[𝘀𝗽𝗿] changes introduced through rebaseusers/Prabhuk/sprmain.callgraph-make-flag-experimental

Created using spr 1.3.6-beta.1

[skip ci]

286 files changed, 5607 insertions, 2057 deletions

diff --git a/clang-tools-extra/clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp b/clang-tools-extra/clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp
index 062e236..76de8bd 100644
--- a/clang-tools-extra/clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp
+++ b/clang-tools-extra/clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp
@@ -97,8 +97,7 @@ int main(int argc, char **argv) {
   cl::ParseCommandLineOptions(argc, argv);
 
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts);
 
   // Determine a formatting style from options.
   auto FormatStyleOrError = format::getStyle(FormatStyleOpt, FormatStyleConfig,
diff --git a/clang-tools-extra/clang-change-namespace/tool/ClangChangeNamespace.cpp b/clang-tools-extra/clang-change-namespace/tool/ClangChangeNamespace.cpp
index 2a8fe2d..2efdd92 100644
--- a/clang-tools-extra/clang-change-namespace/tool/ClangChangeNamespace.cpp
+++ b/clang-tools-extra/clang-change-namespace/tool/ClangChangeNamespace.cpp
@@ -128,9 +128,8 @@ int main(int argc, const char **argv) {
   LangOptions DefaultLangOptions;
   DiagnosticOptions DiagOpts;
   clang::TextDiagnosticPrinter DiagnosticPrinter(errs(), DiagOpts);
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts,
-      &DiagnosticPrinter, false);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts,
+                                &DiagnosticPrinter, false);
   auto &FileMgr = Tool.getFiles();
   SourceManager Sources(Diagnostics, FileMgr);
   Rewriter Rewrite(Sources, DefaultLangOptions);
diff --git a/clang-tools-extra/clang-include-fixer/tool/ClangIncludeFixer.cpp b/clang-tools-extra/clang-include-fixer/tool/ClangIncludeFixer.cpp
index 9f73f47..568cb2b 100644
--- a/clang-tools-extra/clang-include-fixer/tool/ClangIncludeFixer.cpp
+++ b/clang-tools-extra/clang-include-fixer/tool/ClangIncludeFixer.cpp
@@ -454,7 +454,7 @@ int includeFixerMain(int argc, const char **argv) {
 
   // Set up a new source manager for applying the resulting replacements.
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diagnostics(new DiagnosticIDs, DiagOpts);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts);
   TextDiagnosticPrinter DiagnosticPrinter(outs(), DiagOpts);
   SourceManager SM(Diagnostics, tool.getFiles());
   Diagnostics.setClient(&DiagnosticPrinter, false);
diff --git a/clang-tools-extra/clang-move/tool/ClangMove.cpp b/clang-tools-extra/clang-move/tool/ClangMove.cpp
index 750eb95..1be3cb1 100644
--- a/clang-tools-extra/clang-move/tool/ClangMove.cpp
+++ b/clang-tools-extra/clang-move/tool/ClangMove.cpp
@@ -178,9 +178,8 @@ int main(int argc, const char **argv) {
 
   DiagnosticOptions DiagOpts;
   clang::TextDiagnosticPrinter DiagnosticPrinter(errs(), DiagOpts);
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts,
-      &DiagnosticPrinter, false);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts,
+                                &DiagnosticPrinter, false);
   auto &FileMgr = Tool.getFiles();
   SourceManager SM(Diagnostics, FileMgr);
   Rewriter Rewrite(SM, LangOptions());
diff --git a/clang-tools-extra/clang-reorder-fields/tool/ClangReorderFields.cpp b/clang-tools-extra/clang-reorder-fields/tool/ClangReorderFields.cpp
index 0350252..fbfce07 100644
--- a/clang-tools-extra/clang-reorder-fields/tool/ClangReorderFields.cpp
+++ b/clang-tools-extra/clang-reorder-fields/tool/ClangReorderFields.cpp
@@ -74,9 +74,8 @@ int main(int argc, const char **argv) {
   LangOptions DefaultLangOptions;
   DiagnosticOptions DiagOpts;
   TextDiagnosticPrinter DiagnosticPrinter(errs(), DiagOpts);
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts,
-      &DiagnosticPrinter, false);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts,
+                                &DiagnosticPrinter, false);
 
   auto &FileMgr = Tool.getFiles();
   SourceManager Sources(Diagnostics, FileMgr);
diff --git a/clang-tools-extra/clang-tidy/ClangTidy.cpp b/clang-tools-extra/clang-tidy/ClangTidy.cpp
index e84be04..4ae2864 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.cpp
+++ b/clang-tools-extra/clang-tidy/ClangTidy.cpp
@@ -96,8 +96,7 @@ public:
                 llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> BaseFS)
       : Files(FileSystemOptions(), std::move(BaseFS)),
         DiagPrinter(new TextDiagnosticPrinter(llvm::outs(), DiagOpts)),
-        Diags(IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs), DiagOpts,
-              DiagPrinter),
+        Diags(DiagnosticIDs::create(), DiagOpts, DiagPrinter),
         SourceMgr(Diags, Files), Context(Context), ApplyFixes(ApplyFixes) {
     DiagOpts.ShowColors = Context.getOptions().UseColor.value_or(
         llvm::sys::Process::StandardOutHasColors());
@@ -570,7 +569,7 @@ runClangTidy(clang::tidy::ClangTidyContext &Context,
 
   ClangTidyDiagnosticConsumer DiagConsumer(Context, nullptr, true, ApplyAnyFix);
   auto DiagOpts = std::make_unique<DiagnosticOptions>();
-  DiagnosticsEngine DE(new DiagnosticIDs(), *DiagOpts, &DiagConsumer,
+  DiagnosticsEngine DE(DiagnosticIDs::create(), *DiagOpts, &DiagConsumer,
                        /*ShouldOwnClient=*/false);
   Context.setDiagnosticsEngine(std::move(DiagOpts), &DE);
   Tool.setDiagnosticConsumer(&DiagConsumer);
diff --git a/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.cpp b/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.cpp
index 2c17cd3..5e705f7 100644
--- a/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.cpp
+++ b/clang-tools-extra/clang-tidy/ExpandModularHeadersPPCallbacks.cpp
@@ -71,7 +71,7 @@ ExpandModularHeadersPPCallbacks::ExpandModularHeadersPPCallbacks(
       InMemoryFs(new llvm::vfs::InMemoryFileSystem),
       Sources(Compiler.getSourceManager()),
       // Forward the new diagnostics to the original DiagnosticConsumer.
-      Diags(new DiagnosticIDs, DiagOpts,
+      Diags(DiagnosticIDs::create(), DiagOpts,
             new ForwardingDiagnosticConsumer(Compiler.getDiagnosticClient())),
       LangOpts(Compiler.getLangOpts()), HSOpts(Compiler.getHeaderSearchOpts()) {
   // Add a FileSystem containing the extra files needed in place of modular
diff --git a/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp b/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp
index 651a63b..195418d 100644
--- a/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp
+++ b/clang-tools-extra/clang-tidy/plugin/ClangTidyPlugin.cpp
@@ -41,7 +41,7 @@ public:
         new ClangTidyDiagnosticConsumer(*Context, &Compiler.getDiagnostics());
     auto DiagOpts = std::make_unique<DiagnosticOptions>();
     auto DiagEngine = std::make_unique<DiagnosticsEngine>(
-        new DiagnosticIDs, *DiagOpts, DiagConsumer);
+        DiagnosticIDs::create(), *DiagOpts, DiagConsumer);
     Context->setDiagnosticsEngine(std::move(DiagOpts), DiagEngine.get());
 
     // Create the AST consumer.
diff --git a/clang-tools-extra/clangd/Preamble.cpp b/clang-tools-extra/clangd/Preamble.cpp
index 7b4d63f..8af9e46 100644
--- a/clang-tools-extra/clangd/Preamble.cpp
+++ b/clang-tools-extra/clangd/Preamble.cpp
@@ -659,7 +659,7 @@ buildPreamble(PathRef FileName, CompilerInvocation CI,
   WallTimer PreambleTimer;
   PreambleTimer.startTimer();
   auto BuiltPreamble = PrecompiledPreamble::Build(
-      CI, ContentsBuffer.get(), Bounds, *PreambleDiagsEngine,
+      CI, ContentsBuffer.get(), Bounds, PreambleDiagsEngine,
       Stats ? TimedFS : StatCacheFS, std::make_shared<PCHContainerOperations>(),
       StoreInMemory, /*StoragePath=*/"", CapturedInfo);
 
diff --git a/clang-tools-extra/clangd/SystemIncludeExtractor.cpp b/clang-tools-extra/clangd/SystemIncludeExtractor.cpp
index 0b067e8..106de1b 100644
--- a/clang-tools-extra/clangd/SystemIncludeExtractor.cpp
+++ b/clang-tools-extra/clangd/SystemIncludeExtractor.cpp
@@ -254,7 +254,7 @@ bool isValidTarget(llvm::StringRef Triple) {
   std::shared_ptr<TargetOptions> TargetOpts(new TargetOptions);
   TargetOpts->Triple = Triple.str();
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(new DiagnosticIDs, DiagOpts,
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                           new IgnoringDiagConsumer);
   llvm::IntrusiveRefCntPtr<TargetInfo> Target =
       TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
diff --git a/clang-tools-extra/clangd/unittests/ConfigCompileTests.cpp b/clang-tools-extra/clangd/unittests/ConfigCompileTests.cpp
index 75d0ff2..0e411b2 100644
--- a/clang-tools-extra/clangd/unittests/ConfigCompileTests.cpp
+++ b/clang-tools-extra/clangd/unittests/ConfigCompileTests.cpp
@@ -299,7 +299,7 @@ TEST_F(ConfigCompileTests, DiagnosticSuppression) {
                                    "typecheck_bool_condition",
                                    "unexpected_friend", "warn_alloca"));
   clang::DiagnosticOptions DiagOpts;
-  clang::DiagnosticsEngine DiagEngine(new DiagnosticIDs, DiagOpts,
+  clang::DiagnosticsEngine DiagEngine(DiagnosticIDs::create(), DiagOpts,
                                       new clang::IgnoringDiagConsumer);
 
   using Diag = clang::Diagnostic;
diff --git a/clang-tools-extra/clangd/unittests/tweaks/TweakTests.cpp b/clang-tools-extra/clangd/unittests/tweaks/TweakTests.cpp
index e39b702..b6607e9 100644
--- a/clang-tools-extra/clangd/unittests/tweaks/TweakTests.cpp
+++ b/clang-tools-extra/clangd/unittests/tweaks/TweakTests.cpp
@@ -45,7 +45,7 @@ TEST(FileEdits, AbsolutePath) {
     MemFS->addFile(Path, 0, llvm::MemoryBuffer::getMemBuffer("", Path));
   FileManager FM(FileSystemOptions(), MemFS);
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine DE(new DiagnosticIDs, DiagOpts);
+  DiagnosticsEngine DE(DiagnosticIDs::create(), DiagOpts);
   SourceManager SM(DE, FM);
 
   for (const auto *Path : RelPaths) {
diff --git a/clang-tools-extra/modularize/ModularizeUtilities.cpp b/clang-tools-extra/modularize/ModularizeUtilities.cpp
index 8a24f21..4dd84fe 100644
--- a/clang-tools-extra/modularize/ModularizeUtilities.cpp
+++ b/clang-tools-extra/modularize/ModularizeUtilities.cpp
@@ -47,7 +47,7 @@ ModularizeUtilities::ModularizeUtilities(std::vector<std::string> &InputPaths,
       ProblemFilesPath(ProblemFilesListPath), HasModuleMap(false),
       MissingHeaderCount(0),
       // Init clang stuff needed for loading the module map and preprocessing.
-      LangOpts(new LangOptions()), DiagIDs(new DiagnosticIDs()),
+      LangOpts(new LangOptions()), DiagIDs(DiagnosticIDs::create()),
       DC(llvm::errs(), DiagnosticOpts),
       Diagnostics(new DiagnosticsEngine(DiagIDs, DiagnosticOpts, &DC, false)),
       TargetOpts(new ModuleMapTargetOptions()),
diff --git a/clang-tools-extra/test/clang-apply-replacements/basic.cpp b/clang-tools-extra/test/clang-apply-replacements/basic.cpp
index 4f19a96..2399307 100644
--- a/clang-tools-extra/test/clang-apply-replacements/basic.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/basic.cpp
@@ -1,17 +1,17 @@
-// RUN: mkdir -p %T/Inputs/basic
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/basic/basic.h > %T/Inputs/basic/basic.h
-// RUN: sed "s#\$(path)#%/T/Inputs/basic#" %S/Inputs/basic/file1.yaml > %T/Inputs/basic/file1.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/basic#" %S/Inputs/basic/file2.yaml > %T/Inputs/basic/file2.yaml
-// RUN: clang-apply-replacements %T/Inputs/basic
-// RUN: FileCheck -input-file=%T/Inputs/basic/basic.h %S/Inputs/basic/basic.h
+// RUN: mkdir -p %t.dir/Inputs/basic
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/basic/basic.h > %t.dir/Inputs/basic/basic.h
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/basic#" %S/Inputs/basic/file1.yaml > %t.dir/Inputs/basic/file1.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/basic#" %S/Inputs/basic/file2.yaml > %t.dir/Inputs/basic/file2.yaml
+// RUN: clang-apply-replacements %t.dir/Inputs/basic
+// RUN: FileCheck -input-file=%t.dir/Inputs/basic/basic.h %S/Inputs/basic/basic.h
 //
 // Check that the yaml files are *not* deleted after running clang-apply-replacements without remove-change-desc-files.
-// RUN: ls -1 %T/Inputs/basic | FileCheck %s --check-prefix=YAML
+// RUN: ls -1 %t.dir/Inputs/basic | FileCheck %s --check-prefix=YAML
 //
 // Check that the yaml files *are* deleted after running clang-apply-replacements with remove-change-desc-files.
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/basic/basic.h > %T/Inputs/basic/basic.h
-// RUN: clang-apply-replacements -remove-change-desc-files %T/Inputs/basic
-// RUN: ls -1 %T/Inputs/basic | FileCheck %s --check-prefix=NO_YAML
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/basic/basic.h > %t.dir/Inputs/basic/basic.h
+// RUN: clang-apply-replacements -remove-change-desc-files %t.dir/Inputs/basic
+// RUN: ls -1 %t.dir/Inputs/basic | FileCheck %s --check-prefix=NO_YAML
 //
 // YAML: {{^file.\.yaml$}}
 // NO_YAML-NOT: {{^file.\.yaml$}}
diff --git a/clang-tools-extra/test/clang-apply-replacements/conflict.cpp b/clang-tools-extra/test/clang-apply-replacements/conflict.cpp
index c1f2342..7b0a8c2 100644
--- a/clang-tools-extra/test/clang-apply-replacements/conflict.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/conflict.cpp
@@ -1,17 +1,17 @@
-// RUN: mkdir -p %T/Inputs/conflict
-// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/file1.yaml > %T/Inputs/conflict/file1.yaml
-// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/file2.yaml > %T/Inputs/conflict/file2.yaml
-// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/file3.yaml > %T/Inputs/conflict/file3.yaml
-// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/expected.txt > %T/Inputs/conflict/expected.txt
-// RUN: not clang-apply-replacements %T/Inputs/conflict > %T/Inputs/conflict/output.txt 2>&1
-// RUN: diff -b %T/Inputs/conflict/output.txt %T/Inputs/conflict/expected.txt
+// RUN: mkdir -p %t.dir/Inputs/conflict
+// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/file1.yaml > %t.dir/Inputs/conflict/file1.yaml
+// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/file2.yaml > %t.dir/Inputs/conflict/file2.yaml
+// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/file3.yaml > %t.dir/Inputs/conflict/file3.yaml
+// RUN: sed "s#\$(path)#%/S/Inputs/conflict#" %S/Inputs/conflict/expected.txt > %t.dir/Inputs/conflict/expected.txt
+// RUN: not clang-apply-replacements %t.dir/Inputs/conflict > %t.dir/Inputs/conflict/output.txt 2>&1
+// RUN: diff -b %t.dir/Inputs/conflict/output.txt %t.dir/Inputs/conflict/expected.txt
 //
 // Check that the yaml files are *not* deleted after running clang-apply-replacements without remove-change-desc-files even when there is a failure.
-// RUN: ls -1 %T/Inputs/conflict | FileCheck %s --check-prefix=YAML
+// RUN: ls -1 %t.dir/Inputs/conflict | FileCheck %s --check-prefix=YAML
 //
 // Check that the yaml files *are* deleted after running clang-apply-replacements with remove-change-desc-files even when there is a failure.
-// RUN: not clang-apply-replacements %T/Inputs/conflict -remove-change-desc-files > %T/Inputs/conflict/output.txt 2>&1
-// RUN: ls -1 %T/Inputs/conflict | FileCheck %s --check-prefix=NO_YAML
+// RUN: not clang-apply-replacements %t.dir/Inputs/conflict -remove-change-desc-files > %t.dir/Inputs/conflict/output.txt 2>&1
+// RUN: ls -1 %t.dir/Inputs/conflict | FileCheck %s --check-prefix=NO_YAML
 //
 // YAML: {{^file.\.yaml$}}
 // NO_YAML-NOT: {{^file.\.yaml$}}
diff --git a/clang-tools-extra/test/clang-apply-replacements/crlf.cpp b/clang-tools-extra/test/clang-apply-replacements/crlf.cpp
index 15ba5b5..f40429e 100644
--- a/clang-tools-extra/test/clang-apply-replacements/crlf.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/crlf.cpp
@@ -1,5 +1,5 @@
-// RUN: mkdir -p %T/Inputs/crlf
-// RUN: cat %S/Inputs/crlf/crlf.cpp > %T/Inputs/crlf/crlf.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/crlf#" %S/Inputs/crlf/file1.yaml > %T/Inputs/crlf/file1.yaml
-// RUN: clang-apply-replacements %T/Inputs/crlf
-// RUN: diff %T/Inputs/crlf/crlf.cpp %S/Inputs/crlf/crlf.cpp.expected
+// RUN: mkdir -p %t.dir/Inputs/crlf
+// RUN: cat %S/Inputs/crlf/crlf.cpp > %t.dir/Inputs/crlf/crlf.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/crlf#" %S/Inputs/crlf/file1.yaml > %t.dir/Inputs/crlf/file1.yaml
+// RUN: clang-apply-replacements %t.dir/Inputs/crlf
+// RUN: diff %t.dir/Inputs/crlf/crlf.cpp %S/Inputs/crlf/crlf.cpp.expected
diff --git a/clang-tools-extra/test/clang-apply-replacements/format-header.cpp b/clang-tools-extra/test/clang-apply-replacements/format-header.cpp
index 6a221c4..9d2680e 100644
--- a/clang-tools-extra/test/clang-apply-replacements/format-header.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/format-header.cpp
@@ -1,13 +1,13 @@
-// RUN: mkdir -p %T/Inputs/format_header_yes
-// RUN: mkdir -p %T/Inputs/format_header_no
+// RUN: mkdir -p %t.dir/Inputs/format_header_yes
+// RUN: mkdir -p %t.dir/Inputs/format_header_no
 //
 //
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format_header/yes.cpp > %T/Inputs/format_header_yes/yes.cpp
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format_header/no.cpp > %T/Inputs/format_header_no/no.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/format_header_yes#" %S/Inputs/format_header/yes.yaml > %T/Inputs/format_header_yes/yes.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/format_header_no#" %S/Inputs/format_header/no.yaml > %T/Inputs/format_header_no/no.yaml
-// RUN: clang-apply-replacements -format -style="{BasedOnStyle: llvm, SortIncludes: CaseSensitive}" %T/Inputs/format_header_yes
-// RUN: clang-apply-replacements %T/Inputs/format_header_no
-// RUN: FileCheck --strict-whitespace -input-file=%T/Inputs/format_header_yes/yes.cpp %S/Inputs/format_header/yes.cpp
-// RUN: FileCheck --strict-whitespace -input-file=%T/Inputs/format_header_no/no.cpp %S/Inputs/format_header/no.cpp
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format_header/yes.cpp > %t.dir/Inputs/format_header_yes/yes.cpp
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format_header/no.cpp > %t.dir/Inputs/format_header_no/no.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/format_header_yes#" %S/Inputs/format_header/yes.yaml > %t.dir/Inputs/format_header_yes/yes.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/format_header_no#" %S/Inputs/format_header/no.yaml > %t.dir/Inputs/format_header_no/no.yaml
+// RUN: clang-apply-replacements -format -style="{BasedOnStyle: llvm, SortIncludes: CaseSensitive}" %t.dir/Inputs/format_header_yes
+// RUN: clang-apply-replacements %t.dir/Inputs/format_header_no
+// RUN: FileCheck --strict-whitespace -input-file=%t.dir/Inputs/format_header_yes/yes.cpp %S/Inputs/format_header/yes.cpp
+// RUN: FileCheck --strict-whitespace -input-file=%t.dir/Inputs/format_header_no/no.cpp %S/Inputs/format_header/no.cpp
 //
diff --git a/clang-tools-extra/test/clang-apply-replacements/format.cpp b/clang-tools-extra/test/clang-apply-replacements/format.cpp
index 7de320d..0f40ef62 100644
--- a/clang-tools-extra/test/clang-apply-replacements/format.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/format.cpp
@@ -1,15 +1,15 @@
-// RUN: mkdir -p %T/Inputs/format
+// RUN: mkdir -p %t.dir/Inputs/format
 //
 // yes.cpp requires formatting after replacements are applied. no.cpp does not.
 // The presence of no.cpp ensures that files that don't need formatting still
 // have their new state written to disk after applying replacements.
 //
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format/yes.cpp > %T/Inputs/format/yes.cpp
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format/no.cpp > %T/Inputs/format/no.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/format#" %S/Inputs/format/yes.yaml > %T/Inputs/format/yes.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/format#" %S/Inputs/format/no.yaml > %T/Inputs/format/no.yaml
-// RUN: clang-apply-replacements -format %T/Inputs/format
-// RUN: FileCheck --strict-whitespace -input-file=%T/Inputs/format/yes.cpp %S/Inputs/format/yes.cpp
-// RUN: FileCheck --strict-whitespace -input-file=%T/Inputs/format/no.cpp %S/Inputs/format/no.cpp
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format/yes.cpp > %t.dir/Inputs/format/yes.cpp
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/format/no.cpp > %t.dir/Inputs/format/no.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/format#" %S/Inputs/format/yes.yaml > %t.dir/Inputs/format/yes.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/format#" %S/Inputs/format/no.yaml > %t.dir/Inputs/format/no.yaml
+// RUN: clang-apply-replacements -format %t.dir/Inputs/format
+// RUN: FileCheck --strict-whitespace -input-file=%t.dir/Inputs/format/yes.cpp %S/Inputs/format/yes.cpp
+// RUN: FileCheck --strict-whitespace -input-file=%t.dir/Inputs/format/no.cpp %S/Inputs/format/no.cpp
 //
-// RUN not clang-apply-replacements -format=blah %T/Inputs/format
+// RUN not clang-apply-replacements -format=blah %t.dir/Inputs/format
diff --git a/clang-tools-extra/test/clang-apply-replacements/identical-in-TU.cpp b/clang-tools-extra/test/clang-apply-replacements/identical-in-TU.cpp
index 024db11..df9c4fc 100644
--- a/clang-tools-extra/test/clang-apply-replacements/identical-in-TU.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/identical-in-TU.cpp
@@ -1,10 +1,10 @@
-// RUN: mkdir -p %T/Inputs/identical-in-TU
+// RUN: mkdir -p %t.dir/Inputs/identical-in-TU
 
