opInfo.boundTypes, opInfo.baseType, - reductionOp); + reductionOp, varRecipe.CombinerRecipes); operation.addReduction(builder.getContext(), reductionOp, recipe); } diff --git a/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.cpp b/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.cpp index 24a5fc2..ce14aa8 100644 --- a/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.cpp @@ -527,16 +527,142 @@ void OpenACCRecipeBuilderBase::createFirstprivateRecipeCopy( // doesn't restore it aftewards. void OpenACCRecipeBuilderBase::createReductionRecipeCombiner( mlir::Location loc, mlir::Location locEnd, mlir::Value mainOp, - mlir::acc::ReductionRecipeOp recipe, size_t numBounds) { + mlir::acc::ReductionRecipeOp recipe, size_t numBounds, QualType origType, + llvm::ArrayRef<OpenACCReductionRecipe::CombinerRecipe> combinerRecipes) { mlir::Block *block = createRecipeBlock(recipe.getCombinerRegion(), mainOp.getType(), loc, numBounds, /*isInit=*/false); builder.setInsertionPointToEnd(&recipe.getCombinerRegion().back()); CIRGenFunction::LexicalScope ls(cgf, loc, block); - mlir::BlockArgument lhsArg = block->getArgument(0); + mlir::Value lhsArg = block->getArgument(0); + mlir::Value rhsArg = block->getArgument(1); + llvm::MutableArrayRef<mlir::BlockArgument> boundsRange = + block->getArguments().drop_front(2); + + if (llvm::any_of(combinerRecipes, [](auto &r) { return r.Op == nullptr; })) { + cgf.cgm.errorNYI(loc, "OpenACC Reduction combiner not generated"); + mlir::acc::YieldOp::create(builder, locEnd, block->getArgument(0)); + return; + } + + // apply the bounds so that we can get our bounds emitted correctly. + for (mlir::BlockArgument boundArg : llvm::reverse(boundsRange)) + std::tie(lhsArg, rhsArg) = + createBoundsLoop(lhsArg, rhsArg, boundArg, loc, /*inverse=*/false); + + // Emitter for when we know this isn't a struct or array we have to loop + // through. This should work for the 'field' once the get-element call has + // been made. + auto emitSingleCombiner = + [&](mlir::Value lhsArg, mlir::Value rhsArg, + const OpenACCReductionRecipe::CombinerRecipe &combiner) { + mlir::Type elementTy = + mlir::cast<cir::PointerType>(lhsArg.getType()).getPointee(); + CIRGenFunction::DeclMapRevertingRAII declMapRAIILhs{cgf, combiner.LHS}; + cgf.setAddrOfLocalVar( + combiner.LHS, Address{lhsArg, elementTy, + cgf.getContext().getDeclAlign(combiner.LHS)}); + CIRGenFunction::DeclMapRevertingRAII declMapRAIIRhs{cgf, combiner.RHS}; + cgf.setAddrOfLocalVar( + combiner.RHS, Address{rhsArg, elementTy, + cgf.getContext().getDeclAlign(combiner.RHS)}); + + [[maybe_unused]] mlir::LogicalResult stmtRes = + cgf.emitStmt(combiner.Op, /*useCurrentScope=*/true); + }; + + // Emitter for when we know this is either a non-array or element of an array + // (which also shouldn't be an array type?). This function should generate the + // initialization code for an entire 'array-element'/non-array, including + // diving into each element of a struct (if necessary). + auto emitCombiner = [&](mlir::Value lhsArg, mlir::Value rhsArg, QualType ty) { + assert(!ty->isArrayType() && "Array type shouldn't get here"); + if (const auto *rd = ty->getAsRecordDecl()) { + if (combinerRecipes.size() == 1 && + cgf.getContext().hasSameType(ty, combinerRecipes[0].LHS->getType())) { + // If this is a 'top level' operator on the type we can just emit this + // as a simple one. + emitSingleCombiner(lhsArg, rhsArg, combinerRecipes[0]); + } else { + // else we have to handle each individual field after after a + // get-element. + for (const auto &[field, combiner] : + llvm::zip_equal(rd->fields(), combinerRecipes)) { + mlir::Type fieldType = cgf.convertType(field->getType()); + auto fieldPtr = cir::PointerType::get(fieldType); + + mlir::Value lhsField = builder.createGetMember( + loc, fieldPtr, lhsArg, field->getName(), field->getFieldIndex()); + mlir::Value rhsField = builder.createGetMember( + loc, fieldPtr, rhsArg, field->getName(), field->getFieldIndex()); + + emitSingleCombiner(lhsField, rhsField, combiner); + } + } + + } else { + // if this is a single-thing (because we should know this isn't an array, + // as Sema wouldn't let us get here), we can just do a normal emit call. + emitSingleCombiner(lhsArg, rhsArg, combinerRecipes[0]); + } + }; + + if (const auto *cat = cgf.getContext().getAsConstantArrayType(origType)) { + // If we're in an array, we have to emit the combiner for each element of + // the array. + auto itrTy = mlir::cast<cir::IntType>(cgf.PtrDiffTy); + auto itrPtrTy = cir::PointerType::get(itrTy); + + mlir::Value zero = + builder.getConstInt(loc, mlir::cast<cir::IntType>(cgf.PtrDiffTy), 0); + mlir::Value itr = + cir::AllocaOp::create(builder, loc, itrPtrTy, itrTy, "itr", + cgf.cgm.getSize(cgf.getPointerAlign())); + builder.CIRBaseBuilderTy::createStore(loc, zero, itr); + + builder.setInsertionPointAfter(builder.createFor( + loc, + /*condBuilder=*/ + [&](mlir::OpBuilder &b, mlir::Location loc) { + auto loadItr = cir::LoadOp::create(builder, loc, {itr}); + mlir::Value arraySize = builder.getConstInt( + loc, mlir::cast<cir::IntType>(cgf.PtrDiffTy), cat->getZExtSize()); + auto cmp = builder.createCompare(loc, cir::CmpOpKind::lt, loadItr, + arraySize); + builder.createCondition(cmp); + }, + /*bodyBuilder=*/ + [&](mlir::OpBuilder &b, mlir::Location loc) { + auto loadItr = cir::LoadOp::create(builder, loc, {itr}); + auto lhsElt = builder.getArrayElement( + loc, loc, lhsArg, cgf.convertType(cat->getElementType()), loadItr, + /*shouldDecay=*/true); + auto rhsElt = builder.getArrayElement( + loc, loc, rhsArg, cgf.convertType(cat->getElementType()), loadItr, + /*shouldDecay=*/true); + + emitCombiner(lhsElt, rhsElt, cat->getElementType()); + builder.createYield(loc); + }, + /*stepBuilder=*/ + [&](mlir::OpBuilder &b, mlir::Location loc) { + auto loadItr = cir::LoadOp::create(builder, loc, {itr}); + auto inc = cir::UnaryOp::create(builder, loc, loadItr.getType(), + cir::UnaryOpKind::Inc, loadItr); + builder.CIRBaseBuilderTy::createStore(loc, inc, itr); + builder.createYield(loc); + })); - mlir::acc::YieldOp::create(builder, locEnd, lhsArg); + } else if (origType->isArrayType()) { + cgf.cgm.errorNYI(loc, + "OpenACC Reduction combiner non-constant array recipe"); + } else { + emitCombiner(lhsArg, rhsArg, origType); + } + + builder.setInsertionPointToEnd(&recipe.getCombinerRegion().back()); + mlir::acc::YieldOp::create(builder, locEnd, block->getArgument(0)); } } // namespace clang::CIRGen diff --git a/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.h b/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.h index a5da744..745d424 100644 --- a/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.h +++ b/clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.h @@ -64,10 +64,10 @@ protected: // that this function is not 'insertion point' clean, in that it alters the // insertion point to be inside of the 'combiner' section of the recipe, but // doesn't restore it aftewards. - void createReductionRecipeCombiner(mlir::Location loc, mlir::Location locEnd, - mlir::Value mainOp, - mlir::acc::ReductionRecipeOp recipe, - size_t numBounds); + void createReductionRecipeCombiner( + mlir::Location loc, mlir::Location locEnd, mlir::Value mainOp, + mlir::acc::ReductionRecipeOp recipe, size_t numBounds, QualType origType, + llvm::ArrayRef<OpenACCReductionRecipe::CombinerRecipe> combinerRecipes); void createInitRecipe(mlir::Location loc, mlir::Location locEnd, SourceRange exprRange, mlir::Value mainOp, @@ -169,7 +169,9 @@ public: const Expr *varRef, const VarDecl *varRecipe, const VarDecl *temporary, OpenACCReductionOperator reductionOp, DeclContext *dc, QualType origType, size_t numBounds, llvm::ArrayRef<QualType> boundTypes, QualType baseType, - mlir::Value mainOp) { + mlir::Value mainOp, + llvm::ArrayRef<OpenACCReductionRecipe::CombinerRecipe> + reductionCombinerRecipes) { assert(!varRecipe->getType()->isSpecificBuiltinType( BuiltinType::ArraySection) && "array section shouldn't make it to recipe creation"); @@ -208,7 +210,8 @@ public: createInitRecipe(loc, locEnd, varRef->getSourceRange(), mainOp, recipe.getInitRegion(), numBounds, boundTypes, varRecipe, origType, /*emitInitExpr=*/true); - createReductionRecipeCombiner(loc, locEnd, mainOp, recipe, numBounds); + createReductionRecipeCombiner(loc, locEnd, mainOp, recipe, numBounds, + origType, reductionCombinerRecipes); } else { static_assert(std::is_same_v<RecipeTy, mlir::acc::FirstprivateRecipeOp>); createInitRecipe(loc, locEnd, varRef->getSourceRange(), mainOp, diff --git a/clang/lib/CIR/Dialect/IR/CIRDialect.cpp b/clang/lib/CIR/Dialect/IR/CIRDialect.cpp index 12837d9..7af3dc1 100644 --- a/clang/lib/CIR/Dialect/IR/CIRDialect.cpp +++ b/clang/lib/CIR/Dialect/IR/CIRDialect.cpp @@ -2901,20 +2901,6 @@ mlir::LogicalResult cir::ThrowOp::verify() { } //===----------------------------------------------------------------------===// -// AtomicCmpXchg -//===----------------------------------------------------------------------===// - -LogicalResult cir::AtomicCmpXchg::verify() { - mlir::Type pointeeType = getPtr().getType().getPointee(); - - if (pointeeType != getExpected().getType() || - pointeeType != getDesired().getType()) - return emitOpError("ptr, expected and desired types must match"); - - return success(); -} - -//===----------------------------------------------------------------------===// // TypeInfoAttr //===----------------------------------------------------------------------===// diff --git a/clang/lib/CIR/Lowering/DirectToLLVM/LowerToLLVM.cpp b/clang/lib/CIR/Lowering/DirectToLLVM/LowerToLLVM.cpp index f0d73ac..3abba3d 100644 --- a/clang/lib/CIR/Lowering/DirectToLLVM/LowerToLLVM.cpp +++ b/clang/lib/CIR/Lowering/DirectToLLVM/LowerToLLVM.cpp @@ -694,8 +694,8 @@ getLLVMMemOrder(std::optional<cir::MemOrder> memorder) { llvm_unreachable("unknown memory order"); } -mlir::LogicalResult CIRToLLVMAtomicCmpXchgLowering::matchAndRewrite( - cir::AtomicCmpXchg op, OpAdaptor adaptor, +mlir::LogicalResult CIRToLLVMAtomicCmpXchgOpLowering::matchAndRewrite( + cir::AtomicCmpXchgOp op, OpAdaptor adaptor, mlir::ConversionPatternRewriter &rewriter) const { mlir::Value expected = adaptor.getExpected(); mlir::Value desired = adaptor.getDesired(); @@ -719,8 +719,8 @@ mlir::LogicalResult CIRToLLVMAtomicCmpXchgLowering::matchAndRewrite( return mlir::success(); } -mlir::LogicalResult CIRToLLVMAtomicXchgLowering::matchAndRewrite( - cir::AtomicXchg op, OpAdaptor adaptor, +mlir::LogicalResult CIRToLLVMAtomicXchgOpLowering::matchAndRewrite( + cir::AtomicXchgOp op, OpAdaptor adaptor, mlir::ConversionPatternRewriter &rewriter) const { assert(!cir::MissingFeatures::atomicSyncScopeID()); mlir::LLVM::AtomicOrdering llvmOrder = getLLVMMemOrder(adaptor.getMemOrder()); diff --git a/clang/lib/Format/FormatToken.cpp b/clang/lib/Format/FormatToken.cpp index c2956a1..cb3fc1c 100644 --- a/clang/lib/Format/FormatToken.cpp +++ b/clang/lib/Format/FormatToken.cpp @@ -41,8 +41,7 @@ static constexpr std::array<StringRef, 14> QtPropertyKeywords = { bool FormatToken::isQtProperty() const { assert(llvm::is_sorted(QtPropertyKeywords)); - return std::binary_search(QtPropertyKeywords.begin(), - QtPropertyKeywords.end(), TokenText); + return llvm::binary_search(QtPropertyKeywords, TokenText); } // Sorted common C++ non-keyword types. diff --git a/clang/lib/Headers/avx512ifmaintrin.h b/clang/lib/Headers/avx512ifmaintrin.h index f01b322..625a8ff 100644 --- a/clang/lib/Headers/avx512ifmaintrin.h +++ b/clang/lib/Headers/avx512ifmaintrin.h @@ -15,54 +15,53 @@ #define __IFMAINTRIN_H /* Define the default attributes for the functions in this file. */ +#if defined(__cplusplus) && (__cplusplus >= 201103L) +#define __DEFAULT_FN_ATTRS \ + constexpr \ + __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma"), \ + __min_vector_width__(512))) +#else #define __DEFAULT_FN_ATTRS \ __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma"), \ __min_vector_width__(512))) +#endif static __inline__ __m512i __DEFAULT_FN_ATTRS -_mm512_madd52hi_epu64 (__m512i __X, __m512i __Y, __m512i __Z) -{ - return (__m512i)__builtin_ia32_vpmadd52huq512((__v8di) __X, (__v8di) __Y, - (__v8di) __Z); +_mm512_madd52hi_epu64(__m512i __X, __m512i __Y, __m512i __Z) { + return (__m512i)__builtin_ia32_vpmadd52huq512((__v8di)__X, (__v8di)__Y, + (__v8di)__Z); } -static __inline__ __m512i __DEFAULT_FN_ATTRS -_mm512_mask_madd52hi_epu64 (__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y) -{ - return (__m512i)__builtin_ia32_selectq_512(__M, - (__v8di)_mm512_madd52hi_epu64(__W, __X, __Y), - (__v8di)__W); +static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_mask_madd52hi_epu64( + __m512i __W, __mmask8 __M, __m512i __X, __m512i __Y) { + return (__m512i)__builtin_ia32_selectq_512( + __M, (__v8di)_mm512_madd52hi_epu64(__W, __X, __Y), (__v8di)__W); } -static __inline__ __m512i __DEFAULT_FN_ATTRS -_mm512_maskz_madd52hi_epu64 (__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z) -{ - return (__m512i)__builtin_ia32_selectq_512(__M, - (__v8di)_mm512_madd52hi_epu64(__X, __Y, __Z), - (__v8di)_mm512_setzero_si512()); +static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_maskz_madd52hi_epu64( + __mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z) { + return (__m512i)__builtin_ia32_selectq_512( + __M, (__v8di)_mm512_madd52hi_epu64(__X, __Y, __Z), + (__v8di)_mm512_setzero_si512()); } static __inline__ __m512i __DEFAULT_FN_ATTRS -_mm512_madd52lo_epu64 (__m512i __X, __m512i __Y, __m512i __Z) -{ - return (__m512i)__builtin_ia32_vpmadd52luq512((__v8di) __X, (__v8di) __Y, - (__v8di) __Z); +_mm512_madd52lo_epu64(__m512i __X, __m512i __Y, __m512i __Z) { + return (__m512i)__builtin_ia32_vpmadd52luq512((__v8di)__X, (__v8di)__Y, + (__v8di)__Z); } -static __inline__ __m512i __DEFAULT_FN_ATTRS -_mm512_mask_madd52lo_epu64 (__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y) -{ - return (__m512i)__builtin_ia32_selectq_512(__M, - (__v8di)_mm512_madd52lo_epu64(__W, __X, __Y), - (__v8di)__W); +static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_mask_madd52lo_epu64( + __m512i __W, __mmask8 __M, __m512i __X, __m512i __Y) { + return (__m512i)__builtin_ia32_selectq_512( + __M, (__v8di)_mm512_madd52lo_epu64(__W, __X, __Y), (__v8di)__W); } -static __inline__ __m512i __DEFAULT_FN_ATTRS -_mm512_maskz_madd52lo_epu64 (__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z) -{ - return (__m512i)__builtin_ia32_selectq_512(__M, - (__v8di)_mm512_madd52lo_epu64(__X, __Y, __Z), - (__v8di)_mm512_setzero_si512()); +static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_maskz_madd52lo_epu64( + __mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z) { + return (__m512i)__builtin_ia32_selectq_512( + __M, (__v8di)_mm512_madd52lo_epu64(__X, __Y, __Z), + (__v8di)_mm512_setzero_si512()); } #undef __DEFAULT_FN_ATTRS diff --git a/clang/lib/Headers/avx512ifmavlintrin.h b/clang/lib/Headers/avx512ifmavlintrin.h index a72b561..c4449c7 100644 --- a/clang/lib/Headers/avx512ifmavlintrin.h +++ b/clang/lib/Headers/avx512ifmavlintrin.h @@ -8,13 +8,24 @@ *===-----------------------------------------------------------------------=== */ #ifndef __IMMINTRIN_H -#error "Never use <avx512ifmavlintrin.h> directly; include <immintrin.h> instead." +#error \ + "Never use <avx512ifmavlintrin.h> directly; include <immintrin.h> instead." #endif #ifndef __IFMAVLINTRIN_H #define __IFMAVLINTRIN_H /* Define the default attributes for the functions in this file. */ +#if defined(__cplusplus) && (__cplusplus >= 201103L) +#define __DEFAULT_FN_ATTRS128 \ + constexpr __attribute__((__always_inline__, __nodebug__, \ + __target__("avx512ifma,avx512vl"), \ + __min_vector_width__(128))) +#define __DEFAULT_FN_ATTRS256 \ + constexpr __attribute__((__always_inline__, __nodebug__, \ + __target__("avx512ifma,avx512vl"), \ + __min_vector_width__(256))) +#else #define __DEFAULT_FN_ATTRS128 \ __attribute__((__always_inline__, __nodebug__, \ __target__("avx512ifma,avx512vl"), \ @@ -24,6 +35,8 @@ __target__("avx512ifma,avx512vl"), \ __min_vector_width__(256))) +#endif + #define _mm_madd52hi_epu64(X, Y, Z) \ ((__m128i)__builtin_ia32_vpmadd52huq128((__v2di)(X), (__v2di)(Y), \ (__v2di)(Z))) @@ -41,70 +54,57 @@ (__v4di)(Z))) static __inline__ __m128i __DEFAULT_FN_ATTRS128 -_mm_mask_madd52hi_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y) -{ - return (__m128i)__builtin_ia32_selectq_128(__M, - (__v2di)_mm_madd52hi_epu64(__W, __X, __Y), - (__v2di)__W); +_mm_mask_madd52hi_epu64(__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y) { + return (__m128i)__builtin_ia32_selectq_128( + __M, (__v2di)_mm_madd52hi_epu64(__W, __X, __Y), (__v2di)__W); } static __inline__ __m128i __DEFAULT_FN_ATTRS128 -_mm_maskz_madd52hi_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z) -{ - return (__m128i)__builtin_ia32_selectq_128(__M, - (__v2di)_mm_madd52hi_epu64(__X, __Y, __Z), - (__v2di)_mm_setzero_si128()); +_mm_maskz_madd52hi_epu64(__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z) { + return (__m128i)__builtin_ia32_selectq_128( + __M, (__v2di)_mm_madd52hi_epu64(__X, __Y, __Z), + (__v2di)_mm_setzero_si128()); } -static __inline__ __m256i __DEFAULT_FN_ATTRS256 -_mm256_mask_madd52hi_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y) -{ - return (__m256i)__builtin_ia32_selectq_256(__M, - (__v4di)_mm256_madd52hi_epu64(__W, __X, __Y), - (__v4di)__W); +static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_mask_madd52hi_epu64( + __m256i __W, __mmask8 __M, __m256i __X, __m256i __Y) { + return (__m256i)__builtin_ia32_selectq_256( + __M, (__v4di)_mm256_madd52hi_epu64(__W, __X, __Y), (__v4di)__W); } -static __inline__ __m256i __DEFAULT_FN_ATTRS256 -_mm256_maskz_madd52hi_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z) -{ - return (__m256i)__builtin_ia32_selectq_256(__M, - (__v4di)_mm256_madd52hi_epu64(__X, __Y, __Z), - (__v4di)_mm256_setzero_si256()); +static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_maskz_madd52hi_epu64( + __mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z) { + return (__m256i)__builtin_ia32_selectq_256( + __M, (__v4di)_mm256_madd52hi_epu64(__X, __Y, __Z), + (__v4di)_mm256_setzero_si256()); } static __inline__ __m128i __DEFAULT_FN_ATTRS128 -_mm_mask_madd52lo_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y) -{ - return (__m128i)__builtin_ia32_selectq_128(__M, - (__v2di)_mm_madd52lo_epu64(__W, __X, __Y), - (__v2di)__W); +_mm_mask_madd52lo_epu64(__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y) { + return (__m128i)__builtin_ia32_selectq_128( + __M, (__v2di)_mm_madd52lo_epu64(__W, __X, __Y), (__v2di)__W); } static __inline__ __m128i __DEFAULT_FN_ATTRS128 -_mm_maskz_madd52lo_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z) -{ - return (__m128i)__builtin_ia32_selectq_128(__M, - (__v2di)_mm_madd52lo_epu64(__X, __Y, __Z), - (__v2di)_mm_setzero_si128()); +_mm_maskz_madd52lo_epu64(__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z) { + return (__m128i)__builtin_ia32_selectq_128( + __M, (__v2di)_mm_madd52lo_epu64(__X, __Y, __Z), + (__v2di)_mm_setzero_si128()); } -static __inline__ __m256i __DEFAULT_FN_ATTRS256 -_mm256_mask_madd52lo_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y) -{ - return (__m256i)__builtin_ia32_selectq_256(__M, - (__v4di)_mm256_madd52lo_epu64(__W, __X, __Y), - (__v4di)__W); +static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_mask_madd52lo_epu64( + __m256i __W, __mmask8 __M, __m256i __X, __m256i __Y) { + return (__m256i)__builtin_ia32_selectq_256( + __M, (__v4di)_mm256_madd52lo_epu64(__W, __X, __Y), (__v4di)__W); } -static __inline__ __m256i __DEFAULT_FN_ATTRS256 -_mm256_maskz_madd52lo_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z) -{ - return (__m256i)__builtin_ia32_selectq_256(__M, - (__v4di)_mm256_madd52lo_epu64(__X, __Y, __Z), - (__v4di)_mm256_setzero_si256()); +static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_maskz_madd52lo_epu64( + __mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z) { + return (__m256i)__builtin_ia32_selectq_256( + __M, (__v4di)_mm256_madd52lo_epu64(__X, __Y, __Z), + (__v4di)_mm256_setzero_si256()); } - #undef __DEFAULT_FN_ATTRS128 #undef __DEFAULT_FN_ATTRS256 diff --git a/clang/lib/Headers/avxifmaintrin.h b/clang/lib/Headers/avxifmaintrin.h index 5c782d2a..a2ef601 100644 --- a/clang/lib/Headers/avxifmaintrin.h +++ b/clang/lib/Headers/avxifmaintrin.h @@ -15,12 +15,21 @@ #define __AVXIFMAINTRIN_H /* Define the default attributes for the functions in this file. */ +#if defined(__cplusplus) && (__cplusplus >= 201103L) +#define __DEFAULT_FN_ATTRS128 \ + constexpr __attribute__((__always_inline__, __nodebug__, \ + __target__("avxifma"), __min_vector_width__(128))) +#define __DEFAULT_FN_ATTRS256 \ + constexpr __attribute__((__always_inline__, __nodebug__, \ + __target__("avxifma"), __min_vector_width__(256))) +#else #define __DEFAULT_FN_ATTRS128 \ __attribute__((__always_inline__, __nodebug__, __target__("avxifma"), \ __min_vector_width__(128))) #define __DEFAULT_FN_ATTRS256 \ __attribute__((__always_inline__, __nodebug__, __target__("avxifma"), \ __min_vector_width__(256))) +#endif // must vex-encoding diff --git a/clang/lib/Sema/SemaConcept.cpp b/clang/lib/Sema/SemaConcept.cpp index 9cbd1bd..87dd682 100644 --- a/clang/lib/Sema/SemaConcept.cpp +++ b/clang/lib/Sema/SemaConcept.cpp @@ -606,10 +606,9 @@ ConstraintSatisfactionChecker::SubstitutionInTemplateArguments( Constraint.mappingOccurenceList(); // The empty MLTAL situation should only occur when evaluating non-dependent // constraints. - if (!MLTAL.getNumSubstitutedLevels()) - MLTAL.addOuterTemplateArguments(TD, {}, /*Final=*/false); - SubstitutedOuterMost = - llvm::to_vector_of<TemplateArgument>(MLTAL.getOutermost()); + if (MLTAL.getNumSubstitutedLevels()) + SubstitutedOuterMost = + llvm::to_vector_of<TemplateArgument>(MLTAL.getOutermost()); unsigned Offset = 0; for (unsigned I = 0, MappedIndex = 0; I < Used.size(); I++) { TemplateArgument Arg; @@ -627,8 +626,10 @@ ConstraintSatisfactionChecker::SubstitutionInTemplateArguments( if (Offset < SubstitutedOuterMost.size()) SubstitutedOuterMost.erase(SubstitutedOuterMost.begin() + Offset); - MLTAL.replaceOutermostTemplateArguments(TD, SubstitutedOuterMost); - return std::move(MLTAL); + MultiLevelTemplateArgumentList SubstitutedTemplateArgs; + SubstitutedTemplateArgs.addOuterTemplateArguments(TD, SubstitutedOuterMost, + /*Final=*/false); + return std::move(SubstitutedTemplateArgs); } ExprResult ConstraintSatisfactionChecker::EvaluateSlow( diff --git a/clang/lib/Sema/SemaOpenACCClause.cpp b/clang/lib/Sema/SemaOpenACCClause.cpp index ead9781..17078e8 100644 --- a/clang/lib/Sema/SemaOpenACCClause.cpp +++ b/clang/lib/Sema/SemaOpenACCClause.cpp @@ -1924,7 +1924,7 @@ bool SemaOpenACC::CheckReductionVarType(Expr *VarExpr) { // off here. This will result in CurType being the actual 'type' of the // expression, which is what we are looking to check. QualType CurType = isa<ArraySectionExpr>(VarExpr) - ? ArraySectionExpr::getBaseOriginalType(VarExpr) + ? cast<ArraySectionExpr>(VarExpr)->getElementType() : VarExpr->getType(); // This can happen when we have a dependent type in an array element that the diff --git a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp index 85e3d20..73fd33a 100644 --- a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp +++ b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp @@ -5727,7 +5727,7 @@ void Sema::InstantiateFunctionDefinition(SourceLocation PointOfInstantiation, Function->setDeclarationNameLoc(NameLocPointsToPattern()); EnterExpressionEvaluationContextForFunction EvalContext( - *this, Sema::ExpressionEvaluationContext::PotentiallyEvaluated); + *this, Sema::ExpressionEvaluationContext::PotentiallyEvaluated, Function); Qualifiers ThisTypeQuals; CXXRecordDecl *ThisContext = nullptr; diff --git a/clang/test/CIR/CodeGen/atomic.c b/clang/test/CIR/CodeGen/atomic.c index 76289c5..440010a 100644 --- a/clang/test/CIR/CodeGen/atomic.c +++ b/clang/test/CIR/CodeGen/atomic.c @@ -211,7 +211,7 @@ void c11_atomic_cmpxchg_strong(_Atomic(int) *ptr, int *expected, int desired) { __c11_atomic_compare_exchange_strong(ptr, expected, desired, __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE); - // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg(%{{.+}} : !cir.ptr<!s32i>, %{{.+}} : !s32i, %{{.+}} : !s32i, success = seq_cst, failure = acquire) align(4) : (!s32i, !cir.bool) + // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} align(4) : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) // CIR-NEXT: %[[FAILED:.+]] = cir.unary(not, %[[SUCCESS]]) : !cir.bool, !cir.bool // CIR-NEXT: cir.if %[[FAILED]] { // CIR-NEXT: cir.store align(4) %[[OLD]], %{{.+}} : !s32i, !cir.ptr<!s32i> @@ -249,7 +249,7 @@ void c11_atomic_cmpxchg_weak(_Atomic(int) *ptr, int *expected, int desired) { __c11_atomic_compare_exchange_weak(ptr, expected, desired, __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE); - // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg(%{{.+}} : !cir.ptr<!s32i>, %{{.+}} : !s32i, %{{.+}} : !s32i, success = seq_cst, failure = acquire) align(4) weak : (!s32i, !cir.bool) + // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg weak success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} align(4) : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) // CIR-NEXT: %[[FAILED:.+]] = cir.unary(not, %[[SUCCESS]]) : !cir.bool, !cir.bool // CIR-NEXT: cir.if %[[FAILED]] { // CIR-NEXT: cir.store align(4) %[[OLD]], %{{.+}} : !s32i, !cir.ptr<!s32i> @@ -286,7 +286,7 @@ void atomic_cmpxchg(int *ptr, int *expected, int *desired) { // OGCG-LABEL: @atomic_cmpxchg __atomic_compare_exchange(ptr, expected, desired, /*weak=*/0, __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE); - // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg(%{{.+}} : !cir.ptr<!s32i>, %{{.+}} : !s32i, %{{.+}} : !s32i, success = seq_cst, failure = acquire) align(4) : (!s32i, !cir.bool) + // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} align(4) : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) // CIR-NEXT: %[[FAILED:.+]] = cir.unary(not, %[[SUCCESS]]) : !cir.bool, !cir.bool // CIR-NEXT: cir.if %[[FAILED]] { // CIR-NEXT: cir.store align(4) %[[OLD]], %{{.+}} : !s32i, !cir.ptr<!s32i> @@ -317,7 +317,7 @@ void atomic_cmpxchg(int *ptr, int *expected, int *desired) { // OGCG-NEXT: store i8 %[[SUCCESS_2]], ptr %{{.+}}, align 1 __atomic_compare_exchange(ptr, expected, desired, /*weak=*/1, __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE); - // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg(%{{.+}} : !cir.ptr<!s32i>, %{{.+}} : !s32i, %{{.+}} : !s32i, success = seq_cst, failure = acquire) align(4) weak : (!s32i, !cir.bool) + // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg weak success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} align(4) : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) // CIR-NEXT: %[[FAILED:.+]] = cir.unary(not, %[[SUCCESS]]) : !cir.bool, !cir.bool // CIR-NEXT: cir.if %[[FAILED]] { // CIR-NEXT: cir.store align(4) %[[OLD]], %{{.+}} : !s32i, !cir.ptr<!s32i> @@ -354,7 +354,7 @@ void atomic_cmpxchg_n(int *ptr, int *expected, int desired) { // OGCG-LABEL: @atomic_cmpxchg_n __atomic_compare_exchange_n(ptr, expected, desired, /*weak=*/0, __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE); - // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg(%{{.+}} : !cir.ptr<!s32i>, %{{.+}} : !s32i, %{{.+}} : !s32i, success = seq_cst, failure = acquire) align(4) : (!s32i, !cir.bool) + // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} align(4) : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) // CIR-NEXT: %[[FAILED:.+]] = cir.unary(not, %[[SUCCESS]]) : !cir.bool, !cir.bool // CIR-NEXT: cir.if %[[FAILED]] { // CIR-NEXT: cir.store align(4) %[[OLD]], %{{.+}} : !s32i, !cir.ptr<!s32i> @@ -385,7 +385,7 @@ void atomic_cmpxchg_n(int *ptr, int *expected, int desired) { // OGCG-NEXT: store i8 %[[SUCCESS_2]], ptr %{{.+}}, align 1 __atomic_compare_exchange_n(ptr, expected, desired, /*weak=*/1, __ATOMIC_SEQ_CST, __ATOMIC_ACQUIRE); - // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg(%{{.+}} : !cir.ptr<!s32i>, %{{.+}} : !s32i, %{{.+}} : !s32i, success = seq_cst, failure = acquire) align(4) weak : (!s32i, !cir.bool) + // CIR: %[[OLD:.+]], %[[SUCCESS:.+]] = cir.atomic.cmpxchg weak success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} align(4) : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) // CIR-NEXT: %[[FAILED:.+]] = cir.unary(not, %[[SUCCESS]]) : !cir.bool, !cir.bool // CIR-NEXT: cir.if %[[FAILED]] { // CIR-NEXT: cir.store align(4) %[[OLD]], %{{.+}} : !s32i, !cir.ptr<!s32i> @@ -427,12 +427,12 @@ void c11_atomic_exchange(_Atomic(int) *ptr, int value) { __c11_atomic_exchange(ptr, value, __ATOMIC_RELEASE); __c11_atomic_exchange(ptr, value, __ATOMIC_ACQ_REL); __c11_atomic_exchange(ptr, value, __ATOMIC_SEQ_CST); - // CIR: %{{.+}} = cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg consume %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg acquire %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg release %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg consume %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg acquire %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg release %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i // LLVM: %{{.+}} = atomicrmw xchg ptr %{{.+}}, i32 %{{.+}} monotonic, align 4 // LLVM: %{{.+}} = atomicrmw xchg ptr %{{.+}}, i32 %{{.+}} acquire, align 4 @@ -460,12 +460,12 @@ void atomic_exchange(int *ptr, int *value, int *old) { __atomic_exchange(ptr, value, old, __ATOMIC_RELEASE); __atomic_exchange(ptr, value, old, __ATOMIC_ACQ_REL); __atomic_exchange(ptr, value, old, __ATOMIC_SEQ_CST); - // CIR: %{{.+}} = cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg consume %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg acquire %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg release %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg consume %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg acquire %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg release %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i // LLVM: %{{.+}} = atomicrmw xchg ptr %{{.+}}, i32 %{{.+}} monotonic, align 4 // LLVM: %{{.+}} = atomicrmw xchg ptr %{{.+}}, i32 %{{.+}} acquire, align 4 @@ -493,12 +493,12 @@ void atomic_exchange_n(int *ptr, int value) { __atomic_exchange_n(ptr, value, __ATOMIC_RELEASE); __atomic_exchange_n(ptr, value, __ATOMIC_ACQ_REL); __atomic_exchange_n(ptr, value, __ATOMIC_SEQ_CST); - // CIR: %{{.+}} = cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg consume %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg acquire %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg release %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - // CIR: %{{.+}} = cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg consume %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg acquire %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg release %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CIR: %{{.+}} = cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i // LLVM: %{{.+}} = atomicrmw xchg ptr %{{.+}}, i32 %{{.+}} monotonic, align 4 // LLVM: %{{.+}} = atomicrmw xchg ptr %{{.+}}, i32 %{{.+}} acquire, align 4 diff --git a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-default-ops.cpp b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-default-ops.cpp index 040ddd3..ee4fffe 100644 --- a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-default-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-default-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct DefaultOperators { int i; @@ -43,12 +43,43 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(*:someVar) - // CHECK-NEXT: acc.reduction.recipe @reduction_mul__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_DefaultOperators, !cir.ptr<!rec_DefaultOperators>, ["openacc.reduction.init", init] @@ -71,10 +102,42 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(max:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -101,7 +164,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(min:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_min__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -128,7 +191,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&:someVarNoFloats) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__ZTS24DefaultOperatorsNoFloats : !cir.ptr<!rec_DefaultOperatorsNoFloats> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats>{{.*}}) @@ -146,7 +209,27 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } for(int i = 0; i < 5; ++i); @@ -167,7 +250,27 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } for(int i = 0; i < 5; ++i); @@ -188,10 +291,30 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&&:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -218,7 +341,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(||:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -245,7 +368,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(+:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_add__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <add> init { @@ -286,10 +409,65 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(*:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_mul__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) @@ -388,11 +566,67 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(max:someVarArr) + // CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_DefaultOperators x 5>, !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>, ["openacc.reduction.init", init] @@ -493,7 +727,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(min:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_min__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) @@ -664,10 +898,53 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(|:someVarArrNoFloats) // CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}) @@ -702,7 +979,50 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } for(int i = 0; i < 5; ++i); @@ -739,10 +1059,53 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&&:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) @@ -844,9 +1207,9 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(||:someVarArr) -// CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { +// CHECK: acc.reduction.recipe @reduction_lor__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_DefaultOperators x 5>, !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[TEMP_ITR:.*]] = cir.alloca !cir.ptr<!rec_DefaultOperators>, !cir.ptr<!cir.ptr<!rec_DefaultOperators>>, ["arrayinit.temp"] @@ -888,7 +1251,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(+:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_add__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <add> init { @@ -935,9 +1298,67 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(*:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_mul__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -983,9 +1404,67 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(max:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_max__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1033,7 +1512,7 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(min:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_min__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1081,7 +1560,7 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&:someVarArrNoFloats[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__Bcnt1__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1121,9 +1600,55 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(|:someVarArrNoFloats[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_ior__Bcnt1__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1163,9 +1688,55 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(^:someVarArrNoFloats[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_xor__Bcnt1__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1205,9 +1776,55 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&&:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_land__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1255,7 +1872,7 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(||:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_lor__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1303,26 +1920,26 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(+:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(*:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(max:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(min:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&:someVarArrNoFloats[1:1]) for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(|:someVarArrNoFloats[1:1]) for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(^:someVarArrNoFloats[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(&&:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc parallel loop reduction(||:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); // CHECK-NEXT: cir.func {{.*}}@_Z12acc_combined } diff --git a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-float.cpp b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-float.cpp index 6e885cc..472e4ac 100644 --- a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-float.cpp +++ b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-float.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s template<typename T> void acc_combined() { T someVar; @@ -13,7 +13,10 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -26,7 +29,10 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } @@ -111,7 +117,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -142,7 +172,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -302,6 +356,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -337,6 +420,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -372,6 +484,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -407,6 +520,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -442,6 +556,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -477,6 +592,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); diff --git a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-inline-ops.cpp b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-inline-ops.cpp index 3d46ac7..112ff656 100644 --- a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-inline-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-inline-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct HasOperatorsInline { int i; @@ -48,7 +48,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -79,7 +79,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -88,7 +88,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(max:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <max> init { +// CHECK: acc.reduction.recipe @reduction_max__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -172,7 +172,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -181,7 +181,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(|:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <ior> init { +// CHECK: acc.reduction.recipe @reduction_ior__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -203,7 +203,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -212,7 +212,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(^:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <xor> init { +// CHECK: acc.reduction.recipe @reduction_xor__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -234,7 +234,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -243,7 +243,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(&&:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <land> init { +// CHECK: acc.reduction.recipe @reduction_land__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -344,7 +344,29 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -466,7 +488,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -832,7 +875,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -896,7 +960,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -959,7 +1044,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -1216,6 +1322,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1293,6 +1425,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1524,6 +1682,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1601,6 +1785,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1678,6 +1888,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): diff --git a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-int.cpp b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-int.cpp index e9669d3..7eaa822b 100644 --- a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-int.cpp +++ b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-int.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s template<typename T> void acc_combined() { @@ -14,7 +14,10 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -28,7 +31,10 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -67,7 +73,10 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -81,7 +90,10 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -95,7 +107,10 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } @@ -155,7 +170,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -186,7 +225,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -279,7 +342,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -309,7 +396,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -339,7 +450,31 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -437,6 +572,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -472,6 +636,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -507,6 +700,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -542,6 +736,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -577,6 +772,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -612,6 +836,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -647,6 +900,35 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -682,6 +964,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -717,6 +1000,7 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); diff --git a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-outline-ops.cpp b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-outline-ops.cpp index c99e3c1..c2c0c77 100644 --- a/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-outline-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/combined-reduction-clause-outline-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct HasOperatorsOutline { int i; unsigned u; @@ -47,7 +47,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -55,7 +55,6 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } for(int i=0;i < 5; ++i); - #pragma acc parallel loop reduction(*:someVar) // CHECK: acc.reduction.recipe @reduction_mul__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) @@ -78,17 +77,17 @@ void acc_combined() { // CHECK-NEXT: acc.yield // // CHECK-NEXT: } combiner { - for(int i=0;i < 5; ++i); -#pragma acc parallel loop reduction(max:someVar) // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): // CHECK-NEXT: cir.call @_ZN19HasOperatorsOutlineD1Ev(%[[ARG]]) nothrow : (!cir.ptr<!rec_HasOperatorsOutline>) // CHECK-NEXT: acc.yield // CHECK-NEXT: } -// CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <max> init { + for(int i=0;i < 5; ++i); +#pragma acc parallel loop reduction(max:someVar) +// CHECK: acc.reduction.recipe @reduction_max__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -172,7 +171,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -181,7 +180,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(|:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <ior> init { +// CHECK: acc.reduction.recipe @reduction_ior__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -203,7 +202,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -212,7 +211,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(^:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <xor> init { +// CHECK: acc.reduction.recipe @reduction_xor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -234,7 +233,7 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -243,7 +242,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(&&:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <land> init { +// CHECK: acc.reduction.recipe @reduction_land__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -344,7 +343,29 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -466,7 +487,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -832,7 +874,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -896,7 +959,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -959,7 +1043,28 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -982,6 +1087,7 @@ void acc_combined() { // CHECK-NEXT: } // CHECK-NEXT: acc.yield // CHECK-NEXT: } + for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(&&:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <land> init { @@ -1106,8 +1212,7 @@ void acc_combined() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc parallel loop reduction(||:someVarArr) - -// CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { +// CHECK: acc.reduction.recipe @reduction_lor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[TEMP_ITR:.*]] = cir.alloca !cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!cir.ptr<!rec_HasOperatorsOutline>>, ["arrayinit.temp"] @@ -1216,6 +1321,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1293,6 +1424,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1524,6 +1681,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1601,6 +1784,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1678,6 +1887,32 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.c b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.c index c99dc09..b439623 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.c +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.c @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -std=c23 -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -std=c23 -triple x86_64-linux-pc %s -o - | FileCheck %s struct DefaultOperators { int i; @@ -42,7 +42,39 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } ; @@ -69,7 +101,39 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } ; @@ -144,7 +208,27 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } ; @@ -165,7 +249,27 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } ; @@ -186,7 +290,27 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } ; @@ -271,7 +395,62 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -373,7 +552,62 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -650,7 +884,50 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -680,7 +957,50 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -710,7 +1030,50 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -892,6 +1255,64 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -940,6 +1361,64 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -1056,7 +1535,6 @@ void acc_compute() { // CHECK-NEXT: cir.condition(%[[COND]]) // CHECK-NEXT: } body { // CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i - // CHECK-NEXT: %[[DECAY:.*]] = cir.cast array_to_ptrdecay %[[ALLOCA]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: %[[STRIDE:.*]] = cir.ptr_stride %[[DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> @@ -1079,6 +1557,52 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -1121,6 +1645,52 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -1163,6 +1733,52 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.cpp b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.cpp index b90a2fc..f997902 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-default-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct DefaultOperators { int i; @@ -21,7 +21,6 @@ void acc_compute() { struct DefaultOperatorsNoFloats someVarNoFloats; struct DefaultOperatorsNoFloats someVarArrNoFloats[5]; #pragma acc parallel reduction(+:someVar) - ; // CHECK: acc.reduction.recipe @reduction_add__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <add> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_DefaultOperators, !cir.ptr<!rec_DefaultOperators>, ["openacc.reduction.init", init] @@ -44,11 +43,43 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } + ; #pragma acc parallel reduction(*:someVar) - // CHECK-NEXT: acc.reduction.recipe @reduction_mul__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_DefaultOperators, !cir.ptr<!rec_DefaultOperators>, ["openacc.reduction.init", init] @@ -71,7 +102,39 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } ; @@ -146,7 +209,27 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } ; @@ -167,7 +250,27 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } ; @@ -188,7 +291,27 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } ; @@ -286,7 +409,62 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -388,11 +566,67 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; #pragma acc parallel reduction(max:someVarArr) + // CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_DefaultOperators x 5>, !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>, ["openacc.reduction.init", init] @@ -664,7 +898,50 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -702,7 +979,50 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -739,7 +1059,50 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -846,7 +1209,7 @@ void acc_compute() { // CHECK-NEXT: } ; #pragma acc parallel reduction(||:someVarArr) -// CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { +// CHECK: acc.reduction.recipe @reduction_lor__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_DefaultOperators x 5>, !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[TEMP_ITR:.*]] = cir.alloca !cir.ptr<!rec_DefaultOperators>, !cir.ptr<!cir.ptr<!rec_DefaultOperators>>, ["arrayinit.temp"] @@ -935,6 +1298,64 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -983,6 +1404,64 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } ; @@ -1121,6 +1600,52 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -1163,6 +1688,52 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -1205,6 +1776,52 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } ; @@ -1323,9 +1940,11 @@ void acc_compute() { ; #pragma acc parallel reduction(||:someVarArr[1:1]) ; + // CHECK-NEXT: cir.func {{.*}}@_Z11acc_compute } void uses() { acc_compute<DefaultOperators>(); } + diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.c b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.c index 0f7fd84..3e4583f 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.c +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.c @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s void acc_compute() { float someVar; @@ -13,7 +13,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } ; @@ -26,7 +29,10 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } ; @@ -110,7 +116,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -141,7 +171,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -301,6 +355,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -336,6 +419,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -371,6 +483,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -406,6 +519,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -441,6 +555,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -476,6 +591,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.cpp b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.cpp index 4d99a43..833cfad 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.cpp +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-float.