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/identical-in-TU/identical-in-TU.cpp > %T/Inputs/identical-in-TU/identical-in-TU.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/identical-in-TU#" %S/Inputs/identical-in-TU/file1.yaml > %T/Inputs/identical-in-TU/file1.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/identical-in-TU#" %S/Inputs/identical-in-TU/file2.yaml > %T/Inputs/identical-in-TU/file2.yaml
-// RUN: clang-apply-replacements %T/Inputs/identical-in-TU
-// RUN: FileCheck -input-file=%T/Inputs/identical-in-TU/identical-in-TU.cpp %S/Inputs/identical-in-TU/identical-in-TU.cpp
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/identical-in-TU/identical-in-TU.cpp > %t.dir/Inputs/identical-in-TU/identical-in-TU.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/identical-in-TU#" %S/Inputs/identical-in-TU/file1.yaml > %t.dir/Inputs/identical-in-TU/file1.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/identical-in-TU#" %S/Inputs/identical-in-TU/file2.yaml > %t.dir/Inputs/identical-in-TU/file2.yaml
+// RUN: clang-apply-replacements %t.dir/Inputs/identical-in-TU
+// RUN: FileCheck -input-file=%t.dir/Inputs/identical-in-TU/identical-in-TU.cpp %S/Inputs/identical-in-TU/identical-in-TU.cpp
 
 // Similar to identical test but each yaml file contains the same fix twice. 
 // This check ensures that only the duplicated replacements in a single yaml 
diff --git a/clang-tools-extra/test/clang-apply-replacements/identical.cpp b/clang-tools-extra/test/clang-apply-replacements/identical.cpp
index ffbf2e3..8a2d1e5 100644
--- a/clang-tools-extra/test/clang-apply-replacements/identical.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/identical.cpp
@@ -1,6 +1,6 @@
-// RUN: mkdir -p %T/Inputs/identical
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/identical/identical.cpp > %T/Inputs/identical/identical.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/identical#" %S/Inputs/identical/file1.yaml > %T/Inputs/identical/file1.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/identical#" %S/Inputs/identical/file2.yaml > %T/Inputs/identical/file2.yaml
-// RUN: clang-apply-replacements %T/Inputs/identical
-// RUN: FileCheck -input-file=%T/Inputs/identical/identical.cpp %S/Inputs/identical/identical.cpp
+// RUN: mkdir -p %t.dir/Inputs/identical
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/identical/identical.cpp > %t.dir/Inputs/identical/identical.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/identical#" %S/Inputs/identical/file1.yaml > %t.dir/Inputs/identical/file1.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/identical#" %S/Inputs/identical/file2.yaml > %t.dir/Inputs/identical/file2.yaml
+// RUN: clang-apply-replacements %t.dir/Inputs/identical
+// RUN: FileCheck -input-file=%t.dir/Inputs/identical/identical.cpp %S/Inputs/identical/identical.cpp
diff --git a/clang-tools-extra/test/clang-apply-replacements/ignore-conflict.cpp b/clang-tools-extra/test/clang-apply-replacements/ignore-conflict.cpp
index 4e681dd..e310256 100644
--- a/clang-tools-extra/test/clang-apply-replacements/ignore-conflict.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/ignore-conflict.cpp
@@ -1,5 +1,5 @@
-// RUN: mkdir -p %T/Inputs/ignore-conflict
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/ignore-conflict/ignore-conflict.cpp > %T/Inputs/ignore-conflict/ignore-conflict.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/ignore-conflict#" %S/Inputs/ignore-conflict/file1.yaml > %T/Inputs/ignore-conflict/file1.yaml
-// RUN: clang-apply-replacements --ignore-insert-conflict %T/Inputs/ignore-conflict
-// RUN: FileCheck -input-file=%T/Inputs/ignore-conflict/ignore-conflict.cpp %S/Inputs/ignore-conflict/ignore-conflict.cpp
+// RUN: mkdir -p %t.dir/Inputs/ignore-conflict
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/ignore-conflict/ignore-conflict.cpp > %t.dir/Inputs/ignore-conflict/ignore-conflict.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/ignore-conflict#" %S/Inputs/ignore-conflict/file1.yaml > %t.dir/Inputs/ignore-conflict/file1.yaml
+// RUN: clang-apply-replacements --ignore-insert-conflict %t.dir/Inputs/ignore-conflict
+// RUN: FileCheck -input-file=%t.dir/Inputs/ignore-conflict/ignore-conflict.cpp %S/Inputs/ignore-conflict/ignore-conflict.cpp
diff --git a/clang-tools-extra/test/clang-apply-replacements/invalid-files.cpp b/clang-tools-extra/test/clang-apply-replacements/invalid-files.cpp
index b0eb9ef..09efd4c 100644
--- a/clang-tools-extra/test/clang-apply-replacements/invalid-files.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/invalid-files.cpp
@@ -1,6 +1,6 @@
-// RUN: mkdir -p %T/invalid-files
-// RUN: cp %S/Inputs/invalid-files/invalid-files.yaml %T/invalid-files/invalid-files.yaml
-// RUN: clang-apply-replacements %T/invalid-files
+// RUN: mkdir -p %t.dir/invalid-files
+// RUN: cp %S/Inputs/invalid-files/invalid-files.yaml %t.dir/invalid-files/invalid-files.yaml
+// RUN: clang-apply-replacements %t.dir/invalid-files
 //
 // Check that the yaml files are *not* deleted after running clang-apply-replacements without remove-change-desc-files.
-// RUN: ls %T/invalid-files/invalid-files.yaml
+// RUN: ls %t.dir/invalid-files/invalid-files.yaml
diff --git a/clang-tools-extra/test/clang-apply-replacements/order-dependent.cpp b/clang-tools-extra/test/clang-apply-replacements/order-dependent.cpp
index 769f4f7..32a3bd1 100644
--- a/clang-tools-extra/test/clang-apply-replacements/order-dependent.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/order-dependent.cpp
@@ -1,7 +1,7 @@
-// RUN: mkdir -p %T/Inputs/order-dependent
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/order-dependent/order-dependent.cpp > %T/Inputs/order-dependent/order-dependent.cpp
-// RUN: sed "s#\$(path)#%/T/Inputs/order-dependent#" %S/Inputs/order-dependent/file1.yaml > %T/Inputs/order-dependent/file1.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/order-dependent#" %S/Inputs/order-dependent/file2.yaml > %T/Inputs/order-dependent/file2.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/order-dependent#" %S/Inputs/order-dependent/expected.txt > %T/Inputs/order-dependent/expected.txt
-// RUN: not clang-apply-replacements %T/Inputs/order-dependent > %T/Inputs/order-dependent/output.txt 2>&1
-// RUN: diff -b %T/Inputs/order-dependent/output.txt %T/Inputs/order-dependent/expected.txt
+// RUN: mkdir -p %t.dir/Inputs/order-dependent
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/order-dependent/order-dependent.cpp > %t.dir/Inputs/order-dependent/order-dependent.cpp
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/order-dependent#" %S/Inputs/order-dependent/file1.yaml > %t.dir/Inputs/order-dependent/file1.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/order-dependent#" %S/Inputs/order-dependent/file2.yaml > %t.dir/Inputs/order-dependent/file2.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/order-dependent#" %S/Inputs/order-dependent/expected.txt > %t.dir/Inputs/order-dependent/expected.txt
+// RUN: not clang-apply-replacements %t.dir/Inputs/order-dependent > %t.dir/Inputs/order-dependent/output.txt 2>&1
+// RUN: diff -b %t.dir/Inputs/order-dependent/output.txt %t.dir/Inputs/order-dependent/expected.txt
diff --git a/clang-tools-extra/test/clang-apply-replacements/relative-paths.cpp b/clang-tools-extra/test/clang-apply-replacements/relative-paths.cpp
index 92cde84..36e3e89 100644
--- a/clang-tools-extra/test/clang-apply-replacements/relative-paths.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/relative-paths.cpp
@@ -1,7 +1,7 @@
-// RUN: mkdir -p %T/Inputs/relative-paths
-// RUN: mkdir -p %T/Inputs/relative-paths/subdir
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/relative-paths/basic.h > %T/Inputs/relative-paths/basic.h
-// RUN: sed "s#\$(path)#%/T/Inputs/relative-paths#" %S/Inputs/relative-paths/file1.yaml > %T/Inputs/relative-paths/file1.yaml
-// RUN: sed "s#\$(path)#%/T/Inputs/relative-paths#" %S/Inputs/relative-paths/file2.yaml > %T/Inputs/relative-paths/file2.yaml
-// RUN: clang-apply-replacements %T/Inputs/relative-paths
-// RUN: FileCheck -input-file=%T/Inputs/relative-paths/basic.h %S/Inputs/relative-paths/basic.h
+// RUN: mkdir -p %t.dir/Inputs/relative-paths
+// RUN: mkdir -p %t.dir/Inputs/relative-paths/subdir
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/relative-paths/basic.h > %t.dir/Inputs/relative-paths/basic.h
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/relative-paths#" %S/Inputs/relative-paths/file1.yaml > %t.dir/Inputs/relative-paths/file1.yaml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/relative-paths#" %S/Inputs/relative-paths/file2.yaml > %t.dir/Inputs/relative-paths/file2.yaml
+// RUN: clang-apply-replacements %t.dir/Inputs/relative-paths
+// RUN: FileCheck -input-file=%t.dir/Inputs/relative-paths/basic.h %S/Inputs/relative-paths/basic.h
diff --git a/clang-tools-extra/test/clang-apply-replacements/yml-basic.cpp b/clang-tools-extra/test/clang-apply-replacements/yml-basic.cpp
index e6ee919..e076ff7 100644
--- a/clang-tools-extra/test/clang-apply-replacements/yml-basic.cpp
+++ b/clang-tools-extra/test/clang-apply-replacements/yml-basic.cpp
@@ -1,17 +1,17 @@
-// RUN: mkdir -p %T/Inputs/yml-basic
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/yml-basic/basic.h > %T/Inputs/yml-basic/basic.h
-// RUN: sed "s#\$(path)#%/T/Inputs/yml-basic#" %S/Inputs/yml-basic/file1.yml > %T/Inputs/yml-basic/file1.yml
-// RUN: sed "s#\$(path)#%/T/Inputs/yml-basic#" %S/Inputs/yml-basic/file2.yml > %T/Inputs/yml-basic/file2.yml
-// RUN: clang-apply-replacements %T/Inputs/yml-basic
-// RUN: FileCheck -input-file=%T/Inputs/yml-basic/basic.h %S/Inputs/yml-basic/basic.h
+// RUN: mkdir -p %t.dir/Inputs/yml-basic
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/yml-basic/basic.h > %t.dir/Inputs/yml-basic/basic.h
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/yml-basic#" %S/Inputs/yml-basic/file1.yml > %t.dir/Inputs/yml-basic/file1.yml
+// RUN: sed "s#\$(path)#%/t.dir/Inputs/yml-basic#" %S/Inputs/yml-basic/file2.yml > %t.dir/Inputs/yml-basic/file2.yml
+// RUN: clang-apply-replacements %t.dir/Inputs/yml-basic
+// RUN: FileCheck -input-file=%t.dir/Inputs/yml-basic/basic.h %S/Inputs/yml-basic/basic.h
 //
 // Check that the yml files are *not* deleted after running clang-apply-replacements without remove-change-desc-files.
-// RUN: ls -1 %T/Inputs/yml-basic | FileCheck %s --check-prefix=YML
+// RUN: ls -1 %t.dir/Inputs/yml-basic | FileCheck %s --check-prefix=YML
 //
 // Check that the yml files *are* deleted after running clang-apply-replacements with remove-change-desc-files.
-// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/yml-basic/basic.h > %T/Inputs/yml-basic/basic.h
-// RUN: clang-apply-replacements -remove-change-desc-files %T/Inputs/yml-basic
-// RUN: ls -1 %T/Inputs/yml-basic | FileCheck %s --check-prefix=NO_YML
+// RUN: grep -Ev "// *[A-Z-]+:" %S/Inputs/yml-basic/basic.h > %t.dir/Inputs/yml-basic/basic.h
+// RUN: clang-apply-replacements -remove-change-desc-files %t.dir/Inputs/yml-basic
+// RUN: ls -1 %t.dir/Inputs/yml-basic | FileCheck %s --check-prefix=NO_YML
 //
 // YML: {{^file.\.yml$}}
 // NO_YML-NOT: {{^file.\.yml$}}
diff --git a/clang-tools-extra/test/clang-change-namespace/allow-list.cpp b/clang-tools-extra/test/clang-change-namespace/allow-list.cpp
index 7a941dcb..3b0d5b9 100644
--- a/clang-tools-extra/test/clang-change-namespace/allow-list.cpp
+++ b/clang-tools-extra/test/clang-change-namespace/allow-list.cpp
@@ -1,5 +1,5 @@
-// RUN: echo "^std::.*$" > %T/allow-list.txt
-// RUN: clang-change-namespace -old_namespace "na::nb" -new_namespace "x::y" --file_pattern ".*" --allowed_file %T/allow-list.txt %s -- | sed 's,// CHECK.*,,' | FileCheck %s
+// RUN: echo "^std::.*$" > %t.allow-list.txt
+// RUN: clang-change-namespace -old_namespace "na::nb" -new_namespace "x::y" --file_pattern ".*" --allowed_file %t.allow-list.txt %s -- | sed 's,// CHECK.*,,' | FileCheck %s
 