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s template<typename T> void acc_compute() { @@ -14,7 +14,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } ; @@ -27,7 +30,10 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } @@ -112,7 +118,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -143,7 +173,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -303,6 +357,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -338,6 +421,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -373,6 +485,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -443,6 +556,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; @@ -478,6 +592,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } ; diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-inline-ops.cpp b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-inline-ops.cpp index ea00c07..ec4372d 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-inline-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-inline-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct HasOperatorsInline { int i; @@ -48,7 +48,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -79,7 +79,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -88,7 +88,7 @@ void acc_compute() { // CHECK-NEXT: } ; #pragma acc parallel reduction(max:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <max> init { +// CHECK: acc.reduction.recipe @reduction_max__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -172,7 +172,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -181,7 +181,7 @@ void acc_compute() { // CHECK-NEXT: } ; #pragma acc parallel reduction(|:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <ior> init { +// CHECK: acc.reduction.recipe @reduction_ior__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -203,7 +203,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -212,7 +212,7 @@ void acc_compute() { // CHECK-NEXT: } ; #pragma acc parallel reduction(^:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <xor> init { +// CHECK: acc.reduction.recipe @reduction_xor__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -234,7 +234,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -243,7 +243,7 @@ void acc_compute() { // CHECK-NEXT: } ; #pragma acc parallel reduction(&&:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <land> init { +// CHECK: acc.reduction.recipe @reduction_land__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -344,7 +344,29 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -466,7 +488,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -832,7 +875,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -896,7 +960,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -959,7 +1044,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -1216,6 +1322,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1293,6 +1425,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1524,6 +1682,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1601,6 +1785,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1678,6 +1888,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.c b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.c index b170ed0bf..0cee5c6 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.c +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.c @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s void acc_compute() { int someVar; @@ -13,7 +13,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -27,7 +30,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -66,7 +72,10 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -80,7 +89,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -94,9 +106,13 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } + ; #pragma acc parallel reduction(&&:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTSi : !cir.ptr<!s32i> reduction_operator <land> init { @@ -153,7 +169,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -184,7 +224,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -277,7 +341,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -307,7 +395,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -337,7 +449,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -435,6 +571,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -470,6 +635,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -505,6 +699,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -540,6 +735,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -575,6 +771,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -610,6 +835,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -645,6 +899,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -680,6 +963,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -715,6 +999,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.cpp b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.cpp index c678eae..822dd9f6 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.cpp +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-int.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s template<typename T> void acc_compute() { @@ -14,7 +14,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -28,7 +31,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -67,7 +73,10 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -81,7 +90,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } ; @@ -95,7 +107,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } @@ -155,7 +170,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -186,7 +225,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -279,7 +342,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -309,7 +396,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -339,7 +450,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -437,6 +572,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -472,6 +636,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -507,6 +700,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -542,6 +736,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -577,6 +772,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -612,6 +836,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -647,6 +900,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -682,6 +964,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; @@ -717,6 +1000,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } ; diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-outline-ops.cpp b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-outline-ops.cpp index 9ccaea2..873bf51 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-outline-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-outline-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct HasOperatorsOutline { int i; unsigned u; @@ -24,7 +24,7 @@ template<typename T> void acc_compute() { T someVar; T someVarArr[5]; -#pragma acc parallel reduction(+:someVar) +#pragma acc parallel reduction(+:someVar) // CHECK: acc.reduction.recipe @reduction_add__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <add> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] @@ -47,7 +47,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -55,8 +55,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; - -#pragma acc parallel reduction(*:someVar) +#pragma acc parallel reduction(*:someVar) // CHECK: acc.reduction.recipe @reduction_mul__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] @@ -78,17 +77,17 @@ void acc_compute() { // CHECK-NEXT: acc.yield // // CHECK-NEXT: } combiner { - ; -#pragma acc parallel reduction(max:someVar) // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): // CHECK-NEXT: cir.call @_ZN19HasOperatorsOutlineD1Ev(%[[ARG]]) nothrow : (!cir.ptr<!rec_HasOperatorsOutline>) // CHECK-NEXT: acc.yield // CHECK-NEXT: } -// CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <max> init { + ; +#pragma acc parallel reduction(max:someVar) +// CHECK: acc.reduction.recipe @reduction_max__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -118,7 +117,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(min:someVar) +#pragma acc parallel reduction(min:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_min__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] @@ -149,7 +148,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(&:someVar) +#pragma acc parallel reduction(&:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] @@ -172,7 +171,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -180,8 +179,8 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(|:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <ior> init { +#pragma acc parallel reduction(|:someVar) +// CHECK: acc.reduction.recipe @reduction_ior__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -203,7 +202,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -211,8 +210,8 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(^:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <xor> init { +#pragma acc parallel reduction(^:someVar) +// CHECK: acc.reduction.recipe @reduction_xor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -234,7 +233,7 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -242,8 +241,8 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(&&:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <land> init { +#pragma acc parallel reduction(&&:someVar) +// CHECK: acc.reduction.recipe @reduction_land__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -273,7 +272,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(||:someVar) +#pragma acc parallel reduction(||:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] @@ -305,7 +304,7 @@ void acc_compute() { // CHECK-NEXT: } ; -#pragma acc parallel reduction(+:someVarArr) +#pragma acc parallel reduction(+:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_add__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <add> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -344,7 +343,29 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -368,7 +389,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(*:someVarArr) +#pragma acc parallel reduction(*:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_mul__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -466,7 +487,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -490,7 +532,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(max:someVarArr) +#pragma acc parallel reduction(max:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -612,7 +654,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(min:someVarArr) +#pragma acc parallel reduction(min:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_min__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -734,7 +776,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(&:someVarArr) +#pragma acc parallel reduction(&:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -832,7 +874,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -856,7 +919,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(|:someVarArr) +#pragma acc parallel reduction(|:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -896,7 +959,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -920,7 +1004,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(^:someVarArr) +#pragma acc parallel reduction(^:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -959,7 +1043,28 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -982,8 +1087,9 @@ void acc_compute() { // CHECK-NEXT: } // CHECK-NEXT: acc.yield // CHECK-NEXT: } + ; -#pragma acc parallel reduction(&&:someVarArr) +#pragma acc parallel reduction(&&:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] @@ -1105,9 +1211,8 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(||:someVarArr) - -// CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { +#pragma acc parallel reduction(||:someVarArr) +// CHECK: acc.reduction.recipe @reduction_lor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[TEMP_ITR:.*]] = cir.alloca !cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!cir.ptr<!rec_HasOperatorsOutline>>, ["arrayinit.temp"] @@ -1171,7 +1276,7 @@ void acc_compute() { // CHECK-NEXT: } ; -#pragma acc parallel reduction(+:someVarArr[2]) +#pragma acc parallel reduction(+:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_add__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <add> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1216,6 +1321,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1248,7 +1379,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(*:someVarArr[2]) +#pragma acc parallel reduction(*:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_mul__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1293,6 +1424,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1325,7 +1482,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(max:someVarArr[2]) +#pragma acc parallel reduction(max:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_max__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1402,7 +1559,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(min:someVarArr[2]) +#pragma acc parallel reduction(min:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_min__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1479,7 +1636,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(&:someVarArr[2]) +#pragma acc parallel reduction(&:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1524,6 +1681,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1556,7 +1739,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(|:someVarArr[2]) +#pragma acc parallel reduction(|:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_ior__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1601,6 +1784,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1633,7 +1842,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(^:someVarArr[2]) +#pragma acc parallel reduction(^:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_xor__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1678,6 +1887,32 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1710,7 +1945,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(&&:someVarArr[2]) +#pragma acc parallel reduction(&&:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_land__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] @@ -1787,7 +2022,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } ; -#pragma acc parallel reduction(||:someVarArr[2]) +#pragma acc parallel reduction(||:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_lor__Bcnt1__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init"] diff --git a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-unsigned-int.c b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-unsigned-int.c index 783aa9a..b2d1362 100644 --- a/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-unsigned-int.c +++ b/clang/test/CIR/CodeGenOpenACC/compute-reduction-clause-unsigned-int.c @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s void acc_compute() { unsigned int someVar; @@ -13,7 +13,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!u32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!u32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!u32i> // CHECK-NEXT: } ; @@ -27,7 +30,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!u32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!u32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!u32i> // CHECK-NEXT: } ; @@ -66,7 +72,10 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!u32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!u32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!u32i> // CHECK-NEXT: } ; @@ -80,7 +89,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!u32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!u32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!u32i> // CHECK-NEXT: } ; @@ -94,7 +106,10 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!u32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!u32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!u32i> +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!u32i> // CHECK-NEXT: } ; @@ -153,7 +168,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -184,7 +223,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -277,7 +340,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -307,7 +394,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -337,7 +448,31 @@ void acc_compute() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !s64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -435,6 +570,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -470,6 +634,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -505,6 +698,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -540,6 +734,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -575,6 +770,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -610,6 +834,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -645,6 +898,35 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!u32i>, !u64i) -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -680,6 +962,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; @@ -715,6 +998,7 @@ void acc_compute() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!u32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!u32i x 5>> // CHECK-NEXT: } ; diff --git a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-default-ops.cpp b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-default-ops.cpp index 038afcaa..349e0fb 100644 --- a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-default-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-default-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct DefaultOperators { int i; @@ -15,7 +15,7 @@ struct DefaultOperatorsNoFloats { }; template<typename T> -void acc_combined() { +void acc_loop() { T someVar; T someVarArr[5]; struct DefaultOperatorsNoFloats someVarNoFloats; @@ -43,12 +43,43 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(*:someVar) - // CHECK-NEXT: acc.reduction.recipe @reduction_mul__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_DefaultOperators, !cir.ptr<!rec_DefaultOperators>, ["openacc.reduction.init", init] @@ -71,10 +102,42 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperators> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(max:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -101,7 +164,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(min:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_min__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -128,7 +191,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&:someVarNoFloats) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__ZTS24DefaultOperatorsNoFloats : !cir.ptr<!rec_DefaultOperatorsNoFloats> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats>{{.*}}) @@ -146,7 +209,27 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } for(int i = 0; i < 5; ++i); @@ -167,7 +250,27 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } for(int i = 0; i < 5; ++i); @@ -188,10 +291,30 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_DefaultOperatorsNoFloats> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHSARG]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperatorsNoFloats> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&&:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -218,7 +341,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(||:someVar) // CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTS16DefaultOperators : !cir.ptr<!rec_DefaultOperators> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_DefaultOperators>{{.*}}) @@ -245,7 +368,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_DefaultOperators> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(+:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_add__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <add> init { @@ -286,10 +409,65 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(*:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_mul__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) @@ -388,11 +566,67 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !s64i) -> !cir.ptr<!rec_DefaultOperators> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(max:someVarArr) + // CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_DefaultOperators x 5>, !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>, ["openacc.reduction.init", init] @@ -493,7 +727,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(min:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_min__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) @@ -664,10 +898,53 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(|:someVarArrNoFloats) // CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}) @@ -702,7 +979,50 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } for(int i = 0; i < 5; ++i); @@ -739,10 +1059,53 @@ void acc_combined() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !s64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&&:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) @@ -844,9 +1207,9 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(||:someVarArr) -// CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { +// CHECK: acc.reduction.recipe @reduction_lor__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_DefaultOperators x 5>, !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[TEMP_ITR:.*]] = cir.alloca !cir.ptr<!rec_DefaultOperators>, !cir.ptr<!cir.ptr<!rec_DefaultOperators>>, ["arrayinit.temp"] @@ -888,7 +1251,7 @@ void acc_combined() { // TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(+:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_add__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <add> init { @@ -935,9 +1298,67 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[ADD]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(*:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_mul__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -983,9 +1404,67 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperators>, !u64i) -> !cir.ptr<!rec_DefaultOperators> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "f"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.float, !cir.ptr<!cir.float> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][3] {name = "d"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.double> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.double>, !cir.double +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.double +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[GET_MEM_LHS]] : !cir.double, !cir.ptr<!cir.double> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][4] {name = "b"} : !cir.ptr<!rec_DefaultOperators> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) nsw : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[MUL]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(max:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_max__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1033,7 +1512,7 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(min:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_min__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <min> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1081,7 +1560,7 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&:someVarArrNoFloats[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_iand__Bcnt1__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <iand> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1121,9 +1600,55 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[AND]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(|:someVarArrNoFloats[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_ior__Bcnt1__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1163,9 +1688,55 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[OR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(^:someVarArrNoFloats[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_xor__Bcnt1__ZTSA5_24DefaultOperatorsNoFloats : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1205,9 +1776,55 @@ void acc_combined() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_DefaultOperatorsNoFloats>, !u64i) -> !cir.ptr<!rec_DefaultOperatorsNoFloats> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !s32i, !cir.ptr<!s32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][1] {name = "u"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!u32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!u32i>, !u32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !u32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[GET_MEM_LHS]] : !u32i, !cir.ptr<!u32i> +// CHECK-NEXT: %[[GET_MEM_LHS:.*]] = cir.get_member %[[LHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[GET_MEM_RHS:.*]] = cir.get_member %[[RHS_STRIDE]][2] {name = "b"} : !cir.ptr<!rec_DefaultOperatorsNoFloats> -> !cir.ptr<!cir.bool> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_RHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[RHS_INT_CAST:.*]] = cir.cast bool_to_int %[[RHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load{{.*}} %[[GET_MEM_LHS]] : !cir.ptr<!cir.bool>, !cir.bool +// CHECK-NEXT: %[[LHS_INT_CAST:.*]] = cir.cast bool_to_int %[[LHS_LOAD]] : !cir.bool -> !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_INT_CAST]], %[[RHS_INT_CAST]]) : !s32i +// CHECK-NEXT: %[[RES_TO_BOOL_CAST:.*]] = cir.cast int_to_bool %[[XOR]] : !s32i -> !cir.bool +// CHECK-NEXT: cir.store {{.*}} %[[RES_TO_BOOL_CAST]], %[[GET_MEM_LHS]] : !cir.bool, !cir.ptr<!cir.bool> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperatorsNoFloats x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&&:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_land__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1255,7 +1872,7 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(||:someVarArr[2]) // CHECK-NEXT: acc.reduction.recipe @reduction_lor__Bcnt1__ZTSA5_16DefaultOperators : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>>{{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) @@ -1303,29 +1920,29 @@ void acc_combined() { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_DefaultOperators x 5>> // CHECK-NEXT: } - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(+:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(*:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(max:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(min:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&:someVarArrNoFloats[1:1]) for(int i = 0; i < 5; ++i); #pragma acc loop reduction(|:someVarArrNoFloats[1:1]) for(int i = 0; i < 5; ++i); #pragma acc loop reduction(^:someVarArrNoFloats[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(&&:someVarArr[1:1]) - for(int i=0;i < 5; ++i); + for(int i = 0; i < 5; ++i); #pragma acc loop reduction(||:someVarArr[1:1]) - for(int i=0;i < 5; ++i); - // CHECK-NEXT: cir.func {{.*}}@_Z12acc_combined + for(int i = 0; i < 5; ++i); + // CHECK-NEXT: cir.func {{.*}}@_Z8acc_loop } void uses() { - acc_combined<DefaultOperators>(); + acc_loop<DefaultOperators>(); } diff --git a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-float.cpp b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-float.cpp index 11b7c35..8d9269b 100644 --- a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-float.cpp +++ b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-float.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s template<typename T> void acc_loop() { @@ -14,7 +14,10 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -27,7 +30,10 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.float> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!cir.float> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.float> // CHECK-NEXT: } @@ -112,7 +118,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -143,7 +173,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !s64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -303,6 +357,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -338,6 +421,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!cir.float>, !u64i) -> !cir.ptr<!cir.float> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!cir.float>, !cir.float +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !cir.float +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -373,6 +485,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -408,6 +521,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -443,6 +557,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -478,6 +593,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.float x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!cir.float x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); diff --git a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-inline-ops.cpp b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-inline-ops.cpp index d95da8c..1c89515 100644 --- a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-inline-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-inline-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct HasOperatorsInline { int i; @@ -48,7 +48,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -79,7 +79,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -88,7 +88,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(max:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <max> init { +// CHECK: acc.reduction.recipe @reduction_max__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -172,7 +172,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -181,7 +181,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(|:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <ior> init { +// CHECK: acc.reduction.recipe @reduction_ior__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -203,7 +203,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -212,7 +212,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(^:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <xor> init { +// CHECK: acc.reduction.recipe @reduction_xor__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -234,7 +234,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsInline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline> {{.*}}): @@ -243,7 +243,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(&&:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <land> init { +// CHECK: acc.reduction.recipe @reduction_land__ZTS18HasOperatorsInline : !cir.ptr<!rec_HasOperatorsInline> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsInline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsInline, !cir.ptr<!rec_HasOperatorsInline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsInline> -> !cir.ptr<!s32i> @@ -344,7 +344,29 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -466,7 +488,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -832,7 +875,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -896,7 +960,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -959,7 +1044,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !s64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}): @@ -1216,6 +1322,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinepLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1293,6 +1425,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlinemLERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1524,6 +1682,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineaNERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1601,6 +1785,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineoRERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1678,6 +1888,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsInline>, !u64i) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.call @_ZN18HasOperatorsInlineeOERS_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsInline>, !cir.ptr<!rec_HasOperatorsInline>) -> !cir.ptr<!rec_HasOperatorsInline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsInline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): diff --git a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-int.cpp b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-int.cpp index d207e07b..72e9d1f 100644 --- a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-int.cpp +++ b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-int.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s template<typename T> void acc_loop() { @@ -14,7 +14,10 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -28,7 +31,10 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -67,7 +73,10 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -81,7 +90,10 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -95,7 +107,10 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!s32i> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!s32i> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHSARG]] : !cir.ptr<!s32i> +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHSARG]] // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!s32i> // CHECK-NEXT: } @@ -155,7 +170,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -186,7 +225,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -279,7 +342,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -309,7 +396,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -339,7 +450,31 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !s64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -437,6 +572,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[ADD]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -472,6 +636,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[MUL]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -507,6 +700,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -542,6 +736,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -577,6 +772,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[AND:.*]] = cir.binop(and, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[AND]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -612,6 +836,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[OR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -647,6 +900,35 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!s32i>, !u64i) -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_STRIDE]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[XOR:.*]] = cir.binop(xor, %[[LHS_LOAD]], %[[RHS_LOAD]]) : !s32i +// CHECK-NEXT: cir.store {{.*}} %[[XOR]], %[[LHS_STRIDE]] +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -682,6 +964,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); @@ -717,6 +1000,7 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!s32i x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// TODO OpenACC: Expecting combination operation here // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!s32i x 5>> // CHECK-NEXT: } for(int i=0;i < 5; ++i); diff --git a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-outline-ops.cpp b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-outline-ops.cpp index a33c25a..a36d41c1 100644 --- a/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-outline-ops.cpp +++ b/clang/test/CIR/CodeGenOpenACC/loop-reduction-clause-outline-ops.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s struct HasOperatorsOutline { int i; unsigned u; @@ -47,7 +47,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -55,7 +55,6 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } for(int i=0;i < 5; ++i); - #pragma acc loop reduction(*:someVar) // CHECK: acc.reduction.recipe @reduction_mul__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <mul> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) @@ -78,17 +77,17 @@ void acc_loop() { // CHECK-NEXT: acc.yield // // CHECK-NEXT: } combiner { - for(int i=0;i < 5; ++i); -#pragma acc loop reduction(max:someVar) // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): // CHECK-NEXT: cir.call @_ZN19HasOperatorsOutlineD1Ev(%[[ARG]]) nothrow : (!cir.ptr<!rec_HasOperatorsOutline>) // CHECK-NEXT: acc.yield // CHECK-NEXT: } -// CHECK-NEXT: acc.reduction.recipe @reduction_max__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <max> init { + for(int i=0;i < 5; ++i); +#pragma acc loop reduction(max:someVar) +// CHECK: acc.reduction.recipe @reduction_max__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <max> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -172,7 +171,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -181,7 +180,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(|:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_ior__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <ior> init { +// CHECK: acc.reduction.recipe @reduction_ior__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <ior> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -203,7 +202,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -212,7 +211,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(^:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_xor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <xor> init { +// CHECK: acc.reduction.recipe @reduction_xor__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <xor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -234,7 +233,7 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHSARG]], %[[RHSARG]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!rec_HasOperatorsOutline> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}, %[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline> {{.*}}): @@ -243,7 +242,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(&&:someVar) -// CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <land> init { +// CHECK: acc.reduction.recipe @reduction_land__ZTS19HasOperatorsOutline : !cir.ptr<!rec_HasOperatorsOutline> reduction_operator <land> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!rec_HasOperatorsOutline>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !rec_HasOperatorsOutline, !cir.ptr<!rec_HasOperatorsOutline>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[GET_I:.*]] = cir.get_member %[[ALLOCA]][0] {name = "i"} : !cir.ptr<!rec_HasOperatorsOutline> -> !cir.ptr<!s32i> @@ -344,7 +343,29 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -466,7 +487,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -832,7 +874,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -896,7 +959,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -959,7 +1043,28 @@ void acc_loop() { // // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}) -// TODO OpenACC: Expecting combination operation here +// CHECK-NEXT: %[[ZERO:.*]] = cir.const #cir.int<0> : !s64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !s64i, !cir.ptr<!s64i>, ["itr"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[ZERO]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[END_VAL:.*]] = cir.const #cir.int<5> : !s64i +// CHECK-NEXT: %[[CMP:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[END_VAL]]) : !s64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[CMP]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !s64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!s64i>, !s64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !s64i, !s64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !s64i, !cir.ptr<!s64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}): @@ -982,6 +1087,7 @@ void acc_loop() { // CHECK-NEXT: } // CHECK-NEXT: acc.yield // CHECK-NEXT: } + for(int i=0;i < 5; ++i); #pragma acc loop reduction(&&:someVarArr) // CHECK-NEXT: acc.reduction.recipe @reduction_land__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <land> init { @@ -1106,8 +1212,7 @@ void acc_loop() { // CHECK-NEXT: } for(int i=0;i < 5; ++i); #pragma acc loop reduction(||:someVarArr) - -// CHECK-NEXT: acc.reduction.recipe @reduction_lor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { +// CHECK: acc.reduction.recipe @reduction_lor__ZTSA5_19HasOperatorsOutline : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> reduction_operator <lor> init { // CHECK-NEXT: ^bb0(%[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>{{.*}}) // CHECK-NEXT: %[[ALLOCA:.*]] = cir.alloca !cir.array<!rec_HasOperatorsOutline x 5>, !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>>, ["openacc.reduction.init", init] // CHECK-NEXT: %[[TEMP_ITR:.*]] = cir.alloca !cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!cir.ptr<!rec_HasOperatorsOutline>>, ["arrayinit.temp"] @@ -1216,6 +1321,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZpLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1293,6 +1424,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZmLR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1524,6 +1681,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZaNR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1601,6 +1784,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZoRR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): @@ -1678,6 +1887,32 @@ void acc_loop() { // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { // CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}})) +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB]] : index to !u64i +// CHECK-NEXT: %[[UB:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB]] : index to !u64i +// CHECK-NEXT: %[[ITR:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB_CAST]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR_LOAD]], %[[UB_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR_LOAD:.*]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_DECAY]], %[[ITR_LOAD]] : (!cir.ptr<!rec_HasOperatorsOutline>, !u64i) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.call @_ZeOR19HasOperatorsOutlineS0_(%[[LHS_STRIDE]], %[[RHS_STRIDE]]) : (!cir.ptr<!rec_HasOperatorsOutline>, !cir.ptr<!rec_HasOperatorsOutline>) -> !cir.ptr<!rec_HasOperatorsOutline> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR_LOAD]] = cir.load %[[ITR]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield %[[LHSARG]] : !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[ORIG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[ARG:.*]]: !cir.ptr<!cir.array<!rec_HasOperatorsOutline x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty{{.*}}): diff --git a/clang/test/CIR/CodeGenOpenACC/reduction-clause-recipes.cpp b/clang/test/CIR/CodeGenOpenACC/reduction-clause-recipes.cpp index fc696ff..20ad7a3 100644 --- a/clang/test/CIR/CodeGenOpenACC/reduction-clause-recipes.cpp +++ b/clang/test/CIR/CodeGenOpenACC/reduction-clause-recipes.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s +// RUN: not %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s // Note: unlike the 'private' recipe checks, this is just for spot-checking, // so this test isn't as comprehensive. The same code paths are used for @@ -90,7 +90,88 @@ void do_things(unsigned A, unsigned B) { // CHECK-NEXT: } // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { -// CHECK-NEXT: ^bb0(%[[REF:.*]]: !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> {{.*}}, %[[PRIVATE:.*]]: !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND2:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND3:.*]]: !acc.data_bounds_ty {{.*}}): +// CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND2:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND3:.*]]: !acc.data_bounds_ty {{.*}}): +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB3:.*]] = acc.get_lowerbound %[[BOUND3]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB3_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB3]] : index to !u64i +// CHECK-NEXT: %[[UB3:.*]] = acc.get_upperbound %[[BOUND3]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB3_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB3]] : index to !u64i +// CHECK-NEXT: %[[ITR3:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB3_CAST]], %[[ITR3]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR3_LOAD:.*]] = cir.load %[[ITR3]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR3_LOAD]], %[[UB3_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR3_LOAD:.*]] = cir.load %[[ITR3]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_TLA_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHSARG]] : !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> -> !cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>> +// CHECK-NEXT: %[[LHS_BOUND3_STRIDE:.*]] = cir.ptr_stride %[[LHS_TLA_DECAY]], %[[ITR3_LOAD]] : (!cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>>, !u64i) -> !cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>> +// CHECK-NEXT: %[[RHS_TLA_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHSARG]] : !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> -> !cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>> +// CHECK-NEXT: %[[RHS_BOUND3_STRIDE:.*]] = cir.ptr_stride %[[RHS_TLA_DECAY]], %[[ITR3_LOAD]] : (!cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>>, !u64i) -> !cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>> +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB2:.*]] = acc.get_lowerbound %[[BOUND2]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB2_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB2]] : index to !u64i +// CHECK-NEXT: %[[UB2:.*]] = acc.get_upperbound %[[BOUND2]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB2_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB2]] : index to !u64i +// CHECK-NEXT: %[[ITR2:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB2_CAST]], %[[ITR2]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR2_LOAD:.*]] = cir.load %[[ITR2]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR2_LOAD]], %[[UB2_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR2_LOAD:.*]] = cir.load %[[ITR2]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_BOUND3_STRIDE_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHS_BOUND3_STRIDE]] : !cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>> -> !cir.ptr<!cir.array<!rec_NoOps x 5>> +// CHECK-NEXT: %[[LHS_BOUND2_STRIDE:.*]] = cir.ptr_stride %[[LHS_BOUND3_STRIDE_DECAY]], %[[ITR2_LOAD]] : (!cir.ptr<!cir.array<!rec_NoOps x 5>>, !u64i) -> !cir.ptr<!cir.array<!rec_NoOps x 5>> +// CHECK-NEXT: %[[RHS_BOUND3_STRIDE_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHS_BOUND3_STRIDE]] : !cir.ptr<!cir.array<!cir.array<!rec_NoOps x 5> x 5>> -> !cir.ptr<!cir.array<!rec_NoOps x 5>> +// CHECK-NEXT: %[[RHS_BOUND2_STRIDE:.*]] = cir.ptr_stride %[[RHS_BOUND3_STRIDE_DECAY]], %[[ITR2_LOAD]] : (!cir.ptr<!cir.array<!rec_NoOps x 5>>, !u64i) -> !cir.ptr<!cir.array<!rec_NoOps x 5>> +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB1:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB1_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB1]] : index to !u64i +// CHECK-NEXT: %[[UB1:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB1_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB1]] : index to !u64i +// CHECK-NEXT: %[[ITR1:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB1_CAST]], %[[ITR1]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR1_LOAD:.*]] = cir.load %[[ITR1]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR1_LOAD]], %[[UB1_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR1_LOAD:.*]] = cir.load %[[ITR1]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_BOUND2_STRIDE_DECAY:.*]] = cir.cast array_to_ptrdecay %[[LHS_BOUND2_STRIDE]] : !cir.ptr<!cir.array<!rec_NoOps x 5>> -> !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[LHS_BOUND1_STRIDE:.*]] = cir.ptr_stride %[[LHS_BOUND2_STRIDE_DECAY]], %[[ITR1_LOAD]] : (!cir.ptr<!rec_NoOps>, !u64i) -> !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[RHS_BOUND2_STRIDE_DECAY:.*]] = cir.cast array_to_ptrdecay %[[RHS_BOUND2_STRIDE]] : !cir.ptr<!cir.array<!rec_NoOps x 5>> -> !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[RHS_BOUND1_STRIDE:.*]] = cir.ptr_stride %[[RHS_BOUND2_STRIDE_DECAY]], %[[ITR1_LOAD]] : (!cir.ptr<!rec_NoOps>, !u64i) -> !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[LHS_GET_I:.*]] = cir.get_member %[[LHS_BOUND1_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_NoOps> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_GET_I:.*]] = cir.get_member %[[RHS_BOUND1_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_NoOps> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_GET_I]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_GET_I]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OP:.*]] = cir.binop(add, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw +// CHECK-NEXT: cir.store{{.*}} %[[OP]], %[[LHS_GET_I]] : !s32i +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR1_LOAD]] = cir.load %[[ITR1]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR1_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR1]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR2_LOAD]] = cir.load %[[ITR2]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR2_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR2]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR3_LOAD]] = cir.load %[[ITR3]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR3_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR3]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield // CHECK-NEXT:} destroy { // CHECK-NEXT: ^bb0(%[[REF:.*]]: !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> {{.*}}, %[[PRIVATE:.*]]: !cir.ptr<!cir.array<!cir.array<!cir.array<!rec_NoOps x 5> x 5> x 5>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND2:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND3:.*]]: !acc.data_bounds_ty {{.*}}): @@ -355,7 +436,89 @@ void do_things(unsigned A, unsigned B) { // // CHECK-NEXT: acc.yield // CHECK-NEXT: } combiner { -// CHECK-NEXT: ^bb0(%[[REF:.*]]: !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>> {{.*}}, %[[PRIVATE:.*]]: !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND2:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND3:.*]]: !acc.data_bounds_ty {{.*}}): +// CHECK-NEXT: ^bb0(%[[LHSARG:.*]]: !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>> {{.*}}, %[[RHSARG:.*]]: !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND2:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND3:.*]]: !acc.data_bounds_ty {{.*}}): +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB3:.*]] = acc.get_lowerbound %[[BOUND3]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB3_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB3]] : index to !u64i +// CHECK-NEXT: %[[UB3:.*]] = acc.get_upperbound %[[BOUND3]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB3_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB3]] : index to !u64i +// CHECK-NEXT: %[[ITR3:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB3_CAST]], %[[ITR3]] : !u64i, !cir.ptr<!u64i> + +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR3_LOAD:.*]] = cir.load %[[ITR3]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR3_LOAD]], %[[UB3_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR3_LOAD:.*]] = cir.load %[[ITR3]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_TLA_LOAD:.*]] = cir.load %[[LHSARG]] : !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>>, !cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>> +// CHECK-NEXT: %[[LHS_BOUND3_STRIDE:.*]] = cir.ptr_stride %[[LHS_TLA_LOAD]], %[[ITR3_LOAD]] : (!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>, !u64i) -> !cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>> +// CHECK-NEXT: %[[RHS_TLA_LOAD:.*]] = cir.load %[[RHSARG]] : !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>>, !cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>> +// CHECK-NEXT: %[[RHS_BOUND3_STRIDE:.*]] = cir.ptr_stride %[[RHS_TLA_LOAD]], %[[ITR3_LOAD]] : (!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>, !u64i) -> !cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>> +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB2:.*]] = acc.get_lowerbound %[[BOUND2]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB2_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB2]] : index to !u64i +// CHECK-NEXT: %[[UB2:.*]] = acc.get_upperbound %[[BOUND2]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB2_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB2]] : index to !u64i +// CHECK-NEXT: %[[ITR2:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB2_CAST]], %[[ITR2]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR2_LOAD:.*]] = cir.load %[[ITR2]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR2_LOAD]], %[[UB2_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR2_LOAD:.*]] = cir.load %[[ITR2]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_BOUND3_STRIDE_LOAD:.*]] = cir.load %[[LHS_BOUND3_STRIDE]] : !cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>, !cir.ptr<!cir.ptr<!rec_NoOps>> +// CHECK-NEXT: %[[LHS_BOUND2_STRIDE:.*]] = cir.ptr_stride %[[LHS_BOUND3_STRIDE_LOAD]], %[[ITR2_LOAD]] : (!cir.ptr<!cir.ptr<!rec_NoOps>>, !u64i) -> !cir.ptr<!cir.ptr<!rec_NoOps>> +// CHECK-NEXT: %[[RHS_BOUND3_STRIDE_LOAD:.*]] = cir.load %[[RHS_BOUND3_STRIDE]] : !cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>, !cir.ptr<!cir.ptr<!rec_NoOps>> +// CHECK-NEXT: %[[RHS_BOUND2_STRIDE:.*]] = cir.ptr_stride %[[RHS_BOUND3_STRIDE_LOAD]], %[[ITR2_LOAD]] : (!cir.ptr<!cir.ptr<!rec_NoOps>>, !u64i) -> !cir.ptr<!cir.ptr<!rec_NoOps>> +// CHECK-NEXT: cir.scope { +// CHECK-NEXT: %[[LB1:.*]] = acc.get_lowerbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[LB1_CAST:.*]] = builtin.unrealized_conversion_cast %[[LB1]] : index to !u64i +// CHECK-NEXT: %[[UB1:.*]] = acc.get_upperbound %[[BOUND1]] : (!acc.data_bounds_ty) -> index +// CHECK-NEXT: %[[UB1_CAST:.*]] = builtin.unrealized_conversion_cast %[[UB1]] : index to !u64i +// CHECK-NEXT: %[[ITR1:.*]] = cir.alloca !u64i, !cir.ptr<!u64i>, ["iter"] {alignment = 8 : i64} +// CHECK-NEXT: cir.store %[[LB1_CAST]], %[[ITR1]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.for : cond { +// CHECK-NEXT: %[[ITR1_LOAD:.*]] = cir.load %[[ITR1]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[COND:.*]] = cir.cmp(lt, %[[ITR1_LOAD]], %[[UB1_CAST]]) : !u64i, !cir.bool +// CHECK-NEXT: cir.condition(%[[COND]]) +// CHECK-NEXT: } body { +// CHECK-NEXT: %[[ITR1_LOAD:.*]] = cir.load %[[ITR1]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[LHS_BOUND2_STRIDE_LOAD:.*]] = cir.load %[[LHS_BOUND2_STRIDE]] : !cir.ptr<!cir.ptr<!rec_NoOps>>, !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[LHS_STRIDE:.*]] = cir.ptr_stride %[[LHS_BOUND2_STRIDE_LOAD]], %[[ITR1_LOAD]] : (!cir.ptr<!rec_NoOps>, !u64i) -> !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[RHS_BOUND2_STRIDE_LOAD:.*]] = cir.load %[[RHS_BOUND2_STRIDE]] : !cir.ptr<!cir.ptr<!rec_NoOps>>, !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[RHS_STRIDE:.*]] = cir.ptr_stride %[[RHS_BOUND2_STRIDE_LOAD]], %[[ITR1_LOAD]] : (!cir.ptr<!rec_NoOps>, !u64i) -> !cir.ptr<!rec_NoOps> +// CHECK-NEXT: %[[LHS_GET_I:.*]] = cir.get_member %[[LHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_NoOps> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_GET_I:.*]] = cir.get_member %[[RHS_STRIDE]][0] {name = "i"} : !cir.ptr<!rec_NoOps> -> !cir.ptr<!s32i> +// CHECK-NEXT: %[[RHS_LOAD:.*]] = cir.load {{.*}} %[[RHS_GET_I]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[LHS_LOAD:.*]] = cir.load {{.*}} %[[LHS_GET_I]] : !cir.ptr<!s32i>, !s32i +// CHECK-NEXT: %[[OP:.*]] = cir.binop(mul, %[[LHS_LOAD]], %[[RHS_LOAD]]) nsw +// CHECK-NEXT: cir.store{{.*}} %[[OP]], %[[LHS_GET_I]] : !s32i +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR1_LOAD]] = cir.load %[[ITR1]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR1_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR1]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR2_LOAD]] = cir.load %[[ITR2]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR2_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR2]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } step { +// CHECK-NEXT: %[[ITR3_LOAD]] = cir.load %[[ITR3]] : !cir.ptr<!u64i>, !u64i +// CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[ITR3_LOAD]]) : !u64i, !u64i +// CHECK-NEXT: cir.store %[[INC]], %[[ITR3]] : !u64i, !cir.ptr<!u64i> +// CHECK-NEXT: cir.yield +// CHECK-NEXT: } +// CHECK-NEXT: } // CHECK-NEXT: acc.yield // CHECK-NEXT: } destroy { // CHECK-NEXT: ^bb0(%[[REF:.*]]: !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>> {{.*}}, %[[PRIVATE:.*]]: !cir.ptr<!cir.ptr<!cir.ptr<!cir.ptr<!rec_NoOps>>>> {{.*}}, %[[BOUND1:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND2:.*]]: !acc.data_bounds_ty {{.*}}, %[[BOUND3:.*]]: !acc.data_bounds_ty {{.*}}): diff --git a/clang/test/CIR/IR/atomic.cir b/clang/test/CIR/IR/atomic.cir index 8520763..790297f 100644 --- a/clang/test/CIR/IR/atomic.cir +++ b/clang/test/CIR/IR/atomic.cir @@ -5,17 +5,30 @@ cir.func @atomic_xchg(%ptr: !cir.ptr<!s32i>, %val: !s32i) { // CHECK-LABEL: @atomic_xchg - %0 = cir.atomic.xchg relaxed %ptr, %val : !cir.ptr<!s32i> -> !s32i - // CHECK: cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - %1 = cir.atomic.xchg consume %ptr, %val : !cir.ptr<!s32i> -> !s32i - // CHECK: cir.atomic.xchg consume %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - %2 = cir.atomic.xchg acquire %ptr, %val : !cir.ptr<!s32i> -> !s32i - // CHECK: cir.atomic.xchg acquire %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - %3 = cir.atomic.xchg release %ptr, %val : !cir.ptr<!s32i> -> !s32i - // CHECK: cir.atomic.xchg release %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - %4 = cir.atomic.xchg acq_rel %ptr, %val : !cir.ptr<!s32i> -> !s32i - // CHECK: cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i - %5 = cir.atomic.xchg seq_cst %ptr, %val : !cir.ptr<!s32i> -> !s32i - // CHECK: cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : !cir.ptr<!s32i> -> !s32i + %0 = cir.atomic.xchg relaxed %ptr, %val : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CHECK: cir.atomic.xchg relaxed %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + %1 = cir.atomic.xchg consume %ptr, %val : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CHECK: cir.atomic.xchg consume %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + %2 = cir.atomic.xchg acquire %ptr, %val : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CHECK: cir.atomic.xchg acquire %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + %3 = cir.atomic.xchg release %ptr, %val : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CHECK: cir.atomic.xchg release %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + %4 = cir.atomic.xchg acq_rel %ptr, %val : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CHECK: cir.atomic.xchg acq_rel %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + %5 = cir.atomic.xchg seq_cst %ptr, %val : (!cir.ptr<!s32i>, !s32i) -> !s32i + // CHECK: cir.atomic.xchg seq_cst %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i) -> !s32i + cir.return +} + +cir.func @atomic_cmpxchg(%ptr: !cir.ptr<!s32i>, %expected: !s32i, %desired: !s32i) { + // CHECK-LABEL: @atomic_cmpxchg + %0, %1 = cir.atomic.cmpxchg success(relaxed) failure(relaxed) %ptr, %expected, %desired : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + // CHECK: cir.atomic.cmpxchg success(relaxed) failure(relaxed) %{{.+}}, %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + %2, %3 = cir.atomic.cmpxchg weak success(relaxed) failure(relaxed) %ptr, %expected, %desired : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + // CHECK: cir.atomic.cmpxchg weak success(relaxed) failure(relaxed) %{{.+}}, %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + %4, %5 = cir.atomic.cmpxchg success(seq_cst) failure(acquire) %ptr, %expected, %desired : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + // CHECK: cir.atomic.cmpxchg success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + %6, %7 = cir.atomic.cmpxchg weak success(seq_cst) failure(acquire) %ptr, %expected, %desired : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) + // CHECK: cir.atomic.cmpxchg weak success(seq_cst) failure(acquire) %{{.+}}, %{{.+}}, %{{.+}} : (!cir.ptr<!s32i>, !s32i, !s32i) -> (!s32i, !cir.bool) cir.return } diff --git a/clang/test/CodeGen/X86/avx512ifma-builtins.c b/clang/test/CodeGen/X86/avx512ifma-builtins.c index eebefb0..f90697e 100644 --- a/clang/test/CodeGen/X86/avx512ifma-builtins.c +++ b/clang/test/CodeGen/X86/avx512ifma-builtins.c @@ -8,45 +8,230 @@ // RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s // RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s - #include <immintrin.h> +#include "builtin_test_helpers.h" __m512i test_mm512_madd52hi_epu64(__m512i __X, __m512i __Y, __m512i __Z) { // CHECK-LABEL: test_mm512_madd52hi_epu64 // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52h.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}}) - return _mm512_madd52hi_epu64(__X, __Y, __Z); + return _mm512_madd52hi_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v8di(_mm512_madd52hi_epu64( + (__m512i)(__v8du){100, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){10, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){5, 0, 0, 0, 0, 0, 0, 0}), + 100, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52hi_epu64( + (__m512i)(__v8du){0, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, 0, 0, 0, + 0, 0, 0, 0}, + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, 0, 0, 0, + 0, 0, 0, 0}), + 0xFFFFFFFFFFFFEull, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52hi_epu64( + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull}, + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull}), + 4503599627370495ull, 4503599627370496ull, + 4503599627370497ull, 4503599627370498ull, + 4503599627370499ull, 4503599627370500ull, + 4503599627370501ull, 4503599627370502ull)); + __m512i test_mm512_mask_madd52hi_epu64(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y) { // CHECK-LABEL: test_mm512_mask_madd52hi_epu64 // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52h.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}}) // CHECK: select <8 x i1> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}} - return _mm512_mask_madd52hi_epu64(__W, __M, __X, __Y); + return _mm512_mask_madd52hi_epu64(__W, __M, __X, __Y); } +TEST_CONSTEXPR(match_v8di(_mm512_mask_madd52hi_epu64( + (__m512i)(__v8du){111, 222, 333, 444, 555, 666, + 777, 888}, + 0x00, + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}), + 111, 222, 333, 444, 555, 666, 777, 888)); + +TEST_CONSTEXPR(match_v8di(_mm512_mask_madd52hi_epu64( + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}, + 0xFF, + (__m512i)(__v8du){100, 200, 300, 400, 500, 600, + 700, 800}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}), + 10, 20, 30, 40, 50, 60, 70, 80)); + __m512i test_mm512_maskz_madd52hi_epu64(__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z) { // CHECK-LABEL: test_mm512_maskz_madd52hi_epu64 // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52h.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}}) // CHECK: select <8 x i1> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}} - return _mm512_maskz_madd52hi_epu64(__M, __X, __Y, __Z); + return _mm512_maskz_madd52hi_epu64(__M, __X, __Y, __Z); } +TEST_CONSTEXPR(match_v8di(_mm512_maskz_madd52hi_epu64( + 0x00, + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}, + (__m512i)(__v8du){100, 200, 300, 400, 500, 600, + 700, 800}), + 0, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_maskz_madd52hi_epu64( + 0xFF, + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}, + (__m512i)(__v8du){100, 200, 300, 400, 500, 600, + 700, 800}), + 1, 2, 3, 4, 5, 6, 7, 8)); + __m512i test_mm512_madd52lo_epu64(__m512i __X, __m512i __Y, __m512i __Z) { // CHECK-LABEL: test_mm512_madd52lo_epu64 // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52l.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}}) - return _mm512_madd52lo_epu64(__X, __Y, __Z); + return _mm512_madd52lo_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){0, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){10, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){5, 0, 0, 0, 0, 0, 0, 0}), + 50, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){100, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){20, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){30, 0, 0, 0, 0, 0, 0, 0}), + 700, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){0, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, 0, 0, 0, + 0, 0, 0, 0}, + (__m512i)(__v8du){1, 0, 0, 0, 0, 0, 0, 0}), + 0xFFFFFFFFFFFFFull, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){0, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){0x1F000000000000ull, 0, 0, 0, + 0, 0, 0, 0}, + (__m512i)(__v8du){2, 0, 0, 0, 0, 0, 0, 0}), + 0xE000000000000ull, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}, + (__m512i)(__v8du){2, 3, 4, 5, 6, 7, 8, 9}), + 21, 62, 123, 204, 305, 426, 567, 728)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, 0, 0, 0, + 0, 0, 0, 0}, + (__m512i)(__v8du){10, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){5, 0, 0, 0, 0, 0, 0, 0}), + 4503599627370545ull, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}, + (__m512i)(__v8du){100, 200, 300, 400, 500, 600, + 700, 800}, + (__m512i)(__v8du){2, 3, 4, 5, 6, 7, 8, 9}), + 210, 620, 1230, 2040, 3050, 4260, 5670, 7280)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){0, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){0x1F000000000000ull, + 0x1F000000000000ull, 0, 0, 0, + 0, 0, 0}, + (__m512i)(__v8du){2, 3, 0, 0, 0, 0, 0, 0}), + 0xE000000000000ull, 0xD000000000000ull, 0, 0, 0, 0, + 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_madd52lo_epu64( + (__m512i)(__v8du){0, 0, 0, 0, 0, 0, 0, 0}, + (__m512i)(__v8du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull}, + (__m512i)(__v8du){1, 1, 1, 1, 1, 1, 1, 1}), + 0xFFFFFFFFFFFFFull, 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0xFFFFFFFFFFFFFull)); + __m512i test_mm512_mask_madd52lo_epu64(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y) { // CHECK-LABEL: test_mm512_mask_madd52lo_epu64 // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52l.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}}) // CHECK: select <8 x i1> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}} - return _mm512_mask_madd52lo_epu64(__W, __M, __X, __Y); + return _mm512_mask_madd52lo_epu64(__W, __M, __X, __Y); } +TEST_CONSTEXPR(match_v8di(_mm512_mask_madd52lo_epu64( + (__m512i)(__v8du){111, 222, 333, 444, 555, 666, + 777, 888}, + 0x00, + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}), + 111, 222, 333, 444, 555, 666, 777, 888)); + +TEST_CONSTEXPR(match_v8di(_mm512_mask_madd52lo_epu64( + (__m512i)(__v8du){1000, 2000, 3000, 4000, 5000, + 6000, 7000, 8000}, + 0xFF, + (__m512i)(__v8du){100, 200, 300, 400, 500, 600, + 700, 800}, + (__m512i)(__v8du){20, 30, 40, 50, 60, 70, 80, + 90}), + 3000, 8000, 15000, 24000, 35000, 48000, 63000, + 80000)); + __m512i test_mm512_maskz_madd52lo_epu64(__mmask8 __M, __m512i __X, __m512i __Y, __m512i __Z) { // CHECK-LABEL: test_mm512_maskz_madd52lo_epu64 // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52l.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}}) // CHECK: select <8 x i1> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}} - return _mm512_maskz_madd52lo_epu64(__M, __X, __Y, __Z); + return _mm512_maskz_madd52lo_epu64(__M, __X, __Y, __Z); } + +TEST_CONSTEXPR(match_v8di(_mm512_maskz_madd52lo_epu64( + 0x00, + (__m512i)(__v8du){1, 2, 3, 4, 5, 6, 7, 8}, + (__m512i)(__v8du){10, 20, 30, 40, 50, 60, 70, + 80}, + (__m512i)(__v8du){2, 3, 4, 5, 6, 7, 8, 9}), + 0, 0, 0, 0, 0, 0, 0, 0)); + +TEST_CONSTEXPR(match_v8di(_mm512_maskz_madd52lo_epu64( + 0xFF, + (__m512i)(__v8du){100, 200, 300, 400, 500, 600, + 700, 800}, + (__m512i)(__v8du){20, 30, 40, 50, 60, 70, 80, + 90}, + (__m512i)(__v8du){30, 40, 50, 60, 70, 80, 90, + 100}), + 700, 1400, 2300, 3400, 4700, 6200, 7900, 9800)); diff --git a/clang/test/CodeGen/X86/avx512ifmavl-builtins.c b/clang/test/CodeGen/X86/avx512ifmavl-builtins.c index 89108fc..1cbb580 100644 --- a/clang/test/CodeGen/X86/avx512ifmavl-builtins.c +++ b/clang/test/CodeGen/X86/avx512ifmavl-builtins.c @@ -8,85 +8,241 @@ // RUN: %clang_cc1 -x c++ %s -flax-vector-conversions=none -ffreestanding -triple=x86_64-apple-darwin -target-feature +avx512ifma -target-feature +avx512vl -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s // RUN: %clang_cc1 -x c++ %s -flax-vector-conversions=none -ffreestanding -triple=i386-apple-darwin -target-feature +avx512ifma -target-feature +avx512vl -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s - #include <immintrin.h> +#include "builtin_test_helpers.h" __m128i test_mm_madd52hi_epu64(__m128i __X, __m128i __Y, __m128i __Z) { // CHECK-LABEL: test_mm_madd52hi_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52h.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) - return _mm_madd52hi_epu64(__X, __Y, __Z); + return _mm_madd52hi_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_epu64( + (__m128i)((__v2du){100, 0}), + (__m128i)((__v2du){10, 0}), + (__m128i)((__v2du){5, 0})), + 100, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0})), + 0xFFFFFFFFFFFFEull, 0)); + __m128i test_mm_mask_madd52hi_epu64(__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y) { // CHECK-LABEL: test_mm_mask_madd52hi_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52h.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) // CHECK: select <2 x i1> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}} - return _mm_mask_madd52hi_epu64(__W, __M, __X, __Y); + return _mm_mask_madd52hi_epu64(__W, __M, __X, __Y); } +TEST_CONSTEXPR(match_v2di(_mm_mask_madd52hi_epu64((__m128i)((__v2du){111, 222}), + 0x0, + (__m128i)((__v2du){1, 2}), + (__m128i)((__v2du){10, 20})), + 111, 222)); + +TEST_CONSTEXPR(match_v2di(_mm_mask_madd52hi_epu64((__m128i)((__v2du){10, 20}), + 0x2, + (__m128i)((__v2du){0x1000000000000ULL, 0x1000000000000ULL}), + (__m128i)((__v2du){0x1000000000000ULL, 0x1000000000000ULL})), + 10, 0x100000000014ULL)); + __m128i test_mm_maskz_madd52hi_epu64(__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z) { // CHECK-LABEL: test_mm_maskz_madd52hi_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52h.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) // CHECK: select <2 x i1> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}} - return _mm_maskz_madd52hi_epu64(__M, __X, __Y, __Z); + return _mm_maskz_madd52hi_epu64(__M, __X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_maskz_madd52hi_epu64(0x3, + (__m128i)((__v2du){1, 2}), + (__m128i)((__v2du){10, 20}), + (__m128i)((__v2du){100, 200})), + 1, 2)); + +TEST_CONSTEXPR(match_v2di(_mm_maskz_madd52hi_epu64(0x1, + (__m128i)((__v2du){0x1000000000000ULL, 0x1000000000000ULL}), + (__m128i)((__v2du){0x1000000000000ULL, 0x1000000000000ULL}), + (__m128i)((__v2du){0, 0})), + 0x1000000000000ULL, 0)); + __m256i test_mm256_madd52hi_epu64(__m256i __X, __m256i __Y, __m256i __Z) { // CHECK-LABEL: test_mm256_madd52hi_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52h.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) - return _mm256_madd52hi_epu64(__X, __Y, __Z); + return _mm256_madd52hi_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_epu64( + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){5, 6, 7, 8})), + 100, 200, 300, 400)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, 0, 0, + 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, 0, 0, + 0})), + 0xFFFFFFFFFFFFEull, 0, 0, 0)); + __m256i test_mm256_mask_madd52hi_epu64(__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y) { // CHECK-LABEL: test_mm256_mask_madd52hi_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52h.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) // CHECK: select <4 x i1> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}} - return _mm256_mask_madd52hi_epu64(__W, __M, __X, __Y); + return _mm256_mask_madd52hi_epu64(__W, __M, __X, __Y); } +TEST_CONSTEXPR(match_v4di(_mm256_mask_madd52hi_epu64((__m256i)((__v4du){111, 222, 333, 444}), + 0x0, + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40})), + 111, 222, 333, 444)); + +TEST_CONSTEXPR(match_v4di(_mm256_mask_madd52hi_epu64((__m256i)((__v4du){10, 20, 30, 40}), + 0xA, + (__m256i)((__v4du){0x1000000000000ULL, 0x1000000000000ULL, + 0x1000000000000ULL, 0x1000000000000ULL}), + (__m256i)((__v4du){0x1000000000000ULL, 0x1000000000000ULL, + 0x1000000000000ULL, 0x1000000000000ULL})), + 10, 0x100000000014ULL, 30, 0x100000000028ULL)); + __m256i test_mm256_maskz_madd52hi_epu64(__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z) { // CHECK-LABEL: test_mm256_maskz_madd52hi_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52h.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) // CHECK: select <4 x i1> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}} - return _mm256_maskz_madd52hi_epu64(__M, __X, __Y, __Z); + return _mm256_maskz_madd52hi_epu64(__M, __X, __Y, __Z); } +TEST_CONSTEXPR(match_v4di(_mm256_maskz_madd52hi_epu64(0xF, + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){100, 200, 300, 400})), + 1, 2, 3, 4)); + +TEST_CONSTEXPR(match_v4di(_mm256_maskz_madd52hi_epu64(0x5, + (__m256i)((__v4du){0x1000000000000ULL, 0x1000000000000ULL, + 0x1000000000000ULL, 0x1000000000000ULL}), + (__m256i)((__v4du){0x1000000000000ULL, 0x1000000000000ULL, + 0x1000000000000ULL, 0x1000000000000ULL}), + (__m256i)((__v4du){0, 0, 0, 0})), + 0x1000000000000ULL, 0, 0x1000000000000ULL, 0)); + __m128i test_mm_madd52lo_epu64(__m128i __X, __m128i __Y, __m128i __Z) { // CHECK-LABEL: test_mm_madd52lo_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52l.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) - return _mm_madd52lo_epu64(__X, __Y, __Z); + return _mm_madd52lo_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){10, 0}), + (__m128i)((__v2du){5, 0})), + 50, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_epu64( + (__m128i)((__v2du){100, 0}), + (__m128i)((__v2du){20, 0}), + (__m128i)((__v2du){30, 0})), + 700, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_epu64( + (__m128i)((__v2du){1, 2}), + (__m128i)((__v2du){10, 20}), + (__m128i)((__v2du){2, 3})), + 21, 62)); + __m128i test_mm_mask_madd52lo_epu64(__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y) { // CHECK-LABEL: test_mm_mask_madd52lo_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52l.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) // CHECK: select <2 x i1> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}} - return _mm_mask_madd52lo_epu64(__W, __M, __X, __Y); + return _mm_mask_madd52lo_epu64(__W, __M, __X, __Y); } +TEST_CONSTEXPR(match_v2di(_mm_mask_madd52lo_epu64((__m128i)((__v2du){1000, 2000}), + 0x3, + (__m128i)((__v2du){100, 200}), + (__m128i)((__v2du){20, 30})), + 3000, 8000)); + +TEST_CONSTEXPR(match_v2di(_mm_mask_madd52lo_epu64((__m128i)((__v2du){111, 222}), + 0x0, + (__m128i)((__v2du){1, 2}), + (__m128i)((__v2du){10, 20})), + 111, 222)); + __m128i test_mm_maskz_madd52lo_epu64(__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z) { // CHECK-LABEL: test_mm_maskz_madd52lo_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52l.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) // CHECK: select <2 x i1> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}} - return _mm_maskz_madd52lo_epu64(__M, __X, __Y, __Z); + return _mm_maskz_madd52lo_epu64(__M, __X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_maskz_madd52lo_epu64(0x3, + (__m128i)((__v2du){100, 200}), + (__m128i)((__v2du){20, 30}), + (__m128i)((__v2du){30, 40})), + 700, 1400)); + +TEST_CONSTEXPR(match_v2di(_mm_maskz_madd52lo_epu64(0x1, + (__m128i)((__v2du){100, 0}), + (__m128i)((__v2du){20, 0}), + (__m128i)((__v2du){30, 0})), + 700, 0)); + __m256i test_mm256_madd52lo_epu64(__m256i __X, __m256i __Y, __m256i __Z) { // CHECK-LABEL: test_mm256_madd52lo_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52l.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) - return _mm256_madd52lo_epu64(__X, __Y, __Z); + return _mm256_madd52lo_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_epu64( + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){2, 3, 4, 5})), + 21, 62, 123, 204)); + __m256i test_mm256_mask_madd52lo_epu64(__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y) { // CHECK-LABEL: test_mm256_mask_madd52lo_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52l.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) // CHECK: select <4 x i1> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}} - return _mm256_mask_madd52lo_epu64(__W, __M, __X, __Y); + return _mm256_mask_madd52lo_epu64(__W, __M, __X, __Y); } +TEST_CONSTEXPR(match_v4di(_mm256_mask_madd52lo_epu64((__m256i)((__v4du){1000, 2000, 3000, 4000}), + 0xF, + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){20, 30, 40, 50})), + 3000, 8000, 15000, 24000)); + +TEST_CONSTEXPR(match_v4di(_mm256_mask_madd52lo_epu64((__m256i)((__v4du){111, 222, 333, 444}), + 0x0, + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40})), + 111, 222, 333, 444)); + +TEST_CONSTEXPR(match_v4di(_mm256_mask_madd52lo_epu64((__m256i)((__v4du){11, 22, 33, 44}), + 0x5, + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){10, 20, 30, 40})), + 1011, 22, 9033, 44)); + __m256i test_mm256_maskz_madd52lo_epu64(__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z) { // CHECK-LABEL: test_mm256_maskz_madd52lo_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52l.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) // CHECK: select <4 x i1> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}} - return _mm256_maskz_madd52lo_epu64(__M, __X, __Y, __Z); + return _mm256_maskz_madd52lo_epu64(__M, __X, __Y, __Z); } + +TEST_CONSTEXPR(match_v4di(_mm256_maskz_madd52lo_epu64(0xF, + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){20, 30, 40, 50}), + (__m256i)((__v4du){30, 40, 50, 60})), + 700, 1400, 2300, 3400)); + +TEST_CONSTEXPR(match_v4di(_mm256_maskz_madd52lo_epu64(0x9, + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){5, 10, 15, 20})), + 150, 0, 0, 1200)); diff --git a/clang/test/CodeGen/X86/avxifma-builtins.c b/clang/test/CodeGen/X86/avxifma-builtins.c index aa15159..70531da 100644 --- a/clang/test/CodeGen/X86/avxifma-builtins.c +++ b/clang/test/CodeGen/X86/avxifma-builtins.c @@ -8,8 +8,9 @@ // RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avxifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s // RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=i386-apple-darwin -target-feature +avxifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s - #include <immintrin.h> +#include "builtin_test_helpers.h" + __m128i test_mm_madd52hi_epu64(__m128i __X, __m128i __Y, __m128i __Z) { // CHECK-LABEL: test_mm_madd52hi_epu64 @@ -17,44 +18,207 @@ __m128i test_mm_madd52hi_epu64(__m128i __X, __m128i __Y, __m128i __Z) { return _mm_madd52hi_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_epu64( + (__m128i)((__v2du){50, 100}), + (__m128i)((__v2du){10, 20}), + (__m128i)((__v2du){5, 6})), + 50, 100)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0})), + 0xFFFFFFFFFFFFEull, 0)); + __m256i test_mm256_madd52hi_epu64(__m256i __X, __m256i __Y, __m256i __Z) { // CHECK-LABEL: test_mm256_madd52hi_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52h.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) return _mm256_madd52hi_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_epu64( + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){5, 6, 7, 8})), + 100, 200, 300, 400)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0, 0})), + 0xFFFFFFFFFFFFEull, 0xFFFFFFFFFFFFEull, 0, 0)); + __m128i test_mm_madd52lo_epu64(__m128i __X, __m128i __Y, __m128i __Z) { // CHECK-LABEL: test_mm_madd52lo_epu64 // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52l.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) return _mm_madd52lo_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){10, 0}), + (__m128i)((__v2du){5, 0})), + 50, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_epu64( + (__m128i)((__v2du){1, 2}), + (__m128i)((__v2du){10, 20}), + (__m128i)((__v2du){2, 3})), + 21, 62)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0}), + (__m128i)((__v2du){1, 0})), + 0xFFFFFFFFFFFFFull, 0)); + __m256i test_mm256_madd52lo_epu64(__m256i __X, __m256i __Y, __m256i __Z) { // CHECK-LABEL: test_mm256_madd52lo_epu64 // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52l.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) return _mm256_madd52lo_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_epu64( + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){2, 3, 4, 5})), + 21, 62, 123, 204)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, 0, 0, + 0}), + (__m256i)((__v4du){1, 0, 0, 0})), + 0xFFFFFFFFFFFFFull, 0, 0, 0)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0x1F000000000000ull, 0, 0, + 0}), + (__m256i)((__v4du){2, 0, 0, 0})), + 0xE000000000000ull, 0, 0, 0)); + __m128i test_mm_madd52hi_avx_epu64(__m128i __X, __m128i __Y, __m128i __Z) { -// CHECK-LABEL: test_mm_madd52hi_avx_epu64 -// CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52h.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) + // CHECK-LABEL: test_mm_madd52hi_avx_epu64 + // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52h.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) return _mm_madd52hi_avx_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_avx_epu64( + (__m128i)((__v2du){50, 100}), + (__m128i)((__v2du){10, 20}), + (__m128i)((__v2du){5, 6})), + 50, 100)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_avx_epu64( + (__m128i)((__v2du){100, 0}), + (__m128i)((__v2du){10, 0}), + (__m128i)((__v2du){5, 0})), + 100, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52hi_avx_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0})), + 0xFFFFFFFFFFFFEull, 0)); + __m256i test_mm256_madd52hi_avx_epu64(__m256i __X, __m256i __Y, __m256i __Z) { -// CHECK-LABEL: test_mm256_madd52hi_avx_epu64 -// CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52h.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) + // CHECK-LABEL: test_mm256_madd52hi_avx_epu64 + // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52h.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) return _mm256_madd52hi_avx_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_avx_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, + 0xFFFFFFFFFFFFFull, 0, 0})), + 0xFFFFFFFFFFFFEull, 0xFFFFFFFFFFFFEull, 0, 0)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_avx_epu64( + (__m256i)((__v4du){100, 200, 300, 400}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){5, 6, 7, 8})), + 100, 200, 300, 400)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52hi_avx_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, 0, 0, + 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, 0, 0, + 0})), + 0xFFFFFFFFFFFFEull, 0, 0, 0)); + __m128i test_mm_madd52lo_avx_epu64(__m128i __X, __m128i __Y, __m128i __Z) { -// CHECK-LABEL: test_mm_madd52lo_avx_epu64 -// CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52l.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) + // CHECK-LABEL: test_mm_madd52lo_avx_epu64 + // CHECK: call {{.*}}<2 x i64> @llvm.x86.avx512.vpmadd52l.uq.128(<2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i64> %{{.*}}) return _mm_madd52lo_avx_epu64(__X, __Y, __Z); } +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_avx_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){10, 0}), + (__m128i)((__v2du){5, 0})), + 50, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_avx_epu64( + (__m128i)((__v2du){100, 0}), + (__m128i)((__v2du){20, 0}), + (__m128i)((__v2du){30, 0})), + 700, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_avx_epu64( + (__m128i)((__v2du){1, 2}), + (__m128i)((__v2du){10, 20}), + (__m128i)((__v2du){2, 3})), + 21, 62)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_avx_epu64( + (__m128i)((__v2du){0, 0}), + (__m128i)((__v2du){0xFFFFFFFFFFFFFull, 0}), + (__m128i)((__v2du){1, 0})), + 0xFFFFFFFFFFFFFull, 0)); + __m256i test_mm256_madd52lo_avx_epu64(__m256i __X, __m256i __Y, __m256i __Z) { -// CHECK-LABEL: test_mm256_madd52lo_avx_epu64 -// CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52l.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) + // CHECK-LABEL: test_mm256_madd52lo_avx_epu64 + // CHECK: call {{.*}}<4 x i64> @llvm.x86.avx512.vpmadd52l.uq.256(<4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i64> %{{.*}}) return _mm256_madd52lo_avx_epu64(__X, __Y, __Z); } + +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_avx_epu64( + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){2, 3, 4, 5})), + 21, 62, 123, 204)); + + + +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_avx_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0xFFFFFFFFFFFFFull, 0, 0, + 0}), + (__m256i)((__v4du){1, 0, 0, 0})), + 0xFFFFFFFFFFFFFull, 0, 0, 0)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_avx_epu64( + (__m256i)((__v4du){0, 0, 0, 0}), + (__m256i)((__v4du){0x1F000000000000ull, 0, 0, + 0}), + (__m256i)((__v4du){2, 0, 0, 0})), + 0xE000000000000ull, 0, 0, 0)); + +TEST_CONSTEXPR(match_v2di(_mm_madd52lo_avx_epu64( + (__m128i)((__v2du){5, 10}), + (__m128i)((__v2du){100, 200}), + (__m128i)((__v2du){7, 8})), + 705, 1610)); + +TEST_CONSTEXPR(match_v4di(_mm256_madd52lo_avx_epu64( + (__m256i)((__v4du){1, 2, 3, 4}), + (__m256i)((__v4du){10, 20, 30, 40}), + (__m256i)((__v4du){2, 3, 4, 5})), + 21, 62, 123, 204)); + diff --git a/clang/test/SemaCXX/cxx2b-consteval-propagate.cpp b/clang/test/SemaCXX/cxx2b-consteval-propagate.cpp index 6cf0e02..331fe83 100644 --- a/clang/test/SemaCXX/cxx2b-consteval-propagate.cpp +++ b/clang/test/SemaCXX/cxx2b-consteval-propagate.cpp @@ -626,3 +626,20 @@ void fn() { } } + + +namespace GH109096 { +consteval void undefined(); +template <typename T> +struct scope_exit { + T t; + constexpr ~scope_exit() { t(); } + // expected-error@-1 {{call to immediate function 'GH109096::(anonymous class)::operator()' is not a constant expression}} \ + // expected-note@-1 {{implicit use of 'this' pointer is only allowed within the evaluation}} +}; + +scope_exit guard( // expected-note {{in instantiation of member function}} + []() { undefined(); } +); + +} diff --git a/clang/test/SemaTemplate/concepts.cpp b/clang/test/SemaTemplate/concepts.cpp index 3fbe7c0..aaa20f6 100644 --- a/clang/test/SemaTemplate/concepts.cpp +++ b/clang/test/SemaTemplate/concepts.cpp @@ -1416,6 +1416,31 @@ concept IsEntitySpec = } +namespace case8 { + +template <class T> +struct type_identity { + using type = T; +}; + +template <typename Inner> +struct Cat {}; + +template <typename T> +concept CatConcept = requires { + []<class Inner>(type_identity<Cat<Inner>>) {}(type_identity<T>{}); +}; + +template <typename Dummy> +struct Feeder { + template <CatConcept Dummy2> + void feed() noexcept {} +}; + +void main() { Feeder<int>{}.feed<Cat<int>>(); } + +} + } namespace GH162125 { diff --git a/compiler-rt/test/tsan/Darwin/write-interpose.c b/compiler-rt/test/tsan/Darwin/write-interpose.c index cbd9a08..51ff3ee 100644 --- a/compiler-rt/test/tsan/Darwin/write-interpose.c +++ b/compiler-rt/test/tsan/Darwin/write-interpose.c @@ -7,6 +7,8 @@ // Note that running the below command with out `lock_during_write` should // deadlock (self-lock) // RUN: env DYLD_INSERT_LIBRARIES=%t.dylib TSAN_OPTIONS=verbosity=2:lock_during_write=disable_for_current_process %run %t 2>&1 | FileCheck %s +// +// UNSUPPORTED: ios #include <stdio.h> diff --git a/flang/include/flang/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.h b/flang/include/flang/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.h index 1085393..408f039 100644 --- a/flang/include/flang/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.h +++ b/flang/include/flang/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.h @@ -57,8 +57,11 @@ struct OpenACCMappableModel mlir::Location loc, mlir::TypedValue<mlir::acc::MappableType> var, llvm::StringRef varName, - mlir::ValueRange extents, - mlir::Value initVal) const; + mlir::ValueRange extents, mlir::Value initVal, + bool &needsDestroy) const; + + bool generatePrivateDestroy(mlir::Type type, mlir::OpBuilder &builder, + mlir::Location loc, mlir::Value privatized) const; }; } // namespace fir::acc diff --git a/flang/include/flang/Optimizer/Support/Utils.h b/flang/include/flang/Optimizer/Support/Utils.h index 0b31cfe..bbb7e6e 100644 --- a/flang/include/flang/Optimizer/Support/Utils.h +++ b/flang/include/flang/Optimizer/Support/Utils.h @@ -200,6 +200,12 @@ std::optional<llvm::ArrayRef<int64_t>> getComponentLowerBoundsIfNonDefault( fir::RecordType recordType, llvm::StringRef component, mlir::ModuleOp module, const mlir::SymbolTable *symbolTable = nullptr); +/// Indicate if a derived type has final routine. Returns std::nullopt if that +/// information is not in the IR; +std::optional<bool> +isRecordWithFinalRoutine(fir::RecordType recordType, mlir::ModuleOp module, + const mlir::SymbolTable *symbolTable = nullptr); + /// Generate a LLVM constant value of type `ity`, using the provided offset. mlir::LLVM::ConstantOp genConstantIndex(mlir::Location loc, mlir::Type ity, diff --git a/flang/lib/Lower/OpenACC.cpp b/flang/lib/Lower/OpenACC.cpp index 62e5c0c..cfb1891 100644 --- a/flang/lib/Lower/OpenACC.cpp +++ b/flang/lib/Lower/OpenACC.cpp @@ -978,15 +978,40 @@ static RecipeOp genRecipeOp( auto mappableTy = mlir::dyn_cast<mlir::acc::MappableType>(ty); assert(mappableTy && "Expected that all variable types are considered mappable"); + bool needsDestroy = false; auto retVal = mappableTy.generatePrivateInit( builder, loc, mlir::cast<mlir::TypedValue<mlir::acc::MappableType>>( initBlock->getArgument(0)), initName, initBlock->getArguments().take_back(initBlock->getArguments().size() - 1), - initValue); + initValue, needsDestroy); mlir::acc::YieldOp::create(builder, loc, retVal ? retVal : initBlock->getArgument(0)); + // Create destroy region and generate destruction if requested. + if (needsDestroy) { + llvm::SmallVector<mlir::Type> destroyArgsTy; + llvm::SmallVector<mlir::Location> destroyArgsLoc; + // original and privatized/reduction value + destroyArgsTy.push_back(ty); + destroyArgsTy.push_back(ty); + destroyArgsLoc.push_back(loc); + destroyArgsLoc.push_back(loc); + // Append bounds arguments (if any) in the same order as init region + if (argsTy.size() > 1) { + destroyArgsTy.append(argsTy.begin() + 1, argsTy.end()); + destroyArgsLoc.insert(destroyArgsLoc.end(), argsTy.size() - 1, loc); + } + + builder.createBlock(&recipe.getDestroyRegion(), + recipe.getDestroyRegion().end(), destroyArgsTy, + destroyArgsLoc); + builder.setInsertionPointToEnd(&recipe.getDestroyRegion().back()); + // Call interface on the privatized/reduction value (2nd argument). + (void)mappableTy.generatePrivateDestroy( + builder, loc, recipe.getDestroyRegion().front().getArgument(1)); + mlir::acc::TerminatorOp::create(builder, loc); + } return recipe; } diff --git a/flang/lib/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.cpp b/flang/lib/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.cpp index 89aa010..41383fb 100644 --- a/flang/lib/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.cpp +++ b/flang/lib/Optimizer/OpenACC/Support/FIROpenACCTypeInterfaces.cpp @@ -21,6 +21,7 @@ #include "flang/Optimizer/Dialect/FIRType.h" #include "flang/Optimizer/Dialect/Support/FIRContext.h" #include "flang/Optimizer/Dialect/Support/KindMapping.h" +#include "flang/Optimizer/Support/Utils.h" #include "mlir/Dialect/Arith/IR/Arith.h" #include "mlir/Dialect/OpenACC/OpenACC.h" #include "mlir/IR/BuiltinOps.h" @@ -548,14 +549,27 @@ template <typename Ty> mlir::Value OpenACCMappableModel<Ty>::generatePrivateInit( mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, mlir::TypedValue<mlir::acc::MappableType> var, llvm::StringRef varName, - mlir::ValueRange extents, mlir::Value initVal) const { + mlir::ValueRange extents, mlir::Value initVal, bool &needsDestroy) const { + needsDestroy = false; mlir::Value retVal; mlir::Type unwrappedTy = fir::unwrapRefType(type); mlir::ModuleOp mod = builder.getInsertionBlock() ->getParent() ->getParentOfType<mlir::ModuleOp>(); - fir::FirOpBuilder firBuilder(builder, mod); + if (auto recType = llvm::dyn_cast<fir::RecordType>( + fir::getFortranElementType(unwrappedTy))) { + // Need to make deep copies of allocatable components. + if (fir::isRecordWithAllocatableMember(recType)) + TODO(loc, + "OpenACC: privatizing derived type with allocatable components"); + // Need to decide if user assignment/final routine should be called. + if (fir::isRecordWithFinalRoutine(recType, mod).value_or(false)) + TODO(loc, "OpenACC: privatizing derived type with user assignment or " + "final routine "); + } + + fir::FirOpBuilder firBuilder(builder, mod); auto getDeclareOpForType = [&](mlir::Type ty) -> hlfir::DeclareOp { auto alloca = fir::AllocaOp::create(firBuilder, loc, ty); return hlfir::DeclareOp::create(firBuilder, loc, alloca, varName); @@ -615,9 +629,11 @@ mlir::Value OpenACCMappableModel<Ty>::generatePrivateInit( mlir::Value firClass = fir::EmboxOp::create(builder, loc, boxTy, allocatedScalar); fir::StoreOp::create(builder, loc, firClass, retVal); + needsDestroy = true; } else if (mlir::isa<fir::SequenceType>(innerTy)) { hlfir::Entity source = hlfir::Entity{var}; - auto [temp, cleanup] = hlfir::createTempFromMold(loc, firBuilder, source); + auto [temp, cleanupFlag] = + hlfir::createTempFromMold(loc, firBuilder, source); if (fir::isa_ref_type(type)) { // When the temp is created - it is not a reference - thus we can // end up with a type inconsistency. Therefore ensure storage is created @@ -636,6 +652,9 @@ mlir::Value OpenACCMappableModel<Ty>::generatePrivateInit( } else { retVal = temp; } + // If heap was allocated, a destroy is required later. + if (cleanupFlag) + needsDestroy = true; } else { TODO(loc, "Unsupported boxed type for OpenACC private-like recipe"); } @@ -667,23 +686,61 @@ template mlir::Value OpenACCMappableModel<fir::BaseBoxType>::generatePrivateInit( mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, mlir::TypedValue<mlir::acc::MappableType> var, llvm::StringRef varName, - mlir::ValueRange extents, mlir::Value initVal) const; + mlir::ValueRange extents, mlir::Value initVal, bool &needsDestroy) const; template mlir::Value OpenACCMappableModel<fir::ReferenceType>::generatePrivateInit( mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, mlir::TypedValue<mlir::acc::MappableType> var, llvm::StringRef varName, - mlir::ValueRange extents, mlir::Value initVal) const; + mlir::ValueRange extents, mlir::Value initVal, bool &needsDestroy) const; template mlir::Value OpenACCMappableModel<fir::HeapType>::generatePrivateInit( mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, mlir::TypedValue<mlir::acc::MappableType> var, llvm::StringRef varName, - mlir::ValueRange extents, mlir::Value initVal) const; + mlir::ValueRange extents, mlir::Value initVal, bool &needsDestroy) const; template mlir::Value OpenACCMappableModel<fir::PointerType>::generatePrivateInit( mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, mlir::TypedValue<mlir::acc::MappableType> var, llvm::StringRef varName, - mlir::ValueRange extents, mlir::Value initVal) const; + mlir::ValueRange extents, mlir::Value initVal, bool &needsDestroy) const; + +template <typename Ty> +bool OpenACCMappableModel<Ty>::generatePrivateDestroy( + mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, + mlir::Value privatized) const { + mlir::Type unwrappedTy = fir::unwrapRefType(type); + // For boxed scalars allocated with AllocMem during init, free the heap. + if (auto boxTy = mlir::dyn_cast_or_null<fir::BaseBoxType>(unwrappedTy)) { + mlir::Value boxVal = privatized; + if (fir::isa_ref_type(boxVal.getType())) + boxVal = fir::LoadOp::create(builder, loc, boxVal); + mlir::Value addr = fir::BoxAddrOp::create(builder, loc, boxVal); + // FreeMem only accepts fir.heap and this may not be represented in the box + // type if the privatized entity is not an allocatable. + mlir::Type heapType = + fir::HeapType::get(fir::unwrapRefType(addr.getType())); + if (heapType != addr.getType()) + addr = fir::ConvertOp::create(builder, loc, heapType, addr); + fir::FreeMemOp::create(builder, loc, addr); + return true; + } + + // Nothing to do for other categories by default, they are stack allocated. + return true; +} + +template bool OpenACCMappableModel<fir::BaseBoxType>::generatePrivateDestroy( + mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, + mlir::Value privatized) const; +template bool OpenACCMappableModel<fir::ReferenceType>::generatePrivateDestroy( + mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, + mlir::Value privatized) const; +template bool OpenACCMappableModel<fir::HeapType>::generatePrivateDestroy( + mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, + mlir::Value privatized) const; +template bool OpenACCMappableModel<fir::PointerType>::generatePrivateDestroy( + mlir::Type type, mlir::OpBuilder &builder, mlir::Location loc, + mlir::Value privatized) const; } // namespace fir::acc diff --git a/flang/lib/Optimizer/Support/Utils.cpp b/flang/lib/Optimizer/Support/Utils.cpp index c71642c..92390e4a 100644 --- a/flang/lib/Optimizer/Support/Utils.cpp +++ b/flang/lib/Optimizer/Support/Utils.cpp @@ -51,6 +51,16 @@ std::optional<llvm::ArrayRef<int64_t>> fir::getComponentLowerBoundsIfNonDefault( return std::nullopt; } +std::optional<bool> +fir::isRecordWithFinalRoutine(fir::RecordType recordType, mlir::ModuleOp module, + const mlir::SymbolTable *symbolTable) { + fir::TypeInfoOp typeInfo = + fir::lookupTypeInfoOp(recordType, module, symbolTable); + if (!typeInfo) + return std::nullopt; + return !typeInfo.getNoFinal(); +} + mlir::LLVM::ConstantOp fir::genConstantIndex(mlir::Location loc, mlir::Type ity, mlir::ConversionPatternRewriter &rewriter, diff --git a/flang/test/Lower/OpenACC/acc-firstprivate-derived-allocatable-component.f90 b/flang/test/Lower/OpenACC/acc-firstprivate-derived-allocatable-component.f90 index 429f207..3987f9f 100644 --- a/flang/test/Lower/OpenACC/acc-firstprivate-derived-allocatable-component.f90 +++ b/flang/test/Lower/OpenACC/acc-firstprivate-derived-allocatable-component.f90 @@ -4,6 +4,11 @@ ! RUN: bbc -fopenacc -emit-hlfir %s -o - | FileCheck %s ! RUN: bbc -fopenacc -emit-fir %s -o - | FileCheck %s --check-prefix=FIR-CHECK +! TODO: This test hits a fatal TODO. Deal with allocatable component +! destructions. For arrays, allocatable component allocation may also be +! missing. +! XFAIL: * + module m_firstprivate_derived_alloc_comp type point real, allocatable :: x(:) diff --git a/flang/test/Lower/OpenACC/acc-private.f90 b/flang/test/Lower/OpenACC/acc-private.f90 index d37eb8d..485825d 100644 --- a/flang/test/Lower/OpenACC/acc-private.f90 +++ b/flang/test/Lower/OpenACC/acc-private.f90 @@ -26,6 +26,12 @@ ! CHECK: %[[DES_DST:.*]] = hlfir.designate %[[ARG1]] shape %[[SHAPE]] : (!fir.box<!fir.array<?x?x2xi32>>, !fir.shape<3>) -> !fir.box<!fir.array<?x?x2xi32>> ! CHECK: hlfir.assign %[[DES_SRC]] to %[[DES_DST]] : !fir.box<!fir.array<?x?x2xi32>>, !fir.box<!fir.array<?x?x2xi32>> ! CHECK: acc.terminator +! CHECK: } destroy { +! CHECK: ^bb0(%[[ARG0:.*]]: !fir.box<!fir.array<?x?x2xi32>>, %[[ARG1:.*]]: !fir.box<!fir.array<?x?x2xi32>>): +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[ARG1]] : (!fir.box<!fir.array<?x?x2xi32>>) -> !fir.ref<!fir.array<?x?x2xi32>> +! CHECK: %[[CAST:.*]] = fir.convert %[[ADDR]] : (!fir.ref<!fir.array<?x?x2xi32>>) -> !fir.heap<!fir.array<?x?x2xi32>> +! CHECK: fir.freemem %[[CAST]] : !fir.heap<!fir.array<?x?x2xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.firstprivate.recipe @firstprivatization_section_lb4.ub9_box_Uxi32 : !fir.box<!fir.array<?xi32>> init { @@ -47,6 +53,12 @@ ! CHECK: %[[RIGHT:.*]] = hlfir.designate %[[ARG1]] shape %[[SHAPE]] : (!fir.box<!fir.array<?xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<?xi32>> ! CHECK: hlfir.assign %[[LEFT]] to %[[RIGHT]] : !fir.box<!fir.array<?xi32>>, !fir.box<!fir.array<?xi32>> ! CHECK: acc.terminator +! CHECK: } destroy { +! CHECK: ^bb0(%[[ARG0:.*]]: !fir.box<!fir.array<?xi32>>, %[[ARG1:.*]]: !fir.box<!fir.array<?xi32>>): +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[ARG1]] : (!fir.box<!fir.array<?xi32>>) -> !fir.ref<!fir.array<?xi32>> +! CHECK: %[[CAST:.*]] = fir.convert %[[ADDR]] : (!fir.ref<!fir.array<?xi32>>) -> !fir.heap<!fir.array<?xi32>> +! CHECK: fir.freemem %[[CAST]] : !fir.heap<!fir.array<?xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.firstprivate.recipe @firstprivatization_box_Uxi32 : !fir.box<!fir.array<?xi32>> init { @@ -64,6 +76,12 @@ ! CHECK: %[[DES_V2:.*]] = hlfir.designate %[[ARG1]] shape %[[SHAPE]] : (!fir.box<!fir.array<?xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<?xi32>> ! CHECK: hlfir.assign %[[DES_V1]] to %[[DES_V2]] : !fir.box<!fir.array<?xi32>>, !fir.box<!fir.array<?xi32>> ! CHECK: acc.terminator +! CHECK: } destroy { +! CHECK: ^bb0(%[[ARG0:.*]]: !fir.box<!fir.array<?xi32>>, %[[ARG1:.*]]: !fir.box<!fir.array<?xi32>>): +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[ARG1]] : (!fir.box<!fir.array<?xi32>>) -> !fir.ref<!fir.array<?xi32>> +! CHECK: %[[CAST:.*]] = fir.convert %[[ADDR]] : (!fir.ref<!fir.array<?xi32>>) -> !fir.heap<!fir.array<?xi32>> +! CHECK: fir.freemem %[[CAST]] : !fir.heap<!fir.array<?xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.private.recipe @privatization_box_UxUx2xi32 : !fir.box<!fir.array<?x?x2xi32>> init { @@ -74,6 +92,12 @@ ! CHECK: %[[TEMP:.*]] = fir.allocmem !fir.array<?x?x2xi32>, %[[DIM0]]#1, %[[DIM1]]#1 {bindc_name = ".tmp", uniq_name = ""} ! CHECK: %[[DECL:.*]]:2 = hlfir.declare %[[TEMP]](%[[SHAPE]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<?x?x2xi32>>, !fir.shape<3>) -> (!fir.box<!fir.array<?x?x2xi32>>, !fir.heap<!fir.array<?x?x2xi32>>) ! CHECK: acc.yield %[[DECL]]#0 : !fir.box<!fir.array<?x?x2xi32>> +! CHECK: } destroy { +! CHECK: ^bb0(%[[ARG0:.*]]: !fir.box<!fir.array<?x?x2xi32>>, %[[ARG1:.*]]: !fir.box<!fir.array<?x?x2xi32>>): +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[ARG1]] : (!fir.box<!fir.array<?x?x2xi32>>) -> !fir.ref<!fir.array<?x?x2xi32>> +! CHECK: %[[CAST:.*]] = fir.convert %[[ADDR]] : (!fir.ref<!fir.array<?x?x2xi32>>) -> !fir.heap<!fir.array<?x?x2xi32>> +! CHECK: fir.freemem %[[CAST]] : !fir.heap<!fir.array<?x?x2xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.private.recipe @privatization_ref_box_ptr_Uxi32 : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>> init { @@ -89,6 +113,13 @@ ! CHECK: %[[CONV:.*]] = fir.convert %[[DECLAREBOX]]#0 : (!fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>) -> !fir.ref<!fir.box<!fir.array<?xi32>>> ! CHECK: fir.store %[[DECLARE]]#0 to %[[CONV]] : !fir.ref<!fir.box<!fir.array<?xi32>>> ! CHECK: acc.yield %[[DECLAREBOX]]#0 : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>> +! CHECK: } destroy { +! CHECK: ^bb0(%arg0: !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>, %arg1: !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>): +! CHECK: %[[LOAD:.*]] = fir.load %arg1 : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>> +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[LOAD]] : (!fir.box<!fir.ptr<!fir.array<?xi32>>>) -> !fir.ptr<!fir.array<?xi32>> +! CHECK: %[[CAST:.*]] = fir.convert %[[ADDR]] : (!fir.ptr<!fir.array<?xi32>>) -> !fir.heap<!fir.array<?xi32>> +! CHECK: fir.freemem %[[CAST]] : !fir.heap<!fir.array<?xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: @privatization_ref_box_heap_i32 : !fir.ref<!fir.box<!fir.heap<i32>>> init { @@ -99,6 +130,12 @@ ! CHECK: %[[BOX:.*]] = fir.embox %[[ALLOCMEM]] : (!fir.heap<i32>) -> !fir.box<!fir.heap<i32>> ! CHECK: fir.store %[[BOX]] to %[[DECLARE]]#0 : !fir.ref<!fir.box<!fir.heap<i32>>> ! CHECK: acc.yield %[[DECLARE]]#0 : !fir.ref<!fir.box<!fir.heap<i32>>> +! CHECK: } destroy { +! CHECK: ^bb0(%arg0: !fir.ref<!fir.box<!fir.heap<i32>>>, %arg1: !fir.ref<!fir.box<!fir.heap<i32>>>): +! CHECK: %[[LOAD:.*]] = fir.load %arg1 : !fir.ref<!fir.box<!fir.heap<i32>>> +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[LOAD]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32> +! CHECK: fir.freemem %[[ADDR]] : !fir.heap<i32> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.private.recipe @privatization_ref_box_heap_Uxi32 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>> init { @@ -114,6 +151,12 @@ ! CHECK: %[[CONV:.*]] = fir.convert %[[DECLAREBOX]]#0 : (!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>) -> !fir.ref<!fir.box<!fir.array<?xi32>>> ! CHECK: fir.store %[[DECLARE]]#0 to %[[CONV]] : !fir.ref<!fir.box<!fir.array<?xi32>>> ! CHECK: acc.yield %[[DECLAREBOX]]#0 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>> +! CHECK: } destroy { +! CHECK: ^bb0(%arg0: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>, %arg1: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>): +! CHECK: %[[LOAD:.*]] = fir.load %arg1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>> +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[LOAD]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>) -> !fir.heap<!fir.array<?xi32>> +! CHECK: fir.freemem %[[ADDR]] : !fir.heap<!fir.array<?xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.private.recipe @privatization_box_Uxi32 : !fir.box<!fir.array<?xi32>> init { @@ -124,6 +167,12 @@ ! CHECK: %[[TEMP:.*]] = fir.allocmem !fir.array<?xi32>, %0#1 {bindc_name = ".tmp", uniq_name = ""} ! CHECK: %[[DECLARE:.*]]:2 = hlfir.declare %[[TEMP]](%[[SHAPE]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<?xi32>>, !fir.shape<1>) -> (!fir.box<!fir.array<?xi32>>, !fir.heap<!fir.array<?xi32>>) ! CHECK: acc.yield %[[DECLARE:.*]]#0 : !fir.box<!fir.array<?xi32>> +! CHECK: } destroy { +! CHECK: ^bb0(%[[ARG0:.*]]: !fir.box<!fir.array<?xi32>>, %[[ARG1:.*]]: !fir.box<!fir.array<?xi32>>): +! CHECK: %[[ADDR:.*]] = fir.box_addr %[[ARG1]] : (!fir.box<!fir.array<?xi32>>) -> !fir.ref<!fir.array<?xi32>> +! CHECK: %[[CAST:.*]] = fir.convert %[[ADDR]] : (!fir.ref<!fir.array<?xi32>>) -> !fir.heap<!fir.array<?xi32>> +! CHECK: fir.freemem %[[CAST]] : !fir.heap<!fir.array<?xi32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.firstprivate.recipe @firstprivatization_section_lb50.ub99_ref_50xf32 : !fir.ref<!fir.array<50xf32>> init { @@ -140,6 +189,7 @@ ! CHECK: %[[DES_SRC:.*]] = hlfir.designate %[[DECL_SRC]]#0 shape %[[SHAPE:.*]] : (!fir.ref<!fir.array<50xf32>>, !fir.shape<1>) -> !fir.ref<!fir.array<50xf32>> ! CHECK: %[[DES_DST:.*]] = hlfir.designate %[[DECL_DST]]#0 shape %[[SHAPE:.*]] : (!fir.ref<!fir.array<50xf32>>, !fir.shape<1>) -> !fir.ref<!fir.array<50xf32>> ! CHECK: hlfir.assign %[[DES_SRC]] to %[[DES_DST]] : !fir.ref<!fir.array<50xf32>>, !fir.ref<!fir.array<50xf32>> +! CHECK: acc.terminator ! CHECK: } ! CHECK-LABEL: acc.firstprivate.recipe @firstprivatization_ref_100xf32 : !fir.ref<!fir.array<100xf32>> init { diff --git a/libcxx/include/__cxx03/__atomic/atomic.h b/libcxx/include/__cxx03/__atomic/atomic.h index bc4a393..f2e8493 100644 --- a/libcxx/include/__cxx03/__atomic/atomic.h +++ b/libcxx/include/__cxx03/__atomic/atomic.h @@ -34,9 +34,7 @@ _LIBCPP_BEGIN_NAMESPACE_STD template <class _Tp> struct atomic : public __atomic_base<_Tp> { - using __base = __atomic_base<_Tp>; - using value_type = _Tp; - using difference_type = value_type; + using __base = __atomic_base<_Tp>; _LIBCPP_HIDE_FROM_ABI atomic() _NOEXCEPT = default; @@ -59,8 +57,8 @@ struct atomic : public __atomic_base<_Tp> { template <class _Tp> struct atomic<_Tp*> : public __atomic_base<_Tp*> { - using __base = __atomic_base<_Tp*>; - using value_type = _Tp*; + using __base = __atomic_base<_Tp*>; + using difference_type = ptrdiff_t; _LIBCPP_HIDE_FROM_ABI atomic() _NOEXCEPT = default; diff --git a/libcxx/include/__cxx03/__atomic/atomic_base.h b/libcxx/include/__cxx03/__atomic/atomic_base.h index a2b40c6..d79ef7d 100644 --- a/libcxx/include/__cxx03/__atomic/atomic_base.h +++ b/libcxx/include/__cxx03/__atomic/atomic_base.h @@ -32,6 +32,8 @@ struct __atomic_base // false { mutable __cxx_atomic_impl<_Tp> __a_; + using value_type = _Tp; + _LIBCPP_HIDE_FROM_ABI bool is_lock_free() const volatile _NOEXCEPT { return __cxx_atomic_is_lock_free(sizeof(__cxx_atomic_impl<_Tp>)); } @@ -127,6 +129,8 @@ template <class _Tp> struct __atomic_base<_Tp, true> : public __atomic_base<_Tp, false> { using __base = __atomic_base<_Tp, false>; + using difference_type = typename __base::value_type; + _LIBCPP_HIDE_FROM_ABI __atomic_base() _NOEXCEPT = default; _LIBCPP_HIDE_FROM_ABI __atomic_base(_Tp __d) _NOEXCEPT : __base(__d) {} diff --git a/libcxx/include/__cxx03/sstream b/libcxx/include/__cxx03/sstream index 44c2423..741158a 100644 --- a/libcxx/include/__cxx03/sstream +++ b/libcxx/include/__cxx03/sstream @@ -354,9 +354,15 @@ private: public: // [stringbuf.cons] constructors: - _LIBCPP_HIDE_FROM_ABI basic_stringbuf() : __hm_(nullptr), __mode_(ios_base::in | ios_base::out) {} + _LIBCPP_HIDE_FROM_ABI basic_stringbuf() : __hm_(nullptr), __mode_(ios_base::in | ios_base::out) { + // it is implementation-defined whether we initialize eback() & friends to nullptr, and libc++ doesn't + __init_buf_ptrs(); + } - _LIBCPP_HIDE_FROM_ABI explicit basic_stringbuf(ios_base::openmode __wch) : __hm_(nullptr), __mode_(__wch) {} + _LIBCPP_HIDE_FROM_ABI explicit basic_stringbuf(ios_base::openmode __wch) : __hm_(nullptr), __mode_(__wch) { + // it is implementation-defined whether we initialize eback() & friends to nullptr, and libc++ doesn't + __init_buf_ptrs(); + } _LIBCPP_HIDE_FROM_ABI explicit basic_stringbuf(const string_type& __s, ios_base::openmode __wch = ios_base::in | ios_base::out) diff --git a/libcxx/include/__cxx03/vector b/libcxx/include/__cxx03/vector index 4b62e0b..dbaa33c 100644 --- a/libcxx/include/__cxx03/vector +++ b/libcxx/include/__cxx03/vector @@ -432,10 +432,12 @@ public: template <__enable_if_t<__is_allocator<_Allocator>::value, int> = 0> _LIBCPP_HIDE_FROM_ABI vector(size_type __n, const value_type& __x, const allocator_type& __a) : __end_cap_(nullptr, __a) { + auto __guard = std::__make_exception_guard(__destroy_vector(*this)); if (__n > 0) { __vallocate(__n); __construct_at_end(__n, __x); } + __guard.__complete(); } template <class _InputIterator, @@ -1054,9 +1056,7 @@ inline _LIBCPP_HIDE_FROM_ABI vector<_Tp, _Allocator>::vector(vector&& __x, const __x.__begin_ = __x.__end_ = __x.__end_cap() = nullptr; } else { typedef move_iterator<iterator> _Ip; - auto __guard = std::__make_exception_guard(__destroy_vector(*this)); - assign(_Ip(__x.begin()), _Ip(__x.end())); - __guard.