 #include "Inputs/fake-std.h"
 
diff --git a/clang-tools-extra/test/clang-change-namespace/macro.cpp b/clang-tools-extra/test/clang-change-namespace/macro.cpp
index 40c4caf..f0b134d 100644
--- a/clang-tools-extra/test/clang-change-namespace/macro.cpp
+++ b/clang-tools-extra/test/clang-change-namespace/macro.cpp
@@ -1,15 +1,16 @@
-// RUN: cp %S/macro.cpp %T/macro.cpp
-// RUN: echo "#define USING using na::nc::X" > %T/macro.h
+// RUN: mkdir -p %t.dir
+// RUN: cp %S/macro.cpp %t.dir/macro.cpp
+// RUN: echo "#define USING using na::nc::X" > %t.dir/macro.h
 //
-// RUN: clang-change-namespace -old_namespace "na::nb" -new_namespace "x::y" --file_pattern "macro.cpp$" --i %T/macro.cpp --
-// RUN: FileCheck -input-file=%T/macro.cpp -check-prefix=CHECK-CC %s
-// RUN: FileCheck -input-file=%T/macro.h -check-prefix=CHECK-HEADER %s
+// RUN: clang-change-namespace -old_namespace "na::nb" -new_namespace "x::y" --file_pattern "macro.cpp$" --i %t.dir/macro.cpp --
+// RUN: FileCheck -input-file=%t.dir/macro.cpp -check-prefix=CHECK-CC %s
+// RUN: FileCheck -input-file=%t.dir/macro.h -check-prefix=CHECK-HEADER %s
 //
-// RUN: cp %S/macro.cpp %T/macro.cpp
-// RUN: echo "#define USING using na::nc::X" > %T/macro.h
-// RUN: clang-change-namespace -old_namespace "na::nb" -new_namespace "x::y" --file_pattern ".*" --i %T/macro.cpp --
-// RUN: FileCheck -input-file=%T/macro.cpp -check-prefix=CHECK-CC %s
-// RUN: FileCheck -input-file=%T/macro.h -check-prefix=CHECK-CHANGED-HEADER %s
+// RUN: cp %S/macro.cpp %t.dir/macro.cpp
+// RUN: echo "#define USING using na::nc::X" > %t.dir/macro.h
+// RUN: clang-change-namespace -old_namespace "na::nb" -new_namespace "x::y" --file_pattern ".*" --i %t.dir/macro.cpp --
+// RUN: FileCheck -input-file=%t.dir/macro.cpp -check-prefix=CHECK-CC %s
+// RUN: FileCheck -input-file=%t.dir/macro.h -check-prefix=CHECK-CHANGED-HEADER %s
 #include "macro.h"
 namespace na { namespace nc { class X{}; } }
 
diff --git a/clang-tools-extra/test/clang-include-fixer/include_path.cpp b/clang-tools-extra/test/clang-include-fixer/include_path.cpp
index 9185b7a..a6f4a45 100644
--- a/clang-tools-extra/test/clang-include-fixer/include_path.cpp
+++ b/clang-tools-extra/test/clang-include-fixer/include_path.cpp
@@ -1,18 +1,18 @@
-// RUN: mkdir -p %T/clang-include-fixer/include
-// RUN: mkdir -p %T/clang-include-fixer/symbols
-// RUN: mkdir -p %T/clang-include-fixer/build
-// RUN: mkdir -p %T/clang-include-fixer/src
-// RUN: sed 's|test_dir|%/T/clang-include-fixer|g' %S/Inputs/database_template.json > %T/clang-include-fixer/build/compile_commands.json
-// RUN: echo -e '#include "bar.h"\nb::a::bar f;' > %T/clang-include-fixer/src/bar.cpp
-// RUN: echo 'namespace b { namespace a { class bar {}; } }' > %T/clang-include-fixer/include/bar.h
-// RUN: cd %T/clang-include-fixer/build
-// RUN: find-all-symbols -output-dir=%T/clang-include-fixer/symbols -p=. %T/clang-include-fixer/src/bar.cpp
-// RUN: find-all-symbols -merge-dir=%T/clang-include-fixer/symbols %T/clang-include-fixer/build/find_all_symbols.yaml
-// RUN: FileCheck -input-file=%T/clang-include-fixer/build/find_all_symbols.yaml -check-prefix=CHECK-YAML %s
+// RUN: mkdir -p %t.dir/clang-include-fixer/include
+// RUN: mkdir -p %t.dir/clang-include-fixer/symbols
+// RUN: mkdir -p %t.dir/clang-include-fixer/build
+// RUN: mkdir -p %t.dir/clang-include-fixer/src
+// RUN: sed 's|test_dir|%/t.dir/clang-include-fixer|g' %S/Inputs/database_template.json > %t.dir/clang-include-fixer/build/compile_commands.json
+// RUN: echo -e '#include "bar.h"\nb::a::bar f;' > %t.dir/clang-include-fixer/src/bar.cpp
+// RUN: echo 'namespace b { namespace a { class bar {}; } }' > %t.dir/clang-include-fixer/include/bar.h
+// RUN: cd %t.dir/clang-include-fixer/build
+// RUN: find-all-symbols -output-dir=%t.dir/clang-include-fixer/symbols -p=. %t.dir/clang-include-fixer/src/bar.cpp
+// RUN: find-all-symbols -merge-dir=%t.dir/clang-include-fixer/symbols %t.dir/clang-include-fixer/build/find_all_symbols.yaml
+// RUN: FileCheck -input-file=%t.dir/clang-include-fixer/build/find_all_symbols.yaml -check-prefix=CHECK-YAML %s
 //
-// RUN: echo 'b::a::bar f;' > %T/clang-include-fixer/src/bar.cpp
-// RUN: clang-include-fixer -db=yaml -input=%T/clang-include-fixer/build/find_all_symbols.yaml -minimize-paths=true -p=. %T/clang-include-fixer/src/bar.cpp
-// RUN: FileCheck -input-file=%T/clang-include-fixer/src/bar.cpp %s
+// RUN: echo 'b::a::bar f;' > %t.dir/clang-include-fixer/src/bar.cpp
+// RUN: clang-include-fixer -db=yaml -input=%t.dir/clang-include-fixer/build/find_all_symbols.yaml -minimize-paths=true -p=. %t.dir/clang-include-fixer/src/bar.cpp
+// RUN: FileCheck -input-file=%t.dir/clang-include-fixer/src/bar.cpp %s
 
 // CHECK-YAML: ..{{[/\\]}}include{{[/\\]}}bar.h
 // CHECK: #include "bar.h"
diff --git a/clang-tools-extra/test/clang-include-fixer/multiple_fixes.cpp b/clang-tools-extra/test/clang-include-fixer/multiple_fixes.cpp
index 791417a..6c82e2a 100644
--- a/clang-tools-extra/test/clang-include-fixer/multiple_fixes.cpp
+++ b/clang-tools-extra/test/clang-include-fixer/multiple_fixes.cpp
@@ -1,11 +1,11 @@
 // REQUIRES: shell
 // RUN: sed -e 's#//.*$##' %s > %t.cpp
-// RUN: mkdir -p %T/clang-include-fixer/multiple-fixes
-// RUN: echo 'foo f;' > %T/clang-include-fixer/multiple-fixes/foo.cpp
-// RUN: echo 'bar b;' > %T/clang-include-fixer/multiple-fixes/bar.cpp
-// RUN: clang-include-fixer -db=fixed -input='foo= "foo.h";bar= "bar.h"' %T/clang-include-fixer/multiple-fixes/*.cpp --
-// RUN: FileCheck -input-file=%T/clang-include-fixer/multiple-fixes/bar.cpp %s -check-prefix=CHECK-BAR
-// RUN: FileCheck -input-file=%T/clang-include-fixer/multiple-fixes/foo.cpp %s -check-prefix=CHECK-FOO
+// RUN: mkdir -p %t.dir/clang-include-fixer/multiple-fixes
+// RUN: echo 'foo f;' > %t.dir/clang-include-fixer/multiple-fixes/foo.cpp
+// RUN: echo 'bar b;' > %t.dir/clang-include-fixer/multiple-fixes/bar.cpp
+// RUN: clang-include-fixer -db=fixed -input='foo= "foo.h";bar= "bar.h"' %t.dir/clang-include-fixer/multiple-fixes/*.cpp --
+// RUN: FileCheck -input-file=%t.dir/clang-include-fixer/multiple-fixes/bar.cpp %s -check-prefix=CHECK-BAR
+// RUN: FileCheck -input-file=%t.dir/clang-include-fixer/multiple-fixes/foo.cpp %s -check-prefix=CHECK-FOO
 //
 // CHECK-FOO: #include "foo.h"
 // CHECK-FOO: foo f;
diff --git a/clang-tools-extra/test/clang-include-fixer/yamldb_autodetect.cpp b/clang-tools-extra/test/clang-include-fixer/yamldb_autodetect.cpp
index 1997390..1978e5d 100644
--- a/clang-tools-extra/test/clang-include-fixer/yamldb_autodetect.cpp
+++ b/clang-tools-extra/test/clang-include-fixer/yamldb_autodetect.cpp
@@ -1,6 +1,6 @@
-// RUN: mkdir -p %T/foo/bar
-// RUN: cp %p/Inputs/fake_yaml_db.yaml %T/find_all_symbols_db.yaml
-// RUN: cd %T/foo
+// RUN: mkdir -p %t.dir/foo/bar
+// RUN: cp %p/Inputs/fake_yaml_db.yaml %t.dir/find_all_symbols_db.yaml
+// RUN: cd %t.dir/foo
 // RUN: sed -e 's#//.*$##' %s > bar/test.cpp
 // RUN: clang-include-fixer -db=yaml bar/test.cpp --
 // RUN: FileCheck %s -input-file=bar/test.cpp
diff --git a/clang-tools-extra/test/clang-move/move-class.cpp b/clang-tools-extra/test/clang-move/move-class.cpp
index a30cb4d..5fb0258 100644
--- a/clang-tools-extra/test/clang-move/move-class.cpp
+++ b/clang-tools-extra/test/clang-move/move-class.cpp
@@ -1,25 +1,25 @@
-// RUN: mkdir -p %T/clang-move/build
-// RUN: mkdir -p %T/clang-move/include
-// RUN: mkdir -p %T/clang-move/src
-// RUN: sed 's|$test_dir|%/T/clang-move|g' %S/Inputs/database_template.json > %T/clang-move/compile_commands.json
-// RUN: cp %S/Inputs/test.h  %T/clang-move/include
-// RUN: cp %S/Inputs/test.cpp %T/clang-move/src
-// RUN: touch %T/clang-move/include/test2.h
-// RUN: cd %T/clang-move/build
-// RUN: clang-move -names="a::Foo" -new_cc=%T/clang-move/new_test.cpp -new_header=%T/clang-move/new_test.h -old_cc=../src/test.cpp -old_header=../include/test.h %T/clang-move/src/test.cpp
-// RUN: FileCheck -input-file=%T/clang-move/new_test.cpp -check-prefix=CHECK-NEW-TEST-CPP %s
-// RUN: FileCheck -input-file=%T/clang-move/new_test.h -check-prefix=CHECK-NEW-TEST-H %s
-// RUN: FileCheck -input-file=%T/clang-move/src/test.cpp -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
-// RUN: FileCheck -input-file=%T/clang-move/include/test.h -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
+// RUN: mkdir -p %t.dir/clang-move/build
+// RUN: mkdir -p %t.dir/clang-move/include
+// RUN: mkdir -p %t.dir/clang-move/src
+// RUN: sed 's|$test_dir|%/t.dir/clang-move|g' %S/Inputs/database_template.json > %t.dir/clang-move/compile_commands.json
+// RUN: cp %S/Inputs/test.h  %t.dir/clang-move/include
+// RUN: cp %S/Inputs/test.cpp %t.dir/clang-move/src
+// RUN: touch %t.dir/clang-move/include/test2.h
+// RUN: cd %t.dir/clang-move/build
+// RUN: clang-move -names="a::Foo" -new_cc=%t.dir/clang-move/new_test.cpp -new_header=%t.dir/clang-move/new_test.h -old_cc=../src/test.cpp -old_header=../include/test.h %t.dir/clang-move/src/test.cpp
+// RUN: FileCheck -input-file=%t.dir/clang-move/new_test.cpp -check-prefix=CHECK-NEW-TEST-CPP %s
+// RUN: FileCheck -input-file=%t.dir/clang-move/new_test.h -check-prefix=CHECK-NEW-TEST-H %s
+// RUN: FileCheck -input-file=%t.dir/clang-move/src/test.cpp -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
+// RUN: FileCheck -input-file=%t.dir/clang-move/include/test.h -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
 //
-// RUN: cp %S/Inputs/test.h  %T/clang-move/include
-// RUN: cp %S/Inputs/test.cpp %T/clang-move/src
-// RUN: cd %T/clang-move/build
-// RUN: clang-move -names="a::Foo" -new_cc=%T/clang-move/new_test.cpp -new_header=%T/clang-move/new_test.h -old_cc=%T/clang-move/src/test.cpp -old_header=%T/clang-move/include/test.h %T/clang-move/src/test.cpp
-// RUN: FileCheck -input-file=%T/clang-move/new_test.cpp -check-prefix=CHECK-NEW-TEST-CPP %s
-// RUN: FileCheck -input-file=%T/clang-move/new_test.h -check-prefix=CHECK-NEW-TEST-H %s
-// RUN: FileCheck -input-file=%T/clang-move/src/test.cpp -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
-// RUN: FileCheck -input-file=%T/clang-move/include/test.h -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
+// RUN: cp %S/Inputs/test.h  %t.dir/clang-move/include
+// RUN: cp %S/Inputs/test.cpp %t.dir/clang-move/src
+// RUN: cd %t.dir/clang-move/build
+// RUN: clang-move -names="a::Foo" -new_cc=%t.dir/clang-move/new_test.cpp -new_header=%t.dir/clang-move/new_test.h -old_cc=%t.dir/clang-move/src/test.cpp -old_header=%t.dir/clang-move/include/test.h %t.dir/clang-move/src/test.cpp
+// RUN: FileCheck -input-file=%t.dir/clang-move/new_test.cpp -check-prefix=CHECK-NEW-TEST-CPP %s
+// RUN: FileCheck -input-file=%t.dir/clang-move/new_test.h -check-prefix=CHECK-NEW-TEST-H %s
+// RUN: FileCheck -input-file=%t.dir/clang-move/src/test.cpp -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
+// RUN: FileCheck -input-file=%t.dir/clang-move/include/test.h -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
 //
 //
 // CHECK-NEW-TEST-H: #ifndef TEST_H // comment 1
diff --git a/clang-tools-extra/test/clang-move/move-enum-decl.cpp b/clang-tools-extra/test/clang-move/move-enum-decl.cpp
index 42f6f99..f8fb5f6 100644
--- a/clang-tools-extra/test/clang-move/move-enum-decl.cpp
+++ b/clang-tools-extra/test/clang-move/move-enum-decl.cpp
@@ -1,14 +1,14 @@
-// RUN: mkdir -p %T/move-enum
-// RUN: cp %S/Inputs/enum.h  %T/move-enum/enum.h
-// RUN: echo '#include "enum.h"' > %T/move-enum/enum.cpp
-// RUN: cd %T/move-enum
+// RUN: mkdir -p %t.dir/move-enum
+// RUN: cp %S/Inputs/enum.h  %t.dir/move-enum/enum.h
+// RUN: echo '#include "enum.h"' > %t.dir/move-enum/enum.cpp
+// RUN: cd %t.dir/move-enum
 //
 // -----------------------------------------------------------------------------
 // Test moving enum declarations.
 // -----------------------------------------------------------------------------
-// RUN: clang-move -names="a::E1" -new_cc=%T/move-enum/new_test.cpp -new_header=%T/move-enum/new_test.h -old_cc=%T/move-enum/enum.cpp -old_header=%T/move-enum/enum.h %T/move-enum/enum.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-enum/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
-// RUN: FileCheck -input-file=%T/move-enum/enum.h -check-prefix=CHECK-OLD-TEST-H-CASE1 %s
+// RUN: clang-move -names="a::E1" -new_cc=%t.dir/move-enum/new_test.cpp -new_header=%t.dir/move-enum/new_test.h -old_cc=%t.dir/move-enum/enum.cpp -old_header=%t.dir/move-enum/enum.h %t.dir/move-enum/enum.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-enum/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
+// RUN: FileCheck -input-file=%t.dir/move-enum/enum.h -check-prefix=CHECK-OLD-TEST-H-CASE1 %s
 //
 // CHECK-NEW-TEST-H-CASE1: namespace a {
 // CHECK-NEW-TEST-H-CASE1-NEXT: enum E1 { Green, Red };
@@ -20,11 +20,11 @@
 // -----------------------------------------------------------------------------
 // Test moving scoped enum declarations.
 // -----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/enum.h  %T/move-enum/enum.h
-// RUN: echo '#include "enum.h"' > %T/move-enum/enum.cpp
-// RUN: clang-move -names="a::E2" -new_cc=%T/move-enum/new_test.cpp -new_header=%T/move-enum/new_test.h -old_cc=%T/move-enum/enum.cpp -old_header=%T/move-enum/enum.h %T/move-enum/enum.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-enum/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-enum/enum.h -check-prefix=CHECK-OLD-TEST-H-CASE2 %s
+// RUN: cp %S/Inputs/enum.h  %t.dir/move-enum/enum.h
+// RUN: echo '#include "enum.h"' > %t.dir/move-enum/enum.cpp
+// RUN: clang-move -names="a::E2" -new_cc=%t.dir/move-enum/new_test.cpp -new_header=%t.dir/move-enum/new_test.h -old_cc=%t.dir/move-enum/enum.cpp -old_header=%t.dir/move-enum/enum.h %t.dir/move-enum/enum.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-enum/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-enum/enum.h -check-prefix=CHECK-OLD-TEST-H-CASE2 %s
 
 // CHECK-NEW-TEST-H-CASE2: namespace a {
 // CHECK-NEW-TEST-H-CASE2-NEXT: enum class E2 { Yellow };
@@ -36,9 +36,9 @@
 // -----------------------------------------------------------------------------
 // Test not moving class-insided enum declarations.
 // -----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/enum.h  %T/move-enum/enum.h
-// RUN: echo '#include "enum.h"' > %T/move-enum/enum.cpp
-// RUN: clang-move -names="a::C::E3" -new_cc=%T/move-enum/new_test.cpp -new_header=%T/move-enum/new_test.h -old_cc=%T/move-enum/enum.cpp -old_header=%T/move-enum/enum.h %T/move-enum/enum.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-enum/new_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
+// RUN: cp %S/Inputs/enum.h  %t.dir/move-enum/enum.h
+// RUN: echo '#include "enum.h"' > %t.dir/move-enum/enum.cpp
+// RUN: clang-move -names="a::C::E3" -new_cc=%t.dir/move-enum/new_test.cpp -new_header=%t.dir/move-enum/new_test.h -old_cc=%t.dir/move-enum/enum.cpp -old_header=%t.dir/move-enum/enum.h %t.dir/move-enum/enum.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-enum/new_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
 
 // CHECK-EMPTY: {{^}}{{$}}
diff --git a/clang-tools-extra/test/clang-move/move-function.cpp b/clang-tools-extra/test/clang-move/move-function.cpp
index 0324b80..a52d55c 100644
--- a/clang-tools-extra/test/clang-move/move-function.cpp
+++ b/clang-tools-extra/test/clang-move/move-function.cpp
@@ -1,9 +1,9 @@
-// RUN: mkdir -p %T/move-function
-// RUN: cat %S/Inputs/function_test.h > %T/move-function/function_test.h
-// RUN: cat %S/Inputs/function_test.cpp > %T/move-function/function_test.cpp
-// RUN: cd %T/move-function
-// RUN: clang-move -names="g" -new_header=%T/move-function/new_function_test.h -old_header=../move-function/function_test.h %T/move-function/function_test.cpp --
-// RUN: FileCheck -input-file=%T/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
+// RUN: mkdir -p %t.dir/move-function
+// RUN: cat %S/Inputs/function_test.h > %t.dir/move-function/function_test.h
+// RUN: cat %S/Inputs/function_test.cpp > %t.dir/move-function/function_test.cpp
+// RUN: cd %t.dir/move-function
+// RUN: clang-move -names="g" -new_header=%t.dir/move-function/new_function_test.h -old_header=../move-function/function_test.h %t.dir/move-function/function_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
 //
 // CHECK-NEW-TEST-H-CASE1: #ifndef {{.*}}NEW_FUNCTION_TEST_H
 // CHECK-NEW-TEST-H-CASE1: #define {{.*}}NEW_FUNCTION_TEST_H
@@ -12,9 +12,9 @@
 // CHECK-NEW-TEST-H-CASE1: {{[[:space:]]+}}
 // CHECK-NEW-TEST-H-CASE1: #endif // {{.*}}NEW_FUNCTION_TEST_H
 //
-// RUN: cp %S/Inputs/function_test*  %T/move-function
-// RUN: clang-move -names="h" -new_header=%T/move-function/new_function_test.h -old_header=../move-function/function_test.h %T/move-function/function_test.cpp --
-// RUN: FileCheck -input-file=%T/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
+// RUN: cp %S/Inputs/function_test*  %t.dir/move-function
+// RUN: clang-move -names="h" -new_header=%t.dir/move-function/new_function_test.h -old_header=../move-function/function_test.h %t.dir/move-function/function_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
 //
 // CHECK-NEW-TEST-H-CASE2: #ifndef {{.*}}NEW_FUNCTION_TEST_H
 // CHECK-NEW-TEST-H-CASE2: #define {{.*}}NEW_FUNCTION_TEST_H
@@ -25,10 +25,10 @@
 // CHECK-NEW-TEST-H-CASE2: {{[[:space:]]+}}
 // CHECK-NEW-TEST-H-CASE2: #endif // {{.*}}NEW_FUNCTION_TEST_H
 //
-// RUN: cp %S/Inputs/function_test*  %T/move-function
-// RUN: clang-move -names="f" -new_header=%T/move-function/new_function_test.h -new_cc=%T/move-function/new_function_test.cpp -old_header=../move-function/function_test.h -old_cc=../move-function/function_test.cpp %T/move-function/function_test.cpp --
-// RUN: FileCheck -input-file=%T/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE3 %s
-// RUN: FileCheck -input-file=%T/move-function/new_function_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE3 %s
+// RUN: cp %S/Inputs/function_test*  %t.dir/move-function
+// RUN: clang-move -names="f" -new_header=%t.dir/move-function/new_function_test.h -new_cc=%t.dir/move-function/new_function_test.cpp -old_header=../move-function/function_test.h -old_cc=../move-function/function_test.cpp %t.dir/move-function/function_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE3 %s
+// RUN: FileCheck -input-file=%t.dir/move-function/new_function_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE3 %s
 //
 // CHECK-NEW-TEST-H-CASE3: #ifndef {{.*}}NEW_FUNCTION_TEST_H
 // CHECK-NEW-TEST-H-CASE3: #define {{.*}}NEW_FUNCTION_TEST_H
@@ -40,17 +40,17 @@
 // CHECK-NEW-TEST-CPP-CASE3: {{[[:space:]]+}}
 // CHECK-NEW-TEST-CPP-CASE3: void f() {}
 //
-// RUN: cat %S/Inputs/function_test.h > %T/move-function/function_test.h
-// RUN: cat %S/Inputs/function_test.cpp > %T/move-function/function_test.cpp
-// RUN: clang-move -names="A::f" -new_header=%T/move-function/new_function_test.h -new_cc=%T/move-function/new_function_test.cpp -old_header=../move-function/function_test.h -old_cc=../move-function/function_test.cpp %T/move-function/function_test.cpp -dump_result -- | FileCheck %s -check-prefix=CHECK-EMPTY
+// RUN: cat %S/Inputs/function_test.h > %t.dir/move-function/function_test.h
+// RUN: cat %S/Inputs/function_test.cpp > %t.dir/move-function/function_test.cpp
+// RUN: clang-move -names="A::f" -new_header=%t.dir/move-function/new_function_test.h -new_cc=%t.dir/move-function/new_function_test.cpp -old_header=../move-function/function_test.h -old_cc=../move-function/function_test.cpp %t.dir/move-function/function_test.cpp -dump_result -- | FileCheck %s -check-prefix=CHECK-EMPTY
 //
 // CHECK-EMPTY: [{{[[:space:]]*}}]
 //
-// RUN: cat %S/Inputs/function_test.h > %T/move-function/function_test.h
-// RUN: cat %S/Inputs/function_test.cpp > %T/move-function/function_test.cpp
-// RUN: clang-move -names="f,A" -new_header=%T/move-function/new_function_test.h -new_cc=%T/move-function/new_function_test.cpp -old_header=../move-function/function_test.h -old_cc=../move-function/function_test.cpp %T/move-function/function_test.cpp --
-// RUN: FileCheck -input-file=%T/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE4 %s
-// RUN: FileCheck -input-file=%T/move-function/new_function_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE4 %s
+// RUN: cat %S/Inputs/function_test.h > %t.dir/move-function/function_test.h
+// RUN: cat %S/Inputs/function_test.cpp > %t.dir/move-function/function_test.cpp
+// RUN: clang-move -names="f,A" -new_header=%t.dir/move-function/new_function_test.h -new_cc=%t.dir/move-function/new_function_test.cpp -old_header=../move-function/function_test.h -old_cc=../move-function/function_test.cpp %t.dir/move-function/function_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-function/new_function_test.h -check-prefix=CHECK-NEW-TEST-H-CASE4 %s
+// RUN: FileCheck -input-file=%t.dir/move-function/new_function_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE4 %s
 
 // CHECK-NEW-TEST-H-CASE4: #ifndef {{.*}}NEW_FUNCTION_TEST_H
 // CHECK-NEW-TEST-H-CASE4: #define {{.*}}NEW_FUNCTION_TEST_H
diff --git a/clang-tools-extra/test/clang-move/move-multiple-classes.cpp b/clang-tools-extra/test/clang-move/move-multiple-classes.cpp
index 821d567..513b904 100644
--- a/clang-tools-extra/test/clang-move/move-multiple-classes.cpp
+++ b/clang-tools-extra/test/clang-move/move-multiple-classes.cpp
@@ -1,12 +1,12 @@
-// RUN: mkdir -p %T/move-multiple-classes
-// RUN: cp %S/Inputs/multiple_class_test*  %T/move-multiple-classes/
-// RUN: cd %T/move-multiple-classes
-// RUN: clang-move -names="c::EnclosingMove5::Nested" -new_cc=%T/move-multiple-classes/new_multiple_class_test.cpp -new_header=%T/move-multiple-classes/new_multiple_class_test.h -old_cc=%T/move-multiple-classes/multiple_class_test.cpp -old_header=../move-multiple-classes/multiple_class_test.h -dump_result %T/move-multiple-classes/multiple_class_test.cpp -- -std=c++11| FileCheck %s -check-prefix=CHECK-EMPTY
-// RUN: clang-move -names="a::Move1, b::Move2,c::Move3,c::Move4,c::EnclosingMove5" -new_cc=%T/move-multiple-classes/new_multiple_class_test.cpp -new_header=%T/move-multiple-classes/new_multiple_class_test.h -old_cc=%T/move-multiple-classes/multiple_class_test.cpp -old_header=../move-multiple-classes/multiple_class_test.h %T/move-multiple-classes/multiple_class_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-multiple-classes/new_multiple_class_test.cpp -check-prefix=CHECK-NEW-TEST-CPP %s
-// RUN: FileCheck -input-file=%T/move-multiple-classes/new_multiple_class_test.h -check-prefix=CHECK-NEW-TEST-H %s
-// RUN: FileCheck -input-file=%T/move-multiple-classes/multiple_class_test.cpp -check-prefix=CHECK-OLD-TEST-CPP %s
-// RUN: FileCheck -input-file=%T/move-multiple-classes/multiple_class_test.h -check-prefix=CHECK-OLD-TEST-H %s
+// RUN: mkdir -p %t.dir/move-multiple-classes
+// RUN: cp %S/Inputs/multiple_class_test*  %t.dir/move-multiple-classes/
+// RUN: cd %t.dir/move-multiple-classes
+// RUN: clang-move -names="c::EnclosingMove5::Nested" -new_cc=%t.dir/move-multiple-classes/new_multiple_class_test.cpp -new_header=%t.dir/move-multiple-classes/new_multiple_class_test.h -old_cc=%t.dir/move-multiple-classes/multiple_class_test.cpp -old_header=../move-multiple-classes/multiple_class_test.h -dump_result %t.dir/move-multiple-classes/multiple_class_test.cpp -- -std=c++11| FileCheck %s -check-prefix=CHECK-EMPTY
+// RUN: clang-move -names="a::Move1, b::Move2,c::Move3,c::Move4,c::EnclosingMove5" -new_cc=%t.dir/move-multiple-classes/new_multiple_class_test.cpp -new_header=%t.dir/move-multiple-classes/new_multiple_class_test.h -old_cc=%t.dir/move-multiple-classes/multiple_class_test.cpp -old_header=../move-multiple-classes/multiple_class_test.h %t.dir/move-multiple-classes/multiple_class_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-multiple-classes/new_multiple_class_test.cpp -check-prefix=CHECK-NEW-TEST-CPP %s
+// RUN: FileCheck -input-file=%t.dir/move-multiple-classes/new_multiple_class_test.h -check-prefix=CHECK-NEW-TEST-H %s
+// RUN: FileCheck -input-file=%t.dir/move-multiple-classes/multiple_class_test.cpp -check-prefix=CHECK-OLD-TEST-CPP %s
+// RUN: FileCheck -input-file=%t.dir/move-multiple-classes/multiple_class_test.h -check-prefix=CHECK-OLD-TEST-H %s
 //
 // CHECK-EMPTY: [{{[[:space:]]*}}]
 //
diff --git a/clang-tools-extra/test/clang-move/move-template-class.cpp b/clang-tools-extra/test/clang-move/move-template-class.cpp
index 1a6a60b..29ed65e 100644
--- a/clang-tools-extra/test/clang-move/move-template-class.cpp
+++ b/clang-tools-extra/test/clang-move/move-template-class.cpp
@@ -1,18 +1,18 @@
-// RUN: mkdir -p %T/move-template-class
-// RUN: cp %S/Inputs/template_class_test*  %T/move-template-class
-// RUN: cd %T/move-template-class
-// RUN: clang-move -names="A,B" -new_cc=%T/move-template-class/new_template_class_test.cpp -new_header=%T/move-template-class/new_template_class_test.h -old_cc=%T/move-template-class/template_class_test.cpp -old_header=../move-template-class/template_class_test.h %T/move-template-class/template_class_test.cpp --
-// RUN: FileCheck -input-file=%T/move-template-class/template_class_test.cpp -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
-// RUN: FileCheck -input-file=%T/move-template-class/template_class_test.h -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
-// RUN: FileCheck -input-file=%T/move-template-class/new_template_class_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE1 %s
-// RUN: FileCheck -input-file=%T/move-template-class/new_template_class_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
+// RUN: mkdir -p %t.dir/move-template-class
+// RUN: cp %S/Inputs/template_class_test*  %t.dir/move-template-class
+// RUN: cd %t.dir/move-template-class
+// RUN: clang-move -names="A,B" -new_cc=%t.dir/move-template-class/new_template_class_test.cpp -new_header=%t.dir/move-template-class/new_template_class_test.h -old_cc=%t.dir/move-template-class/template_class_test.cpp -old_header=../move-template-class/template_class_test.h %t.dir/move-template-class/template_class_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-template-class/template_class_test.cpp -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
+// RUN: FileCheck -input-file=%t.dir/move-template-class/template_class_test.h -check-prefix=CHECK-OLD-TEST-EMPTY -allow-empty %s
+// RUN: FileCheck -input-file=%t.dir/move-template-class/new_template_class_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE1 %s
+// RUN: FileCheck -input-file=%t.dir/move-template-class/new_template_class_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
 //
-// RUN: cp %S/Inputs/template_class_test*  %T/move-template-class
-// RUN: clang-move -names="A" -new_cc=%T/move-template-class/new_template_class_test.cpp -new_header=%T/move-template-class/new_template_class_test.h -old_cc=%T/move-template-class/template_class_test.cpp -old_header=../move-template-class/template_class_test.h %T/move-template-class/template_class_test.cpp --
-// RUN: FileCheck -input-file=%T/move-template-class/template_class_test.h -check-prefix=CHECK-OLD-TEST-H-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-template-class/template_class_test.cpp -check-prefix=CHECK-OLD-TEST-CPP-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-template-class/new_template_class_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-template-class/new_template_class_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE2 %s
+// RUN: cp %S/Inputs/template_class_test*  %t.dir/move-template-class
+// RUN: clang-move -names="A" -new_cc=%t.dir/move-template-class/new_template_class_test.cpp -new_header=%t.dir/move-template-class/new_template_class_test.h -old_cc=%t.dir/move-template-class/template_class_test.cpp -old_header=../move-template-class/template_class_test.h %t.dir/move-template-class/template_class_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-template-class/template_class_test.h -check-prefix=CHECK-OLD-TEST-H-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-template-class/template_class_test.cpp -check-prefix=CHECK-OLD-TEST-CPP-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-template-class/new_template_class_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-template-class/new_template_class_test.cpp -check-prefix=CHECK-NEW-TEST-CPP-CASE2 %s
 //
 //
 // CHECK-OLD-TEST-EMPTY: {{^}}{{$}}
diff --git a/clang-tools-extra/test/clang-move/move-type-alias.cpp b/clang-tools-extra/test/clang-move/move-type-alias.cpp
index ab70237..54d2b0e 100644
--- a/clang-tools-extra/test/clang-move/move-type-alias.cpp
+++ b/clang-tools-extra/test/clang-move/move-type-alias.cpp
@@ -1,14 +1,14 @@
-// RUN: mkdir -p %T/move-type-alias
-// RUN: cp %S/Inputs/type_alias.h  %T/move-type-alias/type_alias.h
-// RUN: echo '#include "type_alias.h"' > %T/move-type-alias/type_alias.cpp
-// RUN: cd %T/move-type-alias
+// RUN: mkdir -p %t.dir/move-type-alias
+// RUN: cp %S/Inputs/type_alias.h  %t.dir/move-type-alias/type_alias.h
+// RUN: echo '#include "type_alias.h"' > %t.dir/move-type-alias/type_alias.cpp
+// RUN: cd %t.dir/move-type-alias
 //
 // -----------------------------------------------------------------------------
 // Test moving typedef declarations.
 // -----------------------------------------------------------------------------
-// RUN: clang-move -names="Int1" -new_cc=%T/move-type-alias/new_test.cpp -new_header=%T/move-type-alias/new_test.h -old_cc=%T/move-type-alias/type_alias.cpp -old_header=%T/move-type-alias/type_alias.h %T/move-type-alias/type_alias.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-type-alias/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
-// RUN: FileCheck -input-file=%T/move-type-alias/type_alias.h -check-prefix=CHECK-OLD-TEST-H-CASE1 %s
+// RUN: clang-move -names="Int1" -new_cc=%t.dir/move-type-alias/new_test.cpp -new_header=%t.dir/move-type-alias/new_test.h -old_cc=%t.dir/move-type-alias/type_alias.cpp -old_header=%t.dir/move-type-alias/type_alias.h %t.dir/move-type-alias/type_alias.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-type-alias/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE1 %s
+// RUN: FileCheck -input-file=%t.dir/move-type-alias/type_alias.h -check-prefix=CHECK-OLD-TEST-H-CASE1 %s
 