__complete(); + __init_with_size(_Ip(__x.begin()), _Ip(__x.end()), __x.size()); } } diff --git a/libcxx/test/benchmarks/bitset.bench.cpp b/libcxx/test/benchmarks/bitset.bench.cpp index 8fcf52e..b4c7e6f 100644 --- a/libcxx/test/benchmarks/bitset.bench.cpp +++ b/libcxx/test/benchmarks/bitset.bench.cpp @@ -103,7 +103,7 @@ BENCHMARK(BM_BitsetToString<262144>)->Arg(50)->Name("BM_BitsetToString<262144>/U BENCHMARK(BM_BitsetToString<524288>)->Arg(50)->Name("BM_BitsetToString<524288>/Uniform (50%)"); BENCHMARK(BM_BitsetToString<1048576>)->Arg(50)->Name("BM_BitsetToString<1048576>/Uniform (50%)"); // 1 << 20 -static void BM_ctor_ull(benchmark::State& state) { +static void BM_Bitset_ctor_ull(benchmark::State& state) { unsigned long long val = (1ULL << state.range(0)) - 1; for (auto _ : state) { std::bitset<128> b(val); @@ -111,6 +111,6 @@ static void BM_ctor_ull(benchmark::State& state) { } } -BENCHMARK(BM_ctor_ull)->DenseRange(1, 63); +BENCHMARK(BM_Bitset_ctor_ull)->DenseRange(1, 63); BENCHMARK_MAIN(); diff --git a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add.verify.cpp b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add.verify.cpp index 320ef57..1bd5792 100644 --- a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add.verify.cpp +++ b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add.verify.cpp @@ -6,8 +6,6 @@ // //===----------------------------------------------------------------------===// -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - // <atomic> // template <class T> @@ -48,12 +46,12 @@ void pointer_to_incomplete_type() { void function_pointer() { { volatile std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_add(&fun, 0); } { std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_add(&fun, 0); } } diff --git a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add_explicit.verify.cpp b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add_explicit.verify.cpp index bdd8089..bdd4a83 100644 --- a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add_explicit.verify.cpp +++ b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_add_explicit.verify.cpp @@ -6,8 +6,6 @@ // //===----------------------------------------------------------------------===// -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - // <atomic> // template <class T> @@ -51,12 +49,12 @@ void pointer_to_incomplete_type() { void function_pointer() { { volatile std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_add_explicit(&fun, 0, std::memory_order_relaxed); } { std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_add_explicit(&fun, 0, std::memory_order_relaxed); } } diff --git a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub.verify.cpp b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub.verify.cpp index 2c9f898..105a010 100644 --- a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub.verify.cpp +++ b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub.verify.cpp @@ -6,8 +6,6 @@ // //===----------------------------------------------------------------------===// -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - // <atomic> // template <class T> @@ -48,12 +46,12 @@ void pointer_to_incomplete_type() { void function_pointer() { { volatile std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_sub(&fun, 0); } { std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_sub(&fun, 0); } } diff --git a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub_explicit.verify.cpp b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub_explicit.verify.cpp index 88c4275..1647ed3 100644 --- a/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub_explicit.verify.cpp +++ b/libcxx/test/libcxx-03/atomics/atomics.types.operations/atomics.types.operations.req/atomic_fetch_sub_explicit.verify.cpp @@ -6,8 +6,6 @@ // //===----------------------------------------------------------------------===// -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - // <atomic> // template <class T> @@ -51,12 +49,12 @@ void pointer_to_incomplete_type() { void function_pointer() { { volatile std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_sub_explicit(&fun, 0, std::memory_order_relaxed); } { std::atomic<void (*)(int)> fun; - // expected-error-re@*:* {{static assertion failed due to requirement '!is_function<void (int)>::value'{{.*}}Pointer to function isn't allowed}} + // expected-error-re@*:* {{static assertion failed due to requirement {{.+}}Pointer to function isn't allowed}} std::atomic_fetch_sub_explicit(&fun, 0, std::memory_order_relaxed); } } diff --git a/libcxx/test/libcxx-03/input.output/string.streams/stringbuf/const_sso_buffer.pass.cpp b/libcxx/test/libcxx-03/input.output/string.streams/stringbuf/const_sso_buffer.pass.cpp index 2b6c380..d6caa33 100644 --- a/libcxx/test/libcxx-03/input.output/string.streams/stringbuf/const_sso_buffer.pass.cpp +++ b/libcxx/test/libcxx-03/input.output/string.streams/stringbuf/const_sso_buffer.pass.cpp @@ -8,8 +8,6 @@ // <sstream> -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - // How the constructors of basic_stringbuf initialize the buffer pointers is // not specified. For some constructors it's implementation defined whether the // pointers are set to nullptr. Libc++'s implementation directly uses the SSO diff --git a/libcxx/test/std/atomics/types.pass.cpp b/libcxx/test/std/atomics/types.pass.cpp index c979392..b1edec5 100644 --- a/libcxx/test/std/atomics/types.pass.cpp +++ b/libcxx/test/std/atomics/types.pass.cpp @@ -17,9 +17,6 @@ // typedef T value_type; // }; -// atomic still has a difference_type in the C++03 frozen headers -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - #include <atomic> #include <chrono> #include <cstdint> diff --git a/libcxx/test/std/containers/sequences/vector/vector.cons/exceptions.pass.cpp b/libcxx/test/std/containers/sequences/vector/vector.cons/exceptions.pass.cpp index 00de053..679eec2 100644 --- a/libcxx/test/std/containers/sequences/vector/vector.cons/exceptions.pass.cpp +++ b/libcxx/test/std/containers/sequences/vector/vector.cons/exceptions.pass.cpp @@ -11,8 +11,6 @@ // (bug report: https://llvm.org/PR58392) // Check that vector constructors don't leak memory when an operation inside the constructor throws an exception -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - #include <cstddef> #include <memory> #include <type_traits> diff --git a/libcxx/test/std/input.output/string.streams/stringbuf/stringbuf.cons/default.pass.cpp b/libcxx/test/std/input.output/string.streams/stringbuf/stringbuf.cons/default.pass.cpp index dac4396..d131f5c 100644 --- a/libcxx/test/std/input.output/string.streams/stringbuf/stringbuf.cons/default.pass.cpp +++ b/libcxx/test/std/input.output/string.streams/stringbuf/stringbuf.cons/default.pass.cpp @@ -15,8 +15,6 @@ // basic_stringbuf() : basic_stringbuf(ios_base::in | ios_base::out) {} // C++20 // explicit basic_stringbuf(ios_base::openmode which); // C++20 -// XFAIL: FROZEN-CXX03-HEADERS-FIXME - #include <sstream> #include <cassert> diff --git a/libcxx/utils/ci/docker-compose.yml b/libcxx/utils/ci/docker-compose.yml index cac97a9..9367a8f 100644 --- a/libcxx/utils/ci/docker-compose.yml +++ b/libcxx/utils/ci/docker-compose.yml @@ -23,7 +23,7 @@ services: dockerfile: Dockerfile target: actions-builder args: - GITHUB_RUNNER_VERSION: "2.328.0" + GITHUB_RUNNER_VERSION: "2.329.0" <<: [*image_versions, *compiler_versions] android-buildkite-builder: diff --git a/lldb/packages/Python/lldbsuite/test/gdbclientutils.py b/lldb/packages/Python/lldbsuite/test/gdbclientutils.py index 53e991a..1a2860a 100644 --- a/lldb/packages/Python/lldbsuite/test/gdbclientutils.py +++ b/lldb/packages/Python/lldbsuite/test/gdbclientutils.py @@ -1,3 +1,4 @@ +from abc import ABC, abstractmethod import ctypes import errno import io @@ -5,6 +6,7 @@ import threading import socket import traceback from lldbsuite.support import seven +from typing import Optional, List, Tuple def checksum(message): @@ -86,7 +88,7 @@ class MockGDBServerResponder: handles any packet not recognized in the common packet handling code. """ - registerCount = 40 + registerCount: int = 40 class RESPONSE_DISCONNECT: pass @@ -95,7 +97,7 @@ class MockGDBServerResponder: pass def __init__(self): - self.packetLog = [] + self.packetLog: List[str] = [] def respond(self, packet): """ @@ -241,7 +243,7 @@ class MockGDBServerResponder: def qHostInfo(self): return "ptrsize:8;endian:little;" - def qEcho(self): + def qEcho(self, num: int): return "E04" def qQueryGDBServer(self): @@ -262,10 +264,10 @@ class MockGDBServerResponder: def D(self, packet): return "OK" - def readRegisters(self): + def readRegisters(self) -> str: return "00000000" * self.registerCount - def readRegister(self, register): + def readRegister(self, register: int) -> str: return "00000000" def writeRegisters(self, registers_hex): @@ -305,7 +307,9 @@ class MockGDBServerResponder: # SIGINT is 2, return type is 2 digit hex string return "S02" - def qXferRead(self, obj, annex, offset, length): + def qXferRead( + self, obj: str, annex: str, offset: int, length: int + ) -> Tuple[Optional[str], bool]: return None, False def _qXferResponse(self, data, has_more): @@ -373,15 +377,17 @@ class MockGDBServerResponder: pass -class ServerChannel: +class ServerChannel(ABC): """ A wrapper class for TCP or pty-based server. """ - def get_connect_address(self): + @abstractmethod + def get_connect_address(self) -> str: """Get address for the client to connect to.""" - def get_connect_url(self): + @abstractmethod + def get_connect_url(self) -> str: """Get URL suitable for process connect command.""" def close_server(self): @@ -393,10 +399,12 @@ class ServerChannel: def close_connection(self): """Close all resources used by the accepted connection.""" - def recv(self): + @abstractmethod + def recv(self) -> bytes: """Receive a data packet from the connected client.""" - def sendall(self, data): + @abstractmethod + def sendall(self, data: bytes) -> None: """Send the data to the connected client.""" @@ -427,11 +435,11 @@ class ServerSocket(ServerChannel): self._connection.close() self._connection = None - def recv(self): + def recv(self) -> bytes: assert self._connection is not None return self._connection.recv(4096) - def sendall(self, data): + def sendall(self, data: bytes) -> None: assert self._connection is not None return self._connection.sendall(data) @@ -443,10 +451,10 @@ class TCPServerSocket(ServerSocket): )[0] super().__init__(family, type, proto, addr) - def get_connect_address(self): + def get_connect_address(self) -> str: return "[{}]:{}".format(*self._server_socket.getsockname()) - def get_connect_url(self): + def get_connect_url(self) -> str: return "connect://" + self.get_connect_address() @@ -454,10 +462,10 @@ class UnixServerSocket(ServerSocket): def __init__(self, addr): super().__init__(socket.AF_UNIX, socket.SOCK_STREAM, 0, addr) - def get_connect_address(self): + def get_connect_address(self) -> str: return self._server_socket.getsockname() - def get_connect_url(self): + def get_connect_url(self) -> str: return "unix-connect://" + self.get_connect_address() @@ -471,7 +479,7 @@ class PtyServerSocket(ServerChannel): self._primary = io.FileIO(primary, "r+b") self._secondary = io.FileIO(secondary, "r+b") - def get_connect_address(self): + def get_connect_address(self) -> str: libc = ctypes.CDLL(None) libc.ptsname.argtypes = (ctypes.c_int,) libc.ptsname.restype = ctypes.c_char_p @@ -484,7 +492,7 @@ class PtyServerSocket(ServerChannel): self._secondary.close() self._primary.close() - def recv(self): + def recv(self) -> bytes: try: return self._primary.read(4096) except OSError as e: @@ -493,8 +501,8 @@ class PtyServerSocket(ServerChannel): return b"" raise - def sendall(self, data): - return self._primary.write(data) + def sendall(self, data: bytes) -> None: + self._primary.write(data) class MockGDBServer: @@ -527,18 +535,21 @@ class MockGDBServer: self._thread.join() self._thread = None - def get_connect_address(self): + def get_connect_address(self) -> str: + assert self._socket is not None return self._socket.get_connect_address() - def get_connect_url(self): + def get_connect_url(self) -> str: + assert self._socket is not None return self._socket.get_connect_url() def run(self): + assert self._socket is not None # For testing purposes, we only need to worry about one client # connecting just one time. try: self._socket.accept() - except: + except Exception: traceback.print_exc() return self._shouldSendAck = True @@ -553,7 +564,7 @@ class MockGDBServer: self._receive(data) except self.TerminateConnectionException: pass - except Exception as e: + except Exception: print( "An exception happened when receiving the response from the gdb server. Closing the client..." ) @@ -586,7 +597,9 @@ class MockGDBServer: Once a complete packet is found at the front of self._receivedData, its data is removed form self._receivedData. """ + assert self._receivedData is not None data = self._receivedData + assert self._receivedDataOffset is not None i = self._receivedDataOffset data_len = len(data) if data_len == 0: @@ -639,10 +652,13 @@ class MockGDBServer: self._receivedDataOffset = 0 return packet - def _sendPacket(self, packet): - self._socket.sendall(seven.bitcast_to_bytes(frame_packet(packet))) + def _sendPacket(self, packet: str): + assert self._socket is not None + framed_packet = seven.bitcast_to_bytes(frame_packet(packet)) + self._socket.sendall(framed_packet) def _handlePacket(self, packet): + assert self._socket is not None if packet is self.PACKET_ACK: # Ignore ACKs from the client. For the future, we can consider # adding validation code to make sure the client only sends ACKs diff --git a/lldb/source/Host/freebsd/Host.cpp b/lldb/source/Host/freebsd/Host.cpp index fa7efad..dfdbfea 100644 --- a/lldb/source/Host/freebsd/Host.cpp +++ b/lldb/source/Host/freebsd/Host.cpp @@ -18,8 +18,6 @@ #include <dlfcn.h> #include <execinfo.h> -#include "llvm/Object/ELF.h" - #include "lldb/Host/FileSystem.h" #include "lldb/Host/Host.h" #include "lldb/Host/HostInfo.h" @@ -32,6 +30,7 @@ #include "lldb/Utility/Status.h" #include "lldb/Utility/StreamString.h" +#include "llvm/Object/ELF.h" #include "llvm/TargetParser/Host.h" namespace lldb_private { diff --git a/lldb/source/Plugins/Language/CPlusPlus/LibCxxUnorderedMap.cpp b/lldb/source/Plugins/Language/CPlusPlus/LibCxxUnorderedMap.cpp index 4b183a8..5588208 100644 --- a/lldb/source/Plugins/Language/CPlusPlus/LibCxxUnorderedMap.cpp +++ b/lldb/source/Plugins/Language/CPlusPlus/LibCxxUnorderedMap.cpp @@ -52,7 +52,7 @@ private: ValueObject *m_tree = nullptr; size_t m_num_elements = 0; ValueObject *m_next_element = nullptr; - std::vector<std::pair<ValueObject *, uint64_t>> m_elements_cache; + std::vector<ValueObject *> m_elements_cache; }; class LibCxxUnorderedMapIteratorSyntheticFrontEnd @@ -192,26 +192,25 @@ lldb::ValueObjectSP lldb_private::formatters:: return nullptr; } } - m_elements_cache.push_back( - {value_sp.get(), hash_sp->GetValueAsUnsigned(0)}); + m_elements_cache.push_back(value_sp.get()); m_next_element = node_sp->GetChildMemberWithName("__next_").get(); if (!m_next_element || m_next_element->GetValueAsUnsigned(0) == 0) m_next_element = nullptr; } - std::pair<ValueObject *, uint64_t> val_hash = m_elements_cache[idx]; - if (!val_hash.first) + ValueObject *val_hash = m_elements_cache[idx]; + if (!val_hash) return lldb::ValueObjectSP(); StreamString stream; stream.Printf("[%" PRIu64 "]", (uint64_t)idx); DataExtractor data; Status error; - val_hash.first->GetData(data, error); + val_hash->GetData(data, error); if (error.Fail()) return lldb::ValueObjectSP(); const bool thread_and_frame_only_if_stopped = true; - ExecutionContext exe_ctx = val_hash.first->GetExecutionContextRef().Lock( - thread_and_frame_only_if_stopped); + ExecutionContext exe_ctx = + val_hash->GetExecutionContextRef().Lock(thread_and_frame_only_if_stopped); return CreateValueObjectFromData(stream.GetString(), data, exe_ctx, m_element_type); } diff --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst index a4d110f..402fd05 100644 --- a/llvm/docs/AMDGPUUsage.rst +++ b/llvm/docs/AMDGPUUsage.rst @@ -4172,7 +4172,7 @@ non-AMD key names should be prefixed by "*vendor-name*.". "Image", or "Pipe". This may be more restrictive than indicated by "AccQual" to reflect what the - kernel actual does. If not + kernel actually does. If not present then the runtime must assume what is implied by "AccQual" and "IsConst". Values @@ -5436,8 +5436,8 @@ The fields used by CP for code objects before V3 also match those specified in ``COMPUTE_PGM_RSRC1.PRIORITY``. 13:12 2 bits FLOAT_ROUND_MODE_32 Wavefront starts execution with specified rounding - mode for single (32 - bit) floating point + mode for single (32-bit) + floating point precision floating point operations. @@ -5769,7 +5769,7 @@ The fields used by CP for code objects before V3 also match those specified in Wavefront starts execution with memory violation - exceptions exceptions + exceptions enabled which are generated when a memory violation has occurred for this wavefront from @@ -6005,7 +6005,7 @@ The fields used by CP for code objects before V3 also match those specified in FLOAT_DENORM_MODE_FLUSH_NONE 3 No Flush ====================================== ===== ==================================== - Denormal flushing is sign respecting. i.e. the behavior expected by + Denormal flushing is sign respecting, i.e., the behavior expected by ``"denormal-fp-math"="preserve-sign"``. The behavior is undefined with ``"denormal-fp-math"="positive-zero"`` @@ -16831,7 +16831,7 @@ For GFX125x: * Some memory operations contain a ``nv`` bit, for "non-volatile", which indicates memory that is not expected to change during a kernel's execution. This information is propagated to the cache lines for that address - (refered to as ``$nv``). + (referred to as ``$nv``). * When ``nv=0`` reads hit dirty ``$nv=1`` data in cache, the hardware will writeback the data to the next level in the hierarchy and then subsequently read @@ -18970,7 +18970,7 @@ On entry to a function: #. All other registers are unspecified. #. Any necessary ``s_waitcnt`` has been performed to ensure memory is available to the function. -#. Use pass-by-reference (byref) in stead of pass-by-value (byval) for struct +#. Use pass-by-reference (byref) instead of pass-by-value (byval) for struct arguments in C ABI. Callee is responsible for allocating stack memory and copying the value of the struct if modified. Note that the backend still supports byval for struct arguments. @@ -20214,7 +20214,7 @@ from the value of the ``-mcpu`` option that is passed to the assembler. .amdgpu_hsa_kernel (name) +++++++++++++++++++++++++ -This directives specifies that the symbol with given name is a kernel entry +This directive specifies that the symbol with given name is a kernel entry point (label) and the object should contain corresponding symbol of type STT_AMDGPU_HSA_KERNEL. diff --git a/llvm/docs/GettingStartedVS.rst b/llvm/docs/GettingStartedVS.rst index bc5746d..e65fd8f 100644 --- a/llvm/docs/GettingStartedVS.rst +++ b/llvm/docs/GettingStartedVS.rst @@ -126,6 +126,15 @@ These instructions were tested with Visual Studio 2019 and Python 3.9.6: cmake -S llvm\llvm -B build -DLLVM_ENABLE_PROJECTS=clang -DLLVM_TARGETS_TO_BUILD=X86 -Thost=x64 exit + .. note:: + By default, the Visual Studio project files generated by CMake use the + 32-bit toolset. If you are developing on a 64-bit version of Windows and + want to use the 64-bit toolset, pass the ``-Thost=x64`` flag when + generating the Visual Studio solution. This requires CMake 3.8.0 or later. + + For Windows on Arm the equivalent is ``-Thost=ARM64``, but this the default + for those hosts, so you do not have to use this option. + ``LLVM_ENABLE_PROJECTS`` specifies any additional LLVM projects you want to build while ``LLVM_TARGETS_TO_BUILD`` selects the compiler targets. If ``LLVM_TARGETS_TO_BUILD`` is omitted by default all targets are built @@ -149,10 +158,6 @@ These instructions were tested with Visual Studio 2019 and Python 3.9.6: * CMake generates project files for all build types. To select a specific build type, use the Configuration manager from the VS IDE or the ``/property:Configuration`` command-line option when using MSBuild. - * By default, the Visual Studio project files generated by CMake use the - 32-bit toolset. If you are developing on a 64-bit version of Windows and - want to use the 64-bit toolset, pass the ``-Thost=x64`` flag when - generating the Visual Studio solution. This requires CMake 3.8.0 or later. 13. Start Visual Studio and select configuration: diff --git a/llvm/include/llvm/ADT/StringSwitch.h b/llvm/include/llvm/ADT/StringSwitch.h index a96535c..1bb07e39 100644 --- a/llvm/include/llvm/ADT/StringSwitch.h +++ b/llvm/include/llvm/ADT/StringSwitch.h @@ -54,17 +54,14 @@ public: explicit StringSwitch(StringRef S) : Str(S), Result() { } + StringSwitch(StringSwitch &&) = default; + // StringSwitch is not copyable. StringSwitch(const StringSwitch &) = delete; // StringSwitch is not assignable due to 'Str' being 'const'. void operator=(const StringSwitch &) = delete; - void operator=(StringSwitch &&other) = delete; - - StringSwitch(StringSwitch &&other) - : Str(other.Str), Result(std::move(other.Result)) { } - - ~StringSwitch() = default; + void operator=(StringSwitch &&) = delete; // Case-sensitive case matchers StringSwitch &Case(StringLiteral S, T Value) { diff --git a/llvm/include/llvm/BinaryFormat/ELFRelocs/AArch64.def b/llvm/include/llvm/BinaryFormat/ELFRelocs/AArch64.def index 8dcc292..1cfcdbf 100644 --- a/llvm/include/llvm/BinaryFormat/ELFRelocs/AArch64.def +++ b/llvm/include/llvm/BinaryFormat/ELFRelocs/AArch64.def @@ -62,6 +62,7 @@ ELF_RELOC(R_AARCH64_LD64_GOTPAGE_LO15, 0x139) ELF_RELOC(R_AARCH64_PLT32, 0x13a) ELF_RELOC(R_AARCH64_GOTPCREL32, 0x13b) ELF_RELOC(R_AARCH64_PATCHINST, 0x13c) +ELF_RELOC(R_AARCH64_FUNCINIT64, 0x13d) // General dynamic TLS relocations ELF_RELOC(R_AARCH64_TLSGD_ADR_PREL21, 0x200) ELF_RELOC(R_AARCH64_TLSGD_ADR_PAGE21, 0x201) diff --git a/llvm/include/llvm/CAS/CASID.h b/llvm/include/llvm/CAS/CASID.h index 8820994..f508ed3 100644 --- a/llvm/include/llvm/CAS/CASID.h +++ b/llvm/include/llvm/CAS/CASID.h @@ -95,8 +95,7 @@ public: } friend hash_code hash_value(const CASID &ID) { - ArrayRef<uint8_t> Hash = ID.getHash(); - return hash_combine_range(Hash.begin(), Hash.end()); + return hash_combine_range(ID.getHash()); } const CASContext &getContext() const { diff --git a/llvm/include/llvm/CodeGen/LiveIntervals.h b/llvm/include/llvm/CodeGen/LiveIntervals.h index 1050b3d..c252f9d 100644 --- a/llvm/include/llvm/CodeGen/LiveIntervals.h +++ b/llvm/include/llvm/CodeGen/LiveIntervals.h @@ -229,8 +229,8 @@ public: /// doing something wrong if you call pruneValue directly on a /// LiveInterval. Indeed, you are supposed to call pruneValue on the main /// LiveRange and all the LiveRanges of the subranges if any. - LLVM_ATTRIBUTE_UNUSED void pruneValue(LiveInterval &, SlotIndex, - SmallVectorImpl<SlotIndex> *) { + [[maybe_unused]] void pruneValue(LiveInterval &, SlotIndex, + SmallVectorImpl<SlotIndex> *) { llvm_unreachable( "Use pruneValue on the main LiveRange and on each subrange"); } diff --git a/llvm/include/llvm/Object/ELFTypes.h b/llvm/include/llvm/Object/ELFTypes.h index 5a26e2f..e9a417d 100644 --- a/llvm/include/llvm/Object/ELFTypes.h +++ b/llvm/include/llvm/Object/ELFTypes.h @@ -833,6 +833,7 @@ struct BBAddrMap { bool MultiBBRange : 1; bool OmitBBEntries : 1; bool CallsiteEndOffsets : 1; + bool BBHash : 1; bool hasPGOAnalysis() const { return FuncEntryCount || BBFreq || BrProb; } @@ -845,7 +846,8 @@ struct BBAddrMap { (static_cast<uint8_t>(BrProb) << 2) | (static_cast<uint8_t>(MultiBBRange) << 3) | (static_cast<uint8_t>(OmitBBEntries) << 4) | - (static_cast<uint8_t>(CallsiteEndOffsets) << 5); + (static_cast<uint8_t>(CallsiteEndOffsets) << 5) | + (static_cast<uint8_t>(BBHash) << 6); } // Decodes from minimum bit width representation and validates no @@ -854,7 +856,8 @@ struct BBAddrMap { Features Feat{ static_cast<bool>(Val & (1 << 0)), static_cast<bool>(Val & (1 << 1)), static_cast<bool>(Val & (1 << 2)), static_cast<bool>(Val & (1 << 3)), - static_cast<bool>(Val & (1 << 4)), static_cast<bool>(Val & (1 << 5))}; + static_cast<bool>(Val & (1 << 4)), static_cast<bool>(Val & (1 << 5)), + static_cast<bool>(Val & (1 << 6))}; if (Feat.encode() != Val) return createStringError( std::error_code(), "invalid encoding for BBAddrMap::Features: 0x%x", @@ -864,10 +867,10 @@ struct BBAddrMap { bool operator==(const Features &Other) const { return std::tie(FuncEntryCount, BBFreq, BrProb, MultiBBRange, - OmitBBEntries, CallsiteEndOffsets) == + OmitBBEntries, CallsiteEndOffsets, BBHash) == std::tie(Other.FuncEntryCount, Other.BBFreq, Other.BrProb, Other.MultiBBRange, Other.OmitBBEntries, - Other.CallsiteEndOffsets); + Other.CallsiteEndOffsets, Other.BBHash); } }; @@ -920,17 +923,19 @@ struct BBAddrMap { false}; // Metdata for this basic block. // Offsets of end of call instructions, relative to the basic block start. SmallVector<uint32_t, 1> CallsiteEndOffsets; + uint64_t Hash = 0; // Hash for this basic block. BBEntry(uint32_t ID, uint32_t Offset, uint32_t Size, Metadata MD, - SmallVector<uint32_t, 1> CallsiteEndOffsets) + SmallVector<uint32_t, 1> CallsiteEndOffsets, uint64_t Hash) : ID(ID), Offset(Offset), Size(Size), MD(MD), - CallsiteEndOffsets(std::move(CallsiteEndOffsets)) {} + CallsiteEndOffsets(std::move(CallsiteEndOffsets)), Hash(Hash) {} UniqueBBID getID() const { return {ID, 0}; } bool operator==(const BBEntry &Other) const { return ID == Other.ID && Offset == Other.Offset && Size == Other.Size && - MD == Other.MD && CallsiteEndOffsets == Other.CallsiteEndOffsets; + MD == Other.MD && CallsiteEndOffsets == Other.CallsiteEndOffsets && + Hash == Other.Hash; } bool hasReturn() const { return MD.HasReturn; } diff --git a/llvm/include/llvm/ObjectYAML/ELFYAML.h b/llvm/include/llvm/ObjectYAML/ELFYAML.h index c90591d..a7c7c7c 100644 --- a/llvm/include/llvm/ObjectYAML/ELFYAML.h +++ b/llvm/include/llvm/ObjectYAML/ELFYAML.h @@ -163,6 +163,7 @@ struct BBAddrMapEntry { llvm::yaml::Hex64 Size; llvm::yaml::Hex64 Metadata; std::optional<std::vector<llvm::yaml::Hex64>> CallsiteEndOffsets; + std::optional<llvm::yaml::Hex64> Hash; }; uint8_t Version; llvm::yaml::Hex8 Feature; diff --git a/llvm/lib/Analysis/InstructionSimplify.cpp b/llvm/lib/Analysis/InstructionSimplify.cpp index 4e38626..e08ef60 100644 --- a/llvm/lib/Analysis/InstructionSimplify.cpp +++ b/llvm/lib/Analysis/InstructionSimplify.cpp @@ -6644,7 +6644,7 @@ Value *llvm::simplifyBinaryIntrinsic(Intrinsic::ID IID, Type *ReturnType, "Invalid mask width"); // If index-width (mask size) is less than pointer-size then mask is // 1-extended. - if (match(Op1, m_PtrToInt(m_Specific(Op0)))) + if (match(Op1, m_PtrToIntOrAddr(m_Specific(Op0)))) return Op0; // NOTE: We may have attributes associated with the return value of the diff --git a/llvm/lib/Analysis/MemorySSA.cpp b/llvm/lib/Analysis/MemorySSA.cpp index ab37338..0b2e3fc 100644 --- a/llvm/lib/Analysis/MemorySSA.cpp +++ b/llvm/lib/Analysis/MemorySSA.cpp @@ -393,7 +393,7 @@ static bool isUseTriviallyOptimizableToLiveOnEntry(AliasAnalysisType &AA, /// \param AA The AliasAnalysis we used for our search. /// \param AllowImpreciseClobber Always false, unless we do relaxed verify. -LLVM_ATTRIBUTE_UNUSED static void +[[maybe_unused]] static void checkClobberSanity(const MemoryAccess *Start, MemoryAccess *ClobberAt, const MemoryLocation &StartLoc, const MemorySSA &MSSA, const UpwardsMemoryQuery &Query, BatchAAResults &AA, diff --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp index 3fab6b0..4bafd3f 100644 --- a/llvm/lib/Analysis/ScalarEvolution.cpp +++ b/llvm/lib/Analysis/ScalarEvolution.cpp @@ -15761,6 +15761,21 @@ void ScalarEvolution::LoopGuards::collectFromBlock( const SCEV *OneAlignedUp = GetNextSCEVDividesByDivisor(One, DividesBy); To = SE.getUMaxExpr(FromRewritten, OneAlignedUp); + } else { + if (LHS->getType()->isPointerTy()) { + LHS = SE.getLosslessPtrToIntExpr(LHS); + RHS = SE.getLosslessPtrToIntExpr(RHS); + if (isa<SCEVCouldNotCompute>(LHS) || isa<SCEVCouldNotCompute>(RHS)) + break; + } + auto AddSubRewrite = [&](const SCEV *A, const SCEV *B) { + const SCEV *Sub = SE.getMinusSCEV(A, B); + AddRewrite(Sub, Sub, + SE.getUMaxExpr(Sub, SE.getOne(From->getType()))); + }; + AddSubRewrite(LHS, RHS); + AddSubRewrite(RHS, LHS); + continue; } break; default: diff --git a/llvm/lib/CodeGen/AsmPrinter/AsmPrinter.cpp b/llvm/lib/CodeGen/AsmPrinter/AsmPrinter.cpp index 219bbc9..05fffe9 100644 --- a/llvm/lib/CodeGen/AsmPrinter/AsmPrinter.cpp +++ b/llvm/lib/CodeGen/AsmPrinter/AsmPrinter.cpp @@ -1437,7 +1437,8 @@ getBBAddrMapFeature(const MachineFunction &MF, int NumMBBSectionRanges, BrProbEnabled, MF.hasBBSections() && NumMBBSectionRanges > 1, static_cast<bool>(BBAddrMapSkipEmitBBEntries), - HasCalls}; + HasCalls, + false}; } void AsmPrinter::emitBBAddrMapSection(const MachineFunction &MF) { diff --git a/llvm/lib/CodeGen/AtomicExpandPass.cpp b/llvm/lib/CodeGen/AtomicExpandPass.cpp index 4931403..53f1cfe2 100644 --- a/llvm/lib/CodeGen/AtomicExpandPass.cpp +++ b/llvm/lib/CodeGen/AtomicExpandPass.cpp @@ -770,7 +770,7 @@ struct PartwordMaskValues { Value *Inv_Mask = nullptr; }; -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &operator<<(raw_ostream &O, const PartwordMaskValues &PMV) { auto PrintObj = [&O](auto *V) { if (V) diff --git a/llvm/lib/CodeGen/GlobalISel/GISelValueTracking.cpp b/llvm/lib/CodeGen/GlobalISel/GISelValueTracking.cpp index 9b9a6a2..04d9309 100644 --- a/llvm/lib/CodeGen/GlobalISel/GISelValueTracking.cpp +++ b/llvm/lib/CodeGen/GlobalISel/GISelValueTracking.cpp @@ -112,7 +112,7 @@ APInt GISelValueTracking::getKnownOnes(Register R) { return getKnownBits(R).One; } -LLVM_ATTRIBUTE_UNUSED static void +[[maybe_unused]] static void dumpResult(const MachineInstr &MI, const KnownBits &Known, unsigned Depth) { dbgs() << "[" << Depth << "] Compute known bits: " << MI << "[" << Depth << "] Computed for: " << MI << "[" << Depth << "] Known: 0x" diff --git a/llvm/lib/CodeGen/ScheduleDAGInstrs.cpp b/llvm/lib/CodeGen/ScheduleDAGInstrs.cpp index 3268c26..9662511 100644 --- a/llvm/lib/CodeGen/ScheduleDAGInstrs.cpp +++ b/llvm/lib/CodeGen/ScheduleDAGInstrs.cpp @@ -1551,7 +1551,7 @@ LLVM_DUMP_METHOD void ILPValue::dump() const { dbgs() << *this << '\n'; } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &llvm::operator<<(raw_ostream &OS, const ILPValue &Val) { Val.print(OS); return OS; diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp index 876066e9..90edaf3 100644 --- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp +++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp @@ -6405,8 +6405,9 @@ static SDValue foldCONCAT_VECTORS(const SDLoc &DL, EVT VT, if (VT.isScalableVector()) return SDValue(); - // A CONCAT_VECTOR with all UNDEF/BUILD_VECTOR operands can be - // simplified to one big BUILD_VECTOR. + // A CONCAT_VECTOR of scalar sources, such as UNDEF, BUILD_VECTOR and + // single-element INSERT_VECTOR_ELT operands can be simplified to one big + // BUILD_VECTOR. // FIXME: Add support for SCALAR_TO_VECTOR as well. EVT SVT = VT.getScalarType(); SmallVector<SDValue, 16> Elts; @@ -6416,6 +6417,10 @@ static SDValue foldCONCAT_VECTORS(const SDLoc &DL, EVT VT, Elts.append(OpVT.getVectorNumElements(), DAG.getUNDEF(SVT)); else if (Op.getOpcode() == ISD::BUILD_VECTOR) Elts.append(Op->op_begin(), Op->op_end()); + else if (Op.getOpcode() == ISD::INSERT_VECTOR_ELT && + OpVT.getVectorNumElements() == 1 && + isNullConstant(Op.getOperand(2))) + Elts.push_back(Op.getOperand(1)); else return SDValue(); } diff --git a/llvm/lib/ObjCopy/ConfigManager.cpp b/llvm/lib/ObjCopy/ConfigManager.cpp index eef8a21..6b7b4f1 100644 --- a/llvm/lib/ObjCopy/ConfigManager.cpp +++ b/llvm/lib/ObjCopy/ConfigManager.cpp @@ -122,14 +122,14 @@ ConfigManager::getDXContainerConfig() const { if (!Common.AddGnuDebugLink.empty() || !Common.SplitDWO.empty() || !Common.AllocSectionsPrefix.empty() || Common.DiscardMode != DiscardType::None || !Common.AddSection.empty() || - !Common.DumpSection.empty() || !Common.KeepSection.empty() || - !Common.SectionsToRename.empty() || !Common.SetSectionAlignment.empty() || - !Common.SetSectionFlags.empty() || !Common.SetSectionType.empty() || - Common.ExtractDWO || Common.OnlyKeepDebug || Common.StripAllGNU || - Common.StripDWO || Common.StripDebug || Common.StripNonAlloc || - Common.StripSections || Common.StripUnneeded || - Common.DecompressDebugSections || Common.GapFill != 0 || - Common.PadTo != 0 || Common.ChangeSectionLMAValAll != 0 || + !Common.KeepSection.empty() || !Common.SectionsToRename.empty() || + !Common.SetSectionAlignment.empty() || !Common.SetSectionFlags.empty() || + !Common.SetSectionType.empty() || Common.ExtractDWO || + Common.OnlyKeepDebug || Common.StripAllGNU || Common.StripDWO || + Common.StripDebug || Common.StripNonAlloc || Common.StripSections || + Common.StripUnneeded || Common.DecompressDebugSections || + Common.GapFill != 0 || Common.PadTo != 0 || + Common.ChangeSectionLMAValAll != 0 || !Common.ChangeSectionAddress.empty()) { return createStringError(llvm::errc::invalid_argument, "option is not supported for DXContainer"); diff --git a/llvm/lib/ObjCopy/DXContainer/DXContainerObjcopy.cpp b/llvm/lib/ObjCopy/DXContainer/DXContainerObjcopy.cpp index d7f3c0d..95ab3d9 100644 --- a/llvm/lib/ObjCopy/DXContainer/DXContainerObjcopy.cpp +++ b/llvm/lib/ObjCopy/DXContainer/DXContainerObjcopy.cpp @@ -9,8 +9,10 @@ #include "llvm/ObjCopy/DXContainer/DXContainerObjcopy.h" #include "DXContainerReader.h" #include "DXContainerWriter.h" +#include "llvm/BinaryFormat/DXContainer.h" #include "llvm/ObjCopy/CommonConfig.h" #include "llvm/ObjCopy/DXContainer/DXContainerConfig.h" +#include "llvm/Support/FileOutputBuffer.h" #include "llvm/Support/raw_ostream.h" namespace llvm { @@ -42,7 +44,47 @@ static Error extractPartAsObject(StringRef PartName, StringRef OutFilename, "part '%s' not found", PartName.str().c_str()); } +static Error dumpPartToFile(StringRef PartName, StringRef Filename, + StringRef InputFilename, Object &Obj) { + auto PartIter = llvm::find_if( + Obj.Parts, [&PartName](const Part &P) { return P.Name == PartName; }); + if (PartIter == Obj.Parts.end()) + return createFileError(Filename, + std::make_error_code(std::errc::invalid_argument), + "part '%s' not found", PartName.str().c_str()); + ArrayRef<uint8_t> Contents = PartIter->Data; + // The DXContainer format is a bit odd because the part-specific headers are + // contained inside the part data itself. For parts that contain LLVM bitcode + // when we dump the part we want to skip the part-specific header so that we + // get a valid .bc file that we can inspect. All the data contained inside the + // program header is pulled out of the bitcode, so the header can be + // reconstructed if needed from the bitcode itself. More comprehensive + // documentation on the DXContainer format can be found at + // https://llvm.org/docs/DirectX/DXContainer.html. + + if (PartName == "DXIL" || PartName == "STAT") + Contents = Contents.drop_front(sizeof(llvm::dxbc::ProgramHeader)); + if (Contents.empty()) + return createFileError(Filename, object_error::parse_failed, + "part '%s' is empty", PartName.str().c_str()); + Expected<std::unique_ptr<FileOutputBuffer>> BufferOrErr = + FileOutputBuffer::create(Filename, Contents.size()); + if (!BufferOrErr) + return createFileError(Filename, BufferOrErr.takeError()); + std::unique_ptr<FileOutputBuffer> Buf = std::move(*BufferOrErr); + llvm::copy(Contents, Buf->getBufferStart()); + if (Error E = Buf->commit()) + return createFileError(Filename, std::move(E)); + return Error::success(); +} + static Error handleArgs(const CommonConfig &Config, Object &Obj) { + for (StringRef Flag : Config.DumpSection) { + auto [SecName, FileName] = Flag.split("="); + if (Error E = dumpPartToFile(SecName, FileName, Config.InputFilename, Obj)) + return E; + } + // Extract all sections before any modifications. for (StringRef Flag : Config.ExtractSection) { StringRef SectionName; diff --git a/llvm/lib/Object/ELF.cpp b/llvm/lib/Object/ELF.cpp index 53699ce0..f256e7b 100644 --- a/llvm/lib/Object/ELF.cpp +++ b/llvm/lib/Object/ELF.cpp @@ -837,7 +837,7 @@ decodeBBAddrMapImpl(const ELFFile<ELFT> &EF, Version = Data.getU8(Cur); if (!Cur) break; - if (Version < 2 || Version > 3) + if (Version < 2 || Version > 4) return createError("unsupported SHT_LLVM_BB_ADDR_MAP version: " + Twine(static_cast<int>(Version))); Feature = Data.getU8(Cur); // Feature byte @@ -852,6 +852,11 @@ decodeBBAddrMapImpl(const ELFFile<ELFT> &EF, "callsite offsets feature is enabled: version = " + Twine(static_cast<int>(Version)) + " feature = " + Twine(static_cast<int>(Feature))); + if (FeatEnable.BBHash && Version < 4) + return createError("version should be >= 4 for SHT_LLVM_BB_ADDR_MAP when " + "basic block hash feature is enabled: version = " + + Twine(static_cast<int>(Version)) + + " feature = " + Twine(static_cast<int>(Feature))); uint32_t NumBlocksInBBRange = 0; uint32_t NumBBRanges = 1; typename ELFFile<ELFT>::uintX_t RangeBaseAddress = 0; @@ -907,6 +912,7 @@ decodeBBAddrMapImpl(const ELFFile<ELFT> &EF, uint32_t Size = readULEB128As<uint32_t>(Data, Cur, ULEBSizeErr) + LastCallsiteEndOffset; uint32_t MD = readULEB128As<uint32_t>(Data, Cur, ULEBSizeErr); + uint64_t Hash = FeatEnable.BBHash ? Data.getU64(Cur) : 0; Expected<BBAddrMap::BBEntry::Metadata> MetadataOrErr = BBAddrMap::BBEntry::Metadata::decode(MD); if (!MetadataOrErr) { @@ -914,7 +920,7 @@ decodeBBAddrMapImpl(const ELFFile<ELFT> &EF, break; } BBEntries.push_back({ID, Offset + PrevBBEndOffset, Size, - *MetadataOrErr, CallsiteEndOffsets}); + *MetadataOrErr, CallsiteEndOffsets, Hash}); PrevBBEndOffset += Offset + Size; } TotalNumBlocks += BBEntries.size(); diff --git a/llvm/lib/ObjectYAML/ELFEmitter.cpp b/llvm/lib/ObjectYAML/ELFEmitter.cpp index faeeab3..8b75fbe 100644 --- a/llvm/lib/ObjectYAML/ELFEmitter.cpp +++ b/llvm/lib/ObjectYAML/ELFEmitter.cpp @@ -1465,7 +1465,7 @@ void ELFState<ELFT>::writeSectionContent( for (const auto &[Idx, E] : llvm::enumerate(*Section.Entries)) { // Write version and feature values. if (Section.Type == llvm::ELF::SHT_LLVM_BB_ADDR_MAP) { - if (E.Version > 3) + if (E.Version > 4) WithColor::warning() << "unsupported SHT_LLVM_BB_ADDR_MAP version: " << static_cast<int>(E.Version) << "; encoding using the most recent version"; @@ -1526,6 +1526,12 @@ void ELFState<ELFT>::writeSectionContent( } SHeader.sh_size += CBA.writeULEB128(BBE.Size); SHeader.sh_size += CBA.writeULEB128(BBE.Metadata); + if (FeatureOrErr->BBHash || BBE.Hash.has_value()) { + uint64_t Hash = + BBE.Hash.has_value() ? BBE.Hash.value() : llvm::yaml::Hex64(0); + CBA.write<uint64_t>(Hash, ELFT::Endianness); + SHeader.sh_size += 8; + } } } if (!PGOAnalyses) diff --git a/llvm/lib/ObjectYAML/ELFYAML.cpp b/llvm/lib/ObjectYAML/ELFYAML.cpp index d9cce1e..421d6603 100644 --- a/llvm/lib/ObjectYAML/ELFYAML.cpp +++ b/llvm/lib/ObjectYAML/ELFYAML.cpp @@ -1887,6 +1887,7 @@ void MappingTraits<ELFYAML::BBAddrMapEntry::BBEntry>::mapping( IO.mapRequired("Size", E.Size); IO.mapRequired("Metadata", E.Metadata); IO.mapOptional("CallsiteEndOffsets", E.CallsiteEndOffsets); + IO.mapOptional("Hash", E.Hash); } void MappingTraits<ELFYAML::PGOAnalysisMapEntry>::mapping( diff --git a/llvm/lib/Support/TextEncoding.cpp b/llvm/lib/Support/TextEncoding.cpp index 804ff07..41f5187 100644 --- a/llvm/lib/Support/TextEncoding.cpp +++ b/llvm/lib/Support/TextEncoding.cpp @@ -54,9 +54,9 @@ static std::optional<TextEncoding> getKnownEncoding(StringRef Name) { return std::nullopt; } -LLVM_ATTRIBUTE_UNUSED static void -HandleOverflow(size_t &Capacity, char *&Output, size_t &OutputLength, - SmallVectorImpl<char> &Result) { +[[maybe_unused]] static void HandleOverflow(size_t &Capacity, char *&Output, + size_t &OutputLength, + SmallVectorImpl<char> &Result) { // No space left in output buffer. Double the size of the underlying // memory in the SmallVectorImpl, adjust pointer and length and continue // the conversion. diff --git a/llvm/lib/Support/UnicodeNameToCodepoint.cpp b/llvm/lib/Support/UnicodeNameToCodepoint.cpp index 8d66348..6f8e091 100644 --- a/llvm/lib/Support/UnicodeNameToCodepoint.cpp +++ b/llvm/lib/Support/UnicodeNameToCodepoint.cpp @@ -476,7 +476,7 @@ nearestMatchesForCodepointName(StringRef Pattern, std::size_t MaxMatchesCount) { std::min(NormalizedName.size(), UnicodeNameToCodepointLargestNameSize) + 1; - LLVM_ATTRIBUTE_UNUSED static std::size_t Rows = + [[maybe_unused]] static std::size_t Rows = UnicodeNameToCodepointLargestNameSize + 1; std::vector<char> Distances( diff --git a/llvm/lib/Target/AArch64/AsmParser/AArch64AsmParser.cpp b/llvm/lib/Target/AArch64/AsmParser/AArch64AsmParser.cpp index 2c3870c..636d4f8a 100644 --- a/llvm/lib/Target/AArch64/AsmParser/AArch64AsmParser.cpp +++ b/llvm/lib/Target/AArch64/AsmParser/AArch64AsmParser.cpp @@ -8217,6 +8217,8 @@ bool AArch64AsmParser::parseDataExpr(const MCExpr *&Res) { Spec = AArch64::S_GOTPCREL; else if (Identifier == "plt") Spec = AArch64::S_PLT; + else if (Identifier == "funcinit") + Spec = AArch64::S_FUNCINIT; } if (Spec == AArch64::S_None) return Error(Loc, "invalid relocation specifier"); diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp index a388216..892b8da 100644 --- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp +++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp @@ -232,6 +232,8 @@ unsigned AArch64ELFObjectWriter::getRelocType(const MCFixup &Fixup, } if (RefKind == AArch64::S_AUTH || RefKind == AArch64::S_AUTHADDR) return ELF::R_AARCH64_AUTH_ABS64; + if (RefKind == AArch64::S_FUNCINIT) + return ELF::R_AARCH64_FUNCINIT64; return ELF::R_AARCH64_ABS64; } case AArch64::fixup_aarch64_add_imm12: diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.cpp b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.cpp index 2b5cf34..bc090c6 100644 --- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.cpp +++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.cpp @@ -40,6 +40,7 @@ const MCAsmInfo::AtSpecifier ELFAtSpecifiers[] = { {AArch64::S_GOT, "GOT"}, {AArch64::S_GOTPCREL, "GOTPCREL"}, {AArch64::S_PLT, "PLT"}, + {AArch64::S_FUNCINIT, "FUNCINIT"}, }; const MCAsmInfo::AtSpecifier MachOAtSpecifiers[] = { diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.h b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.h index 0dfa61b..f2acff5 100644 --- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.h +++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCAsmInfo.h @@ -164,6 +164,7 @@ enum { // ELF relocation specifiers in data directives: S_PLT = 0x400, S_GOTPCREL, + S_FUNCINIT, // Mach-O @ relocation specifiers: S_MACHO_GOT, diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp index 80e985d..a2841c11 100644 --- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp +++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp @@ -18168,7 +18168,7 @@ Align SITargetLowering::getPrefLoopAlignment(MachineLoop *ML) const { return CacheLineAlign; } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] static bool isCopyFromRegOfInlineAsm(const SDNode *N) { assert(N->getOpcode() == ISD::CopyFromReg); do { diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp index a44a247..d516330 100644 --- a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp +++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp @@ -865,22 +865,16 @@ void SIInstrInfo::copyPhysReg(MachineBasicBlock &MBB, return; } - if (DestReg == AMDGPU::VCC_LO) { - if (AMDGPU::SReg_32RegClass.contains(SrcReg)) { - BuildMI(MBB, MI, DL, get(AMDGPU::S_MOV_B32), AMDGPU::VCC_LO) - .addReg(SrcReg, getKillRegState(KillSrc)); - } else { + if (!AMDGPU::SReg_32RegClass.contains(SrcReg)) { + if (DestReg == AMDGPU::VCC_LO) { // FIXME: Hack until VReg_1 removed. assert(AMDGPU::VGPR_32RegClass.contains(SrcReg)); BuildMI(MBB, MI, DL, get(AMDGPU::V_CMP_NE_U32_e32)) - .addImm(0) - .addReg(SrcReg, getKillRegState(KillSrc)); + .addImm(0) + .addReg(SrcReg, getKillRegState(KillSrc)); + return; } - return; - } - - if (!AMDGPU::SReg_32RegClass.contains(SrcReg)) { reportIllegalCopy(this, MBB, MI, DL, DestReg, SrcReg, KillSrc); return; } @@ -898,22 +892,16 @@ void SIInstrInfo::copyPhysReg(MachineBasicBlock &MBB, return; } - if (DestReg == AMDGPU::VCC) { - if (AMDGPU::SReg_64_EncodableRegClass.contains(SrcReg)) { - BuildMI(MBB, MI, DL, get(AMDGPU::S_MOV_B64), AMDGPU::VCC) - .addReg(SrcReg, getKillRegState(KillSrc)); - } else { + if (!AMDGPU::SReg_64_EncodableRegClass.contains(SrcReg)) { + if (DestReg == AMDGPU::VCC) { // FIXME: Hack until VReg_1 removed. assert(AMDGPU::VGPR_32RegClass.contains(SrcReg)); BuildMI(MBB, MI, DL, get(AMDGPU::V_CMP_NE_U32_e32)) - .addImm(0) - .addReg(SrcReg, getKillRegState(KillSrc)); + .addImm(0) + .addReg(SrcReg, getKillRegState(KillSrc)); + return; } - return; - } - - if (!AMDGPU::SReg_64_EncodableRegClass.contains(SrcReg)) { reportIllegalCopy(this, MBB, MI, DL, DestReg, SrcReg, KillSrc); return; } diff --git a/llvm/lib/Target/CSKY/Disassembler/CSKYDisassembler.cpp b/llvm/lib/Target/CSKY/Disassembler/CSKYDisassembler.cpp index 39e651d..8945ec3 100644 --- a/llvm/lib/Target/CSKY/Disassembler/CSKYDisassembler.cpp +++ b/llvm/lib/Target/CSKY/Disassembler/CSKYDisassembler.cpp @@ -166,7 +166,7 @@ static DecodeStatus DecodeFPR64RegisterClass(MCInst &Inst, uint64_t RegNo, } // TODO -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] static DecodeStatus DecodesFPR128RegisterClass(MCInst &Inst, uint64_t RegNo, uint64_t Address, const MCDisassembler *Decoder) { diff --git a/llvm/lib/Target/Hexagon/Disassembler/HexagonDisassembler.cpp b/llvm/lib/Target/Hexagon/Disassembler/HexagonDisassembler.cpp index 974f653..5f180d6 100644 --- a/llvm/lib/Target/Hexagon/Disassembler/HexagonDisassembler.cpp +++ b/llvm/lib/Target/Hexagon/Disassembler/HexagonDisassembler.cpp @@ -667,11 +667,10 @@ static DecodeStatus DecodeHvxWRRegisterClass(MCInst &Inst, unsigned RegNo, return DecodeRegisterClass(Inst, RegNo, HvxWRDecoderTable); } -LLVM_ATTRIBUTE_UNUSED // Suppress warning temporarily. - static DecodeStatus - DecodeHvxVQRRegisterClass(MCInst &Inst, unsigned RegNo, - uint64_t /*Address*/, - const MCDisassembler *Decoder) { +[[maybe_unused]] // Suppress warning temporarily. +static DecodeStatus DecodeHvxVQRRegisterClass(MCInst &Inst, unsigned RegNo, + uint64_t /*Address*/, + const MCDisassembler *Decoder) { static const MCPhysReg HvxVQRDecoderTable[] = { Hexagon::VQ0, Hexagon::VQ1, Hexagon::VQ2, Hexagon::VQ3, Hexagon::VQ4, Hexagon::VQ5, Hexagon::VQ6, Hexagon::VQ7}; diff --git a/llvm/lib/Target/Hexagon/HexagonConstExtenders.cpp b/llvm/lib/Target/Hexagon/HexagonConstExtenders.cpp index 5dde47a..a3296e0 100644 --- a/llvm/lib/Target/Hexagon/HexagonConstExtenders.cpp +++ b/llvm/lib/Target/Hexagon/HexagonConstExtenders.cpp @@ -419,8 +419,8 @@ namespace { using HCE = HexagonConstExtenders; - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const OffsetRange &OR) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const OffsetRange &OR) { if (OR.Min > OR.Max) OS << '!'; OS << '[' << OR.Min << ',' << OR.Max << "]a" << unsigned(OR.Align) @@ -435,8 +435,8 @@ namespace { const HexagonRegisterInfo &HRI; }; - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const PrintRegister &P) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const PrintRegister &P) { if (P.Rs.Reg != 0) OS << printReg(P.Rs.Reg, &P.HRI, P.Rs.Sub); else @@ -451,8 +451,8 @@ namespace { const HexagonRegisterInfo &HRI; }; - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const PrintExpr &P) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const PrintExpr &P) { OS << "## " << (P.Ex.Neg ? "- " : "+ "); if (P.Ex.Rs.Reg != 0) OS << printReg(P.Ex.Rs.Reg, &P.HRI, P.Ex.Rs.Sub); @@ -469,15 +469,15 @@ namespace { const HexagonRegisterInfo &HRI; }; - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const PrintInit &P) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const PrintInit &P) { OS << '[' << P.ExtI.first << ", " << PrintExpr(P.ExtI.second, P.HRI) << ']'; return OS; } - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const HCE::ExtDesc &ED) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const HCE::ExtDesc &ED) { assert(ED.OpNum != -1u); const MachineBasicBlock &MBB = *ED.getOp().getParent()->getParent(); const MachineFunction &MF = *MBB.getParent(); @@ -493,8 +493,8 @@ namespace { return OS; } - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const HCE::ExtRoot &ER) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const HCE::ExtRoot &ER) { switch (ER.Kind) { case MachineOperand::MO_Immediate: OS << "imm:" << ER.V.ImmVal; @@ -527,8 +527,8 @@ namespace { return OS; } - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const HCE::ExtValue &EV) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const HCE::ExtValue &EV) { OS << HCE::ExtRoot(EV) << " off:" << EV.Offset; return OS; } @@ -540,8 +540,8 @@ namespace { const HexagonRegisterInfo &HRI; }; - LLVM_ATTRIBUTE_UNUSED - raw_ostream &operator<< (raw_ostream &OS, const PrintIMap &P) { + [[maybe_unused]] + raw_ostream &operator<<(raw_ostream &OS, const PrintIMap &P) { OS << "{\n"; for (const std::pair<const HCE::ExtenderInit, HCE::IndexList> &Q : P.IMap) { OS << " " << PrintInit(Q.first, P.HRI) << " -> {"; diff --git a/llvm/lib/Target/Hexagon/HexagonISelDAGToDAGHVX.cpp b/llvm/lib/Target/Hexagon/HexagonISelDAGToDAGHVX.cpp index 4d96cfa..c7a4f68 100644 --- a/llvm/lib/Target/Hexagon/HexagonISelDAGToDAGHVX.cpp +++ b/llvm/lib/Target/Hexagon/HexagonISelDAGToDAGHVX.cpp @@ -789,7 +789,7 @@ struct ShuffleMask { } }; -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const ShuffleMask &SM) { SM.print(OS); return OS; diff --git a/llvm/lib/Target/Hexagon/HexagonVectorCombine.cpp b/llvm/lib/Target/Hexagon/HexagonVectorCombine.cpp index 87d052b..e4c0a16 100644 --- a/llvm/lib/Target/Hexagon/HexagonVectorCombine.cpp +++ b/llvm/lib/Target/Hexagon/HexagonVectorCombine.cpp @@ -364,7 +364,7 @@ private: const HexagonVectorCombine &HVC; }; -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const AlignVectors::AddrInfo &AI) { OS << "Inst: " << AI.Inst << " " << *AI.Inst << '\n'; OS << "Addr: " << *AI.Addr << '\n'; @@ -375,7 +375,7 @@ raw_ostream &operator<<(raw_ostream &OS, const AlignVectors::AddrInfo &AI) { return OS; } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const AlignVectors::MoveGroup &MG) { OS << "IsLoad:" << (MG.IsLoad ? "yes" : "no"); OS << ", IsHvx:" << (MG.IsHvx ? "yes" : "no") << '\n'; @@ -394,7 +394,7 @@ raw_ostream &operator<<(raw_ostream &OS, const AlignVectors::MoveGroup &MG) { return OS; } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const AlignVectors::ByteSpan::Block &B) { OS << " @" << B.Pos << " [" << B.Seg.Start << ',' << B.Seg.Size << "] "; @@ -408,7 +408,7 @@ raw_ostream &operator<<(raw_ostream &OS, return OS; } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const AlignVectors::ByteSpan &BS) { OS << "ByteSpan[size=" << BS.size() << ", extent=" << BS.extent() << '\n'; for (const AlignVectors::ByteSpan::Block &B : BS) diff --git a/llvm/lib/Target/Hexagon/HexagonVectorLoopCarriedReuse.cpp b/llvm/lib/Target/Hexagon/HexagonVectorLoopCarriedReuse.cpp index fa8ae60..2ff5843 100644 --- a/llvm/lib/Target/Hexagon/HexagonVectorLoopCarriedReuse.cpp +++ b/llvm/lib/Target/Hexagon/HexagonVectorLoopCarriedReuse.cpp @@ -111,7 +111,7 @@ namespace { friend raw_ostream &operator<< (raw_ostream &OS, const DepChain &D); }; - LLVM_ATTRIBUTE_UNUSED + [[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const DepChain &D) { const ChainOfDependences &CD = D.Chain; int ChainSize = CD.size(); @@ -144,7 +144,7 @@ namespace { bool isDefined() { return Inst2Replace != nullptr; } }; - LLVM_ATTRIBUTE_UNUSED + [[maybe_unused]] raw_ostream &operator<<(raw_ostream &OS, const ReuseValue &RU) { OS << "** ReuseValue ***\n"; OS << "Instruction to Replace: " << *(RU.Inst2Replace) << "\n"; diff --git a/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonBaseInfo.h b/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonBaseInfo.h index ca98269..e3094b4 100644 --- a/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonBaseInfo.h +++ b/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonBaseInfo.h @@ -275,7 +275,7 @@ namespace HexagonII { INST_ICLASS_ALU32_3 = 0xf0000000 }; - LLVM_ATTRIBUTE_UNUSED + [[maybe_unused]] static unsigned getMemAccessSizeInBytes(MemAccessSize S) { switch (S) { case ByteAccess: return 1; diff --git a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp index 96ad5c6..0a8838c 100644 --- a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp +++ b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp @@ -156,13 +156,13 @@ FunctionPass *llvm::createRISCVVLOptimizerPass() { return new RISCVVLOptimizer(); } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] static raw_ostream &operator<<(raw_ostream &OS, const OperandInfo &OI) { OI.print(OS); return OS; } -LLVM_ATTRIBUTE_UNUSED +[[maybe_unused]] static raw_ostream &operator<<(raw_ostream &OS, const std::optional<OperandInfo> &OI) { if (OI) diff --git a/llvm/lib/Target/X86/GISel/X86LegalizerInfo.cpp b/llvm/lib/Target/X86/GISel/X86LegalizerInfo.cpp index 28fa2cd..b81641f 100644 --- a/llvm/lib/Target/X86/GISel/X86LegalizerInfo.cpp +++ b/llvm/lib/Target/X86/GISel/X86LegalizerInfo.cpp @@ -414,6 +414,8 @@ X86LegalizerInfo::X86LegalizerInfo(const X86Subtarget &STI, getActionDefinitionsBuilder(G_SEXT_INREG).lower(); + getActionDefinitionsBuilder(G_IS_FPCLASS).lower(); + // fp constants getActionDefinitionsBuilder(G_FCONSTANT) .legalFor({s32, s64}) diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp index eea84a2..c32b1a6 100644 --- a/llvm/lib/Target/X86/X86ISelLowering.cpp +++ b/llvm/lib/Target/X86/X86ISelLowering.cpp @@ -3624,6 +3624,16 @@ X86TargetLowering::getJumpConditionMergingParams(Instruction::BinaryOps Opc, match(Lhs, m_SpecificICmp(ICmpInst::ICMP_EQ, m_Value(), m_Value())) && match(Rhs, m_SpecificICmp(ICmpInst::ICMP_EQ, m_Value(), m_Value()))) BaseCost += 1; + + // For OR conditions with EQ comparisons, prefer splitting into branches + // (unless CCMP is available). OR+EQ cannot be optimized via bitwise ops, + // unlike OR+NE which becomes (P|Q)!