 // CHECK-NEW-TEST-H-CASE1: typedef int Int1;
 
@@ -18,11 +18,11 @@
 // -----------------------------------------------------------------------------
 // Test moving type alias declarations.
 // -----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/type_alias.h  %T/move-type-alias/type_alias.h
-// RUN: echo '#include "type_alias.h"' > %T/move-type-alias/type_alias.cpp
-// RUN: clang-move -names="Int2" -new_cc=%T/move-type-alias/new_test.cpp -new_header=%T/move-type-alias/new_test.h -old_cc=%T/move-type-alias/type_alias.cpp -old_header=%T/move-type-alias/type_alias.h %T/move-type-alias/type_alias.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-type-alias/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-type-alias/type_alias.h -check-prefix=CHECK-OLD-TEST-H-CASE2 %s
+// RUN: cp %S/Inputs/type_alias.h  %t.dir/move-type-alias/type_alias.h
+// RUN: echo '#include "type_alias.h"' > %t.dir/move-type-alias/type_alias.cpp
+// RUN: clang-move -names="Int2" -new_cc=%t.dir/move-type-alias/new_test.cpp -new_header=%t.dir/move-type-alias/new_test.h -old_cc=%t.dir/move-type-alias/type_alias.cpp -old_header=%t.dir/move-type-alias/type_alias.h %t.dir/move-type-alias/type_alias.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-type-alias/new_test.h -check-prefix=CHECK-NEW-TEST-H-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-type-alias/type_alias.h -check-prefix=CHECK-OLD-TEST-H-CASE2 %s
 
 // CHECK-NEW-TEST-H-CASE2: using Int2 = int;
 
@@ -32,10 +32,10 @@
 // -----------------------------------------------------------------------------
 // Test moving template type alias declarations.
 // -----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/type_alias.h  %T/move-type-alias/type_alias.h
-// RUN: echo '#include "type_alias.h"' > %T/move-type-alias/type_alias.cpp
-// RUN: clang-move -names="B" -new_cc=%T/move-type-alias/new_test.cpp -new_header=%T/move-type-alias/new_test.h -old_cc=%T/move-type-alias/type_alias.cpp -old_header=%T/move-type-alias/type_alias.h %T/move-type-alias/type_alias.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-type-alias/new_test.h -check-prefix=CHECK-OLD-TEST-H-CASE3 %s
+// RUN: cp %S/Inputs/type_alias.h  %t.dir/move-type-alias/type_alias.h
+// RUN: echo '#include "type_alias.h"' > %t.dir/move-type-alias/type_alias.cpp
+// RUN: clang-move -names="B" -new_cc=%t.dir/move-type-alias/new_test.cpp -new_header=%t.dir/move-type-alias/new_test.h -old_cc=%t.dir/move-type-alias/type_alias.cpp -old_header=%t.dir/move-type-alias/type_alias.h %t.dir/move-type-alias/type_alias.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-type-alias/new_test.h -check-prefix=CHECK-OLD-TEST-H-CASE3 %s
 
 // CHECK-NEW-TEST-H-CASE3: template<class T> using B = A<T>;
 // CHECK-OLD-TEST-H-CASE3-NOT: template<class T> using B = A<T>;
@@ -44,9 +44,9 @@
 // -----------------------------------------------------------------------------
 // Test not moving class-insided typedef declarations.
 // -----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/type_alias.h  %T/move-type-alias/type_alias.h
-// RUN: echo '#include "type_alias.h"' > %T/move-type-alias/type_alias.cpp
-// RUN: clang-move -names="C::Int3" -new_cc=%T/move-type-alias/new_test.cpp -new_header=%T/move-type-alias/new_test.h -old_cc=%T/move-type-alias/type_alias.cpp -old_header=%T/move-type-alias/type_alias.h %T/move-type-alias/type_alias.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/move-type-alias/new_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
+// RUN: cp %S/Inputs/type_alias.h  %t.dir/move-type-alias/type_alias.h
+// RUN: echo '#include "type_alias.h"' > %t.dir/move-type-alias/type_alias.cpp
+// RUN: clang-move -names="C::Int3" -new_cc=%t.dir/move-type-alias/new_test.cpp -new_header=%t.dir/move-type-alias/new_test.h -old_cc=%t.dir/move-type-alias/type_alias.cpp -old_header=%t.dir/move-type-alias/type_alias.h %t.dir/move-type-alias/type_alias.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/move-type-alias/new_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
 
 // CHECK-EMPTY: {{^}}{{$}}
diff --git a/clang-tools-extra/test/clang-move/move-used-helper-decls.cpp b/clang-tools-extra/test/clang-move/move-used-helper-decls.cpp
index b4aed2c..3092976 100644
--- a/clang-tools-extra/test/clang-move/move-used-helper-decls.cpp
+++ b/clang-tools-extra/test/clang-move/move-used-helper-decls.cpp
@@ -1,13 +1,13 @@
-// RUN: mkdir -p %T/used-helper-decls
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: cd %T/used-helper-decls
+// RUN: mkdir -p %t.dir/used-helper-decls
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: cd %t.dir/used-helper-decls
 
 // ----------------------------------------------------------------------------
 // Test moving used helper function and its transitively used functions.
 // ----------------------------------------------------------------------------
-// RUN: clang-move -names="a::Class1" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS1-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS1-CPP %s
+// RUN: clang-move -names="a::Class1" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS1-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS1-CPP %s
 
 // CHECK-NEW-CLASS1-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS1-CPP-SAME: {{[[:space:]]}}
@@ -31,10 +31,10 @@
 // ----------------------------------------------------------------------------
 // Test moving used helper function and its transitively used static variables.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class2" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS2-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS2-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class2" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS2-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS2-CPP %s
 
 // CHECK-NEW-CLASS2-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS2-CPP-SAME: {{[[:space:]]}}
@@ -67,10 +67,10 @@
 // ----------------------------------------------------------------------------
 // Test using a static member variable of a helper class.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class3" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS3-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS3-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class3" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS3-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS3-CPP %s
 
 // CHECK-NEW-CLASS3-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS3-CPP-SAME: {{[[:space:]]}}
@@ -99,10 +99,10 @@
 // ----------------------------------------------------------------------------
 // Test moving helper classes.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class4" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS4-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS4-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class4" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS4-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS4-CPP %s
 
 // CHECK-NEW-CLASS4-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS4-CPP-SAME: {{[[:space:]]}}
@@ -120,10 +120,10 @@
 // ----------------------------------------------------------------------------
 // Test moving helper variables and helper functions together.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class5" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS5-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS5-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class5" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS5-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS5-CPP %s
 
 // CHECK-NEW-CLASS5-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS5-CPP-SAME: {{[[:space:]]}}
@@ -153,10 +153,10 @@
 // ----------------------------------------------------------------------------
 // Test moving helper variables and their transitively used helper classes.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class6" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS6-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS6-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class6" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS6-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS6-CPP %s
 
 // CHECK-NEW-CLASS6-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS6-CPP-SAME: {{[[:space:]]}}
@@ -186,10 +186,10 @@
 // ----------------------------------------------------------------------------
 // Test moving classes where its methods use helpers.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class7" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS7-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS7-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class7" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CLASS7-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-CLASS7-CPP %s
 
 // CHECK-NEW-CLASS7-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-CLASS7-CPP-SAME: {{[[:space:]]}}
@@ -218,10 +218,10 @@
 // ----------------------------------------------------------------------------
 // Test moving helper function and its transitively used helper variables.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Fun1" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-FUN1-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-FUN1-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Fun1" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-FUN1-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-FUN1-CPP %s
 
 // CHECK-NEW-FUN1-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-FUN1-CPP-SAME: {{[[:space:]]}}
@@ -244,11 +244,11 @@
 // ----------------------------------------------------------------------------
 // Test no moving helpers when moving inline functions in header.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Fun2" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-FUN2-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.h -check-prefix=CHECK-NEW-FUN2-H %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.h -check-prefix=CHECK-OLD-FUN2-H %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Fun2" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-FUN2-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.h -check-prefix=CHECK-NEW-FUN2-H %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.h -check-prefix=CHECK-OLD-FUN2-H %s
 
 // CHECK-NEW-FUN2-H: namespace a {
 // CHECK-NEW-FUN2-H-NEXT: inline void Fun2() {}
@@ -262,10 +262,10 @@
 // ----------------------------------------------------------------------------
 // Test moving used helper function and its transitively used functions.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="b::Fun3" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-FUN3-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-FUN3-CPP %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="b::Fun3" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-FUN3-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -check-prefix=CHECK-OLD-FUN3-CPP %s
 
 // CHECK-NEW-FUN3-CPP: #include "{{.*}}new_helper_decls_test.h"
 // CHECK-NEW-FUN3-CPP-SAME: {{[[:space:]]}}
@@ -294,12 +294,12 @@
 // ----------------------------------------------------------------------------
 // Test moving all symbols in headers.
 // ----------------------------------------------------------------------------
-// RUN: cp %S/Inputs/helper_decls_test*  %T/used-helper-decls/
-// RUN: clang-move -names="a::Class1, a::Class2, a::Class3, a::Class4, a::Class5, a::Class5, a::Class6, a::Class7, a::Fun1, a::Fun2, b::Fun3" -new_cc=%T/used-helper-decls/new_helper_decls_test.cpp -new_header=%T/used-helper-decls/new_helper_decls_test.h -old_cc=%T/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %T/used-helper-decls/helper_decls_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.h -check-prefix=CHECK-NEW-H %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CPP %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
-// RUN: FileCheck -input-file=%T/used-helper-decls/helper_decls_test.cpp -allow-empty -check-prefix=CHECK-EMPTY %s
+// RUN: cp %S/Inputs/helper_decls_test*  %t.dir/used-helper-decls/
+// RUN: clang-move -names="a::Class1, a::Class2, a::Class3, a::Class4, a::Class5, a::Class5, a::Class6, a::Class7, a::Fun1, a::Fun2, b::Fun3" -new_cc=%t.dir/used-helper-decls/new_helper_decls_test.cpp -new_header=%t.dir/used-helper-decls/new_helper_decls_test.h -old_cc=%t.dir/used-helper-decls/helper_decls_test.cpp -old_header=../used-helper-decls/helper_decls_test.h %t.dir/used-helper-decls/helper_decls_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.h -check-prefix=CHECK-NEW-H %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/new_helper_decls_test.cpp -check-prefix=CHECK-NEW-CPP %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
+// RUN: FileCheck -input-file=%t.dir/used-helper-decls/helper_decls_test.cpp -allow-empty -check-prefix=CHECK-EMPTY %s
 
 
 // CHECK-NEW-H: namespace a {
diff --git a/clang-tools-extra/test/clang-move/move-var.cpp b/clang-tools-extra/test/clang-move/move-var.cpp
index 4a3554c..bacba0b 100644
--- a/clang-tools-extra/test/clang-move/move-var.cpp
+++ b/clang-tools-extra/test/clang-move/move-var.cpp
@@ -1,11 +1,11 @@
-// RUN: mkdir -p %T/move-var
-// RUN: cp %S/Inputs/var_test*  %T/move-var
-// RUN: cd %T/move-var
-// RUN: clang-move -names="a::kGlobalInt" -new_header=%T/move-var/new_var_test.h -old_header=../move-var/var_test.h -old_cc=../move-var/var_test.cpp -new_cc=%T/move-var/new_var_test.cpp %T/move-var/var_test.cpp --
-// RUN: FileCheck -input-file=%T/move-var/var_test.h -check-prefix=CHECK-OLD-VAR-H-CASE1 %s
-// RUN: FileCheck -input-file=%T/move-var/var_test.cpp -check-prefix=CHECK-OLD-VAR-CPP-CASE1 %s
-// RUN: FileCheck -input-file=%T/move-var/new_var_test.h -check-prefix=CHECK-NEW-VAR-H-CASE1 %s
-// RUN: FileCheck -input-file=%T/move-var/new_var_test.cpp -check-prefix=CHECK-NEW-VAR-CPP-CASE1 %s
+// RUN: mkdir -p %t.dir/move-var
+// RUN: cp %S/Inputs/var_test*  %t.dir/move-var
+// RUN: cd %t.dir/move-var
+// RUN: clang-move -names="a::kGlobalInt" -new_header=%t.dir/move-var/new_var_test.h -old_header=../move-var/var_test.h -old_cc=../move-var/var_test.cpp -new_cc=%t.dir/move-var/new_var_test.cpp %t.dir/move-var/var_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-var/var_test.h -check-prefix=CHECK-OLD-VAR-H-CASE1 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/var_test.cpp -check-prefix=CHECK-OLD-VAR-CPP-CASE1 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/new_var_test.h -check-prefix=CHECK-NEW-VAR-H-CASE1 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/new_var_test.cpp -check-prefix=CHECK-NEW-VAR-CPP-CASE1 %s
 
 // CHECK-OLD-VAR-H-CASE1-NOT: extern int kGlobalInt;
 // CHECK-OLD-VAR-H-CASE1: int kGlobalInt = 3;
@@ -18,12 +18,12 @@
 // CHECK-NEW-VAR-CPP-CASE1: int kGlobalInt = 1;
 
 
-// RUN: cp %S/Inputs/var_test*  %T/move-var
-// RUN: clang-move -names="a::kGlobalStr" -new_header=%T/move-var/new_var_test.h -old_header=../move-var/var_test.h -old_cc=../move-var/var_test.cpp -new_cc=%T/move-var/new_var_test.cpp %T/move-var/var_test.cpp --
-// RUN: FileCheck -input-file=%T/move-var/var_test.h -check-prefix=CHECK-OLD-VAR-H-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-var/var_test.cpp -check-prefix=CHECK-OLD-VAR-CPP-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-var/new_var_test.h -check-prefix=CHECK-NEW-VAR-H-CASE2 %s
-// RUN: FileCheck -input-file=%T/move-var/new_var_test.cpp -check-prefix=CHECK-NEW-VAR-CPP-CASE2 %s
+// RUN: cp %S/Inputs/var_test*  %t.dir/move-var
+// RUN: clang-move -names="a::kGlobalStr" -new_header=%t.dir/move-var/new_var_test.h -old_header=../move-var/var_test.h -old_cc=../move-var/var_test.cpp -new_cc=%t.dir/move-var/new_var_test.cpp %t.dir/move-var/var_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-var/var_test.h -check-prefix=CHECK-OLD-VAR-H-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/var_test.cpp -check-prefix=CHECK-OLD-VAR-CPP-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/new_var_test.h -check-prefix=CHECK-NEW-VAR-H-CASE2 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/new_var_test.cpp -check-prefix=CHECK-NEW-VAR-CPP-CASE2 %s
 
 // CHECK-OLD-VAR-H-CASE2-NOT: extern const char *const kGlobalStr;
 // CHECK-OLD-VAR-H-CASE2: const char *const kGlobalStr = "Hello2";
@@ -36,10 +36,10 @@
 // CHECK-NEW-VAR-CPP-CASE2: const char *const kGlobalStr = "Hello";
 
 
-// RUN: cp %S/Inputs/var_test*  %T/move-var
-// RUN: clang-move -names="kEvilInt" -new_header=%T/move-var/new_var_test.h -old_header=../move-var/var_test.h -old_cc=../move-var/var_test.cpp -new_cc=%T/move-var/new_var_test.cpp %T/move-var/var_test.cpp --
-// RUN: FileCheck -input-file=%T/move-var/var_test.h -check-prefix=CHECK-OLD-VAR-H-CASE3 %s
-// RUN: FileCheck -input-file=%T/move-var/new_var_test.h -check-prefix=CHECK-NEW-VAR-H-CASE3 %s
+// RUN: cp %S/Inputs/var_test*  %t.dir/move-var
+// RUN: clang-move -names="kEvilInt" -new_header=%t.dir/move-var/new_var_test.h -old_header=../move-var/var_test.h -old_cc=../move-var/var_test.cpp -new_cc=%t.dir/move-var/new_var_test.cpp %t.dir/move-var/var_test.cpp --
+// RUN: FileCheck -input-file=%t.dir/move-var/var_test.h -check-prefix=CHECK-OLD-VAR-H-CASE3 %s
+// RUN: FileCheck -input-file=%t.dir/move-var/new_var_test.h -check-prefix=CHECK-NEW-VAR-H-CASE3 %s
 
 // CHECK-OLD-VAR-H-CASE3-NOT: int kEvilInt = 2;
 
diff --git a/clang-tools-extra/test/clang-move/no-move-macro-helpers.cpp b/clang-tools-extra/test/clang-move/no-move-macro-helpers.cpp
index 282eee0..e7a5b4e 100644
--- a/clang-tools-extra/test/clang-move/no-move-macro-helpers.cpp
+++ b/clang-tools-extra/test/clang-move/no-move-macro-helpers.cpp
@@ -1,16 +1,16 @@
-// RUN: mkdir -p %T/no-move-macro-helper
-// RUN: cat %S/Inputs/macro_helper_test.h > %T/no-move-macro-helper/macro_helper_test.h
-// RUN: cat %S/Inputs/macro_helper_test.cpp > %T/no-move-macro-helper/macro_helper_test.cpp
-// RUN: cd %T/no-move-macro-helper
+// RUN: mkdir -p %t.dir/no-move-macro-helper
+// RUN: cat %S/Inputs/macro_helper_test.h > %t.dir/no-move-macro-helper/macro_helper_test.h
+// RUN: cat %S/Inputs/macro_helper_test.cpp > %t.dir/no-move-macro-helper/macro_helper_test.cpp
+// RUN: cd %t.dir/no-move-macro-helper
 //
 // -----------------------------------------------------------------------------
 // Test no moving helpers in macro.
 // -----------------------------------------------------------------------------
-// RUN: clang-move -names="A" -new_cc=%T/no-move-macro-helper/new_test.cpp -new_header=%T/no-move-macro-helper/new_test.h -old_cc=%T/no-move-macro-helper/macro_helper_test.cpp -old_header=%T/no-move-macro-helper/macro_helper_test.h %T/no-move-macro-helper/macro_helper_test.cpp -- -std=c++11
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/new_test.h -check-prefix=CHECK-NEW-TEST-CASE1-H %s
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/new_test.cpp -check-prefix=CHECK-NEW-TEST-CASE1-CPP %s
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/macro_helper_test.h -check-prefix=CHECK-OLD-TEST-CASE1-H %s
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/macro_helper_test.cpp -check-prefix=CHECK-OLD-TEST-CASE1-CPP %s
+// RUN: clang-move -names="A" -new_cc=%t.dir/no-move-macro-helper/new_test.cpp -new_header=%t.dir/no-move-macro-helper/new_test.h -old_cc=%t.dir/no-move-macro-helper/macro_helper_test.cpp -old_header=%t.dir/no-move-macro-helper/macro_helper_test.h %t.dir/no-move-macro-helper/macro_helper_test.cpp -- -std=c++11
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/new_test.h -check-prefix=CHECK-NEW-TEST-CASE1-H %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/new_test.cpp -check-prefix=CHECK-NEW-TEST-CASE1-CPP %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/macro_helper_test.h -check-prefix=CHECK-OLD-TEST-CASE1-H %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/macro_helper_test.cpp -check-prefix=CHECK-OLD-TEST-CASE1-CPP %s
 
 // CHECK-NEW-TEST-CASE1-H: class A {};
 
@@ -24,14 +24,14 @@
 // -----------------------------------------------------------------------------
 // Test moving all.
 // -----------------------------------------------------------------------------
-// RUN: cat %S/Inputs/macro_helper_test.h > %T/no-move-macro-helper/macro_helper_test.h
-// RUN: cat %S/Inputs/macro_helper_test.cpp > %T/no-move-macro-helper/macro_helper_test.cpp
-// RUN: clang-move -names="A, f1" -new_cc=%T/no-move-macro-helper/new_test.cpp -new_header=%T/no-move-macro-helper/new_test.h -old_cc=%T/no-move-macro-helper/macro_helper_test.cpp -old_header=%T/no-move-macro-helper/macro_helper_test.h %T/no-move-macro-helper/macro_helper_test.cpp -- -std=c++11
+// RUN: cat %S/Inputs/macro_helper_test.h > %t.dir/no-move-macro-helper/macro_helper_test.h
+// RUN: cat %S/Inputs/macro_helper_test.cpp > %t.dir/no-move-macro-helper/macro_helper_test.cpp
+// RUN: clang-move -names="A, f1" -new_cc=%t.dir/no-move-macro-helper/new_test.cpp -new_header=%t.dir/no-move-macro-helper/new_test.h -old_cc=%t.dir/no-move-macro-helper/macro_helper_test.cpp -old_header=%t.dir/no-move-macro-helper/macro_helper_test.h %t.dir/no-move-macro-helper/macro_helper_test.cpp -- -std=c++11
 //
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/new_test.h -check-prefix=CHECK-NEW-TEST-CASE2-H %s
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/new_test.cpp -check-prefix=CHECK-NEW-TEST-CASE2-CPP %s
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/macro_helper_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
-// RUN: FileCheck -input-file=%T/no-move-macro-helper/macro_helper_test.cpp -allow-empty -check-prefix=CHECK-EMPTY %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/new_test.h -check-prefix=CHECK-NEW-TEST-CASE2-H %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/new_test.cpp -check-prefix=CHECK-NEW-TEST-CASE2-CPP %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/macro_helper_test.h -allow-empty -check-prefix=CHECK-EMPTY %s
+// RUN: FileCheck -input-file=%t.dir/no-move-macro-helper/macro_helper_test.cpp -allow-empty -check-prefix=CHECK-EMPTY %s
 
 // CHECK-NEW-TEST-CASE2-H: class A {};
 // CHECK-NEW-TEST-CASE2-H-NEXT:void f1();
diff --git a/clang-tools-extra/test/clang-tidy/checkers/misc/header-include-cycle.cpp b/clang-tools-extra/test/clang-tidy/checkers/misc/header-include-cycle.cpp
index d3c71ad..3694bdd 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/misc/header-include-cycle.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/misc/header-include-cycle.cpp
@@ -1,15 +1,15 @@
-// RUN: rm -rf %T/misc-header-include-cycle-headers
-// RUN: mkdir %T/misc-header-include-cycle-headers
-// RUN: cp -r %S/Inputs/header-include-cycle* %T/misc-header-include-cycle-headers/
-// RUN: mkdir %T/misc-header-include-cycle-headers/system
-// RUN: cp -r %S/Inputs/system/header-include-cycle* %T/misc-header-include-cycle-headers/system
-// RUN: cp %s %T/header-include-cycle.cpp
-// RUN: clang-tidy %T%{fs-sep}header-include-cycle.cpp -checks='-*,misc-header-include-cycle' -header-filter=.* \
+// RUN: rm -rf %t.dir/misc-header-include-cycle-headers
+// RUN: mkdir -p %t.dir/misc-header-include-cycle-headers
+// RUN: cp -r %S/Inputs/header-include-cycle* %t.dir/misc-header-include-cycle-headers/
+// RUN: mkdir %t.dir/misc-header-include-cycle-headers/system
+// RUN: cp -r %S/Inputs/system/header-include-cycle* %t.dir/misc-header-include-cycle-headers/system
+// RUN: cp %s %t.dir/header-include-cycle.cpp
+// RUN: clang-tidy %t.dir%{fs-sep}header-include-cycle.cpp -checks='-*,misc-header-include-cycle' -header-filter=.* \
 // RUN: -config="{CheckOptions: {misc-header-include-cycle.IgnoredFilesList: 'header-include-cycle.self-e.hpp'}}" \
-// RUN: -- -I%T%{fs-sep}misc-header-include-cycle-headers -isystem %T%{fs-sep}misc-header-include-cycle-headers%{fs-sep}system \
-// RUN: --include %T%{fs-sep}misc-header-include-cycle-headers%{fs-sep}header-include-cycle.self-i.hpp | FileCheck %s \
+// RUN: -- -I%t.dir%{fs-sep}misc-header-include-cycle-headers -isystem %t.dir%{fs-sep}misc-header-include-cycle-headers%{fs-sep}system \
+// RUN: --include %t.dir%{fs-sep}misc-header-include-cycle-headers%{fs-sep}header-include-cycle.self-i.hpp | FileCheck %s \
 // RUN: -check-prefix=CHECK-MESSAGES "-implicit-check-not={{note|warning|error}}:" --dump-input=fail
-// RUN: rm -rf %T/misc-header-include-cycle-headers
+// RUN: rm -rf %t.dir/misc-header-include-cycle-headers
 
 #ifndef MAIN_GUARD
 #define MAIN_GUARD
diff --git a/clang-tools-extra/test/clang-tidy/checkers/misc/unused-parameters.cpp b/clang-tools-extra/test/clang-tidy/checkers/misc/unused-parameters.cpp
index f1918e9..c963cb5 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/misc/unused-parameters.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/misc/unused-parameters.cpp
@@ -1,11 +1,11 @@
-// RUN: echo "static void staticFunctionHeader(int i) {;}" > %T/header.h
-// RUN: echo "static void staticFunctionHeader(int  /*i*/) {;}" > %T/header-fixed.h
-// RUN: %check_clang_tidy  --match-partial-fixes -std=c++11 %s misc-unused-parameters %t -- -header-filter='.*' -- -fno-delayed-template-parsing
-// RUN: diff %T/header.h %T/header-fixed.h
+// RUN: mkdir -p %t.dir
+// RUN: echo "static void staticFunctionHeader(int i) {;}" > %t.dir/header.h
+// RUN: echo "static void staticFunctionHeader(int  /*i*/) {;}" > %t.dir/header-fixed.h
+// RUN: %check_clang_tidy  --match-partial-fixes -std=c++11 %s misc-unused-parameters %t.dir/code -- -header-filter='.*' -- -fno-delayed-template-parsing
+// RUN: diff %t.dir/header.h %t.dir/header-fixed.h
 // FIXME: Make the test work in all language modes.
 
 #include "header.h"
-// CHECK-MESSAGES: header.h:1:38: warning
 
 // Basic removal
 // =============
@@ -306,3 +306,5 @@ void test() {
 // Do not warn on naked functions.
 [[gnu::naked]] int nakedFunction(int a, float b, const char *c) { ; }
 __attribute__((naked)) void nakedFunction(int a, int b) { ; }
+
+// CHECK-MESSAGES: header.h:1:38: warning
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/concat-nested-namespaces.cpp b/clang-tools-extra/test/clang-tidy/checkers/modernize/concat-nested-namespaces.cpp
index a4f50dd..78adbeb 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/modernize/concat-nested-namespaces.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/concat-nested-namespaces.cpp
@@ -1,10 +1,11 @@
-// RUN: cp %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h %T/modernize-concat-nested-namespaces.h
-// RUN: %check_clang_tidy --match-partial-fixes -std=c++17 -check-suffix=NORMAL %s modernize-concat-nested-namespaces %t -- -header-filter=".*" -- -I %T
-// RUN: FileCheck -input-file=%T/modernize-concat-nested-namespaces.h %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h -check-prefix=CHECK-FIXES
+// RUN: mkdir -p %t.dir
+// RUN: cp %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h %t.dir/modernize-concat-nested-namespaces.h
+// RUN: %check_clang_tidy --match-partial-fixes -std=c++17 -check-suffix=NORMAL %s modernize-concat-nested-namespaces %t.dir/code -- -header-filter=".*" -- -I %t.dir
+// RUN: FileCheck -input-file=%t.dir/modernize-concat-nested-namespaces.h %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h -check-prefix=CHECK-FIXES
 // Restore header file and re-run with c++20:
-// RUN: cp %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h %T/modernize-concat-nested-namespaces.h
-// RUN: %check_clang_tidy --match-partial-fixes -std=c++20 -check-suffixes=NORMAL,CPP20 %s modernize-concat-nested-namespaces %t -- -header-filter=".*" -- -I %T
-// RUN: FileCheck -input-file=%T/modernize-concat-nested-namespaces.h %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h -check-prefix=CHECK-FIXES
+// RUN: cp %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h %t.dir/modernize-concat-nested-namespaces.h
+// RUN: %check_clang_tidy --match-partial-fixes -std=c++20 -check-suffixes=NORMAL,CPP20 %s modernize-concat-nested-namespaces %t.dir/code -- -header-filter=".*" -- -I %t.dir
+// RUN: FileCheck -input-file=%t.dir/modernize-concat-nested-namespaces.h %S/Inputs/concat-nested-namespaces/modernize-concat-nested-namespaces.h -check-prefix=CHECK-FIXES
 
 #include "modernize-concat-nested-namespaces.h"
 // CHECK-MESSAGES-NORMAL-DAG: modernize-concat-nested-namespaces.h:1:1: warning: nested namespaces can be concatenated [modernize-concat-nested-namespaces]
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-header.cpp b/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-header.cpp
index 5d2f1af..461a637 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-header.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-header.cpp
@@ -1,6 +1,7 @@
-// RUN: cp %S/Inputs/pass-by-value/header.h %T/pass-by-value-header.h
-// RUN: clang-tidy %s -checks='-*,modernize-pass-by-value' -header-filter='.*' -fix -- -std=c++11 -I %T | FileCheck %s -check-prefix=CHECK-MESSAGES -implicit-check-not="{{warning|error}}:"
-// RUN: FileCheck -input-file=%T/pass-by-value-header.h %s -check-prefix=CHECK-FIXES
+// RUN: mkdir -p %t.dir
+// RUN: cp %S/Inputs/pass-by-value/header.h %t.dir/pass-by-value-header.h
+// RUN: clang-tidy %s -checks='-*,modernize-pass-by-value' -header-filter='.*' -fix -- -std=c++11 -I %t.dir | FileCheck %s -check-prefix=CHECK-MESSAGES -implicit-check-not="{{warning|error}}:"
+// RUN: FileCheck -input-file=%t.dir/pass-by-value-header.h %s -check-prefix=CHECK-FIXES
 // FIXME: Make the test work in all language modes.
 
 #include "pass-by-value-header.h"
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-multi-fixes.cpp b/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-multi-fixes.cpp
index 238e445..b77c74b 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-multi-fixes.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/pass-by-value-multi-fixes.cpp
@@ -1,12 +1,13 @@
-// RUN: cat %S/Inputs/pass-by-value/header-with-fix.h > %T/pass-by-value-header-with-fix.h
-// RUN: sed -e 's#//.*$##' %s > %t.cpp
-// RUN: clang-tidy %t.cpp -checks='-*,modernize-pass-by-value' -header-filter='.*' -fix -- -std=c++11 -I %T | FileCheck %s -check-prefix=CHECK-MESSAGES -implicit-check-not="{{warning|error}}:"
-// RUN: FileCheck -input-file=%t.cpp %s -check-prefix=CHECK-FIXES
-// RUN: FileCheck -input-file=%T/pass-by-value-header-with-fix.h %s -check-prefix=CHECK-HEADER-FIXES
+// RUN: mkdir -p %t.dir
+// RUN: cat %S/Inputs/pass-by-value/header-with-fix.h > %t.dir/pass-by-value-header-with-fix.h
+// RUN: sed -e 's#//.*$##' %s > %t.dir/code.cpp
+// RUN: clang-tidy %t.dir/code.cpp -checks='-*,modernize-pass-by-value' -header-filter='.*' -fix -- -std=c++11 -I %t.dir | FileCheck %s -check-prefix=CHECK-MESSAGES -implicit-check-not="{{warning|error}}:"
+// RUN: FileCheck -input-file=%t.dir/code.cpp %s -check-prefix=CHECK-FIXES
+// RUN: FileCheck -input-file=%t.dir/pass-by-value-header-with-fix.h %s -check-prefix=CHECK-HEADER-FIXES
 
 #include "pass-by-value-header-with-fix.h"
 // CHECK-HEADER-FIXES: Foo(S s);
 Foo::Foo(const S &s) : s(s) {}
-// CHECK-MESSAGES: :9:10: warning: pass by value and use std::move [modernize-pass-by-value]
+// CHECK-MESSAGES: :10:10: warning: pass by value and use std::move [modernize-pass-by-value]
 // CHECK-FIXES: #include <utility>
 // CHECK-FIXES: Foo::Foo(S s) : s(std::move(s)) {}
diff --git a/clang-tools-extra/test/clang-tidy/checkers/portability/restrict-system-includes-transitive.cpp b/clang-tools-extra/test/clang-tidy/checkers/portability/restrict-system-includes-transitive.cpp
index 744f45f..ff7c66b 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/portability/restrict-system-includes-transitive.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/portability/restrict-system-includes-transitive.cpp
@@ -1,12 +1,12 @@
-// RUN: rm -rf %T/Headers
-// RUN: mkdir %T/Headers
-// RUN: cp -r %S/Inputs/restrict-system-includes %T/Headers/portability-restrict-system-includes
+// RUN: rm -rf %t.dir/Headers
+// RUN: mkdir -p %t.dir/Headers
+// RUN: cp -r %S/Inputs/restrict-system-includes %t.dir/Headers/portability-restrict-system-includes
 // RUN: %check_clang_tidy -std=c++11 %s portability-restrict-system-includes %t \
 // RUN:   -- -config="{CheckOptions: {portability-restrict-system-includes.Includes: 'transitive.h,s.h'}}" \
 // RUN:   -system-headers -header-filter=.* \
-// RUN:   -- -I %T/Headers/portability-restrict-system-includes -isystem %T/Headers/portability-restrict-system-includes/system
-// RUN: FileCheck -input-file=%T/Headers/portability-restrict-system-includes/transitive2.h %s -check-prefix=CHECK-FIXES
-// RUN: rm -rf %T/Headers
+// RUN:   -- -I %t.dir/Headers/portability-restrict-system-includes -isystem %t.dir/Headers/portability-restrict-system-includes/system
+// RUN: FileCheck -input-file=%t.dir/Headers/portability-restrict-system-includes/transitive2.h %s -check-prefix=CHECK-FIXES
+// RUN: rm -rf %t.dir/Headers
 // FIXME: Make the test work in all language modes.
 