=0. Similarly, don't split signed + // comparisons (SLT, SGT) that can be optimized. + if (BaseCost >= 0 && !Subtarget.hasCCMP() && Opc == Instruction::Or && + match(Lhs, m_SpecificICmp(ICmpInst::ICMP_EQ, m_Value(), m_Value())) && + match(Rhs, m_SpecificICmp(ICmpInst::ICMP_EQ, m_Value(), m_Value()))) + return {-1, -1, -1}; + return {BaseCost, BrMergingLikelyBias.getValue(), BrMergingUnlikelyBias.getValue()}; } @@ -3787,7 +3797,7 @@ static bool isUndefOrZeroOrInRange(ArrayRef<int> Mask, int Low, int Hi) { /// Return true if every element in Mask, is an in-place blend/select mask or is /// undef. -LLVM_ATTRIBUTE_UNUSED static bool isBlendOrUndef(ArrayRef<int> Mask) { +[[maybe_unused]] static bool isBlendOrUndef(ArrayRef<int> Mask) { unsigned NumElts = Mask.size(); for (auto [I, M] : enumerate(Mask)) if (!isUndefOrEqual(M, I) && !isUndefOrEqual(M, I + NumElts)) @@ -8096,7 +8106,7 @@ static SDValue LowerBUILD_VECTORvXi1(SDValue Op, const SDLoc &dl, return DstVec; } -LLVM_ATTRIBUTE_UNUSED static bool isHorizOp(unsigned Opcode) { +[[maybe_unused]] static bool isHorizOp(unsigned Opcode) { switch (Opcode) { case X86ISD::PACKSS: case X86ISD::PACKUS: @@ -20813,7 +20823,7 @@ SDValue X86TargetLowering::FP_TO_INTHelper(SDValue Op, SelectionDAG &DAG, // for DAG type consistency we have to match the FP operand type. APFloat Thresh(APFloat::IEEEsingle(), APInt(32, 0x5f000000)); - LLVM_ATTRIBUTE_UNUSED APFloat::opStatus Status = APFloat::opOK; + [[maybe_unused]] APFloat::opStatus Status = APFloat::opOK; bool LosesInfo = false; if (TheVT == MVT::f64) // The rounding mode is irrelevant as the conversion should be exact. @@ -22856,7 +22866,7 @@ static SDValue combineVectorSizedSetCCEquality(EVT VT, SDValue X, SDValue Y, // be generated by the memcmp expansion pass with oversized integer compares // (see PR33325). bool IsOrXorXorTreeCCZero = isNullConstant(Y) && isOrXorXorTree(X); - if (isNullConstant(Y) && !IsOrXorXorTreeCCZero) + if (isNullConstant(Y) && OpSize == 128 && !IsOrXorXorTreeCCZero) return SDValue(); // Don't perform this combine if constructing the vector will be expensive. diff --git a/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp b/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp index 6b67b48..09cb225 100644 --- a/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp +++ b/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp @@ -2979,10 +2979,14 @@ Instruction *InstCombinerImpl::foldAndOrOfSelectUsingImpliedCond(Value *Op, "Op must be either i1 or vector of i1."); if (SI.getCondition()->getType() != Op->getType()) return nullptr; - if (Value *V = simplifyNestedSelectsUsingImpliedCond(SI, Op, IsAnd, DL)) - return SelectInst::Create(Op, - IsAnd ? V : ConstantInt::getTrue(Op->getType()), - IsAnd ? ConstantInt::getFalse(Op->getType()) : V); + if (Value *V = simplifyNestedSelectsUsingImpliedCond(SI, Op, IsAnd, DL)) { + Instruction *MDFrom = nullptr; + if (!ProfcheckDisableMetadataFixes) + MDFrom = &SI; + return SelectInst::Create( + Op, IsAnd ? V : ConstantInt::getTrue(Op->getType()), + IsAnd ? ConstantInt::getFalse(Op->getType()) : V, "", nullptr, MDFrom); + } return nullptr; } diff --git a/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp b/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp index ff5f390..66e45ec 100644 --- a/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp +++ b/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp @@ -266,8 +266,7 @@ void DFAJumpThreading::unfold(DomTreeUpdater *DTU, LoopInfo *LI, if (!ProfcheckDisableMetadataFixes) BI->setMetadata(LLVMContext::MD_prof, SI->getMetadata(LLVMContext::MD_prof)); - DTU->applyUpdates({{DominatorTree::Insert, StartBlock, EndBlock}, - {DominatorTree::Insert, StartBlock, NewBlock}}); + DTU->applyUpdates({{DominatorTree::Insert, StartBlock, NewBlock}}); } else { BasicBlock *EndBlock = SIUse->getParent(); BasicBlock *NewBlockT = BasicBlock::Create( @@ -1479,10 +1478,13 @@ bool DFAJumpThreading::run(Function &F) { DTU->flush(); #ifdef EXPENSIVE_CHECKS - assert(DTU->getDomTree().verify(DominatorTree::VerificationLevel::Full)); verifyFunction(F, &dbgs()); #endif + if (MadeChanges && VerifyDomInfo) + assert(DTU->getDomTree().verify(DominatorTree::VerificationLevel::Full) && + "Failed to maintain validity of domtree!"); + return MadeChanges; } diff --git a/llvm/lib/Transforms/Scalar/FlattenCFGPass.cpp b/llvm/lib/Transforms/Scalar/FlattenCFGPass.cpp index 213d0f3..1335665 100644 --- a/llvm/lib/Transforms/Scalar/FlattenCFGPass.cpp +++ b/llvm/lib/Transforms/Scalar/FlattenCFGPass.cpp @@ -39,10 +39,11 @@ public: private: AliasAnalysis *AA; }; +} // namespace /// iterativelyFlattenCFG - Call FlattenCFG on all the blocks in the function, /// iterating until no more changes are made. -bool iterativelyFlattenCFG(Function &F, AliasAnalysis *AA) { +static bool iterativelyFlattenCFG(Function &F, AliasAnalysis *AA) { bool Changed = false; bool LocalChange = true; @@ -67,7 +68,6 @@ bool iterativelyFlattenCFG(Function &F, AliasAnalysis *AA) { } return Changed; } -} // namespace char FlattenCFGLegacyPass::ID = 0; diff --git a/llvm/lib/Transforms/Scalar/SROA.cpp b/llvm/lib/Transforms/Scalar/SROA.cpp index b9d332b..d2f09e9 100644 --- a/llvm/lib/Transforms/Scalar/SROA.cpp +++ b/llvm/lib/Transforms/Scalar/SROA.cpp @@ -118,9 +118,13 @@ STATISTIC( STATISTIC(NumDeleted, "Number of instructions deleted"); STATISTIC(NumVectorized, "Number of vectorized aggregates"); +namespace llvm { /// Disable running mem2reg during SROA in order to test or debug SROA. static cl::opt<bool> SROASkipMem2Reg("sroa-skip-mem2reg", cl::init(false), cl::Hidden); +extern cl::opt<bool> ProfcheckDisableMetadataFixes; +} // namespace llvm + namespace { class AllocaSliceRewriter; @@ -1777,7 +1781,8 @@ static void speculateSelectInstLoads(SelectInst &SI, LoadInst &LI, } Value *V = IRB.CreateSelect(SI.getCondition(), TL, FL, - LI.getName() + ".sroa.speculated"); + LI.getName() + ".sroa.speculated", + ProfcheckDisableMetadataFixes ? nullptr : &SI); LLVM_DEBUG(dbgs() << " speculated to: " << *V << "\n"); LI.replaceAllUsesWith(V); @@ -4360,10 +4365,13 @@ private: }; Value *Cond, *True, *False; + Instruction *MDFrom = nullptr; if (auto *SI = dyn_cast<SelectInst>(Sel)) { Cond = SI->getCondition(); True = SI->getTrueValue(); False = SI->getFalseValue(); + if (!ProfcheckDisableMetadataFixes) + MDFrom = SI; } else { Cond = Sel->getOperand(0); True = ConstantInt::get(Sel->getType(), 1); @@ -4383,8 +4391,12 @@ private: IRB.CreateGEP(Ty, FalseOps[0], ArrayRef(FalseOps).drop_front(), False->getName() + ".sroa.gep", NW); - Value *NSel = - IRB.CreateSelect(Cond, NTrue, NFalse, Sel->getName() + ".sroa.sel"); + Value *NSel = MDFrom + ? IRB.CreateSelect(Cond, NTrue, NFalse, + Sel->getName() + ".sroa.sel", MDFrom) + : IRB.CreateSelectWithUnknownProfile( + Cond, NTrue, NFalse, DEBUG_TYPE, + Sel->getName() + ".sroa.sel"); Visited.erase(&GEPI); GEPI.replaceAllUsesWith(NSel); GEPI.eraseFromParent(); diff --git a/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp b/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp index 2ee91a9..0f3978f 100644 --- a/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp +++ b/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp @@ -47,6 +47,7 @@ #include "llvm/Transforms/Utils/BasicBlockUtils.h" #include "llvm/Transforms/Utils/Local.h" #include "llvm/Transforms/Utils/SSAUpdater.h" +#include "llvm/Transforms/Utils/SSAUpdaterBulk.h" #include <cassert> #include <utility> @@ -321,7 +322,7 @@ class StructurizeCFG { void collectInfos(); - void insertConditions(bool Loops); + void insertConditions(bool Loops, SSAUpdaterBulk &PhiInserter); void simplifyConditions(); @@ -671,10 +672,9 @@ void StructurizeCFG::collectInfos() { } /// Insert the missing branch conditions -void StructurizeCFG::insertConditions(bool Loops) { +void StructurizeCFG::insertConditions(bool Loops, SSAUpdaterBulk &PhiInserter) { BranchVector &Conds = Loops ? LoopConds : Conditions; Value *Default = Loops ? BoolTrue : BoolFalse; - SSAUpdater PhiInserter; for (BranchInst *Term : Conds) { assert(Term->isConditional()); @@ -683,8 +683,9 @@ void StructurizeCFG::insertConditions(bool Loops) { BasicBlock *SuccTrue = Term->getSuccessor(0); BasicBlock *SuccFalse = Term->getSuccessor(1); - PhiInserter.Initialize(Boolean, ""); - PhiInserter.AddAvailableValue(Loops ? SuccFalse : Parent, Default); + unsigned Variable = PhiInserter.AddVariable("", Boolean); + PhiInserter.AddAvailableValue(Variable, Loops ? SuccFalse : Parent, + Default); BBPredicates &Preds = Loops ? LoopPreds[SuccFalse] : Predicates[SuccTrue]; @@ -697,7 +698,7 @@ void StructurizeCFG::insertConditions(bool Loops) { ParentInfo = PI; break; } - PhiInserter.AddAvailableValue(BB, PI.Pred); + PhiInserter.AddAvailableValue(Variable, BB, PI.Pred); Dominator.addAndRememberBlock(BB); } @@ -706,9 +707,9 @@ void StructurizeCFG::insertConditions(bool Loops) { CondBranchWeights::setMetadata(*Term, ParentInfo.Weights); } else { if (!Dominator.resultIsRememberedBlock()) - PhiInserter.AddAvailableValue(Dominator.result(), Default); + PhiInserter.AddAvailableValue(Variable, Dominator.result(), Default); - Term->setCondition(PhiInserter.GetValueInMiddleOfBlock(Parent)); + PhiInserter.AddUse(Variable, &Term->getOperandUse(0)); } } } @@ -1414,8 +1415,12 @@ bool StructurizeCFG::run(Region *R, DominatorTree *DT, orderNodes(); collectInfos(); createFlow(); - insertConditions(false); - insertConditions(true); + + SSAUpdaterBulk PhiInserter; + insertConditions(false, PhiInserter); + insertConditions(true, PhiInserter); + PhiInserter.RewriteAndOptimizeAllUses(*DT); + setPhiValues(); simplifyHoistedPhis(); simplifyConditions(); diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h index 8ca3bed..23f5623 100644 --- a/llvm/lib/Transforms/Vectorize/VPlan.h +++ b/llvm/lib/Transforms/Vectorize/VPlan.h @@ -24,12 +24,9 @@ #ifndef LLVM_TRANSFORMS_VECTORIZE_VPLAN_H #define LLVM_TRANSFORMS_VECTORIZE_VPLAN_H -#include "VPlanAnalysis.h" #include "VPlanValue.h" #include "llvm/ADT/DenseMap.h" -#include "llvm/ADT/SmallBitVector.h" #include "llvm/ADT/SmallPtrSet.h" -#include "llvm/ADT/SmallSet.h" #include "llvm/ADT/SmallVector.h" #include "llvm/ADT/Twine.h" #include "llvm/ADT/ilist.h" @@ -41,10 +38,11 @@ #include "llvm/IR/Operator.h" #include "llvm/Support/Compiler.h" #include "llvm/Support/InstructionCost.h" -#include <algorithm> #include <cassert> #include <cstddef> +#include <functional> #include <string> +#include <utility> namespace llvm { @@ -346,13 +344,6 @@ public: /// Return the cost of the block. virtual InstructionCost cost(ElementCount VF, VPCostContext &Ctx) = 0; - /// Return true if it is legal to hoist instructions into this block. - bool isLegalToHoistInto() { - // There are currently no constraints that prevent an instruction to be - // hoisted into a VPBlockBase. - return true; - } - #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP) void printAsOperand(raw_ostream &OS, bool PrintType = false) const { OS << getName(); diff --git a/llvm/test/Analysis/ScalarEvolution/ptrtoint.ll b/llvm/test/Analysis/ScalarEvolution/ptrtoint.ll index acac2c9..0c1f37b 100644 --- a/llvm/test/Analysis/ScalarEvolution/ptrtoint.ll +++ b/llvm/test/Analysis/ScalarEvolution/ptrtoint.ll @@ -382,7 +382,7 @@ define void @pr46786_c26_char(ptr %arg, ptr %arg1, ptr %arg2) { ; X64-NEXT: %i9 = ptrtoint ptr %i7 to i64 ; X64-NEXT: --> {(ptrtoint ptr %arg to i64),+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (ptrtoint ptr %arg1 to i64)) LoopDispositions: { %bb6: Computable } ; X64-NEXT: %i10 = sub i64 %i9, %i4 -; X64-NEXT: --> {0,+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) LoopDispositions: { %bb6: Computable } +; X64-NEXT: --> {0,+,1}<nuw><%bb6> U: [0,-1) S: [0,-1) Exits: (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) LoopDispositions: { %bb6: Computable } ; X64-NEXT: %i11 = getelementptr inbounds i8, ptr %arg2, i64 %i10 ; X64-NEXT: --> {%arg2,+,1}<nw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64) + %arg2) LoopDispositions: { %bb6: Computable } ; X64-NEXT: %i12 = load i8, ptr %i11, align 1 @@ -393,7 +393,7 @@ define void @pr46786_c26_char(ptr %arg, ptr %arg1, ptr %arg2) { ; X64-NEXT: --> {(1 + %arg),+,1}<nuw><%bb6> U: full-set S: full-set Exits: ((-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64) + %arg) LoopDispositions: { %bb6: Computable } ; X64-NEXT: Determining loop execution counts for: @pr46786_c26_char ; X64-NEXT: Loop %bb6: backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) -; X64-NEXT: Loop %bb6: constant max backedge-taken count is i64 -1 +; X64-NEXT: Loop %bb6: constant max backedge-taken count is i64 -2 ; X64-NEXT: Loop %bb6: symbolic max backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) ; X64-NEXT: Loop %bb6: Trip multiple is 1 ; @@ -406,9 +406,9 @@ define void @pr46786_c26_char(ptr %arg, ptr %arg1, ptr %arg2) { ; X32-NEXT: %i8 = load i8, ptr %i7, align 1 ; X32-NEXT: --> %i8 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant } ; X32-NEXT: %i9 = ptrtoint ptr %i7 to i64 -; X32-NEXT: --> {(zext i32 (ptrtoint ptr %arg to i32) to i64),+,1}<nuw><%bb6> U: [0,8589934591) S: [0,8589934591) Exits: ((zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) + (zext i32 (ptrtoint ptr %arg to i32) to i64)) LoopDispositions: { %bb6: Computable } +; X32-NEXT: --> {(zext i32 (ptrtoint ptr %arg to i32) to i64),+,1}<nuw><%bb6> U: [0,8589934590) S: [0,8589934590) Exits: ((zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) + (zext i32 (ptrtoint ptr %arg to i32) to i64)) LoopDispositions: { %bb6: Computable } ; X32-NEXT: %i10 = sub i64 %i9, %i4 -; X32-NEXT: --> {0,+,1}<nuw><%bb6> U: [0,4294967296) S: [0,4294967296) Exits: (zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) LoopDispositions: { %bb6: Computable } +; X32-NEXT: --> {0,+,1}<nuw><%bb6> U: [0,4294967295) S: [0,4294967295) Exits: (zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) LoopDispositions: { %bb6: Computable } ; X32-NEXT: %i11 = getelementptr inbounds i8, ptr %arg2, i64 %i10 ; X32-NEXT: --> {%arg2,+,1}<%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32) + %arg2) LoopDispositions: { %bb6: Computable } ; X32-NEXT: %i12 = load i8, ptr %i11, align 1 @@ -419,7 +419,7 @@ define void @pr46786_c26_char(ptr %arg, ptr %arg1, ptr %arg2) { ; X32-NEXT: --> {(1 + %arg),+,1}<nuw><%bb6> U: full-set S: full-set Exits: ((-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32) + %arg) LoopDispositions: { %bb6: Computable } ; X32-NEXT: Determining loop execution counts for: @pr46786_c26_char ; X32-NEXT: Loop %bb6: backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) -; X32-NEXT: Loop %bb6: constant max backedge-taken count is i32 -1 +; X32-NEXT: Loop %bb6: constant max backedge-taken count is i32 -2 ; X32-NEXT: Loop %bb6: symbolic max backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) ; X32-NEXT: Loop %bb6: Trip multiple is 1 ; @@ -459,7 +459,7 @@ define void @pr46786_c26_char_cmp_ops_swapped(ptr %arg, ptr %arg1, ptr %arg2) { ; X64-NEXT: %i9 = ptrtoint ptr %i7 to i64 ; X64-NEXT: --> {(ptrtoint ptr %arg to i64),+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (ptrtoint ptr %arg1 to i64)) LoopDispositions: { %bb6: Computable } ; X64-NEXT: %i10 = sub i64 %i9, %i4 -; X64-NEXT: --> {0,+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) LoopDispositions: { %bb6: Computable } +; X64-NEXT: --> {0,+,1}<nuw><%bb6> U: [0,-1) S: [0,-1) Exits: (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) LoopDispositions: { %bb6: Computable } ; X64-NEXT: %i11 = getelementptr inbounds i8, ptr %arg2, i64 %i10 ; X64-NEXT: --> {%arg2,+,1}<nw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64) + %arg2) LoopDispositions: { %bb6: Computable } ; X64-NEXT: %i12 = load i8, ptr %i11, align 1 @@ -470,7 +470,7 @@ define void @pr46786_c26_char_cmp_ops_swapped(ptr %arg, ptr %arg1, ptr %arg2) { ; X64-NEXT: --> {(1 + %arg),+,1}<nuw><%bb6> U: full-set S: full-set Exits: ((-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64) + %arg) LoopDispositions: { %bb6: Computable } ; X64-NEXT: Determining loop execution counts for: @pr46786_c26_char_cmp_ops_swapped ; X64-NEXT: Loop %bb6: backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) -; X64-NEXT: Loop %bb6: constant max backedge-taken count is i64 -1 +; X64-NEXT: Loop %bb6: constant max backedge-taken count is i64 -2 ; X64-NEXT: Loop %bb6: symbolic max backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i64)) + (ptrtoint ptr %arg1 to i64)) ; X64-NEXT: Loop %bb6: Trip multiple is 1 ; @@ -483,9 +483,9 @@ define void @pr46786_c26_char_cmp_ops_swapped(ptr %arg, ptr %arg1, ptr %arg2) { ; X32-NEXT: %i8 = load i8, ptr %i7, align 1 ; X32-NEXT: --> %i8 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant } ; X32-NEXT: %i9 = ptrtoint ptr %i7 to i64 -; X32-NEXT: --> {(zext i32 (ptrtoint ptr %arg to i32) to i64),+,1}<nuw><%bb6> U: [0,8589934591) S: [0,8589934591) Exits: ((zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) + (zext i32 (ptrtoint ptr %arg to i32) to i64)) LoopDispositions: { %bb6: Computable } +; X32-NEXT: --> {(zext i32 (ptrtoint ptr %arg to i32) to i64),+,1}<nuw><%bb6> U: [0,8589934590) S: [0,8589934590) Exits: ((zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) + (zext i32 (ptrtoint ptr %arg to i32) to i64)) LoopDispositions: { %bb6: Computable } ; X32-NEXT: %i10 = sub i64 %i9, %i4 -; X32-NEXT: --> {0,+,1}<nuw><%bb6> U: [0,4294967296) S: [0,4294967296) Exits: (zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) LoopDispositions: { %bb6: Computable } +; X32-NEXT: --> {0,+,1}<nuw><%bb6> U: [0,4294967295) S: [0,4294967295) Exits: (zext i32 (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) to i64) LoopDispositions: { %bb6: Computable } ; X32-NEXT: %i11 = getelementptr inbounds i8, ptr %arg2, i64 %i10 ; X32-NEXT: --> {%arg2,+,1}<%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32) + %arg2) LoopDispositions: { %bb6: Computable } ; X32-NEXT: %i12 = load i8, ptr %i11, align 1 @@ -496,7 +496,7 @@ define void @pr46786_c26_char_cmp_ops_swapped(ptr %arg, ptr %arg1, ptr %arg2) { ; X32-NEXT: --> {(1 + %arg),+,1}<nuw><%bb6> U: full-set S: full-set Exits: ((-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32) + %arg) LoopDispositions: { %bb6: Computable } ; X32-NEXT: Determining loop execution counts for: @pr46786_c26_char_cmp_ops_swapped ; X32-NEXT: Loop %bb6: backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) -; X32-NEXT: Loop %bb6: constant max backedge-taken count is i32 -1 +; X32-NEXT: Loop %bb6: constant max backedge-taken count is i32 -2 ; X32-NEXT: Loop %bb6: symbolic max backedge-taken count is (-1 + (-1 * (ptrtoint ptr %arg to i32)) + (ptrtoint ptr %arg1 to i32)) ; X32-NEXT: Loop %bb6: Trip multiple is 1 ; diff --git a/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll b/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll index 28b4541..7bdc4e1 100644 --- a/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll +++ b/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll @@ -44,7 +44,7 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: leaq {{[0-9]+}}(%rsp), %rsi ; CHECK-NEXT: callq __ubyte_convert_to_ctype ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: js LBB0_6 +; CHECK-NEXT: js LBB0_4 ; CHECK-NEXT: ## %bb.1: ## %cond_next.i ; CHECK-NEXT: leaq {{[0-9]+}}(%rsp), %rsi ; CHECK-NEXT: movq %rbx, %rdi @@ -53,84 +53,81 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: sarl $31, %ecx ; CHECK-NEXT: andl %eax, %ecx ; CHECK-NEXT: cmpl $-2, %ecx -; CHECK-NEXT: je LBB0_10 +; CHECK-NEXT: je LBB0_8 ; CHECK-NEXT: ## %bb.2: ## %cond_next.i ; CHECK-NEXT: cmpl $-1, %ecx -; CHECK-NEXT: jne LBB0_3 -; CHECK-NEXT: LBB0_8: ## %bb4 +; CHECK-NEXT: jne LBB0_6 +; CHECK-NEXT: LBB0_3: ## %bb4 ; CHECK-NEXT: movq _PyArray_API@GOTPCREL(%rip), %rax ; CHECK-NEXT: movq (%rax), %rax ; CHECK-NEXT: movq 16(%rax), %rax -; CHECK-NEXT: jmp LBB0_9 -; CHECK-NEXT: LBB0_6: ## %_ubyte_convert2_to_ctypes.exit +; CHECK-NEXT: jmp LBB0_10 +; CHECK-NEXT: LBB0_4: ## %_ubyte_convert2_to_ctypes.exit ; CHECK-NEXT: cmpl $-2, %eax -; CHECK-NEXT: je LBB0_10 -; CHECK-NEXT: ## %bb.7: ## %_ubyte_convert2_to_ctypes.exit -; CHECK-NEXT: cmpl $-1, %eax ; CHECK-NEXT: je LBB0_8 -; CHECK-NEXT: LBB0_3: ## %bb35 +; CHECK-NEXT: ## %bb.5: ## %_ubyte_convert2_to_ctypes.exit +; CHECK-NEXT: cmpl $-1, %eax +; CHECK-NEXT: je LBB0_3 +; CHECK-NEXT: LBB0_6: ## %bb35 ; CHECK-NEXT: movq _PyUFunc_API@GOTPCREL(%rip), %r14 ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: callq *216(%rax) ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %edx ; CHECK-NEXT: testb %dl, %dl -; CHECK-NEXT: je LBB0_4 -; CHECK-NEXT: ## %bb.12: ## %cond_false.i -; CHECK-NEXT: setne %dil +; CHECK-NEXT: je LBB0_11 +; CHECK-NEXT: ## %bb.7: ## %cond_false.i ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %esi ; CHECK-NEXT: movzbl %sil, %ecx ; CHECK-NEXT: movl %ecx, %eax ; CHECK-NEXT: divb %dl ; CHECK-NEXT: movl %eax, %r15d ; CHECK-NEXT: testb %cl, %cl -; CHECK-NEXT: setne %al -; CHECK-NEXT: testb %dil, %al -; CHECK-NEXT: jne LBB0_5 -; CHECK-NEXT: LBB0_13: ## %cond_true.i200 -; CHECK-NEXT: testb %dl, %dl -; CHECK-NEXT: jne LBB0_15 -; CHECK-NEXT: ## %bb.14: ## %cond_true14.i -; CHECK-NEXT: movl $4, %edi -; CHECK-NEXT: callq _feraiseexcept -; CHECK-NEXT: LBB0_15: ## %ubyte_ctype_remainder.exit -; CHECK-NEXT: xorl %ebx, %ebx -; CHECK-NEXT: jmp LBB0_16 -; CHECK-NEXT: LBB0_10: ## %bb17 +; CHECK-NEXT: jne LBB0_12 +; CHECK-NEXT: jmp LBB0_14 +; CHECK-NEXT: LBB0_8: ## %bb17 ; CHECK-NEXT: callq _PyErr_Occurred ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: jne LBB0_23 -; CHECK-NEXT: ## %bb.11: ## %cond_next +; CHECK-NEXT: jne LBB0_27 +; CHECK-NEXT: ## %bb.9: ## %cond_next ; CHECK-NEXT: movq _PyArray_API@GOTPCREL(%rip), %rax ; CHECK-NEXT: movq (%rax), %rax ; CHECK-NEXT: movq 80(%rax), %rax -; CHECK-NEXT: LBB0_9: ## %bb4 +; CHECK-NEXT: LBB0_10: ## %bb4 ; CHECK-NEXT: movq 96(%rax), %rax ; CHECK-NEXT: movq %r14, %rdi ; CHECK-NEXT: movq %rbx, %rsi ; CHECK-NEXT: callq *40(%rax) -; CHECK-NEXT: jmp LBB0_24 -; CHECK-NEXT: LBB0_4: ## %cond_true.i +; CHECK-NEXT: jmp LBB0_28 +; CHECK-NEXT: LBB0_11: ## %cond_true.i ; CHECK-NEXT: movl $4, %edi ; CHECK-NEXT: callq _feraiseexcept ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %edx ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %esi +; CHECK-NEXT: xorl %r15d, %r15d ; CHECK-NEXT: testb %sil, %sil -; CHECK-NEXT: sete %al +; CHECK-NEXT: je LBB0_14 +; CHECK-NEXT: LBB0_12: ## %cond_false.i ; CHECK-NEXT: testb %dl, %dl -; CHECK-NEXT: sete %cl -; CHECK-NEXT: xorl %r15d, %r15d -; CHECK-NEXT: orb %al, %cl -; CHECK-NEXT: jne LBB0_13 -; CHECK-NEXT: LBB0_5: ## %cond_next17.i +; CHECK-NEXT: je LBB0_14 +; CHECK-NEXT: ## %bb.13: ## %cond_next17.i ; CHECK-NEXT: movzbl %sil, %eax ; CHECK-NEXT: divb %dl ; CHECK-NEXT: movzbl %ah, %ebx -; CHECK-NEXT: LBB0_16: ## %ubyte_ctype_remainder.exit +; CHECK-NEXT: jmp LBB0_18 +; CHECK-NEXT: LBB0_14: ## %cond_true.i200 +; CHECK-NEXT: testb %dl, %dl +; CHECK-NEXT: jne LBB0_17 +; CHECK-NEXT: ## %bb.16: ## %cond_true14.i +; CHECK-NEXT: movl $4, %edi +; CHECK-NEXT: callq _feraiseexcept +; CHECK-NEXT: LBB0_17: ## %ubyte_ctype_remainder.exit +; CHECK-NEXT: xorl %ebx, %ebx +; CHECK-NEXT: LBB0_18: ## %ubyte_ctype_remainder.exit ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: callq *224(%rax) ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: je LBB0_19 -; CHECK-NEXT: ## %bb.17: ## %cond_true61 +; CHECK-NEXT: je LBB0_21 +; CHECK-NEXT: ## %bb.19: ## %cond_true61 ; CHECK-NEXT: movl %eax, %ebp ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: movq _.str5@GOTPCREL(%rip), %rdi @@ -139,8 +136,8 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: leaq {{[0-9]+}}(%rsp), %rcx ; CHECK-NEXT: callq *200(%rax) ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: js LBB0_23 -; CHECK-NEXT: ## %bb.18: ## %cond_next73 +; CHECK-NEXT: js LBB0_27 +; CHECK-NEXT: ## %bb.20: ## %cond_next73 ; CHECK-NEXT: movl $1, {{[0-9]+}}(%rsp) ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: movq {{[0-9]+}}(%rsp), %rsi @@ -149,13 +146,13 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: movl %ebp, %edx ; CHECK-NEXT: callq *232(%rax) ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: jne LBB0_23 -; CHECK-NEXT: LBB0_19: ## %cond_next89 +; CHECK-NEXT: jne LBB0_27 +; CHECK-NEXT: LBB0_21: ## %cond_next89 ; CHECK-NEXT: movl $2, %edi ; CHECK-NEXT: callq _PyTuple_New ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: je LBB0_23 -; CHECK-NEXT: ## %bb.20: ## %cond_next97 +; CHECK-NEXT: je LBB0_27 +; CHECK-NEXT: ## %bb.22: ## %cond_next97 ; CHECK-NEXT: movq %rax, %r14 ; CHECK-NEXT: movq _PyArray_API@GOTPCREL(%rip), %r12 ; CHECK-NEXT: movq (%r12), %rax @@ -163,8 +160,8 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: xorl %esi, %esi ; CHECK-NEXT: callq *304(%rdi) ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: je LBB0_21 -; CHECK-NEXT: ## %bb.25: ## %cond_next135 +; CHECK-NEXT: je LBB0_25 +; CHECK-NEXT: ## %bb.23: ## %cond_next135 ; CHECK-NEXT: movb %r15b, 16(%rax) ; CHECK-NEXT: movq %rax, 24(%r14) ; CHECK-NEXT: movq (%r12), %rax @@ -172,22 +169,22 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: xorl %esi, %esi ; CHECK-NEXT: callq *304(%rdi) ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: je LBB0_21 -; CHECK-NEXT: ## %bb.26: ## %cond_next182 +; CHECK-NEXT: je LBB0_25 +; CHECK-NEXT: ## %bb.24: ## %cond_next182 ; CHECK-NEXT: movb %bl, 16(%rax) ; CHECK-NEXT: movq %rax, 32(%r14) ; CHECK-NEXT: movq %r14, %rax -; CHECK-NEXT: jmp LBB0_24 -; CHECK-NEXT: LBB0_21: ## %cond_true113 +; CHECK-NEXT: jmp LBB0_28 +; CHECK-NEXT: LBB0_25: ## %cond_true113 ; CHECK-NEXT: decq (%r14) -; CHECK-NEXT: jne LBB0_23 -; CHECK-NEXT: ## %bb.22: ## %cond_true126 +; CHECK-NEXT: jne LBB0_27 +; CHECK-NEXT: ## %bb.26: ## %cond_true126 ; CHECK-NEXT: movq 8(%r14), %rax ; CHECK-NEXT: movq %r14, %rdi ; CHECK-NEXT: callq *48(%rax) -; CHECK-NEXT: LBB0_23: ## %UnifiedReturnBlock +; CHECK-NEXT: LBB0_27: ## %UnifiedReturnBlock ; CHECK-NEXT: xorl %eax, %eax -; CHECK-NEXT: LBB0_24: ## %UnifiedReturnBlock +; CHECK-NEXT: LBB0_28: ## %UnifiedReturnBlock ; CHECK-NEXT: addq $32, %rsp ; CHECK-NEXT: popq %rbx ; CHECK-NEXT: popq %r12 diff --git a/llvm/test/CodeGen/X86/isel-fpclass.ll b/llvm/test/CodeGen/X86/isel-fpclass.ll index df04b67..c2b7068 100644 --- a/llvm/test/CodeGen/X86/isel-fpclass.ll +++ b/llvm/test/CodeGen/X86/isel-fpclass.ll @@ -1,10 +1,10 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5 -; RUN: llc < %s -mtriple=i686-linux | FileCheck %s -check-prefixes=X86 +; RUN: llc < %s -mtriple=i686-linux | FileCheck %s -check-prefixes=X86,X86-SDAGISEL ; RUN: llc < %s -mtriple=x86_64-linux | FileCheck %s -check-prefixes=X64,X64-SDAGISEL ; RUN: llc < %s -mtriple=i686-linux -fast-isel -fast-isel-abort=1 | FileCheck %s -check-prefixes=X86-FASTISEL ; RUN: llc < %s -mtriple=x86_64-linux -fast-isel -fast-isel-abort=1 | FileCheck %s -check-prefixes=X64,X64-FASTISEL -; RUN: llc < %s -mtriple=i686-linux -global-isel -global-isel-abort=2 | FileCheck %s -check-prefixes=X86 -; RUN: llc < %s -mtriple=x86_64-linux -global-isel -global-isel-abort=2 | FileCheck %s -check-prefixes=X64,X64-GISEL +; RUN: llc < %s -mtriple=i686-linux -global-isel -global-isel-abort=1 | FileCheck %s -check-prefixes=X86,X86-GISEL +; RUN: llc < %s -mtriple=x86_64-linux -global-isel -global-isel-abort=1 | FileCheck %s -check-prefixes=X64-GISEL define i1 @isnone_f(float %x) nounwind { ; X86-LABEL: isnone_f: @@ -23,6 +23,11 @@ define i1 @isnone_f(float %x) nounwind { ; X86-FASTISEL-NEXT: fstp %st(0) ; X86-FASTISEL-NEXT: xorl %eax, %eax ; X86-FASTISEL-NEXT: retl +; +; X64-GISEL-LABEL: isnone_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: xorl %eax, %eax +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 0) ret i1 %0 @@ -45,22 +50,27 @@ define i1 @isany_f(float %x) nounwind { ; X86-FASTISEL-NEXT: fstp %st(0) ; X86-FASTISEL-NEXT: movb $1, %al ; X86-FASTISEL-NEXT: retl +; +; X64-GISEL-LABEL: isany_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movb $1, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 1023) ret i1 %0 } define i1 @issignaling_f(float %x) nounwind { -; X86-LABEL: issignaling_f: -; X86: # %bb.0: -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 -; X86-NEXT: setl %cl -; X86-NEXT: cmpl $2139095041, %eax # imm = 0x7F800001 -; X86-NEXT: setge %al -; X86-NEXT: andb %cl, %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: issignaling_f: +; X86-SDAGISEL: # %bb.0: +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X86-SDAGISEL-NEXT: setl %cl +; X86-SDAGISEL-NEXT: cmpl $2139095041, %eax # imm = 0x7F800001 +; X86-SDAGISEL-NEXT: setge %al +; X86-SDAGISEL-NEXT: andb %cl, %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: issignaling_f: ; X64: # %bb.0: @@ -87,18 +97,44 @@ define i1 @issignaling_f(float %x) nounwind { ; X86-FASTISEL-NEXT: andb %cl, %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: issignaling_f: +; X86-GISEL: # %bb.0: +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: seta %dl +; X86-GISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X86-GISEL-NEXT: setb %al +; X86-GISEL-NEXT: andb %dl, %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: issignaling_f: +; X64-GISEL: # %bb.0: +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: seta %dl +; X64-GISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X64-GISEL-NEXT: setb %al +; X64-GISEL-NEXT: andb %dl, %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq %a0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 1) ; "snan" ret i1 %a0 } define i1 @isquiet_f(float %x) nounwind { -; X86-LABEL: isquiet_f: -; X86: # %bb.0: # %entry -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 -; X86-NEXT: setge %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: isquiet_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X86-SDAGISEL-NEXT: setge %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: isquiet_f: ; X64: # %bb.0: # %entry @@ -119,19 +155,39 @@ define i1 @issignaling_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setge %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: isquiet_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X86-GISEL-NEXT: setae %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: isquiet_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X64-GISEL-NEXT: setae %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 2) ; "qnan" ret i1 %0 } define i1 @not_isquiet_f(float %x) nounwind { -; X86-LABEL: not_isquiet_f: -; X86: # %bb.0: # %entry -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 -; X86-NEXT: setl %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: not_isquiet_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X86-SDAGISEL-NEXT: setl %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: not_isquiet_f: ; X64: # %bb.0: # %entry @@ -152,19 +208,57 @@ define i1 @not_isquiet_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setl %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: not_isquiet_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: setb %dl +; X86-GISEL-NEXT: orb %cl, %dl +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: sete %cl +; X86-GISEL-NEXT: orb %dl, %cl +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: seta %dl +; X86-GISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X86-GISEL-NEXT: setb %al +; X86-GISEL-NEXT: andb %dl, %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: not_isquiet_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: setb %dl +; X64-GISEL-NEXT: orb %cl, %dl +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: sete %cl +; X64-GISEL-NEXT: orb %dl, %cl +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: seta %dl +; X64-GISEL-NEXT: cmpl $2143289344, %eax # imm = 0x7FC00000 +; X64-GISEL-NEXT: setb %al +; X64-GISEL-NEXT: andb %dl, %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 1021) ; ~"qnan" ret i1 %0 } define i1 @isinf_f(float %x) nounwind { -; X86-LABEL: isinf_f: -; X86: # %bb.0: # %entry -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 -; X86-NEXT: sete %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: isinf_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-SDAGISEL-NEXT: sete %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: isinf_f: ; X64: # %bb.0: # %entry @@ -185,19 +279,39 @@ define i1 @isinf_f(float %x) nounwind { ; X86-FASTISEL-NEXT: sete %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: isinf_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: sete %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: isinf_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: sete %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 516) ; 0x204 = "inf" ret i1 %0 } define i1 @not_isinf_f(float %x) nounwind { -; X86-LABEL: not_isinf_f: -; X86: # %bb.0: # %entry -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 -; X86-NEXT: setne %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: not_isinf_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-SDAGISEL-NEXT: setne %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: not_isinf_f: ; X64: # %bb.0: # %entry @@ -218,17 +332,43 @@ define i1 @not_isinf_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setne %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: not_isinf_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: setb %dl +; X86-GISEL-NEXT: orb %cl, %dl +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: seta %al +; X86-GISEL-NEXT: orb %dl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: not_isinf_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: setb %dl +; X64-GISEL-NEXT: orb %cl, %dl +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: seta %al +; X64-GISEL-NEXT: orb %dl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 507) ; ~0x204 = "~inf" ret i1 %0 } define i1 @is_plus_inf_f(float %x) nounwind { -; X86-LABEL: is_plus_inf_f: -; X86: # %bb.0: # %entry -; X86-NEXT: cmpl $2139095040, {{[0-9]+}}(%esp) # imm = 0x7F800000 -; X86-NEXT: sete %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: is_plus_inf_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: cmpl $2139095040, {{[0-9]+}}(%esp) # imm = 0x7F800000 +; X86-SDAGISEL-NEXT: sete %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: is_plus_inf_f: ; X64: # %bb.0: # %entry @@ -246,17 +386,34 @@ define i1 @is_plus_inf_f(float %x) nounwind { ; X86-FASTISEL-NEXT: sete %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: is_plus_inf_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, {{[0-9]+}}(%esp) # imm = 0x7F800000 +; X86-GISEL-NEXT: sete %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: is_plus_inf_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: sete %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 512) ; 0x200 = "+inf" ret i1 %0 } define i1 @is_minus_inf_f(float %x) nounwind { -; X86-LABEL: is_minus_inf_f: -; X86: # %bb.0: # %entry -; X86-NEXT: cmpl $-8388608, {{[0-9]+}}(%esp) # imm = 0xFF800000 -; X86-NEXT: sete %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: is_minus_inf_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: cmpl $-8388608, {{[0-9]+}}(%esp) # imm = 0xFF800000 +; X86-SDAGISEL-NEXT: sete %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: is_minus_inf_f: ; X64: # %bb.0: # %entry @@ -274,17 +431,34 @@ define i1 @is_minus_inf_f(float %x) nounwind { ; X86-FASTISEL-NEXT: sete %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: is_minus_inf_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $-8388608, {{[0-9]+}}(%esp) # imm = 0xFF800000 +; X86-GISEL-NEXT: sete %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: is_minus_inf_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: cmpl $-8388608, %eax # imm = 0xFF800000 +; X64-GISEL-NEXT: sete %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 4) ; "-inf" ret i1 %0 } define i1 @not_is_minus_inf_f(float %x) nounwind { -; X86-LABEL: not_is_minus_inf_f: -; X86: # %bb.0: # %entry -; X86-NEXT: cmpl $-8388608, {{[0-9]+}}(%esp) # imm = 0xFF800000 -; X86-NEXT: setne %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: not_is_minus_inf_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: cmpl $-8388608, {{[0-9]+}}(%esp) # imm = 0xFF800000 +; X86-SDAGISEL-NEXT: setne %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: not_is_minus_inf_f: ; X64: # %bb.0: # %entry @@ -302,19 +476,55 @@ define i1 @not_is_minus_inf_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setne %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: not_is_minus_inf_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: pushl %ebx +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: movl %eax, %ecx +; X86-GISEL-NEXT: andl $2147483647, %ecx # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %edx, %edx +; X86-GISEL-NEXT: cmpl $2139095040, %ecx # imm = 0x7F800000 +; X86-GISEL-NEXT: setb %bl +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: sete %ah +; X86-GISEL-NEXT: orb %dl, %ah +; X86-GISEL-NEXT: orb %bl, %ah +; X86-GISEL-NEXT: cmpl $2139095040, %ecx # imm = 0x7F800000 +; X86-GISEL-NEXT: seta %al +; X86-GISEL-NEXT: orb %ah, %al +; X86-GISEL-NEXT: popl %ebx +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: not_is_minus_inf_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: movl %eax, %ecx +; X64-GISEL-NEXT: andl $2147483647, %ecx # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %edx, %edx +; X64-GISEL-NEXT: cmpl $2139095040, %ecx # imm = 0x7F800000 +; X64-GISEL-NEXT: setb %sil +; X64-GISEL-NEXT: orb %dl, %sil +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: sete %dl +; X64-GISEL-NEXT: cmpl $2139095040, %ecx # imm = 0x7F800000 +; X64-GISEL-NEXT: seta %al +; X64-GISEL-NEXT: orb %dl, %al +; X64-GISEL-NEXT: orb %sil, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 1019) ; ~"-inf" ret i1 %0 } define i1 @isfinite_f(float %x) nounwind { -; X86-LABEL: isfinite_f: -; X86: # %bb.0: # %entry -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 -; X86-NEXT: setl %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: isfinite_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-SDAGISEL-NEXT: setl %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: isfinite_f: ; X64: # %bb.0: # %entry @@ -335,19 +545,39 @@ define i1 @isfinite_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setl %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: isfinite_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: setb %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: isfinite_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: setb %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 504) ; 0x1f8 = "finite" ret i1 %0 } define i1 @not_isfinite_f(float %x) nounwind { -; X86-LABEL: not_isfinite_f: -; X86: # %bb.0: # %entry -; X86-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF -; X86-NEXT: andl {{[0-9]+}}(%esp), %eax -; X86-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 -; X86-NEXT: setge %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: not_isfinite_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: movl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-SDAGISEL-NEXT: andl {{[0-9]+}}(%esp), %eax +; X86-SDAGISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-SDAGISEL-NEXT: setge %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: not_isfinite_f: ; X64: # %bb.0: # %entry @@ -368,17 +598,43 @@ define i1 @not_isfinite_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setge %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: not_isfinite_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: sete %dl +; X86-GISEL-NEXT: orb %cl, %dl +; X86-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X86-GISEL-NEXT: seta %al +; X86-GISEL-NEXT: orb %dl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: not_isfinite_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: andl $2147483647, %eax # imm = 0x7FFFFFFF +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: sete %dl +; X64-GISEL-NEXT: orb %cl, %dl +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: seta %al +; X64-GISEL-NEXT: orb %dl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 519) ; ~0x1f8 = "~finite" ret i1 %0 } define i1 @is_plus_finite_f(float %x) nounwind { -; X86-LABEL: is_plus_finite_f: -; X86: # %bb.0: # %entry -; X86-NEXT: cmpl $2139095040, {{[0-9]+}}(%esp) # imm = 0x7F800000 -; X86-NEXT: setb %al -; X86-NEXT: retl +; X86-SDAGISEL-LABEL: is_plus_finite_f: +; X86-SDAGISEL: # %bb.0: # %entry +; X86-SDAGISEL-NEXT: cmpl $2139095040, {{[0-9]+}}(%esp) # imm = 0x7F800000 +; X86-SDAGISEL-NEXT: setb %al +; X86-SDAGISEL-NEXT: retl ; ; X64-LABEL: is_plus_finite_f: ; X64: # %bb.0: # %entry @@ -396,6 +652,23 @@ define i1 @is_plus_finite_f(float %x) nounwind { ; X86-FASTISEL-NEXT: setb %al ; X86-FASTISEL-NEXT: popl %ecx ; X86-FASTISEL-NEXT: retl +; +; X86-GISEL-LABEL: is_plus_finite_f: +; X86-GISEL: # %bb.0: # %entry +; X86-GISEL-NEXT: xorl %ecx, %ecx +; X86-GISEL-NEXT: cmpl $2139095040, {{[0-9]+}}(%esp) # imm = 0x7F800000 +; X86-GISEL-NEXT: setb %al +; X86-GISEL-NEXT: orb %cl, %al +; X86-GISEL-NEXT: retl +; +; X64-GISEL-LABEL: is_plus_finite_f: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: xorl %ecx, %ecx +; X64-GISEL-NEXT: movd %xmm0, %eax +; X64-GISEL-NEXT: cmpl $2139095040, %eax # imm = 0x7F800000 +; X64-GISEL-NEXT: setb %al +; X64-GISEL-NEXT: orb %cl, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f32(float %x, i32 448) ; 0x1c0 = "+finite" ret i1 %0 @@ -418,6 +691,11 @@ define i1 @isnone_d(double %x) nounwind { ; X86-FASTISEL-NEXT: fstp %st(0) ; X86-FASTISEL-NEXT: xorl %eax, %eax ; X86-FASTISEL-NEXT: retl +; +; X64-GISEL-LABEL: isnone_d: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: xorl %eax, %eax +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f64(double %x, i32 0) ret i1 %0 @@ -440,6 +718,11 @@ define i1 @isany_d(double %x) nounwind { ; X86-FASTISEL-NEXT: fstp %st(0) ; X86-FASTISEL-NEXT: movb $1, %al ; X86-FASTISEL-NEXT: retl +; +; X64-GISEL-LABEL: isany_d: +; X64-GISEL: # %bb.0: # %entry +; X64-GISEL-NEXT: movb $1, %al +; X64-GISEL-NEXT: retq entry: %0 = tail call i1 @llvm.is.fpclass.f64(double %x, i32 1023) ret i1 %0 diff --git a/llvm/test/CodeGen/X86/masked_gather_scatter.ll b/llvm/test/CodeGen/X86/masked_gather_scatter.ll index 4cde581..cdf6bdd 100644 --- a/llvm/test/CodeGen/X86/masked_gather_scatter.ll +++ b/llvm/test/CodeGen/X86/masked_gather_scatter.ll @@ -4765,6 +4765,33 @@ define void @scaleidx_scatter_outofrange(<8 x float> %value, ptr %base, <8 x i32 } declare void @llvm.masked.scatter.v8f32.v8p0(<8 x float>, <8 x ptr>, i32 immarg, <8 x i1>) +define <16 x i32> @pr163023(ptr %a0, <16 x i32> %a1) { +; X64-LABEL: pr163023: +; X64: # %bb.0: +; X64-NEXT: kxnorw %k0, %k0, %k1 +; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; X64-NEXT: vpgatherdd (%rdi,%zmm0), %zmm1 {%k1} +; X64-NEXT: vmovdqa64 %zmm1, %zmm0 +; X64-NEXT: retq +; +; X86-LABEL: pr163023: +; X86: # %bb.0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: kxnorw %k0, %k0, %k1 +; X86-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; X86-NEXT: vpgatherdd (%eax,%zmm0), %zmm1 {%k1} +; X86-NEXT: vmovdqa64 %zmm1, %zmm0 +; X86-NEXT: retl + %addr.p = ptrtoint ptr %a0 to i64 + %addr.v = insertelement <1 x i64> poison, i64 %addr.p, i64 0 + %addr.splat = shufflevector <1 x i64> %addr.v, <1 x i64> poison, <16 x i32> zeroinitializer + %ofs = sext <16 x i32> %a1 to <16 x i64> + %addr = add nuw <16 x i64> %addr.splat, %ofs + %ptr = inttoptr <16 x i64> %addr to <16 x ptr> + %gather = tail call fastcc <16 x i32> @llvm.masked.gather.v16i32.v16p0(<16 x ptr> %ptr, i32 4, <16 x i1> splat (i1 true), <16 x i32> poison) + ret <16 x i32> %gather +} + ; ; PR45906 ; This used to cause fast-isel to generate bad copy instructions that would diff --git a/llvm/test/CodeGen/X86/pr160612.ll b/llvm/test/CodeGen/X86/pr160612.ll new file mode 100644 index 0000000..6572c42 --- /dev/null +++ b/llvm/test/CodeGen/X86/pr160612.ll @@ -0,0 +1,74 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -O2 | FileCheck %s + +; Test for issue #160612: OR conditions in branches should use multiple branches +; instead of materializing booleans with SETCC when no special optimizations apply. + +declare void @subroutine_foo() +declare void @subroutine_bar() + +; Original issue: (x == 0 || y == 0) was generating SETCC + TEST + BRANCH +; instead of using two conditional branches directly. +define void @func_a(i32 noundef %x, i32 noundef %y) { +; CHECK-LABEL: func_a: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: testl %edi, %edi +; CHECK-NEXT: je subroutine_foo@PLT # TAILCALL +; CHECK-NEXT: # %bb.1: # %entry +; CHECK-NEXT: testl %esi, %esi +; CHECK-NEXT: jne subroutine_bar@PLT # TAILCALL +; CHECK-NEXT: # %bb.2: # %if.then +; CHECK-NEXT: jmp subroutine_foo@PLT # TAILCALL +entry: + %cmp = icmp eq i32 %x, 0 + %cmp1 = icmp eq i32 %y, 0 + %or.cond = or i1 %cmp, %cmp1 + br i1 %or.cond, label %if.then, label %if.else + +if.then: + tail call void @subroutine_foo() + br label %if.end + +if.else: + tail call void @subroutine_bar() + br label %if.end + +if.end: + ret void +} + +; Reference implementation that already generated optimal code. +; This should continue to generate the same optimal code. +define void @func_b(i32 noundef %x, i32 noundef %y) { +; CHECK-LABEL: func_b: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: testl %edi, %edi +; CHECK-NEXT: je subroutine_foo@PLT # TAILCALL +; CHECK-NEXT: # %bb.1: # %if.else +; CHECK-NEXT: testl %esi, %esi +; CHECK-NEXT: je subroutine_foo@PLT # TAILCALL +; CHECK-NEXT: # %bb.2: # %if.else3 +; CHECK-NEXT: jmp subroutine_bar@PLT # TAILCALL +entry: + %cmp = icmp eq i32 %x, 0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + tail call void @subroutine_foo() + br label %if.end4 + +if.else: + %cmp1 = icmp eq i32 %y, 0 + br i1 %cmp1, label %if.then2, label %if.else3 + +if.then2: + tail call void @subroutine_foo() + br label %if.end4 + +if.else3: + tail call void @subroutine_bar() + br label %if.end4 + +if.end4: + ret void +} diff --git a/llvm/test/CodeGen/X86/setcc-wide-types.ll b/llvm/test/CodeGen/X86/setcc-wide-types.ll index 69abf6e..d018c53 100644 --- a/llvm/test/CodeGen/X86/setcc-wide-types.ll +++ b/llvm/test/CodeGen/X86/setcc-wide-types.ll @@ -1493,15 +1493,23 @@ define i1 @allbits_i128_load_arg(ptr %w) { } define i1 @anybits_i256_load_arg(ptr %w) { -; ANY-LABEL: anybits_i256_load_arg: -; ANY: # %bb.0: -; ANY-NEXT: movq (%rdi), %rax -; ANY-NEXT: movq 8(%rdi), %rcx -; ANY-NEXT: orq 24(%rdi), %rcx -; ANY-NEXT: orq 16(%rdi), %rax -; ANY-NEXT: orq %rcx, %rax -; ANY-NEXT: setne %al -; ANY-NEXT: retq +; SSE-LABEL: anybits_i256_load_arg: +; SSE: # %bb.0: +; SSE-NEXT: movq (%rdi), %rax +; SSE-NEXT: movq 8(%rdi), %rcx +; SSE-NEXT: orq 24(%rdi), %rcx +; SSE-NEXT: orq 16(%rdi), %rax +; SSE-NEXT: orq %rcx, %rax +; SSE-NEXT: setne %al +; SSE-NEXT: retq +; +; AVXANY-LABEL: anybits_i256_load_arg: +; AVXANY: # %bb.0: +; AVXANY-NEXT: vmovdqu (%rdi), %ymm0 +; AVXANY-NEXT: vptest %ymm0, %ymm0 +; AVXANY-NEXT: setne %al +; AVXANY-NEXT: vzeroupper +; AVXANY-NEXT: retq %ld = load i256, ptr %w %cmp = icmp ne i256 %ld, 0 ret i1 %cmp @@ -1552,21 +1560,30 @@ define i1 @allbits_i256_load_arg(ptr %w) { } define i1 @anybits_i512_load_arg(ptr %w) { -; ANY-LABEL: anybits_i512_load_arg: -; ANY: # %bb.0: -; ANY-NEXT: movq 16(%rdi), %rax -; ANY-NEXT: movq (%rdi), %rcx -; ANY-NEXT: movq 8(%rdi), %rdx -; ANY-NEXT: movq 24(%rdi), %rsi -; ANY-NEXT: orq 56(%rdi), %rsi -; ANY-NEXT: orq 40(%rdi), %rdx -; ANY-NEXT: orq %rsi, %rdx -; ANY-NEXT: orq 48(%rdi), %rax -; ANY-NEXT: orq 32(%rdi), %rcx -; ANY-NEXT: orq %rax, %rcx -; ANY-NEXT: orq %rdx, %rcx -; ANY-NEXT: setne %al -; ANY-NEXT: retq +; NO512-LABEL: anybits_i512_load_arg: +; NO512: # %bb.0: +; NO512-NEXT: movq 16(%rdi), %rax +; NO512-NEXT: movq (%rdi), %rcx +; NO512-NEXT: movq 8(%rdi), %rdx +; NO512-NEXT: movq 24(%rdi), %rsi +; NO512-NEXT: orq 56(%rdi), %rsi +; NO512-NEXT: orq 40(%rdi), %rdx +; NO512-NEXT: orq %rsi, %rdx +; NO512-NEXT: orq 48(%rdi), %rax +; NO512-NEXT: orq 32(%rdi), %rcx +; NO512-NEXT: orq %rax, %rcx +; NO512-NEXT: orq %rdx, %rcx +; NO512-NEXT: setne %al +; NO512-NEXT: retq +; +; AVX512-LABEL: anybits_i512_load_arg: +; AVX512: # %bb.0: +; AVX512-NEXT: vmovdqu64 (%rdi), %zmm0 +; AVX512-NEXT: vptestmd %zmm0, %zmm0, %k0 +; AVX512-NEXT: kortestw %k0, %k0 +; AVX512-NEXT: setne %al +; AVX512-NEXT: vzeroupper +; AVX512-NEXT: retq %ld = load i512, ptr %w %cmp = icmp ne i512 %ld, 0 ret i1 %cmp diff --git a/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll b/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll index b2064b1..02d4d88 100644 --- a/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll +++ b/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll @@ -181,40 +181,38 @@ define zeroext i1 @segmentedStack(ptr readonly %vk1, ptr readonly %vk2, i64 %key ; CHECK-LABEL: segmentedStack: ; CHECK: ## %bb.0: ; CHECK-NEXT: cmpq %gs:816, %rsp -; CHECK-NEXT: jbe LBB3_6 +; CHECK-NEXT: jbe LBB3_7 ; CHECK-NEXT: LBB3_1: ## %entry ; CHECK-NEXT: pushq %rax ; CHECK-NEXT: .cfi_def_cfa_offset 16 -; CHECK-NEXT: testq %rdi, %rdi -; CHECK-NEXT: sete %al -; CHECK-NEXT: testq %rsi, %rsi -; CHECK-NEXT: sete %cl -; CHECK-NEXT: orb %al, %cl ; CHECK-NEXT: movq %rdi, %rax ; CHECK-NEXT: orq %rsi, %rax ; CHECK-NEXT: sete %al -; CHECK-NEXT: testb %cl, %cl -; CHECK-NEXT: jne LBB3_4 -; CHECK-NEXT: ## %bb.2: ## %if.end4.i +; CHECK-NEXT: testq %rdi, %rdi +; CHECK-NEXT: je LBB3_5 +; CHECK-NEXT: ## %bb.2: ## %entry +; CHECK-NEXT: testq %rsi, %rsi +; CHECK-NEXT: je LBB3_5 +; CHECK-NEXT: ## %bb.3: ## %if.end4.i ; CHECK-NEXT: movq 8(%rdi), %rdx ; CHECK-NEXT: cmpq 8(%rsi), %rdx -; CHECK-NEXT: jne LBB3_5 -; CHECK-NEXT: ## %bb.3: ## %land.rhs.i.i +; CHECK-NEXT: jne LBB3_6 +; CHECK-NEXT: ## %bb.4: ## %land.rhs.i.i ; CHECK-NEXT: movq (%rsi), %rsi ; CHECK-NEXT: movq (%rdi), %rdi ; CHECK-NEXT: callq _memcmp ; CHECK-NEXT: testl %eax, %eax ; CHECK-NEXT: sete %al -; CHECK-NEXT: LBB3_4: ## %__go_ptr_strings_equal.exit +; CHECK-NEXT: LBB3_5: ## %__go_ptr_strings_equal.exit ; CHECK-NEXT: ## kill: def $al killed $al killed $eax ; CHECK-NEXT: popq %rcx ; CHECK-NEXT: retq -; CHECK-NEXT: LBB3_5: +; CHECK-NEXT: LBB3_6: ; CHECK-NEXT: xorl %eax, %eax ; CHECK-NEXT: ## kill: def $al killed $al killed $eax ; CHECK-NEXT: popq %rcx ; CHECK-NEXT: retq -; CHECK-NEXT: LBB3_6: +; CHECK-NEXT: LBB3_7: ; CHECK-NEXT: movl $8, %r10d ; CHECK-NEXT: movl $0, %r11d ; CHECK-NEXT: callq ___morestack @@ -224,43 +222,41 @@ define zeroext i1 @segmentedStack(ptr readonly %vk1, ptr readonly %vk2, i64 %key ; NOCOMPACTUNWIND-LABEL: segmentedStack: ; NOCOMPACTUNWIND: # %bb.0: ; NOCOMPACTUNWIND-NEXT: cmpq %fs:112, %rsp -; NOCOMPACTUNWIND-NEXT: jbe .LBB3_6 +; NOCOMPACTUNWIND-NEXT: jbe .LBB3_7 ; NOCOMPACTUNWIND-NEXT: .LBB3_1: # %entry ; NOCOMPACTUNWIND-NEXT: pushq %rax ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 16 -; NOCOMPACTUNWIND-NEXT: testq %rdi, %rdi -; NOCOMPACTUNWIND-NEXT: sete %al -; NOCOMPACTUNWIND-NEXT: testq %rsi, %rsi -; NOCOMPACTUNWIND-NEXT: sete %cl -; NOCOMPACTUNWIND-NEXT: orb %al, %cl ; NOCOMPACTUNWIND-NEXT: movq %rdi, %rax ; NOCOMPACTUNWIND-NEXT: orq %rsi, %rax ; NOCOMPACTUNWIND-NEXT: sete %al -; NOCOMPACTUNWIND-NEXT: testb %cl, %cl -; NOCOMPACTUNWIND-NEXT: jne .LBB3_4 -; NOCOMPACTUNWIND-NEXT: # %bb.2: # %if.end4.i +; NOCOMPACTUNWIND-NEXT: testq %rdi, %rdi +; NOCOMPACTUNWIND-NEXT: je .LBB3_5 +; NOCOMPACTUNWIND-NEXT: # %bb.2: # %entry +; NOCOMPACTUNWIND-NEXT: testq %rsi, %rsi +; NOCOMPACTUNWIND-NEXT: je .LBB3_5 +; NOCOMPACTUNWIND-NEXT: # %bb.3: # %if.end4.i ; NOCOMPACTUNWIND-NEXT: movq 8(%rdi), %rdx ; NOCOMPACTUNWIND-NEXT: cmpq 8(%rsi), %rdx -; NOCOMPACTUNWIND-NEXT: jne .LBB3_5 -; NOCOMPACTUNWIND-NEXT: # %bb.3: # %land.rhs.i.i +; NOCOMPACTUNWIND-NEXT: jne .LBB3_6 +; NOCOMPACTUNWIND-NEXT: # %bb.4: # %land.rhs.i.i ; NOCOMPACTUNWIND-NEXT: movq (%rsi), %rsi ; NOCOMPACTUNWIND-NEXT: movq (%rdi), %rdi ; NOCOMPACTUNWIND-NEXT: callq memcmp@PLT ; NOCOMPACTUNWIND-NEXT: testl %eax, %eax ; NOCOMPACTUNWIND-NEXT: sete %al -; NOCOMPACTUNWIND-NEXT: .LBB3_4: # %__go_ptr_strings_equal.exit +; NOCOMPACTUNWIND-NEXT: .LBB3_5: # %__go_ptr_strings_equal.exit ; NOCOMPACTUNWIND-NEXT: # kill: def $al killed $al killed $eax ; NOCOMPACTUNWIND-NEXT: popq %rcx ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 8 ; NOCOMPACTUNWIND-NEXT: retq -; NOCOMPACTUNWIND-NEXT: .LBB3_5: +; NOCOMPACTUNWIND-NEXT: .LBB3_6: ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 16 ; NOCOMPACTUNWIND-NEXT: xorl %eax, %eax ; NOCOMPACTUNWIND-NEXT: # kill: def $al killed $al killed $eax ; NOCOMPACTUNWIND-NEXT: popq %rcx ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 8 ; NOCOMPACTUNWIND-NEXT: retq -; NOCOMPACTUNWIND-NEXT: .LBB3_6: +; NOCOMPACTUNWIND-NEXT: .LBB3_7: ; NOCOMPACTUNWIND-NEXT: movl $8, %r10d ; NOCOMPACTUNWIND-NEXT: movl $0, %r11d ; NOCOMPACTUNWIND-NEXT: callq __morestack diff --git a/llvm/test/MC/AArch64/data-directive-specifier.s b/llvm/test/MC/AArch64/data-directive-specifier.s index 2cb7eb3..2d1ec4f 100644 --- a/llvm/test/MC/AArch64/data-directive-specifier.s +++ b/llvm/test/MC/AArch64/data-directive-specifier.s @@ -12,6 +12,7 @@ l: # CHECK-NEXT: 0x8 R_AARCH64_PLT32 extern 0x4 # CHECK-NEXT: 0xC R_AARCH64_PLT32 g 0x8 # CHECK-NEXT: 0x10 R_AARCH64_PLT32 g 0x18 +# CHECK-NEXT: 0x14 R_AARCH64_FUNCINIT64 .text 0x0 # CHECK-NEXT: } .data .word l@plt - . @@ -21,6 +22,8 @@ l: .word g@plt - . + 8 .word g@plt - .data + 8 +.quad l@funcinit + # CHECK: Section ({{.