 // transitive.h includes <r.h> and <t.h>
diff --git a/clang-tools-extra/test/clang-tidy/checkers/readability/identifier-naming-symlink.cpp b/clang-tools-extra/test/clang-tidy/checkers/readability/identifier-naming-symlink.cpp
index 34dc340..7c99f70 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/readability/identifier-naming-symlink.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/readability/identifier-naming-symlink.cpp
@@ -2,11 +2,12 @@
 // in the header file so it fails if runs multiple times with different
 // `-std` flags as check_clang_tidy doesn by default.
 //
-// RUN: rm -rf %T/symlink
-// RUN: cp -r %S/Inputs/identifier-naming/symlink %T/symlink
-// RUN: mkdir -p %T/symlink/build
-// RUN: ln -s %T/symlink/include/test.h %T/symlink/build/test.h
-// RUN: %check_clang_tidy -std=c++20 %s readability-identifier-naming %t -- --header-filter="test.h" --config-file=%S/Inputs/identifier-naming/symlink/include/.clang-tidy -- -I %T/symlink/build
+// RUN: rm -rf %t.dir
+// RUN: mkdir -p %t.dir
+// RUN: cp -r %S/Inputs/identifier-naming/symlink %t.dir/symlink
+// RUN: mkdir -p %t.dir/symlink/build
+// RUN: ln -s %t.dir/symlink/include/test.h %t.dir/symlink/build/test.h
+// RUN: %check_clang_tidy -std=c++20 %s readability-identifier-naming %t.dir -- --header-filter="test.h" --config-file=%S/Inputs/identifier-naming/symlink/include/.clang-tidy -- -I %t.dir/symlink/build
 // UNSUPPORTED: system-windows
 
 #include "test.h"
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-diff.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-diff.cpp
index 7aa6ce9..25be90f 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-diff.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-diff.cpp
@@ -3,9 +3,9 @@
 // RUN: clang-tidy -checks=-*,modernize-use-override %t.cpp -- -std=c++11 | FileCheck -check-prefix=CHECK-SANITY %s
 // RUN: not diff -U0 %s %t.cpp | %clang_tidy_diff -checks=-*,modernize-use-override -- -std=c++11 2>&1 | FileCheck %s --check-prefixes=CHECK,CHECK-JMAX
 // RUN: not diff -U0 %s %t.cpp | %clang_tidy_diff -checks=-*,modernize-use-override -quiet -- -std=c++11 2>&1 | FileCheck -check-prefix=CHECK-QUIET %s
-// RUN: mkdir -p %T/compilation-database-test/
-// RUN: echo '[{"directory": "%T", "command": "clang++ -o test.o -std=c++11 %t.cpp", "file": "%t.cpp"}]' > %T/compilation-database-test/compile_commands.json
-// RUN: not diff -U0 %s %t.cpp | %clang_tidy_diff -checks=-*,modernize-use-override -path %T/compilation-database-test 2>&1 | FileCheck -check-prefix=CHECK %s
+// RUN: mkdir -p %t.dir/compilation-database-test/
+// RUN: echo '[{"directory": "%t.dir", "command": "clang++ -o test.o -std=c++11 %t.cpp", "file": "%t.cpp"}]' > %t.dir/compilation-database-test/compile_commands.json
+// RUN: not diff -U0 %s %t.cpp | %clang_tidy_diff -checks=-*,modernize-use-override -path %t.dir/compilation-database-test 2>&1 | FileCheck -check-prefix=CHECK %s
 
 // RUN: not diff -U0 %s %t.cpp | %clang_tidy_diff -checks=-*,modernize-use-override -j 1 -- -std=c++11 2>&1 | FileCheck %s --check-prefix=CHECK-J1
 // CHECK-J1: Running clang-tidy in 1 threads...
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-run-with-database.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-run-with-database.cpp
index 3c4e849..9ca0ab3 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-run-with-database.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-run-with-database.cpp
@@ -1,28 +1,28 @@
-// RUN: mkdir -p %T/compilation-database-test/include
-// RUN: mkdir -p %T/compilation-database-test/a
-// RUN: mkdir -p %T/compilation-database-test/b
-// RUN: echo 'int *AA = 0;' > %T/compilation-database-test/a/a.cpp
-// RUN: echo 'int *AB = 0;' > %T/compilation-database-test/a/b.cpp
-// RUN: echo 'int *BB = 0;' > %T/compilation-database-test/b/b.cpp
-// RUN: echo 'int *BC = 0;' > %T/compilation-database-test/b/c.cpp
-// RUN: echo 'int *BD = 0;' > %T/compilation-database-test/b/d.cpp
-// RUN: echo 'int *HP = 0;' > %T/compilation-database-test/include/header.h
-// RUN: echo '#include "header.h"' > %T/compilation-database-test/include.cpp
-// RUN: sed 's|test_dir|%/T/compilation-database-test|g' %S/Inputs/compilation-database/template.json > %T/compile_commands.json
+// RUN: mkdir -p %t.dir/compilation-database-test/include
+// RUN: mkdir -p %t.dir/compilation-database-test/a
+// RUN: mkdir -p %t.dir/compilation-database-test/b
+// RUN: echo 'int *AA = 0;' > %t.dir/compilation-database-test/a/a.cpp
+// RUN: echo 'int *AB = 0;' > %t.dir/compilation-database-test/a/b.cpp
+// RUN: echo 'int *BB = 0;' > %t.dir/compilation-database-test/b/b.cpp
+// RUN: echo 'int *BC = 0;' > %t.dir/compilation-database-test/b/c.cpp
+// RUN: echo 'int *BD = 0;' > %t.dir/compilation-database-test/b/d.cpp
+// RUN: echo 'int *HP = 0;' > %t.dir/compilation-database-test/include/header.h
+// RUN: echo '#include "header.h"' > %t.dir/compilation-database-test/include.cpp
+// RUN: sed 's|test_dir|%/t.dir/compilation-database-test|g' %S/Inputs/compilation-database/template.json > %t.dir/compile_commands.json
 
 // Regression test: shouldn't crash.
-// RUN: not clang-tidy --checks=-*,modernize-use-nullptr -p %T %T/compilation-database-test/b/not-exist -header-filter=.* 2>&1 | FileCheck %s -check-prefix=CHECK-NOT-EXIST
+// RUN: not clang-tidy --checks=-*,modernize-use-nullptr -p %t.dir %t.dir/compilation-database-test/b/not-exist -header-filter=.* 2>&1 | FileCheck %s -check-prefix=CHECK-NOT-EXIST
 // CHECK-NOT-EXIST: Error while processing {{.*[/\\]}}not-exist.
 // CHECK-NOT-EXIST: unable to handle compilation
 // CHECK-NOT-EXIST: Found compiler error
 
-// RUN: clang-tidy --checks=-*,modernize-use-nullptr -p %T %T/compilation-database-test/a/a.cpp %T/compilation-database-test/a/b.cpp %T/compilation-database-test/b/b.cpp %T/compilation-database-test/b/c.cpp %T/compilation-database-test/b/d.cpp %T/compilation-database-test/include.cpp -header-filter=.* -fix
-// RUN: FileCheck -input-file=%T/compilation-database-test/a/a.cpp %s -check-prefix=CHECK-FIX1
-// RUN: FileCheck -input-file=%T/compilation-database-test/a/b.cpp %s -check-prefix=CHECK-FIX2
-// RUN: FileCheck -input-file=%T/compilation-database-test/b/b.cpp %s -check-prefix=CHECK-FIX3
-// RUN: FileCheck -input-file=%T/compilation-database-test/b/c.cpp %s -check-prefix=CHECK-FIX4
-// RUN: FileCheck -input-file=%T/compilation-database-test/b/d.cpp %s -check-prefix=CHECK-FIX5
-// RUN: FileCheck -input-file=%T/compilation-database-test/include/header.h %s -check-prefix=CHECK-FIX6
+// RUN: clang-tidy --checks=-*,modernize-use-nullptr -p %t.dir %t.dir/compilation-database-test/a/a.cpp %t.dir/compilation-database-test/a/b.cpp %t.dir/compilation-database-test/b/b.cpp %t.dir/compilation-database-test/b/c.cpp %t.dir/compilation-database-test/b/d.cpp %t.dir/compilation-database-test/include.cpp -header-filter=.* -fix
+// RUN: FileCheck -input-file=%t.dir/compilation-database-test/a/a.cpp %s -check-prefix=CHECK-FIX1
+// RUN: FileCheck -input-file=%t.dir/compilation-database-test/a/b.cpp %s -check-prefix=CHECK-FIX2
+// RUN: FileCheck -input-file=%t.dir/compilation-database-test/b/b.cpp %s -check-prefix=CHECK-FIX3
+// RUN: FileCheck -input-file=%t.dir/compilation-database-test/b/c.cpp %s -check-prefix=CHECK-FIX4
+// RUN: FileCheck -input-file=%t.dir/compilation-database-test/b/d.cpp %s -check-prefix=CHECK-FIX5
+// RUN: FileCheck -input-file=%t.dir/compilation-database-test/include/header.h %s -check-prefix=CHECK-FIX6
 
 // CHECK-FIX1: int *AA = nullptr;
 // CHECK-FIX2: int *AB = nullptr;
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-store-check-profile-one-tu.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-store-check-profile-one-tu.cpp
index f0939f7..192fbf5 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-store-check-profile-one-tu.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/clang-tidy-store-check-profile-one-tu.cpp
@@ -1,9 +1,9 @@
-// RUN: rm -rf %T/out
-// RUN: clang-tidy -enable-check-profile -checks='-*,readability-function-size' -store-check-profile=%T/out %s -- 2>&1 | not FileCheck --match-full-lines -implicit-check-not='{{warning:|error:}}' -check-prefix=CHECK-CONSOLE %s
-// RUN: cat %T/out/*-clang-tidy-store-check-profile-one-tu.cpp.json | FileCheck --match-full-lines -implicit-check-not='{{warning:|error:}}' -check-prefix=CHECK-FILE %s
-// RUN: rm -rf %T/out
-// RUN: clang-tidy -enable-check-profile -checks='-*,readability-function-size' -store-check-profile=%T/out %s -- 2>&1
-// RUN: cat %T/out/*-clang-tidy-store-check-profile-one-tu.cpp.json | FileCheck --match-full-lines -implicit-check-not='{{warning:|error:}}' -check-prefix=CHECK-FILE %s
+// RUN: rm -rf %t.dir/out
+// RUN: clang-tidy -enable-check-profile -checks='-*,readability-function-size' -store-check-profile=%t.dir/out %s -- 2>&1 | not FileCheck --match-full-lines -implicit-check-not='{{warning:|error:}}' -check-prefix=CHECK-CONSOLE %s
+// RUN: cat %t.dir/out/*-clang-tidy-store-check-profile-one-tu.cpp.json | FileCheck --match-full-lines -implicit-check-not='{{warning:|error:}}' -check-prefix=CHECK-FILE %s
+// RUN: rm -rf %t.dir/out
+// RUN: clang-tidy -enable-check-profile -checks='-*,readability-function-size' -store-check-profile=%t.dir/out %s -- 2>&1
+// RUN: cat %t.dir/out/*-clang-tidy-store-check-profile-one-tu.cpp.json | FileCheck --match-full-lines -implicit-check-not='{{warning:|error:}}' -check-prefix=CHECK-FILE %s
 
 // CHECK-CONSOLE-NOT: ===-------------------------------------------------------------------------===
 // CHECK-CONSOLE-NOT: {{.*}}  --- Name ---
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/diagnostic.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/diagnostic.cpp
index 57d930b..610d1da 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/diagnostic.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/diagnostic.cpp
@@ -14,15 +14,15 @@
 // Now create a directory with a compilation database file and ensure we don't
 // use it after failing to parse commands from the command line:
 //
-// RUN: mkdir -p %T/diagnostics/
-// RUN: echo '[{"directory": "%/T/diagnostics/","command": "clang++ -fan-option-from-compilation-database -c %/T/diagnostics/input.cpp", "file": "%/T/diagnostics/input.cpp"}]' > %T/diagnostics/compile_commands.json
-// RUN: cat %s > %T/diagnostics/input.cpp
-// RUN: not clang-tidy -checks='-*,modernize-use-override' %T/diagnostics/nonexistent.cpp -- 2>&1 | FileCheck -check-prefix=CHECK1 -implicit-check-not='{{warning:|error:}}' %s
-// RUN: not clang-tidy -checks='-*,clang-diagnostic-*,google-explicit-constructor' %T/diagnostics/input.cpp -- -fan-unknown-option 2>&1 | FileCheck -check-prefix=CHECK2 -implicit-check-not='{{warning:|error:}}' %s
-// RUN: not clang-tidy -checks='-*,google-explicit-constructor,clang-diagnostic-literal-conversion' %T/diagnostics/input.cpp -- -fan-unknown-option 2>&1 | FileCheck -check-prefix=CHECK3 -implicit-check-not='{{warning:|error:}}' %s
-// RUN: clang-tidy -checks='-*,modernize-use-override,clang-diagnostic-macro-redefined' %T/diagnostics/input.cpp -- -DMACRO_FROM_COMMAND_LINE 2>&1 | FileCheck -check-prefix=CHECK4 -implicit-check-not='{{warning:|error:}}' %s
-// RUN: not clang-tidy -checks='-*,clang-diagnostic-*,google-explicit-constructor' %T/diagnostics/input.cpp 2>&1 | FileCheck -check-prefix=CHECK5 -implicit-check-not='{{warning:|error:}}' %s
-// RUN: not clang-tidy -checks='-*,modernize-use-override' %T/diagnostics/input.cpp -- -DCOMPILATION_ERROR 2>&1 | FileCheck -check-prefix=CHECK6 -implicit-check-not='{{warning:|error:}}' %s
+// RUN: mkdir -p %t.dir/diagnostics/
+// RUN: echo '[{"directory": "%/t.dir/diagnostics/","command": "clang++ -fan-option-from-compilation-database -c %/T/diagnostics/input.cpp", "file": "%/T/diagnostics/input.cpp"}]' > %t.dir/diagnostics/compile_commands.json
+// RUN: cat %s > %t.dir/diagnostics/input.cpp
+// RUN: not clang-tidy -checks='-*,modernize-use-override' %t.dir/diagnostics/nonexistent.cpp -- 2>&1 | FileCheck -check-prefix=CHECK1 -implicit-check-not='{{warning:|error:}}' %s
+// RUN: not clang-tidy -checks='-*,clang-diagnostic-*,google-explicit-constructor' %t.dir/diagnostics/input.cpp -- -fan-unknown-option 2>&1 | FileCheck -check-prefix=CHECK2 -implicit-check-not='{{warning:|error:}}' %s
+// RUN: not clang-tidy -checks='-*,google-explicit-constructor,clang-diagnostic-literal-conversion' %t.dir/diagnostics/input.cpp -- -fan-unknown-option 2>&1 | FileCheck -check-prefix=CHECK3 -implicit-check-not='{{warning:|error:}}' %s
+// RUN: clang-tidy -checks='-*,modernize-use-override,clang-diagnostic-macro-redefined' %t.dir/diagnostics/input.cpp -- -DMACRO_FROM_COMMAND_LINE 2>&1 | FileCheck -check-prefix=CHECK4 -implicit-check-not='{{warning:|error:}}' %s
+// RUN: not clang-tidy -checks='-*,clang-diagnostic-*,google-explicit-constructor' %t.dir/diagnostics/input.cpp 2>&1 | FileCheck -check-prefix=CHECK5 -implicit-check-not='{{warning:|error:}}' %s
+// RUN: not clang-tidy -checks='-*,modernize-use-override' %t.dir/diagnostics/input.cpp -- -DCOMPILATION_ERROR 2>&1 | FileCheck -check-prefix=CHECK6 -implicit-check-not='{{warning:|error:}}' %s
 // RUN: clang-tidy -checks='-*,modernize-use-override,clang-diagnostic-macro-redefined' %s -- -DMACRO_FROM_COMMAND_LINE -std=c++20 | FileCheck -check-prefix=CHECK4 -implicit-check-not='{{warning:|error:}}' %s
 // RUN: clang-tidy -checks='-*,modernize-use-override,clang-diagnostic-macro-redefined,clang-diagnostic-literal-conversion' %s -- -DMACRO_FROM_COMMAND_LINE -std=c++20 -Wno-macro-redefined | FileCheck --check-prefix=CHECK7 -implicit-check-not='{{warning:|error:}}' %s
 // RUN: clang-tidy -checks='-*,modernize-use-override' %s -- -std=c++20 -DPR64602
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/export-relpath.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/export-relpath.cpp
index 5bfd41f..5fd7303 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/export-relpath.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/export-relpath.cpp
@@ -1,14 +1,15 @@
-// RUN: rm -rf %T/clang-tidy/export-relpath
-// RUN: mkdir -p %T/clang-tidy/export-relpath/subdir
-// RUN: cp %s %T/clang-tidy/export-relpath/subdir/source.cpp
-// RUN: echo '[{ "directory": "%/T/clang-tidy/export-relpath/subdir", "command": "clang++ source.cpp", "file": "%/T/clang-tidy/export-relpath/subdir/source.cpp"}]' > %T/clang-tidy/export-relpath/subdir/compile_commands.json
+// RUN: mkdir -p %t.dir
+// RUN: rm -rf %t.dir/clang-tidy/export-relpath
+// RUN: mkdir -p %t.dir/clang-tidy/export-relpath/subdir
+// RUN: cp %s %t.dir/clang-tidy/export-relpath/subdir/source.cpp
+// RUN: echo '[{ "directory": "%/t.dir/clang-tidy/export-relpath/subdir", "command": "clang++ source.cpp", "file": "%/T/clang-tidy/export-relpath/subdir/source.cpp"}]' > %t.dir/clang-tidy/export-relpath/subdir/compile_commands.json
 //
 // Check that running clang-tidy in './subdir' and storing results
 // in './fixes.yaml' works as expected.
 //
-// RUN: cd %T/clang-tidy/export-relpath
+// RUN: cd %t.dir/clang-tidy/export-relpath
 // RUN: clang-tidy -p subdir subdir/source.cpp -checks='-*,google-explicit-constructor,llvm-namespace-comment' -export-fixes=./fixes.yaml
-// RUN: FileCheck -input-file=%T/clang-tidy/export-relpath/fixes.yaml -check-prefix=CHECK-YAML %s
+// RUN: FileCheck -input-file=%t.dir/clang-tidy/export-relpath/fixes.yaml -check-prefix=CHECK-YAML %s
 
 namespace i {
 void f(); // So that the namespace isn't empty.
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/list-checks.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/list-checks.cpp
index 674c118..73bb5b7 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/list-checks.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/list-checks.cpp
@@ -1,4 +1,4 @@
-// RUN: mkdir -p %T/clang-tidy/list-checks/
-// RUN: echo '{Checks: "-*,google-*"}' > %T/clang-tidy/.clang-tidy
-// RUN: cd %T/clang-tidy/list-checks
+// RUN: mkdir -p %t.dir/clang-tidy/list-checks/
+// RUN: echo '{Checks: "-*,google-*"}' > %t.dir/clang-tidy/.clang-tidy
+// RUN: cd %t.dir/clang-tidy/list-checks
 // RUN: clang-tidy -list-checks | grep "^ *google-"
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/read_file_config.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/read_file_config.cpp
index 3e39b4d..8b12f45 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/read_file_config.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/read_file_config.cpp
@@ -1,10 +1,10 @@
 // REQUIRES: static-analyzer
-// RUN: mkdir -p %T/read-file-config/
-// RUN: cp %s %T/read-file-config/test.cpp
-// RUN: echo 'Checks: "-*,modernize-use-nullptr"' > %T/read-file-config/.clang-tidy
-// RUN: echo '[{"command": "cc -c -o test.o test.cpp", "directory": "%/T/read-file-config", "file": "%/T/read-file-config/test.cpp"}]' > %T/read-file-config/compile_commands.json
-// RUN: clang-tidy %T/read-file-config/test.cpp | not grep "warning: .*\[clang-analyzer-deadcode.DeadStores\]$"
-// RUN: clang-tidy -checks="-*,clang-analyzer-*" %T/read-file-config/test.cpp | grep "warning: .*\[clang-analyzer-deadcode.DeadStores\]$"
+// RUN: mkdir -p %t.dir/read-file-config/
+// RUN: cp %s %t.dir/read-file-config/test.cpp
+// RUN: echo 'Checks: "-*,modernize-use-nullptr"' > %t.dir/read-file-config/.clang-tidy
+// RUN: echo '[{"command": "cc -c -o test.o test.cpp", "directory": "%/t.dir/read-file-config", "file": "%/t.dir/read-file-config/test.cpp"}]' > %t.dir/read-file-config/compile_commands.json
+// RUN: clang-tidy %t.dir/read-file-config/test.cpp | not grep "warning: .*\[clang-analyzer-deadcode.DeadStores\]$"
+// RUN: clang-tidy -checks="-*,clang-analyzer-*" %t.dir/read-file-config/test.cpp | grep "warning: .*\[clang-analyzer-deadcode.DeadStores\]$"
 
 void f() {
   int x;
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/verify-config.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/verify-config.cpp
index 93f6f9f..30d4933 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/verify-config.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/verify-config.cpp
@@ -19,22 +19,22 @@
 // CHECK-VERIFY: command-line option '-checks': warning: unknown check 'llvm-includeorder'; did you mean 'llvm-include-order' [-verify-config]
 // CHECK-VERIFY: command-line option '-checks': warning: unknown check 'my-made-up-check' [-verify-config]
 