*}}) .rela.data1 { # CHECK-NEXT: 0x0 R_AARCH64_GOTPCREL32 data1 0x0 # CHECK-NEXT: 0x4 R_AARCH64_GOTPCREL32 extern 0x4 diff --git a/llvm/test/Transforms/DFAJumpThreading/dfa-constant-propagation.ll b/llvm/test/Transforms/DFAJumpThreading/dfa-constant-propagation.ll index fdab67a..afc98ce 100644 --- a/llvm/test/Transforms/DFAJumpThreading/dfa-constant-propagation.ll +++ b/llvm/test/Transforms/DFAJumpThreading/dfa-constant-propagation.ll @@ -1,5 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py -; RUN: opt -S -passes=dfa-jump-threading,sccp,simplifycfg %s | FileCheck %s +; RUN: opt -S -passes=dfa-jump-threading,sccp,simplifycfg -verify-dom-info=1 %s | FileCheck %s ; This test checks that a constant propagation is applied for a basic loop. ; Related to bug 44679. diff --git a/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-analysis.ll b/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-analysis.ll index f45798b..5076517 100644 --- a/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-analysis.ll +++ b/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-analysis.ll @@ -1,6 +1,6 @@ ; REQUIRES: asserts -; RUN: opt -S -passes=dfa-jump-threading -debug-only=dfa-jump-threading -disable-output %s 2>&1 | FileCheck %s -; RUN: opt -S -passes=dfa-jump-threading -print-prof-data %s -o - | FileCheck %s --check-prefix=PROFILE +; RUN: opt -S -passes=dfa-jump-threading -verify-dom-info=1 -debug-only=dfa-jump-threading -disable-output %s 2>&1 | FileCheck %s +; RUN: opt -S -passes=dfa-jump-threading -verify-dom-info=1 -print-prof-data %s -o - | FileCheck %s --check-prefix=PROFILE ; This test checks that the analysis identifies all threadable paths in a ; simple CFG. A threadable path includes a list of basic blocks, the exit diff --git a/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-transform.ll b/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-transform.ll index 092c854..426b51e 100644 --- a/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-transform.ll +++ b/llvm/test/Transforms/DFAJumpThreading/dfa-jump-threading-transform.ll @@ -1,5 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals -; RUN: opt -S -passes=dfa-jump-threading %s | FileCheck %s +; RUN: opt -S -passes=dfa-jump-threading -verify-dom-info=1 %s | FileCheck %s ; These tests check that the DFA jump threading transformation is applied ; properly to two CFGs. It checks that blocks are cloned, branches are updated, @@ -445,9 +445,67 @@ bb2: ; preds = %select.unfold unreachable } + +define i16 @DTU_update_crash() { +; CHECK-LABEL: @DTU_update_crash( +; CHECK-NEXT: entry: +; CHECK-NEXT: br label [[FOR_BODY_SELECTBLOCK:%.*]] +; CHECK: for.body.selectblock: +; CHECK-NEXT: br i1 false, label [[SWITCHBLOCK_JT0:%.*]], label [[SEL_SI_UNFOLD_FALSE_JT0:%.*]] +; CHECK: sel.si.unfold.false: +; CHECK-NEXT: br label [[SWITCHBLOCK:%.*]] +; CHECK: sel.si.unfold.false.jt0: +; CHECK-NEXT: [[DOTSI_UNFOLD_PHI_JT0:%.*]] = phi i32 [ 0, [[FOR_BODY_SELECTBLOCK]] ] +; CHECK-NEXT: br label [[SWITCHBLOCK_JT0]] +; CHECK: switchblock: +; CHECK-NEXT: [[SWITCHBLOCK_PHI:%.*]] = phi i32 [ poison, [[SEL_SI_UNFOLD_FALSE:%.*]] ] +; CHECK-NEXT: [[P_24_ADDR_3:%.*]] = phi i32 [ 0, [[SEL_SI_UNFOLD_FALSE]] ] +; CHECK-NEXT: switch i32 [[SWITCHBLOCK_PHI]], label [[CLEANUP:%.*]] [ +; CHECK-NEXT: i32 0, label [[FOR_INC:%.*]] +; CHECK-NEXT: i32 1, label [[CLEANUP]] +; CHECK-NEXT: i32 5, label [[FOR_BODY_SELECTBLOCK]] +; CHECK-NEXT: ] +; CHECK: switchblock.jt0: +; CHECK-NEXT: [[SWITCHBLOCK_PHI_JT0:%.*]] = phi i32 [ 0, [[FOR_BODY_SELECTBLOCK]] ], [ [[DOTSI_UNFOLD_PHI_JT0]], [[SEL_SI_UNFOLD_FALSE_JT0]] ] +; CHECK-NEXT: [[P_24_ADDR_3_JT0:%.*]] = phi i32 [ 0, [[FOR_BODY_SELECTBLOCK]] ], [ 0, [[SEL_SI_UNFOLD_FALSE_JT0]] ] +; CHECK-NEXT: br label [[FOR_INC]] +; CHECK: for.inc: +; CHECK-NEXT: br i1 false, label [[FOR_BODY_SELECTBLOCK]], label [[CLEANUP]] +; CHECK: cleanup: +; CHECK-NEXT: call void (...) @llvm.fake.use(i32 [[P_24_ADDR_3_JT0]]) +; CHECK-NEXT: ret i16 0 +; +entry: + br label %for.body.selectblock + +for.body.selectblock: ; preds = %for.inc, %switchblock, %entry + %sel = select i1 false, i32 0, i32 0 + br label %switchblock + +switchblock: ; preds = %for.body.selectblock + %switchblock.phi = phi i32 [ %sel, %for.body.selectblock ] + %p_24.addr.3 = phi i32 [ 0, %for.body.selectblock ] + switch i32 %switchblock.phi, label %cleanup [ + i32 0, label %for.inc + i32 1, label %cleanup + i32 5, label %for.body.selectblock + ] + +for.inc: ; preds = %switchblock + br i1 false, label %for.body.selectblock, label %cleanup + +cleanup: ; preds = %for.inc, %switchblock, %switchblock + call void (...) @llvm.fake.use(i32 %p_24.addr.3) + ret i16 0 +} + +declare void @llvm.fake.use(...) + !0 = !{!"function_entry_count", i32 10} !1 = !{!"branch_weights", i32 3, i32 5} ;. +; CHECK: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind willreturn memory(inaccessiblemem: readwrite) } +;. ; CHECK: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} ; CHECK: [[PROF1]] = !{!"branch_weights", i32 3, i32 5} ;. diff --git a/llvm/test/Transforms/DFAJumpThreading/dfa-unfold-select.ll b/llvm/test/Transforms/DFAJumpThreading/dfa-unfold-select.ll index de38752..95d3ffa 100644 --- a/llvm/test/Transforms/DFAJumpThreading/dfa-unfold-select.ll +++ b/llvm/test/Transforms/DFAJumpThreading/dfa-unfold-select.ll @@ -1,5 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py -; RUN: opt -S -passes=dfa-jump-threading -dfa-early-exit-heuristic=false %s | FileCheck %s +; RUN: opt -S -passes=dfa-jump-threading -dfa-early-exit-heuristic=false -verify-dom-info=1 %s | FileCheck %s ; These tests check if selects are unfolded properly for jump threading ; opportunities. There are three different patterns to consider: diff --git a/llvm/test/Transforms/DFAJumpThreading/equivalent-states.ll b/llvm/test/Transforms/DFAJumpThreading/equivalent-states.ll index 4555dfb..71a469d 100644 --- a/llvm/test/Transforms/DFAJumpThreading/equivalent-states.ll +++ b/llvm/test/Transforms/DFAJumpThreading/equivalent-states.ll @@ -1,5 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 3 -; RUN: opt -S -passes=dfa-jump-threading %s | FileCheck %s +; RUN: opt -S -passes=dfa-jump-threading -verify-dom-info=1 %s | FileCheck %s declare void @do_something() declare void @user(i32) diff --git a/llvm/test/Transforms/DFAJumpThreading/single_succ_switch.ll b/llvm/test/Transforms/DFAJumpThreading/single_succ_switch.ll index 00500a7..cc117e7 100644 --- a/llvm/test/Transforms/DFAJumpThreading/single_succ_switch.ll +++ b/llvm/test/Transforms/DFAJumpThreading/single_succ_switch.ll @@ -1,5 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 3 -; RUN: opt -S -passes=dfa-jump-threading %s | FileCheck %s +; RUN: opt -S -passes=dfa-jump-threading -verify-dom-info=1 %s | FileCheck %s define void @pr60254() { ; CHECK-LABEL: define void @pr60254() { diff --git a/llvm/test/Transforms/IndVarSimplify/pointer-loop-guards.ll b/llvm/test/Transforms/IndVarSimplify/pointer-loop-guards.ll index 89b132e..9371fe2 100644 --- a/llvm/test/Transforms/IndVarSimplify/pointer-loop-guards.ll +++ b/llvm/test/Transforms/IndVarSimplify/pointer-loop-guards.ll @@ -18,7 +18,7 @@ define i64 @test_ptr_compare_guard(ptr %start, ptr %end) { ; CHECK-NEXT: br i1 [[C_1]], label %[[LOOP_LATCH]], label %[[EXIT_LOOPEXIT:.*]] ; CHECK: [[LOOP_LATCH]]: ; CHECK-NEXT: [[PTR_IV_NEXT]] = getelementptr i8, ptr [[PTR_IV]], i64 1 -; CHECK-NEXT: [[I64_IV_NEXT]] = add i64 [[I64_IV]], 1 +; CHECK-NEXT: [[I64_IV_NEXT]] = add nuw i64 [[I64_IV]], 1 ; CHECK-NEXT: [[C_2:%.*]] = icmp eq ptr [[PTR_IV_NEXT]], [[END]] ; CHECK-NEXT: br i1 [[C_2]], label %[[EXIT_LOOPEXIT]], label %[[LOOP_HEADER]] ; CHECK: [[EXIT_LOOPEXIT]]: diff --git a/llvm/test/Transforms/InstCombine/select-safe-impliedcond-transforms.ll b/llvm/test/Transforms/InstCombine/select-safe-impliedcond-transforms.ll index ba34930..bc988a9 100644 --- a/llvm/test/Transforms/InstCombine/select-safe-impliedcond-transforms.ll +++ b/llvm/test/Transforms/InstCombine/select-safe-impliedcond-transforms.ll @@ -1,4 +1,4 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals ; RUN: opt < %s -passes=instcombine -S | FileCheck %s define i1 @a_true_implies_b_true(i8 %z, i1 %X, i1 %Y) { @@ -34,15 +34,15 @@ define <2 x i1> @a_true_implies_b_true_vec(i8 %z0, <2 x i1> %X, <2 x i1> %Y) { ret <2 x i1> %res } -define i1 @a_true_implies_b_true2(i8 %z, i1 %X, i1 %Y) { +define i1 @a_true_implies_b_true2(i8 %z, i1 %X, i1 %Y) !prof !0 { ; CHECK-LABEL: @a_true_implies_b_true2( ; CHECK-NEXT: [[A:%.*]] = icmp ugt i8 [[Z:%.*]], 20 -; CHECK-NEXT: [[RES:%.*]] = select i1 [[A]], i1 [[X:%.*]], i1 false +; CHECK-NEXT: [[RES:%.*]] = select i1 [[A]], i1 [[X:%.*]], i1 false, !prof [[PROF1:![0-9]+]] ; CHECK-NEXT: ret i1 [[RES]] ; %a = icmp ugt i8 %z, 20 %b = icmp ugt i8 %z, 10 - %sel = select i1 %b, i1 %X, i1 %Y + %sel = select i1 %b, i1 %X, i1 %Y, !prof !1 %res = and i1 %a, %sel ret i1 %res } @@ -258,3 +258,10 @@ define i1 @neg_icmp_eq_implies_trunc(i8 %x, i1 %c) { %sel2 = select i1 %cmp, i1 true, i1 %sel1 ret i1 %sel2 } + +!0 = !{!"function_entry_count", i64 1000} +!1 = !{!"branch_weights", i32 2, i32 3} +;. +; CHECK: [[META0:![0-9]+]] = !{!"function_entry_count", i64 1000} +; CHECK: [[PROF1]] = !{!"branch_weights", i32 2, i32 3} +;. diff --git a/llvm/test/Transforms/InstSimplify/ptrmask.ll b/llvm/test/Transforms/InstSimplify/ptrmask.ll index 5e7c636..a3483af 100644 --- a/llvm/test/Transforms/InstSimplify/ptrmask.ll +++ b/llvm/test/Transforms/InstSimplify/ptrmask.ll @@ -158,6 +158,26 @@ define ptr addrspace(1) @ptrmask_simplify_ptrmask_i32(ptr addrspace(1) %p) { ret ptr addrspace(1) %r } +define ptr @ptrmask_simplify_ptrtoaddr(ptr %p) { +; CHECK-LABEL: define ptr @ptrmask_simplify_ptrtoaddr +; CHECK-SAME: (ptr [[P:%.*]]) { +; CHECK-NEXT: ret ptr [[P]] +; + %m = ptrtoaddr ptr %p to i64 + %r = call ptr @llvm.ptrmask.p0.i64(ptr %p, i64 %m) + ret ptr %r +} + +define ptr addrspace(1) @ptrmask_simplify_ptrtoaddr_i32(ptr addrspace(1) %p) { +; CHECK-LABEL: define ptr addrspace(1) @ptrmask_simplify_ptrtoaddr_i32 +; CHECK-SAME: (ptr addrspace(1) [[P:%.*]]) { +; CHECK-NEXT: ret ptr addrspace(1) [[P]] +; + %m = ptrtoaddr ptr addrspace(1) %p to i32 + %r = call ptr addrspace(1) @llvm.ptrmask.p1.i32(ptr addrspace(1) %p, i32 %m) + ret ptr addrspace(1) %r +} + define ptr @ptrmask_simplify_aligned_unused(ptr align 64 %p) { ; CHECK-LABEL: define ptr @ptrmask_simplify_aligned_unused ; CHECK-SAME: (ptr align 64 [[P:%.*]]) { diff --git a/llvm/test/Transforms/LoopUnroll/scevunroll.ll b/llvm/test/Transforms/LoopUnroll/scevunroll.ll index fa55eab..bc63f79 100644 --- a/llvm/test/Transforms/LoopUnroll/scevunroll.ll +++ b/llvm/test/Transforms/LoopUnroll/scevunroll.ll @@ -465,8 +465,7 @@ define void @peel_int_eq_condition(i32 %start) { ; CHECK-NEXT: br label [[LOOP:%.*]] ; CHECK: loop: ; CHECK-NEXT: [[IV:%.*]] = phi i32 [ [[IV_NEXT_PEEL]], [[ENTRY_PEEL_NEWPH]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ] -; CHECK-NEXT: [[C_0:%.*]] = icmp eq i32 [[IV]], [[START]] -; CHECK-NEXT: br i1 [[C_0]], label [[IF_THEN:%.*]], label [[LOOP_LATCH]] +; CHECK-NEXT: br i1 false, label [[IF_THEN:%.*]], label [[LOOP_LATCH]] ; CHECK: if.then: ; CHECK-NEXT: call void @fn(i32 [[IV]]) ; CHECK-NEXT: br label [[LOOP_LATCH]] diff --git a/llvm/test/Transforms/SROA/phi-and-select.ll b/llvm/test/Transforms/SROA/phi-and-select.ll index 616617b..5d5a610 100644 --- a/llvm/test/Transforms/SROA/phi-and-select.ll +++ b/llvm/test/Transforms/SROA/phi-and-select.ll @@ -1,4 +1,4 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals ; RUN: opt < %s -passes='sroa<preserve-cfg>' -S | FileCheck %s --check-prefixes=CHECK,CHECK-PRESERVE-CFG ; RUN: opt < %s -passes='sroa<modify-cfg>' -S | FileCheck %s --check-prefixes=CHECK,CHECK-MODIFY-CFG target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-n8:16:32:64" @@ -36,11 +36,11 @@ exit: ret i32 %result } -define i32 @test2() { +define i32 @test2() !prof !0 { ; CHECK-LABEL: @test2( ; CHECK-NEXT: entry: ; CHECK-NEXT: [[COND:%.*]] = icmp sle i32 0, 1 -; CHECK-NEXT: [[RESULT_SROA_SPECULATED:%.*]] = select i1 [[COND]], i32 1, i32 0 +; CHECK-NEXT: [[RESULT_SROA_SPECULATED:%.*]] = select i1 [[COND]], i32 1, i32 0, !prof [[PROF1:![0-9]+]] ; CHECK-NEXT: ret i32 [[RESULT_SROA_SPECULATED]] ; entry: @@ -53,7 +53,7 @@ entry: %v1 = load i32, ptr %a1 %cond = icmp sle i32 %v0, %v1 - %select = select i1 %cond, ptr %a1, ptr %a + %select = select i1 %cond, ptr %a1, ptr %a, !prof !1 %result = load i32, ptr %select ret i32 %result @@ -870,3 +870,17 @@ define i8 @volatile_select(ptr %p, i1 %b) { %v2 = load i8, ptr %px ret i8 %v2 } + +!0 = !{!"function_entry_count", i32 10} +!1 = !{!"branch_weights", i32 3, i32 5} +;. +; CHECK-PRESERVE-CFG: attributes #[[ATTR0:[0-9]+]] = { sanitize_address } +;. +; CHECK-MODIFY-CFG: attributes #[[ATTR0:[0-9]+]] = { sanitize_address } +;. +; CHECK-PRESERVE-CFG: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} +; CHECK-PRESERVE-CFG: [[PROF1]] = !{!"branch_weights", i32 3, i32 5} +;. +; CHECK-MODIFY-CFG: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} +; CHECK-MODIFY-CFG: [[PROF1]] = !{!"branch_weights", i32 3, i32 5} +;. diff --git a/llvm/test/Transforms/SROA/phi-gep.ll b/llvm/test/Transforms/SROA/phi-gep.ll index 776624c..45c3bbd 100644 --- a/llvm/test/Transforms/SROA/phi-gep.ll +++ b/llvm/test/Transforms/SROA/phi-gep.ll @@ -1,9 +1,12 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals ; RUN: opt -S -passes='sroa<preserve-cfg>' < %s | FileCheck %s --check-prefixes=CHECK,CHECK-PRESERVE-CFG ; RUN: opt -S -passes='sroa<modify-cfg>' < %s | FileCheck %s --check-prefixes=CHECK,CHECK-MODIFY-CFG %pair = type { i32, i32 } +;. +; CHECK: @g = global %pair zeroinitializer, align 4 +;. define i32 @test_sroa_phi_gep(i1 %cond) { ; CHECK-LABEL: @test_sroa_phi_gep( ; CHECK-NEXT: entry: @@ -334,18 +337,18 @@ exit: unreachable } -define void @test_sroa_gep_phi_select_same_block(i1 %c1, i1 %c2, ptr %ptr) { +define void @test_sroa_gep_phi_select_same_block(i1 %c1, i1 %c2, ptr %ptr) !prof !0 { ; CHECK-LABEL: @test_sroa_gep_phi_select_same_block( ; CHECK-NEXT: entry: ; CHECK-NEXT: [[ALLOCA:%.*]] = alloca [[PAIR:%.*]], align 8 ; CHECK-NEXT: br label [[WHILE_BODY:%.*]] ; CHECK: while.body: ; CHECK-NEXT: [[PHI:%.*]] = phi ptr [ [[ALLOCA]], [[ENTRY:%.*]] ], [ [[SELECT:%.*]], [[WHILE_BODY]] ] -; CHECK-NEXT: [[SELECT]] = select i1 [[C1:%.*]], ptr [[PHI]], ptr [[PTR:%.*]] +; CHECK-NEXT: [[SELECT]] = select i1 [[C1:%.*]], ptr [[PHI]], ptr [[PTR:%.*]], !prof [[PROF1:![0-9]+]] ; CHECK-NEXT: [[PHI_SROA_GEP:%.*]] = getelementptr inbounds [[PAIR]], ptr [[PHI]], i64 1 ; CHECK-NEXT: [[PTR_SROA_GEP:%.*]] = getelementptr inbounds [[PAIR]], ptr [[PTR]], i64 1 -; CHECK-NEXT: [[SELECT_SROA_SEL:%.*]] = select i1 [[C1]], ptr [[PHI_SROA_GEP]], ptr [[PTR_SROA_GEP]] -; CHECK-NEXT: br i1 [[C2:%.*]], label [[EXIT:%.*]], label [[WHILE_BODY]] +; CHECK-NEXT: [[SELECT_SROA_SEL:%.*]] = select i1 [[C1]], ptr [[PHI_SROA_GEP]], ptr [[PTR_SROA_GEP]], !prof [[PROF1]] +; CHECK-NEXT: br i1 [[C2:%.*]], label [[EXIT:%.*]], label [[WHILE_BODY]], !prof [[PROF2:![0-9]+]] ; CHECK: exit: ; CHECK-NEXT: ret void ; @@ -355,9 +358,9 @@ entry: while.body: %phi = phi ptr [ %alloca, %entry ], [ %select, %while.body ] - %select = select i1 %c1, ptr %phi, ptr %ptr + %select = select i1 %c1, ptr %phi, ptr %ptr, !prof !1 %gep = getelementptr inbounds %pair, ptr %select, i64 1 - br i1 %c2, label %exit, label %while.body + br i1 %c2, label %exit, label %while.body, !prof !2 exit: ret void @@ -747,6 +750,18 @@ declare ptr @foo() declare i32 @__gxx_personality_v0(...) declare void @llvm.memcpy.p0.p0.i64(ptr noalias nocapture writeonly, ptr noalias nocapture readonly, i64, i1 immarg) + +!0 = !{!"function_entry_count", i32 10} +!1 = !{!"branch_weights", i32 3, i32 5} +!2 = !{!"branch_weights", i32 7, i32 11} + +;. +; CHECK: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) } +;. +; CHECK: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} +; CHECK: [[PROF1]] = !{!"branch_weights", i32 3, i32 5} +; CHECK: [[PROF2]] = !{!"branch_weights", i32 7, i32 11} +;. ;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line: ; CHECK-MODIFY-CFG: {{.*}} ; CHECK-PRESERVE-CFG: {{.*}} diff --git a/llvm/test/Transforms/SROA/select-gep.ll b/llvm/test/Transforms/SROA/select-gep.ll index b48b0f7..a701d78 100644 --- a/llvm/test/Transforms/SROA/select-gep.ll +++ b/llvm/test/Transforms/SROA/select-gep.ll @@ -1,4 +1,4 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals ; RUN: opt -S -passes='sroa<preserve-cfg>' < %s | FileCheck %s --check-prefixes=CHECK,CHECK-PRESERVE-CFG ; RUN: opt -S -passes='sroa<modify-cfg>' < %s | FileCheck %s --check-prefixes=CHECK,CHECK-MODIFY-CFG @@ -203,10 +203,10 @@ define i32 @test_select_idx_mem2reg(i1 %c) { ; Test gep with a select-like zext index unfolding on an alloca that is ; splittable and promotable. -define i64 @test_select_like_zext_idx_mem2reg(i1 %c) { +define i64 @test_select_like_zext_idx_mem2reg(i1 %c) !prof !0 { ; CHECK-LABEL: @test_select_like_zext_idx_mem2reg( ; CHECK-NEXT: [[IDX:%.*]] = zext i1 [[C:%.*]] to i64 -; CHECK-NEXT: [[RES:%.*]] = select i1 [[C]], i64 2, i64 1 +; CHECK-NEXT: [[RES:%.*]] = select i1 [[C]], i64 2, i64 1, !prof [[PROF1:![0-9]+]] ; CHECK-NEXT: ret i64 [[RES]] ; %alloca = alloca [2 x i64], align 8 @@ -352,3 +352,16 @@ define i32 @test_select_idx_not_constant3(i1 %c, ptr %p, i64 %arg) { %res = load i32, ptr %gep, align 4 ret i32 %res } + +!0 = !{!"function_entry_count", i32 10} +;. +; CHECK-PRESERVE-CFG: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) } +;. +; CHECK-MODIFY-CFG: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) } +;. +; CHECK-PRESERVE-CFG: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} +; CHECK-PRESERVE-CFG: [[PROF1]] = !{!"unknown", !"sroa"} +;. +; CHECK-MODIFY-CFG: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} +; CHECK-MODIFY-CFG: [[PROF1]] = !{!"unknown", !"sroa"} +;. diff --git a/llvm/test/tools/llvm-objcopy/DXContainer/dump-section-errors.yaml b/llvm/test/tools/llvm-objcopy/DXContainer/dump-section-errors.yaml new file mode 100644 index 0000000..e748eecf --- /dev/null +++ b/llvm/test/tools/llvm-objcopy/DXContainer/dump-section-errors.yaml @@ -0,0 +1,27 @@ +# RUN: yaml2obj %s -o %t.dxbc +# RUN: not llvm-objcopy --dump-section=FKE0=%t.fek0 %t.dxbc 2>&1 | FileCheck %s --check-prefix=CHECK-ZEROSIZE -DFILE=%t.fek0 +# RUN: not llvm-objcopy --dump-section=FKE3=%t.fek1 %t.dxbc 2>&1 | FileCheck %s --check-prefix=CHECK-MISSING -DFILE=%t.fek1 +# RUN: not llvm-objcopy --dump-section=FKE2=%t/does_not_exist/.fek2 %t.dxbc 2>&1 | FileCheck %s --check-prefix=CHECK-BAD-PATH -DFILE=%t/does_not_exist/.fek2 -DMSG=%errc_ENOENT + +# CHECK-ZEROSIZE: error: '[[FILE]]': part 'FKE0' is empty +# CHECK-MISSING: error: '[[FILE]]': part 'FKE3' not found +# CHECK-BAD-PATH: error: '[[FILE]]': [[MSG]] + +--- !dxcontainer +Header: + Hash: [ 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ] + Version: + Major: 1 + Minor: 0 + FileSize: 108 + PartCount: 3 + PartOffsets: [ 60, 68, 76 ] +Parts: + - Name: FKE0 + Size: 0 + - Name: FKE1 + Size: 0 + - Name: FKE2 + Size: 8 +... diff --git a/llvm/test/tools/llvm-objcopy/DXContainer/dump-section.yaml b/llvm/test/tools/llvm-objcopy/DXContainer/dump-section.yaml new file mode 100644 index 0000000..7d80a2c --- /dev/null +++ b/llvm/test/tools/llvm-objcopy/DXContainer/dump-section.yaml @@ -0,0 +1,278 @@ +# RUN: yaml2obj %s -o %t.dxbc +# RUN: llvm-objcopy --dump-section=DXIL=%t.bc %t.dxbc +# RUN: llvm-dis %t.bc -o - | FileCheck %s --check-prefix=BITCODE +# RUN: wc -c %t.bc | FileCheck %s --check-prefix=DXIL-SIZE + +## Verify that when dumping the DXIL part we get a valid bitcode file. +# BITCODE: define void @main() +## Verify the size of the bitcode data. +# DXIL-SIZE: 1708 + +## Dump the PSV0 part and verify its size. +# RUN: llvm-objcopy --dump-section=PSV0=%t.psv0 %t.dxbc +# RUN: wc -c %t.psv0 | FileCheck %s --check-prefix=PSV0-SIZE +# RUN: od -v -Ax -t x1 %t.psv0 | FileCheck %s --check-prefix=PSV0-CONTENTS +# PSV0-SIZE: 76 + +# For a compute shader the structure size is encoded followed by a bunch of 00'd +# bytes until you get to the unused wave size min and max (0xffff), followed by +# the shader stage (5 for compute). +# TODO: Update this test to use objdump or obj2yaml once we support +# --add-section in objcopy. See issue: +# https://github.com/llvm/llvm-project/issues/162159. +# PSV0-CONTENTS: 0000 34 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 +# PSV0-CONTENTS: 0010 00 00 00 00 00 00 00 00 ff ff ff ff 05 00 00 00 + +--- !dxcontainer +Header: + Hash: [ 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ] + Version: + Major: 1 + Minor: 0 + FileSize: 1872 + PartCount: 2 + PartOffsets: [ 40, 1780 ] +Parts: + - Name: DXIL + Size: 1732 + Program: + MajorVersion: 6 + MinorVersion: 0 + ShaderKind: 5 + Size: 433 + DXILMajorVersion: 1 + DXILMinorVersion: 0 + DXILSize: 1708 + DXIL: [ 0x42, 0x43, 0xC0, 0xDE, 0x21, 0xC, 0x0, 0x0, 0xA8, + 0x1, 0x0, 0x0, 0xB, 0x82, 0x20, 0x0, 0x2, 0x0, + 0x0, 0x0, 0x13, 0x0, 0x0, 0x0, 0x7, 0x81, 0x23, + 0x91, 0x41, 0xC8, 0x4, 0x49, 0x6, 0x10, 0x32, + 0x39, 0x92, 0x1, 0x84, 0xC, 0x25, 0x5, 0x8, 0x19, + 0x1E, 0x4, 0x8B, 0x62, 0x80, 0x10, 0x45, 0x2, + 0x42, 0x92, 0xB, 0x42, 0x84, 0x10, 0x32, 0x14, + 0x38, 0x8, 0x18, 0x4B, 0xA, 0x32, 0x42, 0x88, + 0x48, 0x90, 0x14, 0x20, 0x43, 0x46, 0x88, 0xA5, + 0x0, 0x19, 0x32, 0x42, 0xE4, 0x48, 0xE, 0x90, + 0x11, 0x22, 0xC4, 0x50, 0x41, 0x51, 0x81, 0x8C, + 0xE1, 0x83, 0xE5, 0x8A, 0x4, 0x21, 0x46, 0x6, + 0x51, 0x18, 0x0, 0x0, 0x4, 0x0, 0x0, 0x0, 0x1B, + 0x90, 0xE0, 0xFF, 0xFF, 0xFF, 0xFF, 0x7, 0xC0, + 0x1, 0x24, 0x80, 0x2, 0x0, 0x0, 0x0, 0x49, 0x18, + 0x0, 0x0, 0x1, 0x0, 0x0, 0x0, 0x13, 0x82, 0x0, + 0x0, 0x89, 0x20, 0x0, 0x0, 0x11, 0x0, 0x0, 0x0, + 0x32, 0x22, 0x8, 0x9, 0x20, 0x64, 0x85, 0x4, 0x13, + 0x22, 0xA4, 0x84, 0x4, 0x13, 0x22, 0xE3, 0x84, + 0xA1, 0x90, 0x14, 0x12, 0x4C, 0x88, 0x8C, 0xB, + 0x84, 0x84, 0x4C, 0x10, 0x20, 0x73, 0x4, 0x8, + 0xC1, 0x65, 0xC3, 0x85, 0x2C, 0xE8, 0x3, 0x40, + 0x14, 0x91, 0x4E, 0xD1, 0x4A, 0x48, 0x44, 0x54, + 0x11, 0xC3, 0x9, 0x30, 0xC4, 0x18, 0x1, 0x30, + 0x2, 0x50, 0x82, 0x21, 0x1A, 0x8, 0x98, 0x23, + 0x0, 0x3, 0x0, 0x13, 0x14, 0x72, 0xC0, 0x87, 0x74, + 0x60, 0x87, 0x36, 0x68, 0x87, 0x79, 0x68, 0x3, + 0x72, 0xC0, 0x87, 0xD, 0xAE, 0x50, 0xE, 0x6D, + 0xD0, 0xE, 0x7A, 0x50, 0xE, 0x6D, 0x0, 0xF, 0x7A, + 0x30, 0x7, 0x72, 0xA0, 0x7, 0x73, 0x20, 0x7, 0x6D, + 0x90, 0xE, 0x71, 0xA0, 0x7, 0x73, 0x20, 0x7, 0x6D, + 0x90, 0xE, 0x78, 0xA0, 0x7, 0x78, 0xD0, 0x6, 0xE9, + 0x10, 0x7, 0x76, 0xA0, 0x7, 0x71, 0x60, 0x7, 0x6D, + 0x90, 0xE, 0x73, 0x20, 0x7, 0x7A, 0x30, 0x7, 0x72, + 0xD0, 0x6, 0xE9, 0x60, 0x7, 0x74, 0xA0, 0x7, 0x76, + 0x40, 0x7, 0x6D, 0x60, 0xE, 0x71, 0x60, 0x7, 0x7A, + 0x10, 0x7, 0x76, 0xD0, 0x6, 0xE6, 0x30, 0x7, 0x72, + 0xA0, 0x7, 0x73, 0x20, 0x7, 0x6D, 0x60, 0xE, 0x76, + 0x40, 0x7, 0x7A, 0x60, 0x7, 0x74, 0xD0, 0x6, 0xEE, + 0x80, 0x7, 0x7A, 0x10, 0x7, 0x76, 0xA0, 0x7, 0x73, + 0x20, 0x7, 0x7A, 0x60, 0x7, 0x74, 0x30, 0xE4, + 0x21, 0x0, 0x0, 0x8, 0x0, 0x0, 0x0, 0x2, 0x0, + 0x0, 0x0, 0x20, 0xB, 0x4, 0x7, 0x0, 0x0, 0x0, + 0x32, 0x1E, 0x98, 0xC, 0x19, 0x11, 0x4C, 0x90, + 0x8C, 0x9, 0x26, 0x47, 0xC6, 0x4, 0x43, 0xBA, + 0x12, 0x28, 0x88, 0x62, 0x18, 0x1, 0x28, 0x84, + 0x22, 0x0, 0x0, 0x0, 0x79, 0x18, 0x0, 0x0, 0xE5, + 0x0, 0x0, 0x0, 0x33, 0x8, 0x80, 0x1C, 0xC4, 0xE1, + 0x1C, 0x66, 0x14, 0x1, 0x3D, 0x88, 0x43, 0x38, + 0x84, 0xC3, 0x8C, 0x42, 0x80, 0x7, 0x79, 0x78, + 0x7, 0x73, 0x98, 0x71, 0xC, 0xE6, 0x0, 0xF, 0xED, + 0x10, 0xE, 0xF4, 0x80, 0xE, 0x33, 0xC, 0x42, 0x1E, + 0xC2, 0xC1, 0x1D, 0xCE, 0xA1, 0x1C, 0x66, 0x30, + 0x5, 0x3D, 0x88, 0x43, 0x38, 0x84, 0x83, 0x1B, + 0xCC, 0x3, 0x3D, 0xC8, 0x43, 0x3D, 0x8C, 0x3, + 0x3D, 0xCC, 0x78, 0x8C, 0x74, 0x70, 0x7, 0x7B, + 0x8, 0x7, 0x79, 0x48, 0x87, 0x70, 0x70, 0x7, 0x7A, + 0x70, 0x3, 0x76, 0x78, 0x87, 0x70, 0x20, 0x87, + 0x19, 0xCC, 0x11, 0xE, 0xEC, 0x90, 0xE, 0xE1, + 0x30, 0xF, 0x6E, 0x30, 0xF, 0xE3, 0xF0, 0xE, 0xF0, + 0x50, 0xE, 0x33, 0x10, 0xC4, 0x1D, 0xDE, 0x21, + 0x1C, 0xD8, 0x21, 0x1D, 0xC2, 0x61, 0x1E, 0x66, + 0x30, 0x89, 0x3B, 0xBC, 0x83, 0x3B, 0xD0, 0x43, + 0x39, 0xB4, 0x3, 0x3C, 0xBC, 0x83, 0x3C, 0x84, + 0x3, 0x3B, 0xCC, 0xF0, 0x14, 0x76, 0x60, 0x7, + 0x7B, 0x68, 0x7, 0x37, 0x68, 0x87, 0x72, 0x68, + 0x7, 0x37, 0x80, 0x87, 0x70, 0x90, 0x87, 0x70, + 0x60, 0x7, 0x76, 0x28, 0x7, 0x76, 0xF8, 0x5, 0x76, + 0x78, 0x87, 0x77, 0x80, 0x87, 0x5F, 0x8, 0x87, + 0x71, 0x18, 0x87, 0x72, 0x98, 0x87, 0x79, 0x98, + 0x81, 0x2C, 0xEE, 0xF0, 0xE, 0xEE, 0xE0, 0xE, + 0xF5, 0xC0, 0xE, 0xEC, 0x30, 0x3, 0x62, 0xC8, + 0xA1, 0x1C, 0xE4, 0xA1, 0x1C, 0xCC, 0xA1, 0x1C, + 0xE4, 0xA1, 0x1C, 0xDC, 0x61, 0x1C, 0xCA, 0x21, + 0x1C, 0xC4, 0x81, 0x1D, 0xCA, 0x61, 0x6, 0xD6, + 0x90, 0x43, 0x39, 0xC8, 0x43, 0x39, 0x98, 0x43, + 0x39, 0xC8, 0x43, 0x39, 0xB8, 0xC3, 0x38, 0x94, + 0x43, 0x38, 0x88, 0x3, 0x3B, 0x94, 0xC3, 0x2F, + 0xBC, 0x83, 0x3C, 0xFC, 0x82, 0x3B, 0xD4, 0x3, + 0x3B, 0xB0, 0xC3, 0xC, 0xC7, 0x69, 0x87, 0x70, + 0x58, 0x87, 0x72, 0x70, 0x83, 0x74, 0x68, 0x7, + 0x78, 0x60, 0x87, 0x74, 0x18, 0x87, 0x74, 0xA0, + 0x87, 0x19, 0xCE, 0x53, 0xF, 0xEE, 0x0, 0xF, 0xF2, + 0x50, 0xE, 0xE4, 0x90, 0xE, 0xE3, 0x40, 0xF, 0xE1, + 0x20, 0xE, 0xEC, 0x50, 0xE, 0x33, 0x20, 0x28, + 0x1D, 0xDC, 0xC1, 0x1E, 0xC2, 0x41, 0x1E, 0xD2, + 0x21, 0x1C, 0xDC, 0x81, 0x1E, 0xDC, 0xE0, 0x1C, + 0xE4, 0xE1, 0x1D, 0xEA, 0x1, 0x1E, 0x66, 0x18, + 0x51, 0x38, 0xB0, 0x43, 0x3A, 0x9C, 0x83, 0x3B, + 0xCC, 0x50, 0x24, 0x76, 0x60, 0x7, 0x7B, 0x68, + 0x7, 0x37, 0x60, 0x87, 0x77, 0x78, 0x7, 0x78, + 0x98, 0x51, 0x4C, 0xF4, 0x90, 0xF, 0xF0, 0x50, + 0xE, 0x33, 0x1E, 0x6A, 0x1E, 0xCA, 0x61, 0x1C, + 0xE8, 0x21, 0x1D, 0xDE, 0xC1, 0x1D, 0x7E, 0x1, + 0x1E, 0xE4, 0xA1, 0x1C, 0xCC, 0x21, 0x1D, 0xF0, + 0x61, 0x6, 0x54, 0x85, 0x83, 0x38, 0xCC, 0xC3, + 0x3B, 0xB0, 0x43, 0x3D, 0xD0, 0x43, 0x39, 0xFC, + 0xC2, 0x3C, 0xE4, 0x43, 0x3B, 0x88, 0xC3, 0x3B, + 0xB0, 0xC3, 0x8C, 0xC5, 0xA, 0x87, 0x79, 0x98, + 0x87, 0x77, 0x18, 0x87, 0x74, 0x8, 0x7, 0x7A, + 0x28, 0x7, 0x72, 0x98, 0x81, 0x5C, 0xE3, 0x10, + 0xE, 0xEC, 0xC0, 0xE, 0xE5, 0x50, 0xE, 0xF3, 0x30, + 0x23, 0xC1, 0xD2, 0x41, 0x1E, 0xE4, 0xE1, 0x17, + 0xD8, 0xE1, 0x1D, 0xDE, 0x1, 0x1E, 0x66, 0x48, + 0x19, 0x3B, 0xB0, 0x83, 0x3D, 0xB4, 0x83, 0x1B, + 0x84, 0xC3, 0x38, 0x8C, 0x43, 0x39, 0xCC, 0xC3, + 0x3C, 0xB8, 0xC1, 0x39, 0xC8, 0xC3, 0x3B, 0xD4, + 0x3, 0x3C, 0xCC, 0x48, 0xB4, 0x71, 0x8, 0x7, 0x76, + 0x60, 0x7, 0x71, 0x8, 0x87, 0x71, 0x58, 0x87, + 0x19, 0xDB, 0xC6, 0xE, 0xEC, 0x60, 0xF, 0xED, + 0xE0, 0x6, 0xF0, 0x20, 0xF, 0xE5, 0x30, 0xF, 0xE5, + 0x20, 0xF, 0xF6, 0x50, 0xE, 0x6E, 0x10, 0xE, 0xE3, + 0x30, 0xE, 0xE5, 0x30, 0xF, 0xF3, 0xE0, 0x6, 0xE9, + 0xE0, 0xE, 0xE4, 0x50, 0xE, 0xF8, 0x30, 0x23, + 0xE2, 0xEC, 0x61, 0x1C, 0xC2, 0x81, 0x1D, 0xD8, + 0xE1, 0x17, 0xEC, 0x21, 0x1D, 0xE6, 0x21, 0x1D, + 0xC4, 0x21, 0x1D, 0xD8, 0x21, 0x1D, 0xE8, 0x21, + 0x1F, 0x66, 0x20, 0x9D, 0x3B, 0xBC, 0x43, 0x3D, + 0xB8, 0x3, 0x39, 0x94, 0x83, 0x39, 0xCC, 0x58, + 0xBC, 0x70, 0x70, 0x7, 0x77, 0x78, 0x7, 0x7A, + 0x8, 0x7, 0x7A, 0x48, 0x87, 0x77, 0x70, 0x87, + 0x19, 0xCB, 0xE7, 0xE, 0xEF, 0x30, 0xF, 0xE1, + 0xE0, 0xE, 0xE9, 0x40, 0xF, 0xE9, 0xA0, 0xF, 0xE5, + 0x30, 0xC3, 0x1, 0x3, 0x73, 0xA8, 0x7, 0x77, 0x18, + 0x87, 0x5F, 0x98, 0x87, 0x70, 0x70, 0x87, 0x74, + 0xA0, 0x87, 0x74, 0xD0, 0x87, 0x72, 0x98, 0x81, + 0x84, 0x41, 0x39, 0xE0, 0xC3, 0x38, 0xB0, 0x43, + 0x3D, 0x90, 0x43, 0x39, 0xCC, 0x40, 0xC4, 0xA0, + 0x1D, 0xCA, 0xA1, 0x1D, 0xE0, 0x41, 0x1E, 0xDE, + 0xC1, 0x1C, 0x66, 0x24, 0x63, 0x30, 0xE, 0xE1, + 0xC0, 0xE, 0xEC, 0x30, 0xF, 0xE9, 0x40, 0xF, 0xE5, + 0x30, 0x43, 0x21, 0x83, 0x75, 0x18, 0x7, 0x73, + 0x48, 0x87, 0x5F, 0xA0, 0x87, 0x7C, 0x80, 0x87, + 0x72, 0x98, 0xB1, 0x94, 0x1, 0x3C, 0x8C, 0xC3, + 0x3C, 0x94, 0xC3, 0x38, 0xD0, 0x43, 0x3A, 0xBC, + 0x83, 0x3B, 0xCC, 0xC3, 0x8C, 0xC5, 0xC, 0x48, + 0x21, 0x15, 0x42, 0x61, 0x1E, 0xE6, 0x21, 0x1D, + 0xCE, 0xC1, 0x1D, 0x52, 0x81, 0x14, 0x66, 0x4C, + 0x67, 0x30, 0xE, 0xEF, 0x20, 0xF, 0xEF, 0xE0, + 0x6, 0xEF, 0x50, 0xF, 0xF4, 0x30, 0xF, 0xE9, 0x40, + 0xE, 0xE5, 0xE0, 0x6, 0xE6, 0x20, 0xF, 0xE1, 0xD0, + 0xE, 0xE5, 0x30, 0xA3, 0x40, 0x83, 0x76, 0x68, + 0x7, 0x79, 0x8, 0x87, 0x19, 0x52, 0x1A, 0xB8, + 0xC3, 0x3B, 0x84, 0x3, 0x3B, 0xA4, 0x43, 0x38, + 0xCC, 0x83, 0x1B, 0x84, 0x3, 0x39, 0x90, 0x83, + 0x3C, 0xCC, 0x3, 0x3C, 0x84, 0xC3, 0x38, 0x94, + 0xC3, 0xC, 0x46, 0xD, 0xC6, 0x21, 0x1C, 0xD8, + 0x81, 0x1D, 0xCA, 0xA1, 0x1C, 0x7E, 0x81, 0x1E, + 0xF2, 0x1, 0x1E, 0xCA, 0x61, 0xC6, 0xB1, 0x6, + 0xEE, 0xF0, 0xE, 0xE6, 0x20, 0xF, 0xE5, 0x50, + 0xE, 0x33, 0x1C, 0x36, 0x20, 0x7, 0x7C, 0x70, + 0x3, 0x77, 0x78, 0x7, 0x77, 0xA8, 0x7, 0x77, 0x48, + 0x7, 0x73, 0x78, 0x7, 0x79, 0x68, 0x87, 0x19, + 0x55, 0x1B, 0x90, 0x3, 0x3E, 0xB8, 0xC1, 0x38, + 0xBC, 0x83, 0x3B, 0xD0, 0x83, 0x3C, 0xBC, 0x3, + 0x3B, 0x98, 0x3, 0x3B, 0xBC, 0xC3, 0x3D, 0xB8, + 0x1, 0x3A, 0xA4, 0x83, 0x3B, 0xD0, 0xC3, 0x3C, + 0xCC, 0x58, 0xDC, 0x80, 0x1C, 0xF0, 0xC1, 0xD, + 0xE0, 0x41, 0x1E, 0xCA, 0x61, 0x1C, 0xD2, 0x61, + 0x1E, 0xCA, 0x1, 0x0, 0x79, 0x28, 0x0, 0x0, 0x52, + 0x0, 0x0, 0x0, 0xC2, 0x3C, 0x90, 0x40, 0x86, 0x10, + 0x19, 0x32, 0xE2, 0x64, 0x90, 0x40, 0x46, 0x2, + 0x19, 0x23, 0x23, 0x46, 0x2, 0x13, 0x24, 0xC6, + 0x0, 0x13, 0x74, 0xCE, 0x61, 0x8C, 0x2D, 0xCC, + 0xED, 0xC, 0xC4, 0xAE, 0x4C, 0x6E, 0x2E, 0xED, + 0xCD, 0xD, 0x44, 0x46, 0xC6, 0x5, 0xC6, 0x5, 0xE6, + 0x2C, 0x8D, 0xE, 0x4, 0xE5, 0x2C, 0x8D, 0xE, 0xE8, + 0x2C, 0x8D, 0xE, 0xAD, 0x4E, 0xCC, 0x65, 0xEC, + 0xAD, 0x4D, 0x87, 0x8D, 0xCD, 0xAE, 0xED, 0x85, + 0x8D, 0xCD, 0xAE, 0xAD, 0x5, 0x4E, 0xEE, 0x4D, + 0xAD, 0x6C, 0x8C, 0xCE, 0xE5, 0x2C, 0x8D, 0xE, + 0xA4, 0xEC, 0xC6, 0x86, 0xA6, 0x2C, 0x26, 0x7, + 0xA6, 0xAC, 0xC, 0x26, 0x26, 0xE7, 0x46, 0x6C, + 0x2C, 0xA6, 0xC, 0x66, 0xA6, 0x6C, 0xC6, 0x4C, + 0x6, 0x86, 0x6C, 0x4C, 0x6, 0x46, 0xCC, 0x66, + 0x6C, 0x2C, 0xC, 0x27, 0x46, 0x6C, 0x86, 0x6C, + 0x2C, 0xE5, 0x8, 0x63, 0x73, 0x87, 0x68, 0xB, + 0x4B, 0x73, 0x3B, 0xCA, 0xDD, 0x18, 0x5A, 0x98, + 0xDC, 0xD7, 0x5C, 0x9A, 0x5E, 0xD9, 0x69, 0xCC, + 0xE4, 0xC2, 0xDA, 0xCA, 0x5A, 0xE0, 0xDE, 0xD2, + 0xDC, 0xE8, 0xCA, 0xE4, 0x86, 0x20, 0x1C, 0xC1, + 0x10, 0x84, 0x43, 0x18, 0x82, 0x70, 0xC, 0x43, + 0x10, 0xE, 0x62, 0x8, 0x42, 0x1, 0xC, 0x41, 0x38, + 0x8A, 0x21, 0x8, 0x87, 0x31, 0x6, 0xC1, 0x38, + 0xC6, 0x10, 0x4, 0x63, 0x18, 0x4, 0x24, 0x19, + 0x83, 0x60, 0x24, 0x63, 0x18, 0xC, 0xC3, 0x18, + 0x83, 0xB0, 0x44, 0x63, 0x28, 0x94, 0x1, 0x0, + 0xA4, 0x31, 0xC, 0x6, 0xB1, 0x8C, 0x61, 0x60, + 0xA, 0xC6, 0x24, 0x64, 0x78, 0x2E, 0x76, 0x61, + 0x6C, 0x76, 0x65, 0x72, 0x43, 0x9, 0x18, 0xA3, + 0xB0, 0xB1, 0xD9, 0xB5, 0xB9, 0xA4, 0x91, 0x95, + 0xB9, 0xD1, 0xD, 0x25, 0x68, 0x8C, 0x43, 0x86, + 0xE7, 0x32, 0x87, 0x16, 0x46, 0x56, 0x26, 0xD7, + 0xF4, 0x46, 0x56, 0xC6, 0x36, 0x94, 0xC0, 0x31, + 0xA, 0x19, 0x9E, 0x8B, 0x5D, 0x99, 0xDC, 0x5C, + 0xDA, 0x9B, 0xDB, 0x50, 0x82, 0xC7, 0x38, 0x64, + 0x78, 0x2E, 0x65, 0x6E, 0x74, 0x72, 0x79, 0x50, + 0x6F, 0x69, 0x6E, 0x74, 0x73, 0x43, 0x9, 0x24, + 0x13, 0xB1, 0xB1, 0xD9, 0xB5, 0xB9, 0xB4, 0xBD, + 0x91, 0xD5, 0xB1, 0x95, 0xB9, 0x98, 0xB1, 0x85, + 0x9D, 0xCD, 0xD, 0x45, 0x98, 0x28, 0x0, 0x71, + 0x20, 0x0, 0x0, 0x2, 0x0, 0x0, 0x0, 0x6, 0x40, + 0x30, 0x0, 0xD2, 0x0, 0x0, 0x0, 0x61, 0x20, 0x0, + 0x0, 0x6, 0x0, 0x0, 0x0, 0x13, 0x4, 0x1, 0x86, + 0x3, 0x1, 0x0, 0x0, 0x2, 0x0, 0x0, 0x0, 0x7, 0x50, + 0x10, 0xCD, 0x14, 0x61, 0x0, 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, 0x0 ] + - Name: PSV0 + Size: 76 + PSVInfo: + Version: 3 + ShaderStage: 5 + MinimumWaveLaneCount: 0 + MaximumWaveLaneCount: 4294967295 + UsesViewID: 0 + SigInputVectors: 0 + SigOutputVectors: [ 0, 0, 0, 0 ] + NumThreadsX: 1 + NumThreadsY: 1 + NumThreadsZ: 1 + EntryName: main + ResourceStride: 24 + Resources: [] + SigInputElements: [] + SigOutputElements: [] + SigPatchOrPrimElements: [] + InputOutputMap: + - [ ] + - [ ] + - [ ] + - [ ] +... diff --git a/llvm/test/tools/llvm-readobj/ELF/bb-addr-map.test b/llvm/test/tools/llvm-readobj/ELF/bb-addr-map.test index fd1492f..bcffd40 100644 --- a/llvm/test/tools/llvm-readobj/ELF/bb-addr-map.test +++ b/llvm/test/tools/llvm-readobj/ELF/bb-addr-map.test @@ -34,6 +34,7 @@ # CHECK-NEXT: { # CHECK-NEXT: ID: 0 # CHECK-NEXT: Offset: 0x0 +# CHECK-NEXT: Hash: 0x0 # CHECK-NEXT: Size: 0x1 # CHECK-NEXT: HasReturn: No # CHECK-NEXT: HasTailCall: Yes @@ -50,6 +51,7 @@ # CHECK-NEXT: ID: 2 # CHECK-NEXT: Offset: 0x3 # CHECK-NEXT: Callsite End Offsets: [1, 3] +# CHECK-NEXT: Hash: 0x123 # CHECK-NEXT: Size: 0x7 # CHECK-NEXT: HasReturn: Yes # CHECK-NEXT: HasTailCall: No @@ -144,8 +146,8 @@ Sections: ShSize: [[SIZE=<none>]] Link: .text Entries: - - Version: 3 - Feature: 0x28 + - Version: 4 + Feature: 0x68 BBRanges: - BaseAddress: [[ADDR=0x11111]] BBEntries: @@ -160,6 +162,7 @@ Sections: Size: 0x4 Metadata: 0x15 CallsiteEndOffsets: [ 0x1 , 0x2 ] + Hash: 0x123 - Version: 2 BBRanges: - BaseAddress: 0x22222 diff --git a/llvm/test/tools/obj2yaml/ELF/bb-addr-map.yaml b/llvm/test/tools/obj2yaml/ELF/bb-addr-map.yaml index dc14025..7a22efe 100644 --- a/llvm/test/tools/obj2yaml/ELF/bb-addr-map.yaml +++ b/llvm/test/tools/obj2yaml/ELF/bb-addr-map.yaml @@ -162,6 +162,92 @@ Sections: BBRanges: - BaseAddress: 0x20 +## Check that obj2yaml can dump basic block hash in the .llvm_bb_addr_map section. + +# RUN: yaml2obj --docnum=4 %s -o %t4 +# RUN: obj2yaml %t4 | FileCheck %s --check-prefix=BBHASH + +# BBHASH: --- !ELF +# BBHASH-NEXT: FileHeader: +# BBHASH-NEXT: Class: ELFCLASS64 +# BBHASH-NEXT: Data: ELFDATA2LSB +# BBHASH-NEXT: Type: ET_EXEC +# BBHASH-NEXT: Sections: +# BBHASH-NEXT: - Name: .llvm_bb_addr_map +# BBHASH-NEXT: Type: SHT_LLVM_BB_ADDR_MAP +# BBHASH-NEXT: Entries: +# BBHASH-NEXT: - Version: 4 +# BBHASH-NEXT: Feature: 0x40 +# BBHASH-NEXT: BBRanges: +# BBHASH-NEXT: - BBEntries: +# BBHASH-NEXT: - ID: 0 +# BBHASH-NEXT: AddressOffset: 0x1 +# BBHASH-NEXT: Size: 0x2 +# BBHASH-NEXT: Metadata: 0x3 +# BBHASH-NEXT: Hash: 0x1 +# BBHASH-NEXT: - ID: 2 +# BBHASH-NEXT: AddressOffset: 0x4 +# BBHASH-NEXT: Size: 0x5 +# BBHASH-NEXT: Metadata: 0x6 +# BBHASH-NEXT: Hash: 0x2 +# BBHASH-NEXT: - ID: 4 +# BBHASH-NEXT: AddressOffset: 0xFFFFFFFFFFFFFFF7 +# BBHASH-NEXT: Size: 0xFFFFFFFFFFFFFFF8 +# BBHASH-NEXT: Metadata: 0xFFFFFFFFFFFFFFF9 +# BBHASH-NEXT: Hash: 0x3 +# BBHASH-NEXT: - Version: 4 +# BBHASH-NEXT: Feature: 0x68 +# BBHASH-NEXT: BBRanges: +# BBHASH-NEXT: - BaseAddress: 0xFFFFFFFFFFFFFF20 +# BBHASH-NEXT: BBEntries: +# BBHASH-NEXT: - ID: 6 +# BBHASH-NEXT: AddressOffset: 0xA +# BBHASH-NEXT: Size: 0xB +# BBHASH-NEXT: Metadata: 0xC +# BBHASH-NEXT: CallsiteEndOffsets: [ 0x1, 0x2 ] +# BBHASH-NEXT: Hash: 0x123 + +--- !ELF +FileHeader: + Class: ELFCLASS64 + Data: ELFDATA2LSB + Type: ET_EXEC +Sections: + - Name: .llvm_bb_addr_map + Type: SHT_LLVM_BB_ADDR_MAP + Entries: + - Version: 4 + Feature: 0x40 + BBRanges: + - BaseAddress: 0x0 + BBEntries: + - ID: 0 + AddressOffset: 0x1 + Size: 0x2 + Metadata: 0x3 + Hash: 0x1 + - ID: 2 + AddressOffset: 0x4 + Size: 0x5 + Metadata: 0x6 + Hash: 0x2 + - ID: 4 + AddressOffset: 0xFFFFFFFFFFFFFFF7 + Size: 0xFFFFFFFFFFFFFFF8 + Metadata: 0xFFFFFFFFFFFFFFF9 + Hash: 0x3 + - Version: 4 + Feature: 0x68 + BBRanges: + - BaseAddress: 0xFFFFFFFFFFFFFF20 + BBEntries: + - ID: 6 + AddressOffset: 0xA + Size: 0xB + Metadata: 0xC + CallsiteEndOffsets: [ 0x1, 0x2 ] + Hash: 0x123 + ## Check that obj2yaml uses the "Content" tag to describe an .llvm_bb_addr_map section ## when it can't extract the entries, for example, when the section is truncated, or ## when an invalid 'NumBlocks' or 'NumBBRanges` field is specified. diff --git a/llvm/test/tools/yaml2obj/ELF/bb-addr-map.yaml b/llvm/test/tools/yaml2obj/ELF/bb-addr-map.yaml index 418f90f..339e419 100644 --- a/llvm/test/tools/yaml2obj/ELF/bb-addr-map.yaml +++ b/llvm/test/tools/yaml2obj/ELF/bb-addr-map.yaml @@ -72,6 +72,13 @@ # CHECK-NEXT: 0000: 03202000 00000000 0000010E 01000203 # CHECK-NEXT: ) +# Case 10: Specify basic block hash. +# CHECK: Name: .llvm_bb_addr_map (1) +# CHECK: SectionData ( +# CHECK-NEXT: 0000: 04602000 00000000 0000010E 01000203 +# CHECK-NEXT: 0010: 23010000 00000000 +# CHECK-NEXT: ) + --- !ELF FileHeader: @@ -176,6 +183,22 @@ Sections: Metadata: 0x00000003 CallsiteEndOffsets: [] +## 10) We can produce a SHT_LLVM_BB_ADDR_MAP section with basic block hash. + - Name: '.llvm_bb_addr_map (10)' + Type: SHT_LLVM_BB_ADDR_MAP + Entries: + - Version: 4 + Feature: 0x60 + BBRanges: + - BaseAddress: 0x0000000000000020 + BBEntries: + - ID: 14 + AddressOffset: 0x00000001 + Size: 0x00000002 + Metadata: 0x00000003 + CallsiteEndOffsets: [] + Hash: 0x123 + ## Check we can't use Entries at the same time as either Content or Size. # RUN: not yaml2obj --docnum=2 -DCONTENT="00" %s 2>&1 | FileCheck %s --check-prefix=INVALID # RUN: not yaml2obj --docnum=2 -DSIZE="0" %s 2>&1 | FileCheck %s --check-prefix=INVALID @@ -197,7 +220,7 @@ Sections: ## Check that yaml2obj generates a warning when we use unsupported versions. # RUN: yaml2obj --docnum=3 %s 2>&1 | FileCheck %s --check-prefix=INVALID-VERSION -# INVALID-VERSION: warning: unsupported SHT_LLVM_BB_ADDR_MAP version: 4; encoding using the most recent version +# INVALID-VERSION: warning: unsupported SHT_LLVM_BB_ADDR_MAP version: 5; encoding using the most recent version --- !ELF FileHeader: @@ -209,4 +232,4 @@ Sections: Type: SHT_LLVM_BB_ADDR_MAP Entries: ## Specify unsupported version - - Version: 4 + - Version: 5 diff --git a/llvm/tools/llvm-readobj/ELFDumper.cpp b/llvm/tools/llvm-readobj/ELFDumper.cpp index ab93316..9c9b2dd 100644 --- a/llvm/tools/llvm-readobj/ELFDumper.cpp +++ b/llvm/tools/llvm-readobj/ELFDumper.cpp @@ -8155,6 +8155,8 @@ void LLVMELFDumper<ELFT>::printBBAddrMaps(bool PrettyPGOAnalysis) { W.printHex("Offset", BBE.Offset); if (!BBE.CallsiteEndOffsets.empty()) W.printList("Callsite End Offsets", BBE.CallsiteEndOffsets); + if (PAM.FeatEnable.BBHash) + W.printHex("Hash", BBE.Hash); W.printHex("Size", BBE.Size); W.printBoolean("HasReturn", BBE.hasReturn()); W.printBoolean("HasTailCall", BBE.hasTailCall()); diff --git a/llvm/tools/obj2yaml/elf2yaml.cpp b/llvm/tools/obj2yaml/elf2yaml.cpp index ef4552f..68e18f6 100644 --- a/llvm/tools/obj2yaml/elf2yaml.cpp +++ b/llvm/tools/obj2yaml/elf2yaml.cpp @@ -900,7 +900,7 @@ ELFDumper<ELFT>::dumpBBAddrMapSection(const Elf_Shdr *Shdr) { while (Cur && Cur.tell() < Content.size()) { if (Shdr->sh_type == ELF::SHT_LLVM_BB_ADDR_MAP) { Version = Data.getU8(Cur); - if (Cur && Version > 3) + if (Cur && Version > 4) return createStringError( errc::invalid_argument, "invalid SHT_LLVM_BB_ADDR_MAP section version: " + @@ -946,8 +946,11 @@ ELFDumper<ELFT>::dumpBBAddrMapSection(const Elf_Shdr *Shdr) { } uint64_t Size = Data.getULEB128(Cur); uint64_t Metadata = Data.getULEB128(Cur); + std::optional<llvm::yaml::Hex64> Hash; + if (FeatureOrErr->BBHash) + Hash = Data.getU64(Cur); BBEntries.push_back( - {ID, Offset, Size, Metadata, std::move(CallsiteEndOffsets)}); + {ID, Offset, Size, Metadata, std::move(CallsiteEndOffsets), Hash}); } TotalNumBlocks += BBEntries.size(); BBRanges.push_back({BaseAddress, /*NumBlocks=*/{}, BBEntries}); diff --git a/llvm/unittests/ADT/TypeTraitsTest.cpp b/llvm/unittests/ADT/TypeTraitsTest.cpp index a56aa7e..f9b8d6d 100644 --- a/llvm/unittests/ADT/TypeTraitsTest.cpp +++ b/llvm/unittests/ADT/TypeTraitsTest.cpp @@ -40,9 +40,7 @@ struct Foo { struct CheckMethodPointer : CheckFunctionTraits<decltype(&Foo::func)> {}; /// Test lambda references. -LLVM_ATTRIBUTE_UNUSED auto lambdaFunc = [](const int &v) -> bool { - return true; -}; +[[maybe_unused]] auto lambdaFunc = [](const int &v) -> bool { return true; }; struct CheckLambda : CheckFunctionTraits<decltype(lambdaFunc)> {}; } // end anonymous namespace diff --git a/llvm/unittests/Object/ELFObjectFileTest.cpp b/llvm/unittests/Object/ELFObjectFileTest.cpp index 17d9f50..d6a3ca5 100644 --- a/llvm/unittests/Object/ELFObjectFileTest.cpp +++ b/llvm/unittests/Object/ELFObjectFileTest.cpp @@ -531,7 +531,7 @@ Sections: // Check that we can detect unsupported versions. SmallString<128> UnsupportedVersionYamlString(CommonYamlString); UnsupportedVersionYamlString += R"( - - Version: 4 + - Version: 5 BBRanges: - BaseAddress: 0x11111 BBEntries: @@ -543,7 +543,7 @@ Sections: { SCOPED_TRACE("unsupported version"); DoCheck(UnsupportedVersionYamlString, - "unsupported SHT_LLVM_BB_ADDR_MAP version: 4"); + "unsupported SHT_LLVM_BB_ADDR_MAP version: 5"); } SmallString<128> ZeroBBRangesYamlString(CommonYamlString); @@ -761,14 +761,14 @@ Sections: BBAddrMap E1 = { {{0x11111, - {{1, 0x0, 0x3, {false, true, false, false, false}, {0x1, 0x2}}}}}}; + {{1, 0x0, 0x3, {false, true, false, false, false}, {0x1, 0x2}, 0}}}}}; BBAddrMap E2 = { - {{0x22222, {{2, 0x0, 0x2, {false, false, true, false, false}, {}}}}, - {0xFFFFF, {{15, 0xF0, 0xF1, {true, true, true, true, true}, {}}}}}}; + {{0x22222, {{2, 0x0, 0x2, {false, false, true, false, false}, {}, 0}}}, + {0xFFFFF, {{15, 0xF0, 0xF1, {true, true, true, true, true}, {}, 0}}}}}; BBAddrMap E3 = { - {{0x33333, {{0, 0x0, 0x3, {false, true, true, false, false}, {}}}}}}; + {{0x33333, {{0, 0x0, 0x3, {false, true, true, false, false}, {}, 0}}}}}; BBAddrMap E4 = { - {{0x44444, {{0, 0x0, 0x4, {false, false, false, true, true}, {}}}}}}; + {{0x44444, {{0, 0x0, 0x4, {false, false, false, true, true}, {}, 0}}}}}; std::vector<BBAddrMap> Section0BBAddrMaps = {E4}; std::vector<BBAddrMap> Section1BBAddrMaps = {E3}; @@ -988,6 +988,123 @@ Sections: } } +// Test for the ELFObjectFile::readBBAddrMap API with BBHash. +TEST(ELFObjectFileTest, ReadBBHash) { + StringRef CommonYamlString(R"( +--- !ELF +FileHeader: + Class: ELFCLASS64 + Data: ELFDATA2LSB + Type: ET_EXEC +Sections: + - Name: .llvm_bb_addr_map_1 + Type: SHT_LLVM_BB_ADDR_MAP + Link: 1 + Entries: + - Version: 4 + Feature: 0x60 + BBRanges: + - BaseAddress: 0x11111 + BBEntries: + - ID: 1 + AddressOffset: 0x0 + Size: 0x1 + Metadata: 0x2 + CallsiteEndOffsets: [ 0x1 , 0x1 ] + Hash: 0x1 + - Name: .llvm_bb_addr_map_2 + Type: SHT_LLVM_BB_ADDR_MAP + Link: 1 + Entries: + - Version: 4 + Feature: 0x48 + BBRanges: + - BaseAddress: 0x22222 + BBEntries: + - ID: 2 + AddressOffset: 0x0 + Size: 0x2 + Metadata: 0x4 + Hash: 0x2 + - BaseAddress: 0xFFFFF + BBEntries: + - ID: 15 + AddressOffset: 0xF0 + Size: 0xF1 + Metadata: 0x1F + Hash: 0xF + - Name: .llvm_bb_addr_map_3 + Type: SHT_LLVM_BB_ADDR_MAP + Link: 2 + Entries: + - Version: 4 + Feature: 0x40 + BBRanges: + - BaseAddress: 0x33333 + BBEntries: + - ID: 0 + AddressOffset: 0x0 + Size: 0x3 + Metadata: 0x6 + Hash: 0x3 + - Name: .llvm_bb_addr_map_4 + Type: SHT_LLVM_BB_ADDR_MAP + # Link: 0 (by default, can be overriden) + Entries: + - Version: 4 + Feature: 0x40 + BBRanges: + - BaseAddress: 0x44444 + BBEntries: + - ID: 0 + AddressOffset: 0x0 + Size: 0x4 + Metadata: 0x18 + Hash: 0x4 +)"); + + BBAddrMap E1 = { + {{0x11111, + {{1, 0x0, 0x3, {false, true, false, false, false}, {0x1, 0x2}, 0x1}}}}}; + BBAddrMap E2 = { + {{0x22222, {{2, 0x0, 0x2, {false, false, true, false, false}, {}, 0x2}}}, + {0xFFFFF, {{15, 0xF0, 0xF1, {true, true, true, true, true}, {}, 0xF}}}}}; + BBAddrMap E3 = { + {{0x33333, {{0, 0x0, 0x3, {false, true, true, false, false}, {}, 0x3}}}}}; + BBAddrMap E4 = { + {{0x44444, {{0, 0x0, 0x4, {false, false, false, true, true}, {}, 0x4}}}}}; + + std::vector<BBAddrMap> Section0BBAddrMaps = {E4}; + std::vector<BBAddrMap> Section1BBAddrMaps = {E3}; + std::vector<BBAddrMap> Section2BBAddrMaps = {E1, E2}; + std::vector<BBAddrMap> AllBBAddrMaps = {E1, E2, E3, E4}; + + auto DoCheckSucceeds = [&](StringRef YamlString, + std::optional<unsigned> TextSectionIndex, + std::vector<BBAddrMap> ExpectedResult) { + SCOPED_TRACE("for TextSectionIndex: " + + (TextSectionIndex ? llvm::Twine(*TextSectionIndex) : "{}") + + " and object yaml:\n" + YamlString); + SmallString<0> Storage; + Expected<ELFObjectFile<ELF64LE>> ElfOrErr = + toBinary<ELF64LE>(Storage, YamlString); + ASSERT_THAT_EXPECTED(ElfOrErr, Succeeded()); + + Expected<const typename ELF64LE::Shdr *> BBAddrMapSecOrErr = + ElfOrErr->getELFFile().getSection(1); + ASSERT_THAT_EXPECTED(BBAddrMapSecOrErr, Succeeded()); + auto BBAddrMaps = ElfOrErr->readBBAddrMap(TextSectionIndex); + ASSERT_THAT_EXPECTED(BBAddrMaps, Succeeded()); + EXPECT_EQ(*BBAddrMaps, ExpectedResult); + }; + + DoCheckSucceeds(CommonYamlString, /*TextSectionIndex=*/std::nullopt, + AllBBAddrMaps); + DoCheckSucceeds(CommonYamlString, /*TextSectionIndex=*/0, Section0BBAddrMaps); + DoCheckSucceeds(CommonYamlString, /*TextSectionIndex=*/2, Section1BBAddrMaps); + DoCheckSucceeds(CommonYamlString, /*TextSectionIndex=*/1, Section2BBAddrMaps); +} + // Test for the ELFObjectFile::readBBAddrMap API with PGOAnalysisMap. TEST(ELFObjectFileTest, ReadPGOAnalysisMap) { StringRef CommonYamlString(R"( @@ -1159,29 +1276,32 @@ Sections: )"); BBAddrMap E1 = { - {{0x11111, {{1, 0x0, 0x1, {false, true, false, false, false}, {}}}}}}; - PGOAnalysisMap P1 = {892, {}, {true, false, false, false, false, false}}; + {{0x11111, {{1, 0x0, 0x1, {false, true, false, false, false}, {}, 0}}}}}; + PGOAnalysisMap P1 = { + 892, {}, {true, false, false, false, false, false, false}}; BBAddrMap E2 = { - {{0x22222, {{2, 0x0, 0x2, {false, false, true, false, false}, {}}}}}}; + {{0x22222, {{2, 0x0, 0x2, {false, false, true, false, false}, {}, 0}}}}}; PGOAnalysisMap P2 = {{}, {{BlockFrequency(343), {}}}, - {false, true, false, false, false, false}}; - BBAddrMap E3 = {{{0x33333, - {{0, 0x0, 0x3, {false, true, true, false, false}, {}}, - {1, 0x3, 0x3, {false, false, true, false, false}, {}}, - {2, 0x6, 0x3, {false, false, false, false, false}, {}}}}}}; + {false, true, false, false, false, false, false}}; + BBAddrMap E3 = { + {{0x33333, + {{0, 0x0, 0x3, {false, true, true, false, false}, {}, 0}, + {1, 0x3, 0x3, {false, false, true, false, false}, {}, 0}, + {2, 0x6, 0x3, {false, false, false, false, false}, {}, 0}}}}}; PGOAnalysisMap P3 = {{}, {{{}, {{1, BranchProbability::getRaw(0x1111'1111)}, {2, BranchProbability::getRaw(0xeeee'eeee)}}}, {{}, {{2, BranchProbability::getRaw(0xffff'ffff)}}}, {{}, {}}}, - {false, false, true, false, false, false}}; - BBAddrMap E4 = {{{0x44444, - {{0, 0x0, 0x4, {false, false, false, true, true}, {}}, - {1, 0x4, 0x4, {false, false, false, false, false}, {}}, - {2, 0x8, 0x4, {false, false, false, false, false}, {}}, - {3, 0xc, 0x4, {false, false, false, false, false}, {}}}}}}; + {false, false, true, false, false, false, false}}; + BBAddrMap E4 = { + {{0x44444, + {{0, 0x0, 0x4, {false, false, false, true, true}, {}, 0}, + {1, 0x4, 0x4, {false, false, false, false, false}, {}, 0}, + {2, 0x8, 0x4, {false, false, false, false, false}, {}, 0}, + {3, 0xc, 0x4, {false, false, false, false, false}, {}, 0}}}}}; PGOAnalysisMap P4 = { 1000, {{BlockFrequency(1000), @@ -1193,22 +1313,24 @@ Sections: {3, BranchProbability::getRaw(0xeeee'eeee)}}}, {BlockFrequency(18), {{3, BranchProbability::getRaw(0xffff'ffff)}}}, {BlockFrequency(1000), {}}}, - {true, true, true, false, false, false}}; + {true, true, true, false, false, false, false}}; BBAddrMap E5 = { - {{0x55555, {{2, 0x0, 0x2, {false, false, true, false, false}, {}}}}}}; - PGOAnalysisMap P5 = {{}, {}, {false, false, false, false, false, false}}; + {{0x55555, {{2, 0x0, 0x2, {false, false, true, false, false}, {}, 0}}}}}; + PGOAnalysisMap P5 = { + {}, {}, {false, false, false, false, false, false, false}}; BBAddrMap E6 = { {{0x66666, - {{0, 0x0, 0x6, {false, true, true, false, false}, {}}, - {1, 0x6, 0x6, {false, false, true, false, false}, {}}}}, - {0x666661, {{2, 0x0, 0x6, {false, false, false, false, false}, {}}}}}}; + {{0, 0x0, 0x6, {false, true, true, false, false}, {}, 0}, + {1, 0x6, 0x6, {false, false, true, false, false}, {}, 0}}}, + {0x666661, + {{2, 0x0, 0x6, {false, false, false, false, false}, {}, 0}}}}}; PGOAnalysisMap P6 = {{}, {{{}, {{1, BranchProbability::getRaw(0x2222'2222)}, {2, BranchProbability::getRaw(0xcccc'cccc)}}}, {{}, {{2, BranchProbability::getRaw(0x8888'8888)}}}, {{}, {}}}, - {false, false, true, true, false, false}}; + {false, false, true, true, false, false, false}}; std::vector<BBAddrMap> Section0BBAddrMaps = {E4, E5, E6}; std::vector<BBAddrMap> Section1BBAddrMaps = {E3}; diff --git a/llvm/unittests/Object/ELFTypesTest.cpp b/llvm/unittests/Object/ELFTypesTest.cpp index f88931b5f..1765e15 100644 --- a/llvm/unittests/Object/ELFTypesTest.cpp +++ b/llvm/unittests/Object/ELFTypesTest.cpp @@ -101,21 +101,22 @@ static_assert( "PGOAnalysisMap should use the same type for basic block ID as BBAddrMap"); TEST(ELFTypesTest, BBAddrMapFeaturesEncodingTest) { - const std::array<BBAddrMap::Features, 11> Decoded = { - {{false, false, false, false, false, false}, - {true, false, false, false, false, false}, - {false, true, false, false, false, false}, - {false, false, true, false, false, false}, - {false, false, false, true, false, false}, - {true, true, false, false, false, false}, - {false, true, true, false, false, false}, - {false, true, true, true, false, false}, - {true, true, true, true, false, false}, - {false, false, false, false, true, false}, - {false, false, false, false, false, true}}}; - const std::array<uint8_t, 11> Encoded = {{0b0000, 0b0001, 0b0010, 0b0100, - 0b1000, 0b0011, 0b0110, 0b1110, - 0b1111, 0b1'0000, 0b10'0000}}; + const std::array<BBAddrMap::Features, 12> Decoded = { + {{false, false, false, false, false, false, false}, + {true, false, false, false, false, false, false}, + {false, true, false, false, false, false, false}, + {false, false, true, false, false, false, false}, + {false, false, false, true, false, false, false}, + {true, true, false, false, false, false, false}, + {false, true, true, false, false, false, false}, + {false, true, true, true, false, false, false}, + {true, true, true, true, false, false, false}, + {false, false, false, false, true, false, false}, + {false, false, false, false, false, true, false}, + {false, false, false, false, false, false, true}}}; + const std::array<uint8_t, 12> Encoded = { + {0b0000, 0b0001, 0b0010, 0b0100, 0b1000, 0b0011, 0b0110, 0b1110, 0b1111, + 0b1'0000, 0b10'0000, 0b100'0000}}; for (const auto &[Feat, EncodedVal] : llvm::zip(Decoded, Encoded)) EXPECT_EQ(Feat.encode(), EncodedVal); for (const auto &[Feat, EncodedVal] : llvm::zip(Decoded, Encoded)) { @@ -128,9 +129,9 @@ TEST(ELFTypesTest, BBAddrMapFeaturesEncodingTest) { TEST(ELFTypesTest, BBAddrMapFeaturesInvalidEncodingTest) { const std::array<std::string, 2> Errors = { - "invalid encoding for BBAddrMap::Features: 0x40", + "invalid encoding for BBAddrMap::Features: 0x80", "invalid encoding for BBAddrMap::Features: 0xf0"}; - const std::array<uint8_t, 2> Values = {{0b100'0000, 0b1111'0000}}; + const std::array<uint8_t, 2> Values = {{0b1000'0000, 0b1111'0000}}; for (const auto &[Val, Error] : llvm::zip(Values, Errors)) { EXPECT_THAT_ERROR(BBAddrMap::Features::decode(Val).takeError(), FailedWithMessage(Error)); diff --git a/llvm/utils/profcheck-xfail.txt b/llvm/utils/profcheck-xfail.txt index 343c2bb71..a5c5426 100644 --- a/llvm/utils/profcheck-xfail.txt +++ b/llvm/utils/profcheck-xfail.txt @@ -906,7 +906,6 @@ Transforms/InstCombine/select_frexp.ll Transforms/InstCombine/select.ll Transforms/InstCombine/select-min-max.ll Transforms/InstCombine/select-of-symmetric-selects.ll -Transforms/InstCombine/select-safe-impliedcond-transforms.ll Transforms/InstCombine/select-safe-transforms.ll Transforms/InstCombine/select-select.ll Transforms/InstCombine/select-with-extreme-eq-cond.ll @@ -1237,7 +1236,6 @@ Transforms/PartiallyInlineLibCalls/X86/good-prototype.ll Transforms/PGOProfile/chr-dead-pred.ll Transforms/PGOProfile/chr-dup-threshold.ll Transforms/PGOProfile/chr-lifetimes.ll -Transforms/PGOProfile/chr.ll Transforms/PGOProfile/chr-poison.ll Transforms/PGOProfile/comdat.ll Transforms/PGOProfile/memop_profile_funclet_wasm.ll @@ -1312,14 +1310,11 @@ Transforms/SimpleLoopUnswitch/pr60736.ll Transforms/SimpleLoopUnswitch/trivial-unswitch-freeze-individual-conditions.ll Transforms/SimpleLoopUnswitch/trivial-unswitch.ll Transforms/SimpleLoopUnswitch/trivial-unswitch-logical-and-or.ll -Transforms/SROA/addrspacecast.ll -Transforms/SROA/phi-and-select.ll Transforms/SROA/phi-gep.ll Transforms/SROA/scalable-vectors-with-known-vscale.ll Transforms/SROA/select-gep.ll Transforms/SROA/select-load.ll Transforms/SROA/slice-width.ll -Transforms/SROA/std-clamp.ll Transforms/SROA/vector-conversion.ll Transforms/SROA/vector-promotion-cannot-tree-structure-merge.ll Transforms/SROA/vector-promotion.ll diff --git a/mlir/include/mlir/Analysis/DataFlow/StridedMetadataRangeAnalysis.h b/mlir/include/mlir/Analysis/DataFlow/StridedMetadataRangeAnalysis.h new file mode 100644 index 0000000..72ac247 --- /dev/null +++ b/mlir/include/mlir/Analysis/DataFlow/StridedMetadataRangeAnalysis.h @@ -0,0 +1,54 @@ +//===- StridedMetadataRange.h - Strided metadata range analysis -*- C++ -*-===// +// +// 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 +// +//===----------------------------------------------------------------------===// + +#ifndef MLIR_ANALYSIS_DATAFLOW_STRIDEDMETADATARANGE_H +#define MLIR_ANALYSIS_DATAFLOW_STRIDEDMETADATARANGE_H + +#include "mlir/Analysis/DataFlow/SparseAnalysis.h" +#include "mlir/Interfaces/InferStridedMetadataInterface.h" + +namespace mlir { +namespace dataflow { + +/// This lattice element represents the strided metadata of an SSA value. +class StridedMetadataRangeLattice : public Lattice<StridedMetadataRange> { +public: + using Lattice::Lattice; +}; + +/// Strided metadata range analysis determines the strided metadata ranges of +/// SSA values using operations that define `InferStridedMetadataInterface`. +/// +/// This analysis depends on DeadCodeAnalysis, SparseConstantPropagation, and +/// IntegerRangeAnalysis, and will be a silent no-op if the analyses are not +/// loaded in the same solver context. +class StridedMetadataRangeAnalysis + : public SparseForwardDataFlowAnalysis<StridedMetadataRangeLattice> { +public: + StridedMetadataRangeAnalysis(DataFlowSolver &solver, + int32_t indexBitwidth = 64); + + /// At an entry point, we cannot reason about strided metadata ranges unless + /// the type also encodes the data. For example, a memref with static layout. + void setToEntryState(StridedMetadataRangeLattice *lattice) override; + + /// Visit an operation. Invoke the transfer function on each operation that + /// implements `InferStridedMetadataInterface`. + LogicalResult + visitOperation(Operation *op, + ArrayRef<const StridedMetadataRangeLattice *> operands, + ArrayRef<StridedMetadataRangeLattice *> results) override; + +private: + /// Index bitwidth to use when operating with the int-ranges. + int32_t indexBitwidth = 64; +}; +} // namespace dataflow +} // end namespace mlir + +#endif // MLIR_ANALYSIS_DATAFLOW_STRIDEDMETADATARANGE_H diff --git a/mlir/include/mlir/Dialect/MemRef/IR/MemRef.h b/mlir/include/mlir/Dialect/MemRef/IR/MemRef.h index 30f33ed..69447f7 100644 --- a/mlir/include/mlir/Dialect/MemRef/IR/MemRef.h +++ b/mlir/include/mlir/Dialect/MemRef/IR/MemRef.h @@ -17,6 +17,7 @@ #include "mlir/Interfaces/CastInterfaces.h" #include "mlir/Interfaces/ControlFlowInterfaces.h" #include "mlir/Interfaces/InferIntRangeInterface.h" +#include "mlir/Interfaces/InferStridedMetadataInterface.h" #include "mlir/Interfaces/InferTypeOpInterface.h" #include "mlir/Interfaces/MemOpInterfaces.h" #include "mlir/Interfaces/MemorySlotInterfaces.h" diff --git a/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td b/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td index 89bd0f1..b39207f 100644 --- a/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td +++ b/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td @@ -14,6 +14,7 @@ include "mlir/Dialect/MemRef/IR/MemRefBase.td" include "mlir/Interfaces/CastInterfaces.td" include "mlir/Interfaces/ControlFlowInterfaces.td" include "mlir/Interfaces/InferIntRangeInterface.td" +include "mlir/Interfaces/InferStridedMetadataInterface.td" include "mlir/Interfaces/InferTypeOpInterface.td" include "mlir/Interfaces/MemOpInterfaces.td" include "mlir/Interfaces/MemorySlotInterfaces.td" @@ -2085,6 +2086,7 @@ def MemRef_StoreOp : MemRef_Op<"store", def SubViewOp : MemRef_OpWithOffsetSizesAndStrides<"subview", [ DeclareOpInterfaceMethods<OpAsmOpInterface, ["getAsmResultNames"]>, + DeclareOpInterfaceMethods<InferStridedMetadataOpInterface>, DeclareOpInterfaceMethods<MemorySpaceCastConsumerOpInterface>, DeclareOpInterfaceMethods<ViewLikeOpInterface>, AttrSizedOperandSegments, diff --git a/mlir/include/mlir/Dialect/OpenACC/OpenACCOps.td b/mlir/include/mlir/Dialect/OpenACC/OpenACCOps.td index 77e833f..fecf81b 100644 --- a/mlir/include/mlir/Dialect/OpenACC/OpenACCOps.td +++ b/mlir/include/mlir/Dialect/OpenACC/OpenACCOps.td @@ -1316,6 +1316,24 @@ def OpenACC_PrivateRecipeOp }]; let hasRegionVerifier = 1; + + let extraClassDeclaration = [{ + /// Creates a PrivateRecipeOp and populates its regions based on the + /// variable type as long as the type implements MappableType or + /// PointerLikeType interface. If a type implements both, the MappableType + /// API will be preferred. Returns std::nullopt if the recipe cannot be + /// created or populated. The builder's current insertion point will be used + /// and it must be a valid place for this operation to be inserted. The + /// `recipeName` must be a unique name to prevent "redefinition of symbol" + /// IR errors. + static std::optional<PrivateRecipeOp> createAndPopulate( + ::mlir::OpBuilder &builder, + ::mlir::Location loc, + ::llvm::StringRef recipeName, + ::mlir::Type varType, + ::llvm::StringRef varName = "", + ::mlir::ValueRange bounds = {}); + }]; } //===----------------------------------------------------------------------===// @@ -1410,6 +1428,24 @@ def OpenACC_FirstprivateRecipeOp }]; let hasRegionVerifier = 1; + + let extraClassDeclaration = [{ + /// Creates a FirstprivateRecipeOp and populates its regions based on the + /// variable type as long as the type implements MappableType or + /// PointerLikeType interface. If a type implements both, the MappableType + /// API will be preferred. Returns std::nullopt if the recipe cannot be + /// created or populated. The builder's current insertion point will be used + /// and it must be a valid place for this operation to be inserted. The + /// `recipeName` must be a unique name to prevent "redefinition of symbol" + /// IR errors. + static std::optional<FirstprivateRecipeOp> createAndPopulate( + ::mlir::OpBuilder &builder, + ::mlir::Location loc, + ::llvm::StringRef recipeName, + ::mlir::Type varType, + ::llvm::StringRef varName = "", + ::mlir::ValueRange bounds = {}); + }]; } //===----------------------------------------------------------------------===// diff --git a/mlir/include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.td b/mlir/include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.td index 0d16255..6736bc8 100644 --- a/mlir/include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.td +++ b/mlir/include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.td @@ -83,7 +83,15 @@ def OpenACC_PointerLikeTypeInterface : TypeInterface<"PointerLikeType"> { The `originalVar` parameter is optional but enables support for dynamic types (e.g., dynamic memrefs). When provided, implementations can extract runtime dimension information from the original variable to create - allocations with matching dynamic sizes. + allocations with matching dynamic sizes. When generating recipe bodies, + `originalVar` should be the block argument representing the original + variable in the recipe region. + + The `needsFree` output parameter indicates whether the allocated memory + requires explicit deallocation. Implementations should set this to true + for heap allocations that need a matching deallocation operation (e.g., + alloc) and false for stack-based allocations (e.g., alloca). During + recipe generation, this determines whether a destroy region is created. Returns a Value representing the result of the allocation. If no value is returned, it means the allocation was not successfully generated. @@ -94,7 +102,8 @@ def OpenACC_PointerLikeTypeInterface : TypeInterface<"PointerLikeType"> { "::mlir::Location":$loc, "::llvm::StringRef":$varName, "::mlir::Type":$varType, - "::mlir::Value":$originalVar), + "::mlir::Value":$originalVar, + "bool &":$needsFree), /*methodBody=*/"", /*defaultImplementation=*/[{ return {}; @@ -102,23 +111,34 @@ def OpenACC_PointerLikeTypeInterface : TypeInterface<"PointerLikeType"> { >, InterfaceMethod< /*description=*/[{ - Generates deallocation operations for the pointer-like type. It deallocates - the instance provided. + Generates deallocation operations for the pointer-like type. - The `varPtr` parameter is required and must represent an instance that was - previously allocated. If the current type is represented in a way that it - does not capture the pointee type, `varType` must be passed in to provide - the necessary type information. Nothing is generated in case the allocate - is `alloca`-like. + The `varToFree` parameter is required and must represent an instance + that was previously allocated. When generating recipe bodies, this + should be the block argument representing the private variable in the + destroy region. + + The `allocRes` parameter is optional and provides the result of the + corresponding allocation from the init region. This allows implementations + to inspect the allocation operation to determine the appropriate + deallocation strategy. This is necessary because in recipe generation, + the allocation and deallocation occur in separate regions. Dialects that + use only one allocation type or can determine deallocation from type + information alone may ignore this parameter. - Returns true if deallocation was successfully generated or successfully - deemed as not needed to be generated, false otherwise. + The `varType` parameter must be provided if the current type does not + capture the pointee type information. No deallocation is generated for + stack-based allocations (e.g., alloca). + + Returns true if deallocation was successfully generated or determined to + be unnecessary, false otherwise. }], /*retTy=*/"bool", /*methodName=*/"genFree", /*args=*/(ins "::mlir::OpBuilder &":$builder, "::mlir::Location":$loc, - "::mlir::TypedValue<::mlir::acc::PointerLikeType>":$varPtr, + "::mlir::TypedValue<::mlir::acc::PointerLikeType>":$varToFree, + "::mlir::Value":$allocRes, "::mlir::Type":$varType), /*methodBody=*/"", /*defaultImplementation=*/[{ @@ -274,6 +294,14 @@ def OpenACC_MappableTypeInterface : TypeInterface<"MappableType"> { The `initVal` can be empty - it is primarily needed for reductions to ensure the variable is also initialized with appropriate value. + The `needsDestroy` out-parameter is set by implementations to indicate + that destruction code must be generated after the returned private + variable usages, typically in the destroy region of recipe operations + (for example, when heap allocations or temporaries requiring cleanup + are created during initialization). When `needsDestroy` is set, callers + should invoke `generatePrivateDestroy` in the recipe's destroy region + with the privatized value returned by this method. + If the return value is empty, it means that recipe body was not successfully generated. }], @@ -284,12 +312,38 @@ def OpenACC_MappableTypeInterface : TypeInterface<"MappableType"> { "::mlir::TypedValue<::mlir::acc::MappableType>":$var, "::llvm::StringRef":$varName, "::mlir::ValueRange":$extents, - "::mlir::Value":$initVal), + "::mlir::Value":$initVal, + "bool &":$needsDestroy), /*methodBody=*/"", /*defaultImplementation=*/[{ return {}; }] >, + InterfaceMethod< + /*description=*/[{ + Generates destruction operations for a privatized value previously + produced by `generatePrivateInit`. This is typically inserted in a + recipe's destroy region, after all uses of the privatized value. + + The `privatized` value is the SSA value yielded by the init region + (and passed as the privatized argument to the destroy region). + Implementations should free heap-allocated storage or perform any + cleanup required for the given type. If no destruction is required, + this function should be a no-op and return `true`. + + Returns true if destruction was successfully generated or deemed not + necessary, false otherwise. + }], + /*retTy=*/"bool", + /*methodName=*/"generatePrivateDestroy", + /*args=*/(ins "::mlir::OpBuilder &":$builder, + "::mlir::Location":$loc, + "::mlir::Value":$privatized), + /*methodBody=*/"", + /*defaultImplementation=*/[{ + return true; + }] + >, ]; } diff --git a/mlir/include/mlir/IR/Remarks.h b/mlir/include/mlir/IR/Remarks.h index 20e84ec..9877926 100644 --- a/mlir/include/mlir/IR/Remarks.h +++ b/mlir/include/mlir/IR/Remarks.h @@ -18,7 +18,6 @@ #include "llvm/Remarks/Remark.h" #include "llvm/Support/FormatVariadic.h" #include "llvm/Support/Regex.h" -#include <optional> #include "mlir/IR/Diagnostics.h" #include "mlir/IR/MLIRContext.h" @@ -144,7 +143,7 @@ public: llvm::StringRef getCategoryName() const { return categoryName; } - llvm::StringRef getFullCategoryName() const { + llvm::StringRef getCombinedCategoryName() const { if (categoryName.empty() && subCategoryName.empty()) return {}; if (subCategoryName.empty()) @@ -318,7 +317,7 @@ private: }; //===----------------------------------------------------------------------===// -// MLIR Remark Streamer +// Pluggable Remark Utilities //===----------------------------------------------------------------------===// /// Base class for MLIR remark streamers that is used to stream @@ -338,6 +337,26 @@ public: virtual void finalize() {} // optional }; +using ReportFn = llvm::unique_function<void(const Remark &)>; + +/// Base class for MLIR remark emitting policies that is used to emit +/// optimization remarks to the underlying remark streamer. The derived classes +/// should implement the `reportRemark` method to provide the actual emitting +/// implementation. +class RemarkEmittingPolicyBase { +protected: + ReportFn reportImpl; + +public: + RemarkEmittingPolicyBase() = default; + virtual ~RemarkEmittingPolicyBase() = default; + + void initialize(ReportFn fn) { reportImpl = std::move(fn); } + + virtual void reportRemark(const Remark &remark) = 0; + virtual void finalize() = 0; +}; + //===----------------------------------------------------------------------===// // Remark Engine (MLIR Context will own this class) //===----------------------------------------------------------------------===// @@ -355,6 +374,8 @@ private: std::optional<llvm::Regex> failedFilter; /// The MLIR remark streamer that will be used to emit the remarks. std::unique_ptr<MLIRRemarkStreamerBase> remarkStreamer; + /// The MLIR remark policy that will be used to emit the remarks. + std::unique_ptr<RemarkEmittingPolicyBase> remarkEmittingPolicy; /// When is enabled, engine also prints remarks as mlir::emitRemarks. bool printAsEmitRemarks = false; @@ -392,6 +413,8 @@ private: InFlightRemark emitIfEnabled(Location loc, RemarkOpts opts, bool (RemarkEngine::*isEnabled)(StringRef) const); + /// Report a remark. + void reportImpl(const Remark &remark); public: /// Default constructor is deleted, use the other constructor. @@ -407,8 +430,15 @@ public: ~RemarkEngine(); /// Setup the remark engine with the given output path and format. - LogicalResult initialize(std::unique_ptr<MLIRRemarkStreamerBase> streamer, - std::string *errMsg); + LogicalResult + initialize(std::unique_ptr<MLIRRemarkStreamerBase> streamer, + std::unique_ptr<RemarkEmittingPolicyBase> remarkEmittingPolicy, + std::string *errMsg); + + /// Get the remark emitting policy. + RemarkEmittingPolicyBase *getRemarkEmittingPolicy() const { + return remarkEmittingPolicy.get(); + } /// Report a remark. void report(const Remark &&remark); @@ -446,6 +476,46 @@ inline InFlightRemark withEngine(Fn fn, Location loc, Args &&...args) { namespace mlir::remark { +//===----------------------------------------------------------------------===// +// Remark Emitting Policies +//===----------------------------------------------------------------------===// + +/// Policy that emits all remarks. +class RemarkEmittingPolicyAll : public detail::RemarkEmittingPolicyBase { +public: + RemarkEmittingPolicyAll(); + + void reportRemark(const detail::Remark &remark) override { + assert(reportImpl && "reportImpl is not set"); + reportImpl(remark); + } + void finalize() override {} +}; + +/// Policy that emits final remarks. +class RemarkEmittingPolicyFinal : public detail::RemarkEmittingPolicyBase { +private: + /// user can intercept them for custom processing via a registered callback, + /// otherwise they will be reported on engine destruction. + llvm::DenseSet<detail::Remark> postponedRemarks; + +public: + RemarkEmittingPolicyFinal(); + + void reportRemark(const detail::Remark &remark) override { + postponedRemarks.erase(remark); + postponedRemarks.insert(remark); + } + + void finalize() override { + assert(reportImpl && "reportImpl is not set"); + for (auto &remark : postponedRemarks) { + if (reportImpl) + reportImpl(remark); + } + } +}; + /// Create a Reason with llvm::formatv formatting. template <class... Ts> inline detail::LazyTextBuild reason(const char *fmt, Ts &&...ts) { @@ -505,16 +575,72 @@ inline detail::InFlightRemark analysis(Location loc, RemarkOpts opts) { /// Setup remarks for the context. This function will enable the remark engine /// and set the streamer to be used for optimization remarks. The remark -/// categories are used to filter the remarks that will be emitted by the remark -/// engine. If a category is not specified, it will not be emitted. If +/// categories are used to filter the remarks that will be emitted by the +/// remark engine. If a category is not specified, it will not be emitted. If /// `printAsEmitRemarks` is true, the remarks will be printed as /// mlir::emitRemarks. 'streamer' must inherit from MLIRRemarkStreamerBase and /// will be used to stream the remarks. LogicalResult enableOptimizationRemarks( MLIRContext &ctx, std::unique_ptr<remark::detail::MLIRRemarkStreamerBase> streamer, + std::unique_ptr<remark::detail::RemarkEmittingPolicyBase> + remarkEmittingPolicy, const remark::RemarkCategories &cats, bool printAsEmitRemarks = false); } // namespace mlir::remark +// DenseMapInfo specialization for Remark +namespace llvm { +template <> +struct DenseMapInfo<mlir::remark::detail::Remark> { + static constexpr StringRef kEmptyKey = "<EMPTY_KEY>"; + static constexpr StringRef kTombstoneKey = "<TOMBSTONE_KEY>"; + + /// Helper to provide a static dummy context for sentinel keys. + static mlir::MLIRContext *getStaticDummyContext() { + static mlir::MLIRContext dummyContext; + return &dummyContext; + } + + /// Create an empty remark + static inline mlir::remark::detail::Remark getEmptyKey() { + return mlir::remark::detail::Remark( + mlir::remark::RemarkKind::RemarkUnknown, mlir::DiagnosticSeverity::Note, + mlir::UnknownLoc::get(getStaticDummyContext()), + mlir::remark::RemarkOpts::name(kEmptyKey)); + } + + /// Create a dead remark + static inline mlir::remark::detail::Remark getTombstoneKey() { + return mlir::remark::detail::Remark( + mlir::remark::RemarkKind::RemarkUnknown, mlir::DiagnosticSeverity::Note, + mlir::UnknownLoc::get(getStaticDummyContext()), + mlir::remark::RemarkOpts::name(kTombstoneKey)); + } + + /// Compute the hash value of the remark + static unsigned getHashValue(const mlir::remark::detail::Remark &remark) { + return llvm::hash_combine( + remark.getLocation().getAsOpaquePointer(), + llvm::hash_value(remark.getRemarkName()), + llvm::hash_value(remark.getCombinedCategoryName())); + } + + static bool isEqual(const mlir::remark::detail::Remark &lhs, + const mlir::remark::detail::Remark &rhs) { + // Check for empty/tombstone keys first + if (lhs.getRemarkName() == kEmptyKey || + lhs.getRemarkName() == kTombstoneKey || + rhs.getRemarkName() == kEmptyKey || + rhs.getRemarkName() == kTombstoneKey) { + return lhs.getRemarkName() == rhs.getRemarkName(); + } + + // For regular remarks, compare key identifying fields + return lhs.getLocation() == rhs.getLocation() && + lhs.getRemarkName() == rhs.getRemarkName() && + lhs.getCombinedCategoryName() == rhs.getCombinedCategoryName(); + } +}; +} // namespace llvm #endif // MLIR_IR_REMARKS_H diff --git a/mlir/include/mlir/Interfaces/CMakeLists.txt b/mlir/include/mlir/Interfaces/CMakeLists.txt index a5feb59..72ed046 100644 --- a/mlir/include/mlir/Interfaces/CMakeLists.txt +++ b/mlir/include/mlir/Interfaces/CMakeLists.txt @@ -6,6 +6,7 @@ add_mlir_interface(DestinationStyleOpInterface) add_mlir_interface(FunctionInterfaces) add_mlir_interface(IndexingMapOpInterface) add_mlir_interface(InferIntRangeInterface) +add_mlir_interface(InferStridedMetadataInterface) add_mlir_interface(InferTypeOpInterface) add_mlir_interface(LoopLikeInterface) add_mlir_interface(MemOpInterfaces) diff --git a/mlir/include/mlir/Interfaces/InferIntRangeInterface.h b/mlir/include/mlir/Interfaces/InferIntRangeInterface.h index 0e107e8..a6de3d1 100644 --- a/mlir/include/mlir/Interfaces/InferIntRangeInterface.h +++ b/mlir/include/mlir/Interfaces/InferIntRangeInterface.h @@ -117,7 +117,8 @@ public: IntegerValueRange(ConstantIntRanges value) : value(std::move(value)) {} /// Create an integer value range lattice value. - IntegerValueRange(std::optional<ConstantIntRanges> value = std::nullopt) + explicit IntegerValueRange( + std::optional<ConstantIntRanges> value = std::nullopt) : value(std::move(value)) {} /// Whether the range is uninitialized. This happens when the state hasn't @@ -167,6 +168,15 @@ using SetIntRangeFn = using SetIntLatticeFn = llvm::function_ref<void(Value, const IntegerValueRange &)>; +/// Helper callback type to get the integer range of a value. +using GetIntRangeFn = function_ref<IntegerValueRange(Value)>; + +/// Helper function to collect the integer range values of an array of op fold +/// results. +SmallVector<IntegerValueRange> getIntValueRanges(ArrayRef<OpFoldResult> values, + GetIntRangeFn getIntRange, + int32_t indexBitwidth); + class InferIntRangeInterface; namespace intrange::detail { diff --git a/mlir/include/mlir/Interfaces/InferStridedMetadataInterface.h b/mlir/include/mlir/Interfaces/InferStridedMetadataInterface.h new file mode 100644 index 0000000..0c572e0 --- /dev/null +++ b/mlir/include/mlir/Interfaces/InferStridedMetadataInterface.h @@ -0,0 +1,145 @@ +//===- InferStridedMetadataInterface.h - Strided Metadata Inference -C++-*-===// +// +// 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 +// +//===----------------------------------------------------------------------===// +// +// This file contains definitions of the strided metadata inference interface +// defined in `InferStridedMetadataInterface.td` +// +//===----------------------------------------------------------------------===// + +#ifndef MLIR_INTERFACES_INFERSTRIDEDMETADATAINTERFACE_H +#define MLIR_INTERFACES_INFERSTRIDEDMETADATAINTERFACE_H + +#include "mlir/Interfaces/InferIntRangeInterface.h" + +namespace mlir { +/// A class that represents the strided metadata range information, including +/// offsets, sizes, and strides as integer ranges. +class StridedMetadataRange { +public: + /// Default constructor creates uninitialized ranges. + StridedMetadataRange() = default; + + /// Returns a ranked strided metadata range. + static StridedMetadataRange + getRanked(SmallVectorImpl<ConstantIntRanges> &&offsets, + SmallVectorImpl<ConstantIntRanges> &&sizes, + SmallVectorImpl<ConstantIntRanges> &&strides) { + return StridedMetadataRange(std::move(offsets), std::move(sizes), + std::move(strides)); + } + + /// Returns a strided metadata range with maximum ranges. + static StridedMetadataRange getMaxRanges(int32_t indexBitwidth, + int32_t offsetsRank, + int32_t sizeRank, + int32_t stridedRank) { + return StridedMetadataRange( + SmallVector<ConstantIntRanges>( + offsetsRank, ConstantIntRanges::maxRange(indexBitwidth)), + SmallVector<ConstantIntRanges>( + sizeRank, ConstantIntRanges::maxRange(indexBitwidth)), + SmallVector<ConstantIntRanges>( + stridedRank, ConstantIntRanges::maxRange(indexBitwidth))); + } + + static StridedMetadataRange getMaxRanges(int32_t indexBitwidth, + int32_t rank) { + return getMaxRanges(indexBitwidth, 1, rank, rank); + } + + /// Returns whether the metadata is uninitialized. + bool isUninitialized() const { return !offsets.has_value(); } + + /// Get the offsets range. + ArrayRef<ConstantIntRanges> getOffsets() const { + return offsets ? *offsets : ArrayRef<ConstantIntRanges>(); + } + MutableArrayRef<ConstantIntRanges> getOffsets() { + return offsets ? *offsets : MutableArrayRef<ConstantIntRanges>(); + } + + /// Get the sizes ranges. + ArrayRef<ConstantIntRanges> getSizes() const { return sizes; } + MutableArrayRef<ConstantIntRanges> getSizes() { return sizes; } + + /// Get the strides ranges. + ArrayRef<ConstantIntRanges> getStrides() const { return strides; } + MutableArrayRef<ConstantIntRanges> getStrides() { return strides; } + + /// Compare two strided metadata ranges. + bool operator==(const StridedMetadataRange &other) const { + return offsets == other.offsets && sizes == other.sizes && + strides == other.strides; + } + + /// Print the strided metadata range. + void print(raw_ostream &os) const; + + /// Join two strided metadata ranges, by taking the element-wise union of the + /// metadata. + static StridedMetadataRange join(const StridedMetadataRange &lhs, + const StridedMetadataRange &rhs) { + if (lhs.isUninitialized()) + return rhs; + if (rhs.isUninitialized()) + return lhs; + + // Helper fuction to compute the range union of constant ranges. + auto rangeUnion = + +[](const std::tuple<ConstantIntRanges, ConstantIntRanges> &lhsRhs) + -> ConstantIntRanges { + return std::get<0>(lhsRhs).rangeUnion(std::get<1>(lhsRhs)); + }; + + // Get the elementwise range union. Note, that `zip_equal` will assert if + // sizes are not equal. + SmallVector<ConstantIntRanges> offsets = llvm::map_to_vector( + llvm::zip_equal(*lhs.offsets, *rhs.offsets), rangeUnion); + SmallVector<ConstantIntRanges> sizes = + llvm::map_to_vector(llvm::zip_equal(lhs.sizes, rhs.sizes), rangeUnion); + SmallVector<ConstantIntRanges> strides = llvm::map_to_vector( + llvm::zip_equal(lhs.strides, rhs.strides), rangeUnion); + + // Return the joined metadata. + return StridedMetadataRange(std::move(offsets), std::move(sizes), + std::move(strides)); + } + +private: + /// Create a strided metadata range with the given offset, sizes, and strides. + StridedMetadataRange(SmallVectorImpl<ConstantIntRanges> &&offsets, + SmallVectorImpl<ConstantIntRanges> &&sizes, + SmallVectorImpl<ConstantIntRanges> &&strides) + : offsets(std::move(offsets)), sizes(std::move(sizes)), + strides(std::move(strides)) {} + + /// The offsets range. + std::optional<SmallVector<ConstantIntRanges>> offsets; + + /// The sizes ranges. + SmallVector<ConstantIntRanges> sizes; + + /// The strides ranges. + SmallVector<ConstantIntRanges> strides; +}; + +/// Print the strided metadata to `os`. +inline raw_ostream &operator<<(raw_ostream &os, + const StridedMetadataRange &range) { + range.print(os); + return os; +} + +/// Callback function type for setting the strided metadata of a value. +using SetStridedMetadataRangeFn = + function_ref<void(Value, const StridedMetadataRange &)>; +} // end namespace mlir + +#include "mlir/Interfaces/InferStridedMetadataInterface.h.inc" + +#endif // MLIR_INTERFACES_INFERSTRIDEDMETADATAINTERFACE_H diff --git a/mlir/include/mlir/Interfaces/InferStridedMetadataInterface.td b/mlir/include/mlir/Interfaces/InferStridedMetadataInterface.td new file mode 100644 index 0000000..ee5b094 --- /dev/null +++ b/mlir/include/mlir/Interfaces/InferStridedMetadataInterface.td @@ -0,0 +1,45 @@ +//===- InferStridedMetadataInterface.td - Strided MD Inference ----------*-===// +// +// 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 +// +//===----------------------------------------------------------------------===// +// +// Defines the interface for strided metadata range analysis +// +//===----------------------------------------------------------------------===// + +#ifndef MLIR_INTERFACES_INFERSTRIDEDMETADATAINTERFACE +#define MLIR_INTERFACES_INFERSTRIDEDMETADATAINTERFACE + +include "mlir/IR/OpBase.td" + +def InferStridedMetadataOpInterface : + OpInterface<"InferStridedMetadataOpInterface"> { + let description = [{ + Allows operations to participate in strided metadata analysis by providing + methods that allow them to specify bounds on offsets, sizes, and strides + of their result(s) given bounds on their input(s) if known. + }]; + let cppNamespace = "::mlir"; + + let methods = [ + InterfaceMethod<[{ + Infer the strided metadata bounds on the results of this op given + the bounds on its operands. + For each result value or block argument of interest, the method should + call `setMetadata` with that `Value` as an argument. + The `operands` parameter contains the strided metadata ranges for all the + operands of the operation in order. + The `getIntRange` callback is provided for obtaining the int-range + analysis result for a given value. + }], + "void", "inferStridedMetadataRanges", + (ins "::llvm::ArrayRef<::mlir::StridedMetadataRange>":$operands, + "::mlir::GetIntRangeFn":$getIntRange, + "::mlir::SetStridedMetadataRangeFn":$setMetadata, + "int32_t":$indexBitwidth)> + ]; +} +#endif // MLIR_INTERFACES_INFERSTRIDEDMETADATAINTERFACE diff --git a/mlir/include/mlir/Remark/RemarkStreamer.h b/mlir/include/mlir/Remark/RemarkStreamer.h index 170d6b4..19a70fa 100644 --- a/mlir/include/mlir/Remark/RemarkStreamer.h +++ b/mlir/include/mlir/Remark/RemarkStreamer.h @@ -45,6 +45,7 @@ namespace mlir::remark { /// mlir::emitRemarks. LogicalResult enableOptimizationRemarksWithLLVMStreamer( MLIRContext &ctx, StringRef filePath, llvm::remarks::Format fmt, + std::unique_ptr<detail::RemarkEmittingPolicyBase> remarkEmittingPolicy, const RemarkCategories &cat, bool printAsEmitRemarks = false); } // namespace mlir::remark diff --git a/mlir/include/mlir/Tools/mlir-opt/MlirOptMain.h b/mlir/include/mlir/Tools/mlir-opt/MlirOptMain.h index 0fbe15f..b739438 100644 --- a/mlir/include/mlir/Tools/mlir-opt/MlirOptMain.h +++ b/mlir/include/mlir/Tools/mlir-opt/MlirOptMain.h @@ -44,6 +44,11 @@ enum class RemarkFormat { REMARK_FORMAT_BITSTREAM, }; +enum class RemarkPolicy { + REMARK_POLICY_ALL, + REMARK_POLICY_FINAL, +}; + /// Configuration options for the mlir-opt tool. /// This is intended to help building tools like mlir-opt by collecting the /// supported options. @@ -242,6 +247,8 @@ public: /// Set the reproducer output filename RemarkFormat getRemarkFormat() const { return remarkFormatFlag; } + /// Set the remark policy to use. + RemarkPolicy getRemarkPolicy() const { return remarkPolicyFlag; } /// Set the remark format to use. std::string getRemarksAllFilter() const { return remarksAllFilterFlag; } /// Set the remark output file. @@ -265,6 +272,8 @@ protected: /// Remark format RemarkFormat remarkFormatFlag = RemarkFormat::REMARK_FORMAT_STDOUT; + /// Remark policy + RemarkPolicy remarkPolicyFlag = RemarkPolicy::REMARK_POLICY_ALL; /// Remark file to output to std::string remarksOutputFileFlag = ""; /// Remark filters diff --git a/mlir/lib/Analysis/CMakeLists.txt b/mlir/lib/Analysis/CMakeLists.txt index 609cb34..db10ebc 100644 --- a/mlir/lib/Analysis/CMakeLists.txt +++ b/mlir/lib/Analysis/CMakeLists.txt @@ -40,6 +40,7 @@ add_mlir_library(MLIRAnalysis DataFlow/IntegerRangeAnalysis.cpp DataFlow/LivenessAnalysis.cpp DataFlow/SparseAnalysis.cpp + DataFlow/StridedMetadataRangeAnalysis.cpp ADDITIONAL_HEADER_DIRS ${MLIR_MAIN_INCLUDE_DIR}/mlir/Analysis @@ -53,6 +54,7 @@ add_mlir_library(MLIRAnalysis MLIRDataLayoutInterfaces MLIRFunctionInterfaces MLIRInferIntRangeInterface + MLIRInferStridedMetadataInterface MLIRInferTypeOpInterface MLIRLoopLikeInterface MLIRPresburger diff --git a/mlir/lib/Analysis/DataFlow/StridedMetadataRangeAnalysis.cpp b/mlir/lib/Analysis/DataFlow/StridedMetadataRangeAnalysis.cpp new file mode 100644 index 0000000..01c9daf --- /dev/null +++ b/mlir/lib/Analysis/DataFlow/StridedMetadataRangeAnalysis.cpp @@ -0,0 +1,127 @@ +//===- StridedMetadataRangeAnalysis.cpp - Integer range analysis --------*- C++ +//-*-===// +// +// 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 +// +//===----------------------------------------------------------------------===// +// +// This file defines the dataflow analysis class for integer range inference +// which is used in transformations over the `arith` dialect such as +// branch elimination or signed->unsigned rewriting +// +//===----------------------------------------------------------------------===// + +#include "mlir/Analysis/DataFlow/StridedMetadataRangeAnalysis.h" +#include "mlir/Analysis/DataFlow/IntegerRangeAnalysis.h" +#include "mlir/Dialect/Utils/IndexingUtils.h" +#include "mlir/IR/Operation.h" +#include "mlir/IR/Value.h" +#include "mlir/Support/DebugStringHelper.h" +#include "llvm/Support/Debug.h" +#include "llvm/Support/DebugLog.h" + +#define DEBUG_TYPE "strided-metadata-range-analysis" + +using namespace mlir; +using namespace mlir::dataflow; + +/// Get the entry state for a value. For any value that is not a ranked memref, +/// this function sets the metadata to a top state with no offsets, sizes, or +/// strides. For `memref` types, this function will use the metadata in the type +/// to try to deduce as much informaiton as possible. +static StridedMetadataRange getEntryStateImpl(Value v, int32_t indexBitwidth) { + // TODO: generalize this method with a type interface. + auto mTy = dyn_cast<BaseMemRefType>(v.getType()); + + // If not a memref or it's un-ranked, don't infer any metadata. + if (!mTy || !mTy.hasRank()) + return StridedMetadataRange::getMaxRanges(indexBitwidth, 0, 0, 0); + + // Get the top state. + auto metadata = + StridedMetadataRange::getMaxRanges(indexBitwidth, mTy.getRank()); + + // Compute the offset and strides. + int64_t offset; + SmallVector<int64_t> strides; + if (failed(cast<MemRefType>(mTy).getStridesAndOffset(strides, offset))) + return metadata; + + // Refine the metadata if we know it from the type. + if (!ShapedType::isDynamic(offset)) { + metadata.getOffsets()[0] = + ConstantIntRanges::constant(APInt(indexBitwidth, offset)); + } + for (auto &&[size, range] : + llvm::zip_equal(mTy.getShape(), metadata.getSizes())) { + if (ShapedType::isDynamic(size)) + continue; + range = ConstantIntRanges::constant(APInt(indexBitwidth, size)); + } + for (auto &&[stride, range] : + llvm::zip_equal(strides, metadata.getStrides())) { + if (ShapedType::isDynamic(stride)) + continue; + range = ConstantIntRanges::constant(APInt(indexBitwidth, stride)); + } + + return metadata; +} + +StridedMetadataRangeAnalysis::StridedMetadataRangeAnalysis( + DataFlowSolver &solver, int32_t indexBitwidth) + : SparseForwardDataFlowAnalysis(solver), indexBitwidth(indexBitwidth) { + assert(indexBitwidth > 0 && "invalid bitwidth"); +} + +void StridedMetadataRangeAnalysis::setToEntryState( + StridedMetadataRangeLattice *lattice) { + propagateIfChanged(lattice, lattice->join(getEntryStateImpl( + lattice->getAnchor(), indexBitwidth))); +} + +LogicalResult StridedMetadataRangeAnalysis::visitOperation( + Operation *op, ArrayRef<const StridedMetadataRangeLattice *> operands, + ArrayRef<StridedMetadataRangeLattice *> results) { + auto inferrable = dyn_cast<InferStridedMetadataOpInterface>(op); + + // Bail if we cannot reason about the op. + if (!inferrable) { + setAllToEntryStates(results); + return success(); + } + + LDBG() << "Inferring metadata for: " + << OpWithFlags(op, OpPrintingFlags().skipRegions()); + + // Helper function to retrieve int range values. + auto getIntRange = [&](Value value) -> IntegerValueRange { + auto lattice = getOrCreateFor<IntegerValueRangeLattice>( + getProgramPointAfter(op), value); + return lattice ? lattice->getValue() : IntegerValueRange(); + }; + + // Convert the arguments lattices to a vector. + SmallVector<StridedMetadataRange> argRanges = llvm::map_to_vector( + operands, [](const StridedMetadataRangeLattice *lattice) { + return lattice->getValue(); + }); + + // Callback to set metadata on a result. + auto joinCallback = [&](Value v, const StridedMetadataRange &md) { + auto result = cast<OpResult>(v); + assert(llvm::is_contained(op->getResults(), result)); + LDBG() << "- Inferred metadata: " << md; + StridedMetadataRangeLattice *lattice = results[result.getResultNumber()]; + ChangeResult changed = lattice->join(md); + LDBG() << "- Joined metadata: " << lattice->getValue(); + propagateIfChanged(lattice, changed); + }; + + // Infer the metadata. + inferrable.inferStridedMetadataRanges(argRanges, getIntRange, joinCallback, + indexBitwidth); + return success(); +} diff --git a/mlir/lib/Dialect/MemRef/IR/CMakeLists.txt b/mlir/lib/Dialect/MemRef/IR/CMakeLists.txt index e25a012..1382c7ac 100644 --- a/mlir/lib/Dialect/MemRef/IR/CMakeLists.txt +++ b/mlir/lib/Dialect/MemRef/IR/CMakeLists.txt @@ -5,7 +5,7 @@ add_mlir_dialect_library(MLIRMemRefDialect ValueBoundsOpInterfaceImpl.cpp ADDITIONAL_HEADER_DIRS - ${PROJECT_SOURCE_DIR}/inlude/mlir/Dialect/MemRefDialect + ${PROJECT_SOURCE_DIR}/inlude/mlir/Dialect/MemRef/IR DEPENDS MLIRMemRefOpsIncGen @@ -18,6 +18,7 @@ add_mlir_dialect_library(MLIRMemRefDialect MLIRDialectUtils MLIRInferIntRangeCommon MLIRInferIntRangeInterface + MLIRInferStridedMetadataInterface MLIRInferTypeOpInterface MLIRIR MLIRMemOpInterfaces diff --git a/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp b/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp index e9bdcda..507597b 100644 --- a/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp +++ b/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp @@ -3437,6 +3437,65 @@ SubViewOp::bubbleDownCasts(OpBuilder &builder) { return bubbleDownCastsPassthroughOpImpl(*this, builder, getSourceMutable()); } +void SubViewOp::inferStridedMetadataRanges( + ArrayRef<StridedMetadataRange> ranges, GetIntRangeFn getIntRange, + SetStridedMetadataRangeFn setMetadata, int32_t indexBitwidth) { + auto isUninitialized = + +[](IntegerValueRange range) { return range.isUninitialized(); }; + + // Bail early if any of the operands metadata is not ready: + SmallVector<IntegerValueRange> offsetOperands = + getIntValueRanges(getMixedOffsets(), getIntRange, indexBitwidth); + if (llvm::any_of(offsetOperands, isUninitialized)) + return; + + SmallVector<IntegerValueRange> sizeOperands = + getIntValueRanges(getMixedSizes(), getIntRange, indexBitwidth); + if (llvm::any_of(sizeOperands, isUninitialized)) + return; + + SmallVector<IntegerValueRange> stridesOperands = + getIntValueRanges(getMixedStrides(), getIntRange, indexBitwidth); + if (llvm::any_of(stridesOperands, isUninitialized)) + return; + + StridedMetadataRange sourceRange = + ranges[getSourceMutable().getOperandNumber()]; + if (sourceRange.isUninitialized()) + return; + + ArrayRef<ConstantIntRanges> srcStrides = sourceRange.getStrides(); + + // Get the dropped dims. + llvm::SmallBitVector droppedDims = getDroppedDims(); + + // Compute the new offset, strides and sizes. + ConstantIntRanges offset = sourceRange.getOffsets()[0]; + SmallVector<ConstantIntRanges> strides, sizes; + + for (size_t i = 0, e = droppedDims.size(); i < e; ++i) { + bool dropped = droppedDims.test(i); + // Compute the new offset. + ConstantIntRanges off = + intrange::inferMul({offsetOperands[i].getValue(), srcStrides[i]}); + offset = intrange::inferAdd({offset, off}); + + // Skip dropped dimensions. + if (dropped) + continue; + // Multiply the strides. + strides.push_back( + intrange::inferMul({stridesOperands[i].getValue(), srcStrides[i]})); + // Get the sizes. + sizes.push_back(sizeOperands[i].getValue()); + } + + setMetadata(getResult(), + StridedMetadataRange::getRanked( + SmallVector<ConstantIntRanges>({std::move(offset)}), + std::move(sizes), std::move(strides))); +} + //===----------------------------------------------------------------------===// // TransposeOp //===----------------------------------------------------------------------===// diff --git a/mlir/lib/Dialect/OpenACC/IR/OpenACC.cpp b/mlir/lib/Dialect/OpenACC/IR/OpenACC.cpp index 6564a4e..642ced9 100644 --- a/mlir/lib/Dialect/OpenACC/IR/OpenACC.cpp +++ b/mlir/lib/Dialect/OpenACC/IR/OpenACC.cpp @@ -17,6 +17,7 @@ #include "mlir/IR/DialectImplementation.h" #include "mlir/IR/Matchers.h" #include "mlir/IR/OpImplementation.h" +#include "mlir/IR/SymbolTable.h" #include "mlir/Support/LLVM.h" #include "mlir/Transforms/DialectConversion.h" #include "llvm/ADT/SmallSet.h" @@ -74,14 +75,16 @@ struct MemRefPointerLikeModel } mlir::Value genAllocate(Type pointer, OpBuilder &builder, Location loc, - StringRef varName, Type varType, - Value originalVar) const { + StringRef varName, Type varType, Value originalVar, + bool &needsFree) const { auto memrefTy = cast<MemRefType>(pointer); // Check if this is a static memref (all dimensions are known) - if yes // then we can generate an alloca operation. - if (memrefTy.hasStaticShape()) + if (memrefTy.hasStaticShape()) { + needsFree = false; // alloca doesn't need deallocation return memref::AllocaOp::create(builder, loc, memrefTy).getResult(); + } // For dynamic memrefs, extract sizes from the original variable if // provided. Otherwise they cannot be handled. @@ -99,6 +102,7 @@ struct MemRefPointerLikeModel // Note: We only add dynamic sizes to the dynamicSizes array // Static dimensions are handled automatically by AllocOp } + needsFree = true; // alloc needs deallocation return memref::AllocOp::create(builder, loc, memrefTy, dynamicSizes) .getResult(); } @@ -108,10 +112,14 @@ struct MemRefPointerLikeModel } bool genFree(Type pointer, OpBuilder &builder, Location loc, - TypedValue<PointerLikeType> varPtr, Type varType) const { - if (auto memrefValue = dyn_cast<TypedValue<MemRefType>>(varPtr)) { + TypedValue<PointerLikeType> varToFree, Value allocRes, + Type varType) const { + if (auto memrefValue = dyn_cast<TypedValue<MemRefType>>(varToFree)) { + // Use allocRes if provided to determine the allocation type + Value valueToInspect = allocRes ? allocRes : memrefValue; + // Walk through casts to find the original allocation - Value currentValue = memrefValue; + Value currentValue = valueToInspect; Operation *originalAlloc = nullptr; // Follow the chain of operations to find the original allocation @@ -150,7 +158,7 @@ struct MemRefPointerLikeModel return true; } if (isa<memref::AllocOp>(originalAlloc)) { - // This is an alloc - generate dealloc + // This is an alloc - generate dealloc on varToFree memref::DeallocOp::create(builder, loc, memrefValue); return true; } @@ -1003,6 +1011,142 @@ struct RemoveConstantIfConditionWithRegion : public OpRewritePattern<OpTy> { } }; +//===----------------------------------------------------------------------===// +// Recipe Region Helpers +//===----------------------------------------------------------------------===// + +/// Create and populate an init region for privatization recipes. +/// Returns the init block on success, or nullptr on failure. +/// Sets needsFree to indicate if the allocated memory requires deallocation. +static std::unique_ptr<Block> createInitRegion(OpBuilder &builder, Location loc, + Type varType, StringRef varName, + ValueRange bounds, + bool &needsFree) { + // Create init block with arguments: original value + bounds + SmallVector<Type> argTypes{varType}; + SmallVector<Location> argLocs{loc}; + for (Value bound : bounds) { + argTypes.push_back(bound.getType()); + argLocs.push_back(loc); + } + + auto initBlock = std::make_unique<Block>(); + initBlock->addArguments(argTypes, argLocs); + builder.setInsertionPointToStart(initBlock.get()); + + Value privatizedValue; + + // Get the block argument that represents the original variable + Value blockArgVar = initBlock->getArgument(0); + + // Generate init region body based on variable type + if (isa<MappableType>(varType)) { + auto mappableTy = cast<MappableType>(varType); + auto typedVar = cast<TypedValue<MappableType>>(blockArgVar); + privatizedValue = mappableTy.generatePrivateInit( + builder, loc, typedVar, varName, bounds, {}, needsFree); + if (!privatizedValue) + return nullptr; + } else { + assert(isa<PointerLikeType>(varType) && "Expected PointerLikeType"); + auto pointerLikeTy = cast<PointerLikeType>(varType); + // Use PointerLikeType's allocation API with the block argument + privatizedValue = pointerLikeTy.genAllocate(builder, loc, varName, varType, + blockArgVar, needsFree); + if (!privatizedValue) + return nullptr; + } + + // Add yield operation to init block + acc::YieldOp::create(builder, loc, privatizedValue); + + return initBlock; +} + +/// Create and populate a copy region for firstprivate recipes. +/// Returns the copy block on success, or nullptr on failure. +/// TODO: Handle MappableType - it does not yet have a copy API. +static std::unique_ptr<Block> createCopyRegion(OpBuilder &builder, Location loc, + Type varType, + ValueRange bounds) { + // Create copy block with arguments: original value + privatized value + + // bounds + SmallVector<Type> copyArgTypes{varType, varType}; + SmallVector<Location> copyArgLocs{loc, loc}; + for (Value bound : bounds) { + copyArgTypes.push_back(bound.getType()); + copyArgLocs.push_back(loc); + } + + auto copyBlock = std::make_unique<Block>(); + copyBlock->addArguments(copyArgTypes, copyArgLocs); + builder.setInsertionPointToStart(copyBlock.get()); + + bool isMappable = isa<MappableType>(varType); + bool isPointerLike = isa<PointerLikeType>(varType); + // TODO: Handle MappableType - it does not yet have a copy API. + // Otherwise, for now just fallback to pointer-like behavior. + if (isMappable && !isPointerLike) + return nullptr; + + // Generate copy region body based on variable type + if (isPointerLike) { + auto pointerLikeTy = cast<PointerLikeType>(varType); + Value originalArg = copyBlock->getArgument(0); + Value privatizedArg = copyBlock->getArgument(1); + + // Generate copy operation using PointerLikeType interface + if (!pointerLikeTy.genCopy( + builder, loc, cast<TypedValue<PointerLikeType>>(privatizedArg), + cast<TypedValue<PointerLikeType>>(originalArg), varType)) + return nullptr; + } + + // Add terminator to copy block + acc::TerminatorOp::create(builder, loc); + + return copyBlock; +} + +/// Create and populate a destroy region for privatization recipes. +/// Returns the destroy block on success, or nullptr if not needed. +static std::unique_ptr<Block> createDestroyRegion(OpBuilder &builder, + Location loc, Type varType, + Value allocRes, + ValueRange bounds) { + // Create destroy block with arguments: original value + privatized value + + // bounds + SmallVector<Type> destroyArgTypes{varType, varType}; + SmallVector<Location> destroyArgLocs{loc, loc}; + for (Value bound : bounds) { + destroyArgTypes.push_back(bound.getType()); + destroyArgLocs.push_back(loc); + } + + auto destroyBlock = std::make_unique<Block>(); + destroyBlock->addArguments(destroyArgTypes, destroyArgLocs); + builder.setInsertionPointToStart(destroyBlock.get()); + + bool isMappable = isa<MappableType>(varType); + bool isPointerLike = isa<PointerLikeType>(varType); + // TODO: Handle MappableType - it does not yet have a deallocation API. + // Otherwise, for now just fallback to pointer-like behavior. + if (isMappable && !isPointerLike) + return nullptr; + + assert(isa<PointerLikeType>(varType) && "Expected PointerLikeType"); + auto pointerLikeTy = cast<PointerLikeType>(varType); + auto privatizedArg = + cast<TypedValue<PointerLikeType>>(destroyBlock->getArgument(1)); + // Pass allocRes to help determine the allocation type + if (!pointerLikeTy.genFree(builder, loc, privatizedArg, allocRes, varType)) + return nullptr; + + acc::TerminatorOp::create(builder, loc); + + return destroyBlock; +} + } // namespace //===----------------------------------------------------------------------===// @@ -1050,6 +1194,55 @@ LogicalResult acc::PrivateRecipeOp::verifyRegions() { return success(); } +std::optional<PrivateRecipeOp> +PrivateRecipeOp::createAndPopulate(OpBuilder &builder, Location loc, + StringRef recipeName, Type varType, + StringRef varName, ValueRange bounds) { + // First, validate that we can handle this variable type + bool isMappable = isa<MappableType>(varType); + bool isPointerLike = isa<PointerLikeType>(varType); + + // Unsupported type + if (!isMappable && !isPointerLike) + return std::nullopt; + + // Create init and destroy blocks using shared helpers + OpBuilder::InsertionGuard guard(builder); + + // Save the original insertion point for creating the recipe operation later + auto originalInsertionPoint = builder.saveInsertionPoint(); + + bool needsFree = false; + auto initBlock = + createInitRegion(builder, loc, varType, varName, bounds, needsFree); + if (!initBlock) + return std::nullopt; + + // Only create destroy region if the allocation needs deallocation + std::unique_ptr<Block> destroyBlock; + if (needsFree) { + // Extract the allocated value from the init block's yield operation + auto yieldOp = cast<acc::YieldOp>(initBlock->getTerminator()); + Value allocRes = yieldOp.getOperand(0); + + destroyBlock = createDestroyRegion(builder, loc, varType, allocRes, bounds); + if (!destroyBlock) + return std::nullopt; + } + + // Now create the recipe operation at the original insertion point and attach + // the blocks + builder.restoreInsertionPoint(originalInsertionPoint); + auto recipe = PrivateRecipeOp::create(builder, loc, recipeName, varType); + + // Move the blocks into the recipe's regions + recipe.getInitRegion().push_back(initBlock.release()); + if (destroyBlock) + recipe.getDestroyRegion().push_back(destroyBlock.release()); + + return recipe; +} + //===----------------------------------------------------------------------===// // FirstprivateRecipeOp //===----------------------------------------------------------------------===// @@ -1080,6 +1273,60 @@ LogicalResult acc::FirstprivateRecipeOp::verifyRegions() { return success(); } +std::optional<FirstprivateRecipeOp> +FirstprivateRecipeOp::createAndPopulate(OpBuilder &builder, Location loc, + StringRef recipeName, Type varType, + StringRef varName, ValueRange bounds) { + // First, validate that we can handle this variable type + bool isMappable = isa<MappableType>(varType); + bool isPointerLike = isa<PointerLikeType>(varType); + + // Unsupported type + if (!isMappable && !isPointerLike) + return std::nullopt; + + // Create init, copy, and destroy blocks using shared helpers + OpBuilder::InsertionGuard guard(builder); + + // Save the original insertion point for creating the recipe operation later + auto originalInsertionPoint = builder.saveInsertionPoint(); + + bool needsFree = false; + auto initBlock = + createInitRegion(builder, loc, varType, varName, bounds, needsFree); + if (!initBlock) + return std::nullopt; + + auto copyBlock = createCopyRegion(builder, loc, varType, bounds); + if (!copyBlock) + return std::nullopt; + + // Only create destroy region if the allocation needs deallocation + std::unique_ptr<Block> destroyBlock; + if (needsFree) { + // Extract the allocated value from the init block's yield operation + auto yieldOp = cast<acc::YieldOp>(initBlock->getTerminator()); + Value allocRes = yieldOp.getOperand(0); + + destroyBlock = createDestroyRegion(builder, loc, varType, allocRes, bounds); + if (!destroyBlock) + return std::nullopt; + } + + // Now create the recipe operation at the original insertion point and attach + // the blocks + builder.restoreInsertionPoint(originalInsertionPoint); + auto recipe = FirstprivateRecipeOp::create(builder, loc, recipeName, varType); + + // Move the blocks into the recipe's regions + recipe.getInitRegion().push_back(initBlock.release()); + recipe.getCopyRegion().push_back(copyBlock.release()); + if (destroyBlock) + recipe.getDestroyRegion().push_back(destroyBlock.release()); + + return recipe; +} + //===----------------------------------------------------------------------===// // ReductionRecipeOp //===----------------------------------------------------------------------===// diff --git a/mlir/lib/IR/MLIRContext.cpp b/mlir/lib/IR/MLIRContext.cpp index 1fa04ed..89b81cf 100644 --- a/mlir/lib/IR/MLIRContext.cpp +++ b/mlir/lib/IR/MLIRContext.cpp @@ -121,6 +121,11 @@ namespace mlir { class MLIRContextImpl { public: //===--------------------------------------------------------------------===// + // Remark + //===--------------------------------------------------------------------===// + std::unique_ptr<remark::detail::RemarkEngine> remarkEngine; + + //===--------------------------------------------------------------------===// // Debugging //===--------------------------------------------------------------------===// @@ -135,11 +140,6 @@ public: DiagnosticEngine diagEngine; //===--------------------------------------------------------------------===// - // Remark - //===--------------------------------------------------------------------===// - std::unique_ptr<remark::detail::RemarkEngine> remarkEngine; - - //===--------------------------------------------------------------------===// // Options //===--------------------------------------------------------------------===// @@ -357,7 +357,10 @@ MLIRContext::MLIRContext(const DialectRegistry ®istry, Threading setting) impl->affineUniquer.registerParametricStorageType<IntegerSetStorage>(); } -MLIRContext::~MLIRContext() = default; +MLIRContext::~MLIRContext() { + // finalize remark engine before destroying anything else. + impl->remarkEngine.reset(); +} /// Copy the specified array of elements into memory managed by the provided /// bump pointer allocator. This assumes the elements are all PODs. diff --git a/mlir/lib/IR/Remarks.cpp b/mlir/lib/IR/Remarks.cpp index a55f61a..031eae2 100644 --- a/mlir/lib/IR/Remarks.cpp +++ b/mlir/lib/IR/Remarks.cpp @@ -16,7 +16,7 @@ #include "llvm/ADT/StringRef.h" using namespace mlir::remark::detail; - +using namespace mlir::remark; //------------------------------------------------------------------------------ // Remark //------------------------------------------------------------------------------ @@ -70,7 +70,7 @@ static void printArgs(llvm::raw_ostream &os, llvm::ArrayRef<Remark::Arg> args) { void Remark::print(llvm::raw_ostream &os, bool printLocation) const { // Header: [Type] pass:remarkName StringRef type = getRemarkTypeString(); - StringRef categoryName = getFullCategoryName(); + StringRef categoryName = getCombinedCategoryName(); StringRef name = remarkName; os << '[' << type << "] "; @@ -81,9 +81,10 @@ void Remark::print(llvm::raw_ostream &os, bool printLocation) const { os << "Function=" << getFunction() << " | "; if (printLocation) { - if (auto flc = mlir::dyn_cast<mlir::FileLineColLoc>(getLocation())) + if (auto flc = mlir::dyn_cast<mlir::FileLineColLoc>(getLocation())) { os << " @" << flc.getFilename() << ":" << flc.getLine() << ":" << flc.getColumn(); + } } printArgs(os, getArgs()); @@ -140,7 +141,7 @@ llvm::remarks::Remark Remark::generateRemark() const { r.RemarkType = getRemarkType(); r.RemarkName = getRemarkName(); // MLIR does not use passes; instead, it has categories and sub-categories. - r.PassName = getFullCategoryName(); + r.PassName = getCombinedCategoryName(); r.FunctionName = getFunction(); r.Loc = locLambda(); for (const Remark::Arg &arg : getArgs()) { @@ -225,26 +226,42 @@ InFlightRemark RemarkEngine::emitOptimizationRemarkAnalysis(Location loc, // RemarkEngine //===----------------------------------------------------------------------===// -void RemarkEngine::report(const Remark &&remark) { +void RemarkEngine::reportImpl(const Remark &remark) { // Stream the remark - if (remarkStreamer) + if (remarkStreamer) { remarkStreamer->streamOptimizationRemark(remark); + } // Print using MLIR's diagnostic if (printAsEmitRemarks) emitRemark(remark.getLocation(), remark.getMsg()); } +void RemarkEngine::report(const Remark &&remark) { + if (remarkEmittingPolicy) + remarkEmittingPolicy->reportRemark(remark); +} + RemarkEngine::~RemarkEngine() { + if (remarkEmittingPolicy) + remarkEmittingPolicy->finalize(); + if (remarkStreamer) remarkStreamer->finalize(); } -llvm::LogicalResult -RemarkEngine::initialize(std::unique_ptr<MLIRRemarkStreamerBase> streamer, - std::string *errMsg) { - // If you need to validate categories/filters, do so here and set errMsg. +llvm::LogicalResult RemarkEngine::initialize( + std::unique_ptr<MLIRRemarkStreamerBase> streamer, + std::unique_ptr<RemarkEmittingPolicyBase> remarkEmittingPolicy, + std::string *errMsg) { + remarkStreamer = std::move(streamer); + + auto reportFunc = + std::bind(&RemarkEngine::reportImpl, this, std::placeholders::_1); + remarkEmittingPolicy->initialize(ReportFn(std::move(reportFunc))); + + this->remarkEmittingPolicy = std::move(remarkEmittingPolicy); return success(); } @@ -301,14 +318,15 @@ RemarkEngine::RemarkEngine(bool printAsEmitRemarks, } llvm::LogicalResult mlir::remark::enableOptimizationRemarks( - MLIRContext &ctx, - std::unique_ptr<remark::detail::MLIRRemarkStreamerBase> streamer, - const remark::RemarkCategories &cats, bool printAsEmitRemarks) { + MLIRContext &ctx, std::unique_ptr<detail::MLIRRemarkStreamerBase> streamer, + std::unique_ptr<detail::RemarkEmittingPolicyBase> remarkEmittingPolicy, + const RemarkCategories &cats, bool printAsEmitRemarks) { auto engine = - std::make_unique<remark::detail::RemarkEngine>(printAsEmitRemarks, cats); + std::make_unique<detail::RemarkEngine>(printAsEmitRemarks, cats); std::string errMsg; - if (failed(engine->initialize(std::move(streamer), &errMsg))) { + if (failed(engine->initialize(std::move(streamer), + std::move(remarkEmittingPolicy), &errMsg))) { llvm::report_fatal_error( llvm::Twine("Failed to initialize remark engine. Error: ") + errMsg); } @@ -316,3 +334,12 @@ llvm::LogicalResult mlir::remark::enableOptimizationRemarks( return success(); } + +//===----------------------------------------------------------------------===// +// Remark emitting policies +//===----------------------------------------------------------------------===// + +namespace mlir::remark { +RemarkEmittingPolicyAll::RemarkEmittingPolicyAll() = default; +RemarkEmittingPolicyFinal::RemarkEmittingPolicyFinal() = default; +} // namespace mlir::remark diff --git a/mlir/lib/Interfaces/CMakeLists.txt b/mlir/lib/Interfaces/CMakeLists.txt index 388de1c..f96af02 100644 --- a/mlir/lib/Interfaces/CMakeLists.txt +++ b/mlir/lib/Interfaces/CMakeLists.txt @@ -9,6 +9,7 @@ set(LLVM_OPTIONAL_SOURCES FunctionInterfaces.cpp IndexingMapOpInterface.cpp InferIntRangeInterface.cpp + InferStridedMetadataInterface.cpp InferTypeOpInterface.cpp LoopLikeInterface.cpp MemOpInterfaces.cpp @@ -64,6 +65,21 @@ add_mlir_library(MLIRFunctionInterfaces add_mlir_interface_library(IndexingMapOpInterface) add_mlir_interface_library(InferIntRangeInterface) + +add_mlir_library(MLIRInferStridedMetadataInterface + InferStridedMetadataInterface.cpp + + ADDITIONAL_HEADER_DIRS + ${MLIR_MAIN_INCLUDE_DIR}/mlir/Interfaces + + DEPENDS + MLIRInferStridedMetadataInterfaceIncGen + + LINK_LIBS PUBLIC + MLIRInferIntRangeInterface + MLIRIR +) + add_mlir_interface_library(InferTypeOpInterface) add_mlir_library(MLIRLoopLikeInterface diff --git a/mlir/lib/Interfaces/InferIntRangeInterface.cpp b/mlir/lib/Interfaces/InferIntRangeInterface.cpp index 9f3e97d..84fc9b8 100644 --- a/mlir/lib/Interfaces/InferIntRangeInterface.cpp +++ b/mlir/lib/Interfaces/InferIntRangeInterface.cpp @@ -146,6 +146,25 @@ raw_ostream &mlir::operator<<(raw_ostream &os, const IntegerValueRange &range) { return os; } +SmallVector<IntegerValueRange> +mlir::getIntValueRanges(ArrayRef<OpFoldResult> values, + GetIntRangeFn getIntRange, int32_t indexBitwidth) { + SmallVector<IntegerValueRange> ranges; + ranges.reserve(values.size()); + for (OpFoldResult ofr : values) { + if (auto value = dyn_cast<Value>(ofr)) { + ranges.push_back(getIntRange(value)); + continue; + } + + // Create a constant range. + auto attr = cast<IntegerAttr>(cast<Attribute>(ofr)); + ranges.emplace_back(ConstantIntRanges::constant( + attr.getValue().sextOrTrunc(indexBitwidth))); + } + return ranges; +} + void mlir::intrange::detail::defaultInferResultRanges( InferIntRangeInterface interface, ArrayRef<IntegerValueRange> argRanges, SetIntLatticeFn setResultRanges) { diff --git a/mlir/lib/Interfaces/InferStridedMetadataInterface.cpp b/mlir/lib/Interfaces/InferStridedMetadataInterface.cpp new file mode 100644 index 0000000..483e9f1 --- /dev/null +++ b/mlir/lib/Interfaces/InferStridedMetadataInterface.cpp @@ -0,0 +1,36 @@ +//===- InferStridedMetadataInterface.cpp - Strided md inference interface -===// +// +// 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 +// +//===----------------------------------------------------------------------===// + +#include "mlir/Interfaces/InferStridedMetadataInterface.h" +#include "mlir/IR/BuiltinTypes.h" +#include "mlir/IR/TypeUtilities.h" +#include <optional> + +using namespace mlir; + +#include "mlir/Interfaces/InferStridedMetadataInterface.cpp.inc" + +void StridedMetadataRange::print(raw_ostream &os) const { + if (isUninitialized()) { + os << "strided_metadata<None>"; + return; + } + os << "strided_metadata<offset = ["; + llvm::interleaveComma(*offsets, os, [&](const ConstantIntRanges &range) { + os << "{" << range << "}"; + }); + os << "], sizes = ["; + llvm::interleaveComma(sizes, os, [&](const ConstantIntRanges &range) { + os << "{" << range << "}"; + }); + os << "], strides = ["; + llvm::interleaveComma(strides, os, [&](const ConstantIntRanges &range) { + os << "{" << range << "}"; + }); + os << "]>"; +} diff --git a/mlir/lib/Remark/RemarkStreamer.cpp b/mlir/lib/Remark/RemarkStreamer.cpp index d213a1a..bf36286 100644 --- a/mlir/lib/Remark/RemarkStreamer.cpp +++ b/mlir/lib/Remark/RemarkStreamer.cpp @@ -60,6 +60,7 @@ void LLVMRemarkStreamer::finalize() { namespace mlir::remark { LogicalResult enableOptimizationRemarksWithLLVMStreamer( MLIRContext &ctx, StringRef path, llvm::remarks::Format fmt, + std::unique_ptr<detail::RemarkEmittingPolicyBase> remarkEmittingPolicy, const RemarkCategories &cat, bool printAsEmitRemarks) { FailureOr<std::unique_ptr<detail::MLIRRemarkStreamerBase>> sOr = @@ -67,7 +68,8 @@ LogicalResult enableOptimizationRemarksWithLLVMStreamer( if (failed(sOr)) return failure(); - return remark::enableOptimizationRemarks(ctx, std::move(*sOr), cat, + return remark::enableOptimizationRemarks(ctx, std::move(*sOr), + std::move(remarkEmittingPolicy), cat, printAsEmitRemarks); } diff --git a/mlir/lib/Target/Wasm/TranslateFromWasm.cpp b/mlir/lib/Target/Wasm/TranslateFromWasm.cpp index 132be4e..51c6077 100644 --- a/mlir/lib/Target/Wasm/TranslateFromWasm.cpp +++ b/mlir/lib/Target/Wasm/TranslateFromWasm.cpp @@ -956,7 +956,7 @@ inline parsed_inst_t ExpressionParser::buildNumericOp( << ", type = " << ty << " ***"; auto tysToPop = SmallVector<Type, numOperands>(); tysToPop.resize(numOperands); - std::fill(tysToPop.begin(), tysToPop.end(), ty); + llvm::fill(tysToPop, ty); auto operands = popOperands(tysToPop); if (failed(operands)) return failure(); diff --git a/mlir/lib/Tools/mlir-opt/MlirOptMain.cpp b/mlir/lib/Tools/mlir-opt/MlirOptMain.cpp index 30fd384..9ef405d 100644 --- a/mlir/lib/Tools/mlir-opt/MlirOptMain.cpp +++ b/mlir/lib/Tools/mlir-opt/MlirOptMain.cpp @@ -37,6 +37,7 @@ #include "llvm/ADT/StringRef.h" #include "llvm/Remarks/RemarkFormat.h" #include "llvm/Support/CommandLine.h" +#include "llvm/Support/Debug.h" #include "llvm/Support/InitLLVM.h" #include "llvm/Support/LogicalResult.h" #include "llvm/Support/ManagedStatic.h" @@ -226,6 +227,18 @@ struct MlirOptMainConfigCLOptions : public MlirOptMainConfig { "bitstream", "Print bitstream file")), llvm::cl::cat(remarkCategory)}; + static llvm::cl::opt<RemarkPolicy, /*ExternalStorage=*/true> remarkPolicy{ + "remark-policy", + llvm::cl::desc("Specify the policy for remark output."), + cl::location(remarkPolicyFlag), + llvm::cl::value_desc("format"), + llvm::cl::init(RemarkPolicy::REMARK_POLICY_ALL), + llvm::cl::values(clEnumValN(RemarkPolicy::REMARK_POLICY_ALL, "all", + "Print all remarks"), + clEnumValN(RemarkPolicy::REMARK_POLICY_FINAL, "final", + "Print final remarks")), + llvm::cl::cat(remarkCategory)}; + static cl::opt<std::string, /*ExternalStorage=*/true> remarksAll( "remarks-filter", cl::desc("Show all remarks: passed, missed, failed, analysis"), @@ -517,18 +530,28 @@ performActions(raw_ostream &os, return failure(); context->enableMultithreading(wasThreadingEnabled); - + // Set the remark categories and policy. remark::RemarkCategories cats{ config.getRemarksAllFilter(), config.getRemarksPassedFilter(), config.getRemarksMissedFilter(), config.getRemarksAnalyseFilter(), config.getRemarksFailedFilter()}; mlir::MLIRContext &ctx = *context; + // Helper to create the appropriate policy based on configuration + auto createPolicy = [&config]() + -> std::unique_ptr<mlir::remark::detail::RemarkEmittingPolicyBase> { + if (config.getRemarkPolicy() == RemarkPolicy::REMARK_POLICY_ALL) + return std::make_unique<mlir::remark::RemarkEmittingPolicyAll>(); + if (config.getRemarkPolicy() == RemarkPolicy::REMARK_POLICY_FINAL) + return std::make_unique<mlir::remark::RemarkEmittingPolicyFinal>(); + + llvm_unreachable("Invalid remark policy"); + }; switch (config.getRemarkFormat()) { case RemarkFormat::REMARK_FORMAT_STDOUT: if (failed(mlir::remark::enableOptimizationRemarks( - ctx, nullptr, cats, true /*printAsEmitRemarks*/))) + ctx, nullptr, createPolicy(), cats, true /*printAsEmitRemarks*/))) return failure(); break; @@ -537,7 +560,7 @@ performActions(raw_ostream &os, ? "mlir-remarks.yaml" : config.getRemarksOutputFile(); if (failed(mlir::remark::enableOptimizationRemarksWithLLVMStreamer( - ctx, file, llvm::remarks::Format::YAML, cats))) + ctx, file, llvm::remarks::Format::YAML, createPolicy(), cats))) return failure(); break; } @@ -547,7 +570,7 @@ performActions(raw_ostream &os, ? "mlir-remarks.bitstream" : config.getRemarksOutputFile(); if (failed(mlir::remark::enableOptimizationRemarksWithLLVMStreamer( - ctx, file, llvm::remarks::Format::Bitstream, cats))) + ctx, file, llvm::remarks::Format::Bitstream, createPolicy(), cats))) return failure(); break; } @@ -593,6 +616,12 @@ performActions(raw_ostream &os, AsmState asmState(op.get(), OpPrintingFlags(), /*locationMap=*/nullptr, &fallbackResourceMap); os << OpWithState(op.get(), asmState) << '\n'; + + // This is required if the remark policy is final. Otherwise, the remarks are + // not emitted. + if (remark::detail::RemarkEngine *engine = ctx.getRemarkEngine()) + engine->getRemarkEmittingPolicy()->finalize(); + return success(); } diff --git a/mlir/lib/Transforms/Utils/LoopInvariantCodeMotionUtils.cpp b/mlir/lib/Transforms/Utils/LoopInvariantCodeMotionUtils.cpp index 111f58e..5f3b04a 100644 --- a/mlir/lib/Transforms/Utils/LoopInvariantCodeMotionUtils.cpp +++ b/mlir/lib/Transforms/Utils/LoopInvariantCodeMotionUtils.cpp @@ -66,7 +66,9 @@ size_t mlir::moveLoopInvariantCode( size_t numMoved = 0; for (Region *region : regions) { - LDBG() << "Original loop:\n" << *region->getParentOp(); + LDBG() << "Original loop:\n" + << OpWithFlags(region->getParentOp(), + OpPrintingFlags().skipRegions()); std::queue<Operation *> worklist; // Add top-level operations in the loop body to the worklist. @@ -90,7 +92,8 @@ size_t mlir::moveLoopInvariantCode( !canBeHoisted(op, definedOutside)) continue; - LDBG() << "Moving loop-invariant op: " << *op; + LDBG() << "Moving loop-invariant op: " + << OpWithFlags(op, OpPrintingFlags().skipRegions()); moveOutOfRegion(op, region); ++numMoved; @@ -111,9 +114,7 @@ size_t mlir::moveLoopInvariantCode(LoopLikeOpInterface loopLike) { [&](Value value, Region *) { return loopLike.isDefinedOutsideOfLoop(value); }, - [&](Operation *op, Region *) { - return isMemoryEffectFree(op) && isSpeculatable(op); - }, + [&](Operation *op, Region *) { return isPure(op); }, [&](Operation *op, Region *) { loopLike.moveOutOfLoop(op); }); } diff --git a/mlir/test/Analysis/DataFlow/test-strided-metadata-range-analysis.mlir b/mlir/test/Analysis/DataFlow/test-strided-metadata-range-analysis.mlir new file mode 100644 index 0000000..808c1c2 --- /dev/null +++ b/mlir/test/Analysis/DataFlow/test-strided-metadata-range-analysis.mlir @@ -0,0 +1,67 @@ +// RUN: mlir-opt -test-strided-metadata-range-analysis %s 2>&1 | FileCheck %s + +func.func @memref_subview(%arg0: memref<8x16x4xf32, strided<[64, 4, 1]>>, %arg1: memref<1x128x1x32x1xf32, strided<[4096, 32, 32, 1, 1]>>, %arg2: memref<8x16x4xf32, strided<[1, 64, 8], offset: 16>>, %arg3: index, %arg4: index, %arg5: index) { + %c0 = arith.constant 0 : index + %c1 = arith.constant 1 : index + %c2 = arith.constant 2 : index + %0 = test.with_bounds {smax = 13 : index, smin = 11 : index, umax = 13 : index, umin = 11 : index} : index + %1 = test.with_bounds {smax = 7 : index, smin = 5 : index, umax = 7 : index, umin = 5 : index} : index + + // Test subview with unknown sizes, and constant offsets and strides. + // CHECK: Op: %[[SV0:.*]] = memref.subview + // CHECK-NEXT: result[0]: strided_metadata< + // CHECK-SAME: offset = [{unsigned : [1, 1] signed : [1, 1]}] + // CHECK-SAME: sizes = [{unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}, {unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}, {unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}] + // CHECK-SAME: strides = [{unsigned : [64, 64] signed : [64, 64]}, {unsigned : [4, 4] signed : [4, 4]}, {unsigned : [1, 1] signed : [1, 1]}] + %subview = memref.subview %arg0[%c0, %c0, %c1] [%arg3, %arg4, %arg5] [%c1, %c1, %c1] : memref<8x16x4xf32, strided<[64, 4, 1]>> to memref<?x?x?xf32, strided<[?, ?, ?], offset: ?>> + + // Test a subview of a subview, with bounded dynamic offsets. + // CHECK: Op: %[[SV1:.*]] = memref.subview + // CHECK-NEXT: result[0]: strided_metadata< + // CHECK-SAME: offset = [{unsigned : [346, 484] signed : [346, 484]}] + // CHECK-SAME: sizes = [{unsigned : [2, 2] signed : [2, 2]}, {unsigned : [2, 2] signed : [2, 2]}, {unsigned : [2, 2] signed : [2, 2]}] + // CHECK-SAME: strides = [{unsigned : [704, 832] signed : [704, 832]}, {unsigned : [44, 52] signed : [44, 52]}, {unsigned : [11, 13] signed : [11, 13]}] + %subview_0 = memref.subview %subview[%1, %1, %1] [%c2, %c2, %c2] [%0, %0, %0] : memref<?x?x?xf32, strided<[?, ?, ?], offset: ?>> to memref<?x?x?xf32, strided<[?, ?, ?], offset: ?>> + + // Test a subview of a subview, with constant operands. + // CHECK: Op: %[[SV2:.*]] = memref.subview + // CHECK-NEXT: result[0]: strided_metadata< + // CHECK-SAME: offset = [{unsigned : [368, 510] signed : [368, 510]}] + // CHECK-SAME: sizes = [{unsigned : [2, 2] signed : [2, 2]}, {unsigned : [2, 2] signed : [2, 2]}, {unsigned : [2, 2] signed : [2, 2]}] + // CHECK-SAME: strides = [{unsigned : [704, 832] signed : [704, 832]}, {unsigned : [44, 52] signed : [44, 52]}, {unsigned : [11, 13] signed : [11, 13]}] + %subview_1 = memref.subview %subview_0[%c0, %c0, %c2] [%c2, %c2, %c2] [%c1, %c1, %c1] : memref<?x?x?xf32, strided<[?, ?, ?], offset: ?>> to memref<?x?x?xf32, strided<[?, ?, ?], offset: ?>> + + // Test a rank-reducing subview. + // CHECK: Op: %[[SV3:.*]] = memref.subview + // CHECK-NEXT: result[0]: strided_metadata< + // CHECK-SAME: offset = [{unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}] + // CHECK-SAME: sizes = [{unsigned : [64, 64] signed : [64, 64]}, {unsigned : [16, 16] signed : [16, 16]}] + // CHECK-SAME: strides = [{unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}, {unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}] + %subview_2 = memref.subview %arg1[%arg4, %arg4, %arg4, %arg4, %arg4] [1, 64, 1, 16, 1] [%arg5, %arg5, %arg5, %arg5, %arg5] : memref<1x128x1x32x1xf32, strided<[4096, 32, 32, 1, 1]>> to memref<64x16xf32, strided<[?, ?], offset: ?>> + + // Test a subview of a rank-reducing subview + // CHECK: Op: %[[SV4:.*]] = memref.subview + // CHECK-NEXT: result[0]: strided_metadata< + // CHECK-SAME: offset = [{unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}] + // CHECK-SAME: sizes = [{unsigned : [5, 7] signed : [5, 7]}] + // CHECK-SAME: strides = [{unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}] + %subview_3 = memref.subview %subview_2[%c0, %0] [1, %1] [%c1, %c2] : memref<64x16xf32, strided<[?, ?], offset: ?>> to memref<?xf32, strided<[?], offset: ?>> + + // Test a subview with mixed bounded and unbound dynamic sizes. + // CHECK: Op: %[[SV5:.*]] = memref.subview + // CHECK-NEXT: result[0]: strided_metadata< + // CHECK-SAME: offset = [{unsigned : [32, 32] signed : [32, 32]}] + // CHECK-SAME: sizes = [{unsigned : [11, 13] signed : [11, 13]}, {unsigned : [5, 7] signed : [5, 7]}, {unsigned : [0, 18446744073709551615] signed : [-9223372036854775808, 9223372036854775807]}] + // CHECK-SAME: strides = [{unsigned : [1, 1] signed : [1, 1]}, {unsigned : [64, 64] signed : [64, 64]}, {unsigned : [8, 8] signed : [8, 8]}] + %subview_4 = memref.subview %arg2[%c0, %c0, %c2] [%0, %1, %arg5] [%c1, %c1, %c1] : memref<8x16x4xf32, strided<[1, 64, 8], offset: 16>> to memref<?x?x?xf32, strided<[?, ?, ?], offset: ?>> + return +} + +// CHECK: func.func @memref_subview +// CHECK: %[[A0:.*]]: memref<8x16x4xf32, strided<[64, 4, 1]>> +// CHECK: %[[SV0]] = memref.subview %[[A0]] +// CHECK-NEXT: %[[SV1]] = memref.subview +// CHECK-NEXT: %[[SV2]] = memref.subview +// CHECK-NEXT: %[[SV3]] = memref.subview +// CHECK-NEXT: %[[SV4]] = memref.subview +// CHECK-NEXT: %[[SV5]] = memref.subview diff --git a/mlir/test/Dialect/OpenACC/recipe-populate-firstprivate.mlir b/mlir/test/Dialect/OpenACC/recipe-populate-firstprivate.mlir new file mode 100644 index 0000000..35355c6 --- /dev/null +++ b/mlir/test/Dialect/OpenACC/recipe-populate-firstprivate.mlir @@ -0,0 +1,102 @@ +// RUN: mlir-opt %s --split-input-file --pass-pipeline="builtin.module(test-acc-recipe-populate{recipe-type=firstprivate})" | FileCheck %s + +// CHECK: acc.firstprivate.recipe @firstprivate_scalar : memref<f32> init { +// CHECK: ^bb0(%{{.*}}: memref<f32>): +// CHECK: %[[ALLOC:.*]] = memref.alloca() : memref<f32> +// CHECK: acc.yield %[[ALLOC]] : memref<f32> +// CHECK: } copy { +// CHECK: ^bb0(%[[SRC:.*]]: memref<f32>, %[[DST:.*]]: memref<f32>): +// CHECK: memref.copy %[[SRC]], %[[DST]] : memref<f32> to memref<f32> +// CHECK: acc.terminator +// CHECK: } +// CHECK-NOT: destroy + +func.func @test_scalar() { + %0 = memref.alloca() {test.var = "scalar"} : memref<f32> + return +} + +// ----- + +// CHECK: acc.firstprivate.recipe @firstprivate_static_2d : memref<10x20xf32> init { +// CHECK: ^bb0(%{{.*}}: memref<10x20xf32>): +// CHECK: %[[ALLOC:.*]] = memref.alloca() : memref<10x20xf32> +// CHECK: acc.yield %[[ALLOC]] : memref<10x20xf32> +// CHECK: } copy { +// CHECK: ^bb0(%[[SRC:.*]]: memref<10x20xf32>, %[[DST:.*]]: memref<10x20xf32>): +// CHECK: memref.copy %[[SRC]], %[[DST]] : memref<10x20xf32> to memref<10x20xf32> +// CHECK: acc.terminator +// CHECK: } +// CHECK-NOT: destroy + +func.func @test_static_2d() { + %0 = memref.alloca() {test.var = "static_2d"} : memref<10x20xf32> + return +} + +// ----- + +// CHECK: acc.firstprivate.recipe @firstprivate_dynamic_2d : memref<?x?xf32> init { +// CHECK: ^bb0(%[[ARG:.*]]: memref<?x?xf32>): +// CHECK: %[[C0:.*]] = arith.constant 0 : index +// CHECK: %[[DIM0:.*]] = memref.dim %[[ARG]], %[[C0]] : memref<?x?xf32> +// CHECK: %[[C1:.*]] = arith.constant 1 : index +// CHECK: %[[DIM1:.*]] = memref.dim %[[ARG]], %[[C1]] : memref<?x?xf32> +// CHECK: %[[ALLOC:.*]] = memref.alloc(%[[DIM0]], %[[DIM1]]) : memref<?x?xf32> +// CHECK: acc.yield %[[ALLOC]] : memref<?x?xf32> +// CHECK: } copy { +// CHECK: ^bb0(%[[SRC:.*]]: memref<?x?xf32>, %[[DST:.*]]: memref<?x?xf32>): +// CHECK: memref.copy %[[SRC]], %[[DST]] : memref<?x?xf32> to memref<?x?xf32> +// CHECK: acc.terminator +// CHECK: } destroy { +// CHECK: ^bb0(%{{.*}}: memref<?x?xf32>, %[[VAL:.*]]: memref<?x?xf32>): +// CHECK: memref.dealloc %[[VAL]] : memref<?x?xf32> +// CHECK: acc.terminator +// CHECK: } + +func.func @test_dynamic_2d(%arg0: index, %arg1: index) { + %0 = memref.alloc(%arg0, %arg1) {test.var = "dynamic_2d"} : memref<?x?xf32> + return +} + +// ----- + +// CHECK: acc.firstprivate.recipe @firstprivate_mixed_dims : memref<10x?xf32> init { +// CHECK: ^bb0(%[[ARG:.*]]: memref<10x?xf32>): +// CHECK: %[[C1:.*]] = arith.constant 1 : index +// CHECK: %[[DIM1:.*]] = memref.dim %[[ARG]], %[[C1]] : memref<10x?xf32> +// CHECK: %[[ALLOC:.*]] = memref.alloc(%[[DIM1]]) : memref<10x?xf32> +// CHECK: acc.yield %[[ALLOC]] : memref<10x?xf32> +// CHECK: } copy { +// CHECK: ^bb0(%[[SRC:.*]]: memref<10x?xf32>, %[[DST:.*]]: memref<10x?xf32>): +// CHECK: memref.copy %[[SRC]], %[[DST]] : memref<10x?xf32> to memref<10x?xf32> +// CHECK: acc.terminator +// CHECK: } destroy { +// CHECK: ^bb0(%{{.*}}: memref<10x?xf32>, %[[VAL:.*]]: memref<10x?xf32>): +// CHECK: memref.dealloc %[[VAL]] : memref<10x?xf32> +// CHECK: acc.terminator +// CHECK: } + +func.func @test_mixed_dims(%arg0: index) { + %0 = memref.alloc(%arg0) {test.var = "mixed_dims"} : memref<10x?xf32> + return +} + +// ----- + +// CHECK: acc.firstprivate.recipe @firstprivate_scalar_int : memref<i32> init { +// CHECK: ^bb0(%{{.*}}: memref<i32>): +// CHECK: %[[ALLOC:.*]] = memref.alloca() : memref<i32> +// CHECK: acc.yield %[[ALLOC]] : memref<i32> +// CHECK: } copy { +// CHECK: ^bb0(%[[SRC:.*]]: memref<i32>, %[[DST:.*]]: memref<i32>): +// CHECK: memref.copy %[[SRC]], %[[DST]] : memref<i32> to memref<i32> +// CHECK: acc.terminator +// CHECK: } +// CHECK-NOT: destroy + +func.func @test_scalar_int() { + %0 = memref.alloca() {test.var = "scalar_int"} : memref<i32> + return +} + diff --git a/mlir/test/Dialect/OpenACC/recipe-populate-private.mlir b/mlir/test/Dialect/OpenACC/recipe-populate-private.mlir new file mode 100644 index 0000000..8403ee8 --- /dev/null +++ b/mlir/test/Dialect/OpenACC/recipe-populate-private.mlir @@ -0,0 +1,82 @@ +// RUN: mlir-opt %s --split-input-file --pass-pipeline="builtin.module(test-acc-recipe-populate{recipe-type=private})" | FileCheck %s + +// CHECK: acc.private.recipe @private_scalar : memref<f32> init { +// CHECK: ^bb0(%{{.*}}: memref<f32>): +// CHECK: %[[ALLOC:.*]] = memref.alloca() : memref<f32> +// CHECK: acc.yield %[[ALLOC]] : memref<f32> +// CHECK: } +// CHECK-NOT: destroy + +func.func @test_scalar() { + %0 = memref.alloca() {test.var = "scalar"} : memref<f32> + return +} + +// ----- + +// CHECK: acc.private.recipe @private_static_2d : memref<10x20xf32> init { +// CHECK: ^bb0(%{{.*}}: memref<10x20xf32>): +// CHECK: %[[ALLOC:.*]] = memref.alloca() : memref<10x20xf32> +// CHECK: acc.yield %[[ALLOC]] : memref<10x20xf32> +// CHECK: } +// CHECK-NOT: destroy + +func.func @test_static_2d() { + %0 = memref.alloca() {test.var = "static_2d"} : memref<10x20xf32> + return +} + +// ----- + +// CHECK: acc.private.recipe @private_dynamic_2d : memref<?x?xf32> init { +// CHECK: ^bb0(%[[ARG:.*]]: memref<?x?xf32>): +// CHECK: %[[C0:.*]] = arith.constant 0 : index +// CHECK: %[[DIM0:.*]] = memref.dim %[[ARG]], %[[C0]] : memref<?x?xf32> +// CHECK: %[[C1:.*]] = arith.constant 1 : index +// CHECK: %[[DIM1:.*]] = memref.dim %[[ARG]], %[[C1]] : memref<?x?xf32> +// CHECK: %[[ALLOC:.*]] = memref.alloc(%[[DIM0]], %[[DIM1]]) : memref<?x?xf32> +// CHECK: acc.yield %[[ALLOC]] : memref<?x?xf32> +// CHECK: } destroy { +// CHECK: ^bb0(%{{.*}}: memref<?x?xf32>, %[[VAL:.*]]: memref<?x?xf32>): +// CHECK: memref.dealloc %[[VAL]] : memref<?x?xf32> +// CHECK: acc.terminator +// CHECK: } + +func.func @test_dynamic_2d(%arg0: index, %arg1: index) { + %0 = memref.alloc(%arg0, %arg1) {test.var = "dynamic_2d"} : memref<?x?xf32> + return +} + +// ----- + +// CHECK: acc.private.recipe @private_mixed_dims : memref<10x?xf32> init { +// CHECK: ^bb0(%[[ARG:.*]]: memref<10x?xf32>): +// CHECK: %[[C1:.*]] = arith.constant 1 : index +// CHECK: %[[DIM1:.*]] = memref.dim %[[ARG]], %[[C1]] : memref<10x?xf32> +// CHECK: %[[ALLOC:.*]] = memref.alloc(%[[DIM1]]) : memref<10x?xf32> +// CHECK: acc.yield %[[ALLOC]] : memref<10x?xf32> +// CHECK: } destroy { +// CHECK: ^bb0(%{{.*}}: memref<10x?xf32>, %[[VAL:.*]]: memref<10x?xf32>): +// CHECK: memref.dealloc %[[VAL]] : memref<10x?xf32> +// CHECK: acc.terminator +// CHECK: } + +func.func @test_mixed_dims(%arg0: index) { + %0 = memref.alloc(%arg0) {test.var = "mixed_dims"} : memref<10x?xf32> + return +} + +// ----- + +// CHECK: acc.private.recipe @private_scalar_int : memref<i32> init { +// CHECK: ^bb0(%{{.*}}: memref<i32>): +// CHECK: %[[ALLOC:.*]] = memref.alloca() : memref<i32> +// CHECK: acc.yield %[[ALLOC]] : memref<i32> +// CHECK: } +// CHECK-NOT: destroy + +func.func @test_scalar_int() { + %0 = memref.alloca() {test.var = "scalar_int"} : memref<i32> + return +} + diff --git a/mlir/test/Integration/GPU/SPIRV/simple_add.mlir b/mlir/test/Integration/GPU/SPIRV/simple_add.mlir index cb16c37..b3154d4 100644 --- a/mlir/test/Integration/GPU/SPIRV/simple_add.mlir +++ b/mlir/test/Integration/GPU/SPIRV/simple_add.mlir @@ -3,7 +3,16 @@ // RUN: | FileCheck %s // CHECK: data = -// CHECK-RAW: [[[7.7, 0, 0], [7.7, 0, 0], [7.7, 0, 0]], [[0, 7.7, 0], [0, 7.7, 0], [0, 7.7, 0]], [[0, 0, 7.7], [0, 0, 7.7], [0, 0, 7.7]]] +// CHECK{LITERAL}: [[[7.7, 0, 0], +// CHECK{LITERAL}: [7.7, 0, 0], +// CHECK{LITERAL}: [7.7, 0, 0]], +// CHECK{LITERAL}: [[0, 7.7, 0], +// CHECK{LITERAL}: [0, 7.7, 0], +// CHECK{LITERAL}: [0, 7.7, 0]], +// CHECK{LITERAL}: [[0, 0, 7.7], +// CHECK{LITERAL}: [0, 0, 7.7], +// CHECK{LITERAL}: [0, 0, 7.7]]] + module attributes { gpu.container_module, spirv.target_env = #spirv.target_env< diff --git a/mlir/test/Pass/remark-final.mlir b/mlir/test/Pass/remark-final.mlir new file mode 100644 index 0000000..325271e --- /dev/null +++ b/mlir/test/Pass/remark-final.mlir @@ -0,0 +1,17 @@ +// RUN: mlir-opt %s --test-remark --remarks-filter="category.*" --remark-policy=final 2>&1 | FileCheck %s +// RUN: mlir-opt %s --test-remark --remarks-filter="category.*" --remark-policy=final --remark-format=yaml --remarks-output-file=%t.yaml +// RUN: FileCheck --check-prefix=CHECK-YAML %s < %t.yaml +module @foo { + "test.op"() : () -> () + +} + +// CHECK-YAML-NOT: This is a test passed remark (should be dropped) +// CHECK-YAML-DAG: !Analysis +// CHECK-YAML-DAG: !Failure +// CHECK-YAML-DAG: !Passed + +// CHECK-NOT: This is a test passed remark (should be dropped) +// CHECK-DAG: remark: [Analysis] test-remark +// CHECK-DAG: remark: [Failure] test-remark | Category:category-2-failed +// CHECK-DAG: remark: [Passed] test-remark | Category:category-1-passed diff --git a/mlir/test/lib/Analysis/CMakeLists.txt b/mlir/test/lib/Analysis/CMakeLists.txt index 9187998..c37671a 100644 --- a/mlir/test/lib/Analysis/CMakeLists.txt +++ b/mlir/test/lib/Analysis/CMakeLists.txt @@ -17,6 +17,7 @@ add_mlir_library(MLIRTestAnalysis DataFlow/TestDenseForwardDataFlowAnalysis.cpp DataFlow/TestLivenessAnalysis.cpp DataFlow/TestSparseBackwardDataFlowAnalysis.cpp + DataFlow/TestStridedMetadataRangeAnalysis.cpp EXCLUDE_FROM_LIBMLIR diff --git a/mlir/test/lib/Analysis/DataFlow/TestStridedMetadataRangeAnalysis.cpp b/mlir/test/lib/Analysis/DataFlow/TestStridedMetadataRangeAnalysis.cpp new file mode 100644 index 0000000..6ac09fd --- /dev/null +++ b/mlir/test/lib/Analysis/DataFlow/TestStridedMetadataRangeAnalysis.cpp @@ -0,0 +1,86 @@ +//===- TestStridedMetadataRangeAnalysis.cpp - Test strided md analysis ----===// +// +// 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 +// +//===----------------------------------------------------------------------===// + +#include "mlir/Analysis/DataFlow/ConstantPropagationAnalysis.h" +#include "mlir/Analysis/DataFlow/DeadCodeAnalysis.h" +#include "mlir/Analysis/DataFlow/IntegerRangeAnalysis.h" +#include "mlir/Analysis/DataFlow/StridedMetadataRangeAnalysis.h" +#include "mlir/Analysis/DataFlowFramework.h" +#include "mlir/IR/BuiltinAttributes.h" +#include "mlir/IR/Operation.h" +#include "mlir/Pass/Pass.h" +#include "mlir/Pass/PassRegistry.h" +#include "llvm/ADT/STLExtras.h" +#include "llvm/Support/raw_ostream.h" + +using namespace mlir; +using namespace mlir::dataflow; + +static void printAnalysisResults(DataFlowSolver &solver, Operation *op, + raw_ostream &os) { + // Collect the strided metadata of the op results. + SmallVector<std::pair<unsigned, const StridedMetadataRangeLattice *>> results; + for (OpResult result : op->getResults()) { + const auto *state = solver.lookupState<StridedMetadataRangeLattice>(result); + // Skip the result if it's uninitialized. + if (!state || state->getValue().isUninitialized()) + continue; + + // Skip the result if the range is empty. + const mlir::StridedMetadataRange &md = state->getValue(); + if (md.getOffsets().empty() && md.getSizes().empty() && + md.getStrides().empty()) + continue; + results.push_back({result.getResultNumber(), state}); + } + + // Early exit if there's no metadata to print. + if (results.empty()) + return; + + // Print the metadata. + os << "Op: " << OpWithFlags(op, OpPrintingFlags().skipRegions()) << "\n"; + for (auto [idx, state] : results) + os << " result[" << idx << "]: " << state->getValue() << "\n"; + os << "\n"; +} + +namespace { +struct TestStridedMetadataRangeAnalysisPass + : public PassWrapper<TestStridedMetadataRangeAnalysisPass, + OperationPass<>> { + MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID( + TestStridedMetadataRangeAnalysisPass) + + StringRef getArgument() const override { + return "test-strided-metadata-range-analysis"; + } + void runOnOperation() override { + Operation *op = getOperation(); + + DataFlowSolver solver; + solver.load<DeadCodeAnalysis>(); + solver.load<SparseConstantPropagation>(); + solver.load<IntegerRangeAnalysis>(); + solver.load<StridedMetadataRangeAnalysis>(); + if (failed(solver.initializeAndRun(op))) + return signalPassFailure(); + + op->walk( + [&](Operation *op) { printAnalysisResults(solver, op, llvm::errs()); }); + } +}; +} // end anonymous namespace + +namespace mlir { +namespace test { +void registerTestStridedMetadataRangeAnalysisPass() { + PassRegistration<TestStridedMetadataRangeAnalysisPass>(); +} +} // end namespace test +} // end namespace mlir diff --git a/mlir/test/lib/Dialect/OpenACC/CMakeLists.txt b/mlir/test/lib/Dialect/OpenACC/CMakeLists.txt index f84055d..1e59338 100644 --- a/mlir/test/lib/Dialect/OpenACC/CMakeLists.txt +++ b/mlir/test/lib/Dialect/OpenACC/CMakeLists.txt @@ -1,6 +1,7 @@ add_mlir_library(MLIROpenACCTestPasses TestOpenACC.cpp TestPointerLikeTypeInterface.cpp + TestRecipePopulate.cpp EXCLUDE_FROM_LIBMLIR ) diff --git a/mlir/test/lib/Dialect/OpenACC/TestOpenACC.cpp b/mlir/test/lib/Dialect/OpenACC/TestOpenACC.cpp index 9886240..bea21b9 100644 --- a/mlir/test/lib/Dialect/OpenACC/TestOpenACC.cpp +++ b/mlir/test/lib/Dialect/OpenACC/TestOpenACC.cpp @@ -15,9 +15,13 @@ namespace test { // Forward declarations of individual test pass registration functions void registerTestPointerLikeTypeInterfacePass(); +void registerTestRecipePopulatePass(); // Unified registration function for all OpenACC tests -void registerTestOpenACC() { registerTestPointerLikeTypeInterfacePass(); } +void registerTestOpenACC() { + registerTestPointerLikeTypeInterfacePass(); + registerTestRecipePopulatePass(); +} } // namespace test } // namespace mlir diff --git a/mlir/test/lib/Dialect/OpenACC/TestPointerLikeTypeInterface.cpp b/mlir/test/lib/Dialect/OpenACC/TestPointerLikeTypeInterface.cpp index 85f9283..027b0a1 100644 --- a/mlir/test/lib/Dialect/OpenACC/TestPointerLikeTypeInterface.cpp +++ b/mlir/test/lib/Dialect/OpenACC/TestPointerLikeTypeInterface.cpp @@ -196,13 +196,15 @@ void TestPointerLikeTypeInterfacePass::testGenAllocate( newBuilder.setInsertionPointAfter(op); // Call the genAllocate API + bool needsFree = false; Value allocRes = pointerType.genAllocate(newBuilder, loc, "test_alloc", - result.getType(), result); + result.getType(), result, needsFree); if (allocRes) { llvm::errs() << "Successfully generated alloc for operation: "; op->print(llvm::errs()); llvm::errs() << "\n"; + llvm::errs() << "\tneeds free: " << (needsFree ? "true" : "false") << "\n"; // Print all operations that were inserted for (Operation *insertedOp : tracker.insertedOps) { @@ -230,8 +232,8 @@ void TestPointerLikeTypeInterfacePass::testGenFree(Operation *op, Value result, // Call the genFree API auto typedResult = cast<TypedValue<PointerLikeType>>(result); - bool success = - pointerType.genFree(newBuilder, loc, typedResult, result.getType()); + bool success = pointerType.genFree(newBuilder, loc, typedResult, result, + result.getType()); if (success) { llvm::errs() << "Successfully generated free for operation: "; diff --git a/mlir/test/lib/Dialect/OpenACC/TestRecipePopulate.cpp b/mlir/test/lib/Dialect/OpenACC/TestRecipePopulate.cpp new file mode 100644 index 0000000..35f092c --- /dev/null +++ b/mlir/test/lib/Dialect/OpenACC/TestRecipePopulate.cpp @@ -0,0 +1,110 @@ +//===- TestRecipePopulate.cpp - Test Recipe Population -------------------===// +// +// 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 +// +//===----------------------------------------------------------------------===// +// +// This file contains test passes for testing the createAndPopulate methods +// of the recipe operations. +// +//===----------------------------------------------------------------------===// + +#include "mlir/Dialect/Arith/IR/Arith.h" +#include "mlir/Dialect/Func/IR/FuncOps.h" +#include "mlir/Dialect/MemRef/IR/MemRef.h" +#include "mlir/Dialect/OpenACC/OpenACC.h" +#include "mlir/IR/Builders.h" +#include "mlir/Pass/Pass.h" +#include "llvm/Support/CommandLine.h" + +using namespace mlir; +using namespace mlir::acc; + +namespace { + +struct TestRecipePopulatePass + : public PassWrapper<TestRecipePopulatePass, OperationPass<ModuleOp>> { + MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID(TestRecipePopulatePass) + + TestRecipePopulatePass() = default; + TestRecipePopulatePass(const TestRecipePopulatePass &pass) + : PassWrapper(pass) { + recipeType = pass.recipeType; + } + + Pass::Option<std::string> recipeType{ + *this, "recipe-type", + llvm::cl::desc("Recipe type: private or firstprivate"), + llvm::cl::init("private")}; + + StringRef getArgument() const override { return "test-acc-recipe-populate"; } + + StringRef getDescription() const override { + return "Test OpenACC recipe population"; + } + + void runOnOperation() override; + + void getDependentDialects(DialectRegistry ®istry) const override { + registry.insert<acc::OpenACCDialect>(); + registry.insert<arith::ArithDialect>(); + registry.insert<memref::MemRefDialect>(); + } +}; + +void TestRecipePopulatePass::runOnOperation() { + auto module = getOperation(); + OpBuilder builder(&getContext()); + + // Collect all test variables + SmallVector<std::tuple<Operation *, Value, std::string>> testVars; + + module.walk([&](Operation *op) { + if (auto varName = op->getAttrOfType<StringAttr>("test.var")) { + for (auto result : op->getResults()) { + testVars.push_back({op, result, varName.str()}); + } + } + }); + + // Generate recipes at module level + builder.setInsertionPoint(&module.getBodyRegion().front(), + module.getBodyRegion().front().begin()); + + for (auto [op, var, varName] : testVars) { + Location loc = op->getLoc(); + + std::string recipeName = recipeType.getValue() + "_" + varName; + ValueRange bounds; // No bounds for memref tests + + if (recipeType == "private") { + auto recipe = PrivateRecipeOp::createAndPopulate( + builder, loc, recipeName, var.getType(), varName, bounds); + + if (!recipe) { + op->emitError("Failed to create private recipe for ") << varName; + } + } else if (recipeType == "firstprivate") { + auto recipe = FirstprivateRecipeOp::createAndPopulate( + builder, loc, recipeName, var.getType(), varName, bounds); + + if (!recipe) { + op->emitError("Failed to create firstprivate recipe for ") << varName; + } + } + } +} + +} // namespace + +namespace mlir { +namespace test { + +void registerTestRecipePopulatePass() { + PassRegistration<TestRecipePopulatePass>(); +} + +} // namespace test +} // namespace mlir diff --git a/mlir/test/lib/Pass/TestRemarksPass.cpp b/mlir/test/lib/Pass/TestRemarksPass.cpp index 3b25686..5ca2d1a 100644 --- a/mlir/test/lib/Pass/TestRemarksPass.cpp +++ b/mlir/test/lib/Pass/TestRemarksPass.cpp @@ -43,7 +43,12 @@ public: << remark::add("This is a test missed remark") << remark::reason("because we are testing the remark pipeline") << remark::suggest("try using the remark pipeline feature"); - + mlir::remark::passed( + loc, + remark::RemarkOpts::name("test-remark").category("category-1-passed")) + << remark::add("This is a test passed remark (should be dropped)") + << remark::reason("because we are testing the remark pipeline") + << remark::suggest("try using the remark pipeline feature"); mlir::remark::passed( loc, remark::RemarkOpts::name("test-remark").category("category-1-passed")) diff --git a/mlir/tools/mlir-opt/mlir-opt.cpp b/mlir/tools/mlir-opt/mlir-opt.cpp index 6432fae..8842180 100644 --- a/mlir/tools/mlir-opt/mlir-opt.cpp +++ b/mlir/tools/mlir-opt/mlir-opt.cpp @@ -151,6 +151,7 @@ void registerTestSliceAnalysisPass(); void registerTestSPIRVCPURunnerPipeline(); void registerTestSPIRVFuncSignatureConversion(); void registerTestSPIRVVectorUnrolling(); +void registerTestStridedMetadataRangeAnalysisPass(); void registerTestTensorCopyInsertionPass(); void registerTestTensorLikeAndBufferLikePass(); void registerTestTensorTransforms(); @@ -299,6 +300,7 @@ void registerTestPasses() { mlir::test::registerTestSPIRVCPURunnerPipeline(); mlir::test::registerTestSPIRVFuncSignatureConversion(); mlir::test::registerTestSPIRVVectorUnrolling(); + mlir::test::registerTestStridedMetadataRangeAnalysisPass(); mlir::test::registerTestTensorCopyInsertionPass(); mlir::test::registerTestTensorLikeAndBufferLikePass(); mlir::test::registerTestTensorTransforms(); diff --git a/mlir/unittests/IR/RemarkTest.cpp b/mlir/unittests/IR/RemarkTest.cpp index bcbda90..09c576c 100644 --- a/mlir/unittests/IR/RemarkTest.cpp +++ b/mlir/unittests/IR/RemarkTest.cpp @@ -53,10 +53,12 @@ TEST(Remark, TestOutputOptimizationRemark) { /*missed=*/categoryUnroll, /*analysis=*/categoryRegister, /*failed=*/categoryInliner}; - + std::unique_ptr<remark::RemarkEmittingPolicyAll> policy = + std::make_unique<remark::RemarkEmittingPolicyAll>(); LogicalResult isEnabled = mlir::remark::enableOptimizationRemarksWithLLVMStreamer( - context, yamlFile, llvm::remarks::Format::YAML, cats); + context, yamlFile, llvm::remarks::Format::YAML, std::move(policy), + cats); ASSERT_TRUE(succeeded(isEnabled)) << "Failed to enable remark engine"; // PASS: something succeeded @@ -202,9 +204,10 @@ TEST(Remark, TestOutputOptimizationRemarkDiagnostic) { /*missed=*/categoryUnroll, /*analysis=*/categoryRegister, /*failed=*/categoryUnroll}; - - LogicalResult isEnabled = - remark::enableOptimizationRemarks(context, nullptr, cats, true); + std::unique_ptr<remark::RemarkEmittingPolicyAll> policy = + std::make_unique<remark::RemarkEmittingPolicyAll>(); + LogicalResult isEnabled = remark::enableOptimizationRemarks( + context, nullptr, std::move(policy), cats, true); ASSERT_TRUE(succeeded(isEnabled)) << "Failed to enable remark engine"; @@ -282,8 +285,11 @@ TEST(Remark, TestCustomOptimizationRemarkDiagnostic) { /*analysis=*/std::nullopt, /*failed=*/categoryLoopunroll}; + std::unique_ptr<remark::RemarkEmittingPolicyAll> policy = + std::make_unique<remark::RemarkEmittingPolicyAll>(); LogicalResult isEnabled = remark::enableOptimizationRemarks( - context, std::make_unique<MyCustomStreamer>(), cats, true); + context, std::make_unique<MyCustomStreamer>(), std::move(policy), cats, + true); ASSERT_TRUE(succeeded(isEnabled)) << "Failed to enable remark engine"; // Remark 1: pass, category LoopUnroll @@ -311,4 +317,66 @@ TEST(Remark, TestCustomOptimizationRemarkDiagnostic) { EXPECT_NE(errOut.find(pass2Msg), std::string::npos); // printed EXPECT_EQ(errOut.find(pass3Msg), std::string::npos); // filtered out } + +TEST(Remark, TestRemarkFinal) { + testing::internal::CaptureStderr(); + const auto *pass1Msg = "I failed"; + const auto *pass2Msg = "I failed too"; + const auto *pass3Msg = "I succeeded"; + const auto *pass4Msg = "I succeeded too"; + + std::string categoryLoopunroll("LoopUnroll"); + + std::string seenMsg = ""; + + { + MLIRContext context; + Location loc = FileLineColLoc::get(&context, "test.cpp", 1, 5); + Location locOther = FileLineColLoc::get(&context, "test.cpp", 55, 5); + + // Setup the remark engine + mlir::remark::RemarkCategories cats{/*all=*/"", + /*passed=*/categoryLoopunroll, + /*missed=*/categoryLoopunroll, + /*analysis=*/categoryLoopunroll, + /*failed=*/categoryLoopunroll}; + + std::unique_ptr<remark::RemarkEmittingPolicyFinal> policy = + std::make_unique<remark::RemarkEmittingPolicyFinal>(); + LogicalResult isEnabled = remark::enableOptimizationRemarks( + context, std::make_unique<MyCustomStreamer>(), std::move(policy), cats, + true); + ASSERT_TRUE(succeeded(isEnabled)) << "Failed to enable remark engine"; + + // Remark 1: failure + remark::failed( + loc, remark::RemarkOpts::name("Unroller").category(categoryLoopunroll)) + << pass1Msg; + + // Remark 2: failure + remark::missed( + loc, remark::RemarkOpts::name("Unroller").category(categoryLoopunroll)) + << remark::reason(pass2Msg); + + // Remark 3: pass + remark::passed( + loc, remark::RemarkOpts::name("Unroller").category(categoryLoopunroll)) + << pass3Msg; + + // Remark 4: pass + remark::passed( + locOther, + remark::RemarkOpts::name("Unroller").category(categoryLoopunroll)) + << pass4Msg; + } + + llvm::errs().flush(); + std::string errOut = ::testing::internal::GetCapturedStderr(); + + // Containment checks for messages. + EXPECT_EQ(errOut.find(pass1Msg), std::string::npos); // dropped + EXPECT_EQ(errOut.find(pass2Msg), std::string::npos); // dropped + EXPECT_NE(errOut.find(pass3Msg), std::string::npos); // shown + EXPECT_NE(errOut.find(pass4Msg), std::string::npos); // shown +} } // namespace diff --git a/mlir/utils/generate-test-checks.py b/mlir/utils/generate-test-checks.py index f80a181..3712a6b 100755 --- a/mlir/utils/generate-test-checks.py +++ b/mlir/utils/generate-test-checks.py @@ -31,13 +31,16 @@ import argparse import os # Used to advertise this file's name ("autogenerated_note"). import re import sys +from collections import Counter ADVERT_BEGIN = "// NOTE: Assertions have been autogenerated by " ADVERT_END = """ -// The script is designed to make adding checks to -// a test case fast, it is *not* designed to be authoritative -// about what constitutes a good test! The CHECK should be -// minimized and named to reflect the test intent. +// This script is intended to make adding checks to a test case quick and easy. +// It is *not* authoritative about what constitutes a good test. After using the +// script, be sure to review and refine the generated checks. For example, +// CHECK lines should be minimized and named to reflect the test’s intent. +// For comprehensive guidelines, see: +// * https://mlir.llvm.org/getting_started/TestingGuide/ """ @@ -45,6 +48,9 @@ ADVERT_END = """ SSA_RE_STR = "[0-9]+|[a-zA-Z$._-][a-zA-Z0-9$._-]*" SSA_RE = re.compile(SSA_RE_STR) +# Regex matching `dialect.op_name` (e.g. `vector.transfer_read`). +SSA_OP_NAME_RE = re.compile(r"\b(?:\s=\s[a-z_]+)[.]([a-z_]+)\b") + # Regex matching the left-hand side of an assignment SSA_RESULTS_STR = r'\s*(%' + SSA_RE_STR + r')(\s*,\s*(%' + SSA_RE_STR + r'))*\s*=' SSA_RESULTS_RE = re.compile(SSA_RESULTS_STR) @@ -63,7 +69,12 @@ ATTR_DEF_RE = re.compile(ATTR_DEF_RE_STR) class VariableNamer: def __init__(self, variable_names): self.scopes = [] + # Counter for generic FileCHeck names, e.g. VAL_#N self.name_counter = 0 + # Counters for FileCheck names derived from Op names, e.g. + # TRANSFER_READ_#N (based on `vector.transfer_read`). Note, there's a + # dedicated counter for every Op type present in the input. + self.op_name_counter = Counter() # Number of variable names to still generate in parent scope self.generate_in_parent_scope_left = 0 @@ -77,17 +88,29 @@ class VariableNamer: self.generate_in_parent_scope_left = n # Generate a substitution name for the given ssa value name. - def generate_name(self, source_variable_name, use_ssa_name): + def generate_name(self, source_variable_name, use_ssa_name, op_name=""): # Compute variable name - variable_name = self.variable_names.pop(0) if len(self.variable_names) > 0 else '' - if variable_name == '': + variable_name = ( + self.variable_names.pop(0) if len(self.variable_names) > 0 else "" + ) + if variable_name == "": # If `use_ssa_name` is set, use the MLIR SSA value name to generate # a FileCHeck substation string. As FileCheck requires these # strings to start with a character, skip MLIR variables starting # with a digit (e.g. `%0`). + # + # The next fallback option is to use the op name, if the + # corresponding match succeeds. + # + # If neither worked, use a generic name: `VAL_#N`. if use_ssa_name and source_variable_name[0].isalpha(): variable_name = source_variable_name.upper() + elif op_name != "": + variable_name = ( + op_name.upper() + "_" + str(self.op_name_counter[op_name]) + ) + self.op_name_counter[op_name] += 1 else: variable_name = "VAL_" + str(self.name_counter) self.name_counter += 1 @@ -123,6 +146,7 @@ class VariableNamer: def clear_names(self): self.name_counter = 0 self.used_variable_names = set() + self.op_name_counter.clear() class AttributeNamer: @@ -170,8 +194,12 @@ def process_line(line_chunks, variable_namer, use_ssa_name=False, strict_name_re # Process the rest that contained an SSA value name. for chunk in line_chunks: - m = SSA_RE.match(chunk) - ssa_name = m.group(0) if m is not None else '' + ssa = SSA_RE.match(chunk) + op_name_with_dialect = SSA_OP_NAME_RE.search(chunk) + ssa_name = ssa.group(0) if ssa is not None else "" + op_name = ( + op_name_with_dialect.group(1) if op_name_with_dialect is not None else "" + ) # Check if an existing variable exists for this name. variable = None @@ -185,7 +213,7 @@ def process_line(line_chunks, variable_namer, use_ssa_name=False, strict_name_re output_line += "%[[" + variable + "]]" else: # Otherwise, generate a new variable. - variable = variable_namer.generate_name(ssa_name, use_ssa_name) + variable = variable_namer.generate_name(ssa_name, use_ssa_name, op_name) if strict_name_re: # Use stricter regexp for the variable name, if requested. # Greedy matching may cause issues with the generic '.*' diff --git a/offload/liboffload/API/Common.td b/offload/liboffload/API/Common.td index ac27d85..b472236 100644 --- a/offload/liboffload/API/Common.td +++ b/offload/liboffload/API/Common.td @@ -140,9 +140,10 @@ def ol_dimensions_t : Struct { } def olInit : Function { - let desc = "Perform initialization of the Offload library and plugins"; + let desc = "Perform initialization of the Offload library"; let details = [ "This must be the first API call made by a user of the Offload library", + "The underlying platforms are lazily initialized on their first use" "Each call will increment an internal reference count that is decremented by `olShutDown`" ]; let params = []; diff --git a/offload/liboffload/src/OffloadImpl.cpp b/offload/liboffload/src/OffloadImpl.cpp index c549ae0..6d22fae 100644 --- a/offload/liboffload/src/OffloadImpl.cpp +++ b/offload/liboffload/src/OffloadImpl.cpp @@ -42,9 +42,7 @@ using namespace error; struct ol_platform_impl_t { ol_platform_impl_t(std::unique_ptr<GenericPluginTy> Plugin, ol_platform_backend_t BackendType) - : Plugin(std::move(Plugin)), BackendType(BackendType) {} - std::unique_ptr<GenericPluginTy> Plugin; - llvm::SmallVector<std::unique_ptr<ol_device_impl_t>> Devices; + : BackendType(BackendType), Plugin(std::move(Plugin)) {} ol_platform_backend_t BackendType; /// Complete all pending work for this platform and perform any needed @@ -53,6 +51,14 @@ struct ol_platform_impl_t { /// After calling this function, no liboffload functions should be called with /// this platform handle. llvm::Error destroy(); + + /// Initialize the associated plugin and devices. + llvm::Error init(); + + /// Direct access to the plugin, may be uninitialized if accessed here. + std::unique_ptr<GenericPluginTy> Plugin; + + llvm::SmallVector<std::unique_ptr<ol_device_impl_t>> Devices; }; // Handle type definitions. Ideally these would be 1:1 with the plugins, but @@ -130,6 +136,28 @@ llvm::Error ol_platform_impl_t::destroy() { return Result; } +llvm::Error ol_platform_impl_t::init() { + if (!Plugin) + return llvm::Error::success(); + + if (llvm::Error Err = Plugin->init()) + return Err; + + for (auto Id = 0, End = Plugin->getNumDevices(); Id != End; Id++) { + if (llvm::Error Err = Plugin->initDevice(Id)) + return Err; + + auto Device = &Plugin->getDevice(Id); + auto Info = Device->obtainInfoImpl(); + if (llvm::Error Err = Info.takeError()) + return Err; + Devices.emplace_back(std::make_unique<ol_device_impl_t>(Id, Device, *this, + std::move(*Info))); + } + + return llvm::Error::success(); +} + struct ol_queue_impl_t { ol_queue_impl_t(__tgt_async_info *AsyncInfo, ol_device_handle_t Device) : AsyncInfo(AsyncInfo), Device(Device), Id(IdCounter++) {} @@ -207,15 +235,11 @@ struct OffloadContext { std::mutex AllocInfoMapMutex{}; // Partitioned list of memory base addresses. Each element in this list is a // key in AllocInfoMap - llvm::SmallVector<void *> AllocBases{}; + SmallVector<void *> AllocBases{}; SmallVector<std::unique_ptr<ol_platform_impl_t>, 4> Platforms{}; + ol_device_handle_t HostDevice; size_t RefCount; - ol_device_handle_t HostDevice() { - // The host platform is always inserted last - return Platforms.back()->Devices[0].get(); - } - static OffloadContext &get() { assert(OffloadContextVal); return *OffloadContextVal; @@ -259,28 +283,21 @@ Error initPlugins(OffloadContext &Context) { } while (false); #include "Shared/Targets.def" - // Preemptively initialize all devices in the plugin + // Eagerly initialize all of the plugins and devices. We need to make sure + // that the platform is initialized at a consistent point to maintain the + // expected teardown order in the vendor libraries. for (auto &Platform : Context.Platforms) { - auto Err = Platform->Plugin->init(); - [[maybe_unused]] std::string InfoMsg = toString(std::move(Err)); - for (auto DevNum = 0; DevNum < Platform->Plugin->number_of_devices(); - DevNum++) { - if (Platform->Plugin->init_device(DevNum) == OFFLOAD_SUCCESS) { - auto Device = &Platform->Plugin->getDevice(DevNum); - auto Info = Device->obtainInfoImpl(); - if (auto Err = Info.takeError()) - return Err; - Platform->Devices.emplace_back(std::make_unique<ol_device_impl_t>( - DevNum, Device, *Platform, std::move(*Info))); - } - } + if (Error Err = Platform->init()) + return Err; } - // Add the special host device + // Add the special host device. auto &HostPlatform = Context.Platforms.emplace_back( std::make_unique<ol_platform_impl_t>(nullptr, OL_PLATFORM_BACKEND_HOST)); - HostPlatform->Devices.emplace_back(std::make_unique<ol_device_impl_t>( - -1, nullptr, *HostPlatform, InfoTreeNode{})); + Context.HostDevice = HostPlatform->Devices + .emplace_back(std::make_unique<ol_device_impl_t>( + -1, nullptr, *HostPlatform, InfoTreeNode{})) + .get(); Context.TracingEnabled = std::getenv("OFFLOAD_TRACE"); Context.ValidationEnabled = !std::getenv("OFFLOAD_DISABLE_VALIDATION"); @@ -312,16 +329,16 @@ Error olShutDown_impl() { if (--OffloadContext::get().RefCount != 0) return Error::success(); - llvm::Error Result = Error::success(); + Error Result = Error::success(); auto *OldContext = OffloadContextVal.exchange(nullptr); - for (auto &P : OldContext->Platforms) { + for (auto &Platform : OldContext->Platforms) { // Host plugin is nullptr and has no deinit - if (!P->Plugin || !P->Plugin->is_initialized()) + if (!Platform->Plugin || !Platform->Plugin->is_initialized()) continue; - if (auto Res = P->destroy()) - Result = llvm::joinErrors(std::move(Result), std::move(Res)); + if (auto Res = Platform->destroy()) + Result = joinErrors(std::move(Result), std::move(Res)); } delete OldContext; @@ -334,6 +351,8 @@ Error olGetPlatformInfoImplDetail(ol_platform_handle_t Platform, InfoWriter Info(PropSize, PropValue, PropSizeRet); bool IsHost = Platform->BackendType == OL_PLATFORM_BACKEND_HOST; + // Note that the plugin is potentially uninitialized here. It will need to be + // initialized once info is added that requires it to be initialized. switch (PropName) { case OL_PLATFORM_INFO_NAME: return Info.writeString(IsHost ? "Host" : Platform->Plugin->getName()); @@ -373,12 +392,12 @@ Error olGetPlatformInfoSize_impl(ol_platform_handle_t Platform, Error olGetDeviceInfoImplDetail(ol_device_handle_t Device, ol_device_info_t PropName, size_t PropSize, void *PropValue, size_t *PropSizeRet) { - assert(Device != OffloadContext::get().HostDevice()); + assert(Device != OffloadContext::get().HostDevice); InfoWriter Info(PropSize, PropValue, PropSizeRet); auto makeError = [&](ErrorCode Code, StringRef Err) { std::string ErrBuffer; - llvm::raw_string_ostream(ErrBuffer) << PropName << ": " << Err; + raw_string_ostream(ErrBuffer) << PropName << ": " << Err; return Plugin::error(ErrorCode::UNIMPLEMENTED, ErrBuffer.c_str()); }; @@ -511,7 +530,7 @@ Error olGetDeviceInfoImplDetail(ol_device_handle_t Device, Error olGetDeviceInfoImplDetailHost(ol_device_handle_t Device, ol_device_info_t PropName, size_t PropSize, void *PropValue, size_t *PropSizeRet) { - assert(Device == OffloadContext::get().HostDevice()); + assert(Device == OffloadContext::get().HostDevice); InfoWriter Info(PropSize, PropValue, PropSizeRet); constexpr auto uint32_max = std::numeric_limits<uint32_t>::max(); @@ -579,7 +598,7 @@ Error olGetDeviceInfoImplDetailHost(ol_device_handle_t Device, Error olGetDeviceInfo_impl(ol_device_handle_t Device, ol_device_info_t PropName, size_t PropSize, void *PropValue) { - if (Device == OffloadContext::get().HostDevice()) + if (Device == OffloadContext::get().HostDevice) return olGetDeviceInfoImplDetailHost(Device, PropName, PropSize, PropValue, nullptr); return olGetDeviceInfoImplDetail(Device, PropName, PropSize, PropValue, @@ -588,7 +607,7 @@ Error olGetDeviceInfo_impl(ol_device_handle_t Device, ol_device_info_t PropName, Error olGetDeviceInfoSize_impl(ol_device_handle_t Device, ol_device_info_t PropName, size_t *PropSizeRet) { - if (Device == OffloadContext::get().HostDevice()) + if (Device == OffloadContext::get().HostDevice) return olGetDeviceInfoImplDetailHost(Device, PropName, 0, nullptr, PropSizeRet); return olGetDeviceInfoImplDetail(Device, PropName, 0, nullptr, PropSizeRet); @@ -598,7 +617,7 @@ Error olIterateDevices_impl(ol_device_iterate_cb_t Callback, void *UserData) { for (auto &Platform : OffloadContext::get().Platforms) { for (auto &Device : Platform->Devices) { if (!Callback(Device.get(), UserData)) { - break; + return Error::success(); } } } @@ -949,7 +968,7 @@ Error olCreateEvent_impl(ol_queue_handle_t Queue, ol_event_handle_t *EventOut) { Error olMemcpy_impl(ol_queue_handle_t Queue, void *DstPtr, ol_device_handle_t DstDevice, const void *SrcPtr, ol_device_handle_t SrcDevice, size_t Size) { - auto Host = OffloadContext::get().HostDevice(); + auto Host = OffloadContext::get().HostDevice; if (DstDevice == Host && SrcDevice == Host) { if (!Queue) { std::memcpy(DstPtr, SrcPtr, Size); @@ -1138,7 +1157,7 @@ Error olGetSymbolInfoImplDetail(ol_symbol_handle_t Symbol, auto CheckKind = [&](ol_symbol_kind_t Required) { if (Symbol->Kind != Required) { std::string ErrBuffer; - llvm::raw_string_ostream(ErrBuffer) + raw_string_ostream(ErrBuffer) << PropName << ": Expected a symbol of Kind " << Required << " but given a symbol of Kind " << Symbol->Kind; return Plugin::error(ErrorCode::SYMBOL_KIND, ErrBuffer.c_str()); |