-// RUN: echo -e 'Checks: |\n bugprone-argument-comment\n bugprone-assert-side-effect,\n bugprone-bool-pointer-implicit-conversion\n readability-use-anyof*' > %T/MyClangTidyConfig
+// RUN: echo -e 'Checks: |\n bugprone-argument-comment\n bugprone-assert-side-effect,\n bugprone-bool-pointer-implicit-conversion\n readability-use-anyof*' > %t.MyClangTidyConfig
 // RUN: clang-tidy -verify-config \
-// RUN: --config-file=%T/MyClangTidyConfig | FileCheck %s -check-prefix=CHECK-VERIFY-BLOCK-OK
+// RUN: --config-file=%t.MyClangTidyConfig | FileCheck %s -check-prefix=CHECK-VERIFY-BLOCK-OK
 // CHECK-VERIFY-BLOCK-OK: No config errors detected.
 
-// RUN: echo -e 'Checks: |\n bugprone-arguments-*\n bugprone-assert-side-effects\n bugprone-bool-pointer-implicit-conversion' > %T/MyClangTidyConfigBad
+// RUN: echo -e 'Checks: |\n bugprone-arguments-*\n bugprone-assert-side-effects\n bugprone-bool-pointer-implicit-conversion' > %t.MyClangTidyConfigBad
 // RUN: not clang-tidy -verify-config \
-// RUN: --config-file=%T/MyClangTidyConfigBad 2>&1 | FileCheck %s -check-prefix=CHECK-VERIFY-BLOCK-BAD
+// RUN: --config-file=%t.MyClangTidyConfigBad 2>&1 | FileCheck %s -check-prefix=CHECK-VERIFY-BLOCK-BAD
 // CHECK-VERIFY-BLOCK-BAD: command-line option '-config': warning: check glob 'bugprone-arguments-*' doesn't match any known check [-verify-config]
 // CHECK-VERIFY-BLOCK-BAD: command-line option '-config': warning: unknown check 'bugprone-assert-side-effects'; did you mean 'bugprone-assert-side-effect' [-verify-config]
 
-// RUN: echo -e 'Checks: "-*,clang-analyzer-optin.cplusplus.UninitializedObject"\nCheckOptions:\n clang-analyzer-optin.cplusplus.UninitializedObject:Pedantic: true' > %T/MyClangTidyConfigCSA
-// RUN: clang-tidy --verify-config --config-file=%T/MyClangTidyConfigCSA 2>&1 | FileCheck %s -check-prefix=CHECK-VERIFY-CSA-OK -implicit-check-not='{{warnings|error}}'
+// RUN: echo -e 'Checks: "-*,clang-analyzer-optin.cplusplus.UninitializedObject"\nCheckOptions:\n clang-analyzer-optin.cplusplus.UninitializedObject:Pedantic: true' > %t.MyClangTidyConfigCSA
+// RUN: clang-tidy --verify-config --config-file=%t.MyClangTidyConfigCSA 2>&1 | FileCheck %s -check-prefix=CHECK-VERIFY-CSA-OK -implicit-check-not='{{warnings|error}}'
 // CHECK-VERIFY-CSA-OK: No config errors detected.
 
-// RUN: echo -e 'Checks: "-*,clang-analyzer-optin.cplusplus.UninitializedObject"\nCheckOptions:\n clang-analyzer-optin.cplusplus.UninitializedObject.Pedantic: true' > %T/MyClangTidyConfigCSABad
-// RUN: not clang-tidy --verify-config --config-file=%T/MyClangTidyConfigCSABad 2>&1 | FileCheck %s -check-prefix=CHECK-VERIFY-CSA-BAD -implicit-check-not='{{warnings|error}}'
+// RUN: echo -e 'Checks: "-*,clang-analyzer-optin.cplusplus.UninitializedObject"\nCheckOptions:\n clang-analyzer-optin.cplusplus.UninitializedObject.Pedantic: true' > %t.MyClangTidyConfigCSABad
+// RUN: not clang-tidy --verify-config --config-file=%t.MyClangTidyConfigCSABad 2>&1 | FileCheck %s -check-prefix=CHECK-VERIFY-CSA-BAD -implicit-check-not='{{warnings|error}}'
 // CHECK-VERIFY-CSA-BAD: command-line option '-config': warning: unknown check option 'clang-analyzer-optin.cplusplus.UninitializedObject.Pedantic'; did you mean 'clang-analyzer-optin.cplusplus.UninitializedObject:Pedantic' [-verify-config]
 
diff --git a/clang-tools-extra/test/modularize/NoProblemsAssistant.modularize b/clang-tools-extra/test/modularize/NoProblemsAssistant.modularize
index 7ddc726..39c06dc 100644
--- a/clang-tools-extra/test/modularize/NoProblemsAssistant.modularize
+++ b/clang-tools-extra/test/modularize/NoProblemsAssistant.modularize
@@ -1,5 +1,7 @@
+# RUN: mkdir -p %t.dir/Output
+# RUN: cd %t.dir
 # RUN: modularize -module-map-path=Output/NoProblemsAssistant.txt -root-module=Root -prefix=%S/Input %s
-# RUN: FileCheck --input-file=%T/NoProblemsAssistant.txt %s
+# RUN: FileCheck --input-file=Output/NoProblemsAssistant.txt %s
 
 SomeTypes.h
 SomeDecls.h
diff --git a/clang-tools-extra/unittests/clang-apply-replacements/ApplyReplacementsTest.cpp b/clang-tools-extra/unittests/clang-apply-replacements/ApplyReplacementsTest.cpp
index 87b0d69..0b92118 100644
--- a/clang-tools-extra/unittests/clang-apply-replacements/ApplyReplacementsTest.cpp
+++ b/clang-tools-extra/unittests/clang-apply-replacements/ApplyReplacementsTest.cpp
@@ -33,8 +33,7 @@ makeTUDiagnostics(const std::string &MainSourceFile, StringRef DiagnosticName,
 // before applying.
 TEST(ApplyReplacementsTest, mergeDiagnosticsWithNoFixes) {
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts);
   FileManager Files((FileSystemOptions()));
   SourceManager SM(Diagnostics, Files);
   TUReplacements TURs;
diff --git a/clang-tools-extra/unittests/clang-tidy/ClangTidyOptionsTest.cpp b/clang-tools-extra/unittests/clang-tidy/ClangTidyOptionsTest.cpp
index d3ca26a..410cebf 100644
--- a/clang-tools-extra/unittests/clang-tidy/ClangTidyOptionsTest.cpp
+++ b/clang-tools-extra/unittests/clang-tidy/ClangTidyOptionsTest.cpp
@@ -318,7 +318,8 @@ TEST(CheckOptionsValidation, MissingOptions) {
       ClangTidyGlobalOptions(), Options));
   ClangTidyDiagnosticConsumer DiagConsumer(Context);
   auto DiagOpts = std::make_unique<DiagnosticOptions>();
-  DiagnosticsEngine DE(new DiagnosticIDs(), *DiagOpts, &DiagConsumer, false);
+  DiagnosticsEngine DE(DiagnosticIDs::create(), *DiagOpts, &DiagConsumer,
+                       false);
   Context.setDiagnosticsEngine(std::move(DiagOpts), &DE);
   TestCheck TestCheck(&Context);
   EXPECT_FALSE(TestCheck.getLocal("Opt"));
@@ -348,7 +349,8 @@ TEST(CheckOptionsValidation, ValidIntOptions) {
       ClangTidyGlobalOptions(), Options));
   ClangTidyDiagnosticConsumer DiagConsumer(Context);
   auto DiagOpts = std::make_unique<DiagnosticOptions>();
-  DiagnosticsEngine DE(new DiagnosticIDs(), *DiagOpts, &DiagConsumer, false);
+  DiagnosticsEngine DE(DiagnosticIDs::create(), *DiagOpts, &DiagConsumer,
+                       false);
   Context.setDiagnosticsEngine(std::move(DiagOpts), &DE);
   TestCheck TestCheck(&Context);
 
@@ -410,7 +412,8 @@ TEST(ValidConfiguration, ValidEnumOptions) {
       ClangTidyGlobalOptions(), Options));
   ClangTidyDiagnosticConsumer DiagConsumer(Context);
   auto DiagOpts = std::make_unique<DiagnosticOptions>();
-  DiagnosticsEngine DE(new DiagnosticIDs(), *DiagOpts, &DiagConsumer, false);
+  DiagnosticsEngine DE(DiagnosticIDs::create(), *DiagOpts, &DiagConsumer,
+                       false);
   Context.setDiagnosticsEngine(std::move(DiagOpts), &DE);
   TestCheck TestCheck(&Context);
 
diff --git a/clang-tools-extra/unittests/clang-tidy/ClangTidyTest.h b/clang-tools-extra/unittests/clang-tidy/ClangTidyTest.h
index 789cc2a..89f0f9f 100644
--- a/clang-tools-extra/unittests/clang-tidy/ClangTidyTest.h
+++ b/clang-tools-extra/unittests/clang-tidy/ClangTidyTest.h
@@ -97,7 +97,8 @@ runCheckOnCode(StringRef Code, std::vector<ClangTidyError> *Errors = nullptr,
       ClangTidyGlobalOptions(), Options));
   ClangTidyDiagnosticConsumer DiagConsumer(Context);
   auto DiagOpts = std::make_unique<DiagnosticOptions>();
-  DiagnosticsEngine DE(new DiagnosticIDs(), *DiagOpts, &DiagConsumer, false);
+  DiagnosticsEngine DE(DiagnosticIDs::create(), *DiagOpts, &DiagConsumer,
+                       false);
   Context.setDiagnosticsEngine(std::move(DiagOpts), &DE);
 
   std::vector<std::string> Args(1, "clang-tidy");
diff --git a/clang-tools-extra/unittests/include/common/VirtualFileHelper.h b/clang-tools-extra/unittests/include/common/VirtualFileHelper.h
index 86991bb3..cb075f8 100644
--- a/clang-tools-extra/unittests/include/common/VirtualFileHelper.h
+++ b/clang-tools-extra/unittests/include/common/VirtualFileHelper.h
@@ -32,8 +32,7 @@ class VirtualFileHelper {
 
 public:
   VirtualFileHelper()
-      : Diagnostics(IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs),
-                    DiagOpts),
+      : Diagnostics(DiagnosticIDs::create(), DiagOpts),
         DiagnosticPrinter(llvm::outs(), DiagOpts),
         Files((FileSystemOptions())) {}
 
diff --git a/clang/docs/ShadowCallStack.rst b/clang/docs/ShadowCallStack.rst
index fc8bea8..9b104cc 100644
--- a/clang/docs/ShadowCallStack.rst
+++ b/clang/docs/ShadowCallStack.rst
@@ -61,7 +61,7 @@ The instrumentation makes use of the platform register ``x18`` on AArch64,
 ``x3`` (``gp``) on RISC-V with software shadow stack and ``ssp`` on RISC-V with
 hardware shadow stack, which needs `Zicfiss`_ and ``-fcf-protection=return``.
 Users can choose between the software and hardware based shadow stack
-implementation on RISC-V backend by passing ``-fsanitize=shadowcallstack``
+implementation on RISC-V backend by passing ``-fsanitize=shadow-call-stack``
 or ``Zicfiss`` with ``-fcf-protection=return``.
 For simplicity we will refer to this as the ``SCSReg``. On some platforms,
 ``SCSReg`` is reserved, and on others, it is designated as a scratch register.
diff --git a/clang/include/clang/Basic/BuiltinsAMDGPU.def b/clang/include/clang/Basic/BuiltinsAMDGPU.def
index 172ac46..5b1c14e 100644
--- a/clang/include/clang/Basic/BuiltinsAMDGPU.def
+++ b/clang/include/clang/Basic/BuiltinsAMDGPU.def
@@ -698,6 +698,7 @@ TARGET_BUILTIN(__builtin_amdgcn_sin_bf16, "yy", "nc", "bf16-trans-insts")
 TARGET_BUILTIN(__builtin_amdgcn_cos_bf16, "yy", "nc", "bf16-trans-insts")
 
 TARGET_BUILTIN(__builtin_amdgcn_cvt_sr_pk_bf16_f32, "V2yffi", "nc", "bf16-cvt-insts")
+TARGET_BUILTIN(__builtin_amdgcn_cvt_sr_pk_f16_f32, "V2hffi", "nc", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_cvt_f16_fp8, "hiIi", "nc", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_cvt_f16_bf8, "hiIi", "nc", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_cvt_pk_f16_fp8, "V2hs", "nc", "gfx1250-insts")
diff --git a/clang/include/clang/Basic/BuiltinsNVPTX.td b/clang/include/clang/Basic/BuiltinsNVPTX.td
index 6e531ef..2d6fa17 100644
--- a/clang/include/clang/Basic/BuiltinsNVPTX.td
+++ b/clang/include/clang/Basic/BuiltinsNVPTX.td
@@ -21,13 +21,17 @@ class SM<string version, list<SMFeatures> newer_list> : SMFeatures {
                         !strconcat(f, "|", newer.Features));
 }
 
+let Features = "sm_121a" in def SM_121a : SMFeatures;
 let Features = "sm_120a" in def SM_120a : SMFeatures;
+let Features = "sm_103a" in def SM_103a : SMFeatures;
 let Features = "sm_101a" in def SM_101a : SMFeatures;
 let Features = "sm_100a" in def SM_100a : SMFeatures;
 let Features = "sm_90a" in def SM_90a : SMFeatures;
 
-def SM_120 : SM<"120", [SM_120a]>;
-def SM_101 : SM<"101", [SM_101a, SM_120]>;
+def SM_121 : SM<"121", [SM_121a]>;
+def SM_120 : SM<"120", [SM_120a, SM_121]>;
+def SM_103 : SM<"103", [SM_103a, SM_120]>;
+def SM_101 : SM<"101", [SM_101a, SM_103]>;
 def SM_100 : SM<"100", [SM_100a, SM_101]>;
 def SM_90 : SM<"90", [SM_90a, SM_100]>;
 def SM_89 : SM<"89", [SM_90]>;
@@ -50,8 +54,9 @@ class PTX<string version, PTXFeatures newer> : PTXFeatures {
   let Features = !strconcat("ptx", version, "|", newer.Features);
 }
 
-let Features = "ptx87" in def PTX87 : PTXFeatures;
+let Features = "ptx88" in def PTX88 : PTXFeatures;
 
+def PTX87 : PTX<"87", PTX88>;
 def PTX86 : PTX<"86", PTX87>;
 def PTX85 : PTX<"85", PTX86>;
 def PTX84 : PTX<"84", PTX85>;
diff --git a/clang/include/clang/Basic/Cuda.h b/clang/include/clang/Basic/Cuda.h
index d6a22a7..81a792d 100644
--- a/clang/include/clang/Basic/Cuda.h
+++ b/clang/include/clang/Basic/Cuda.h
@@ -47,9 +47,10 @@ enum class CudaVersion {
   CUDA_125,
   CUDA_126,
   CUDA_128,
-  FULLY_SUPPORTED = CUDA_123,
+  CUDA_129,
+  FULLY_SUPPORTED = CUDA_128,
   PARTIALLY_SUPPORTED =
-      CUDA_128, // Partially supported. Proceed with a warning.
+      CUDA_129, // Partially supported. Proceed with a warning.
   NEW = 10000,  // Too new. Issue a warning, but allow using it.
 };
 const char *CudaVersionToString(CudaVersion V);
diff --git a/clang/include/clang/Basic/DiagnosticIDs.h b/clang/include/clang/Basic/DiagnosticIDs.h
index f07a003..b21a3b6 100644
--- a/clang/include/clang/Basic/DiagnosticIDs.h
+++ b/clang/include/clang/Basic/DiagnosticIDs.h
@@ -272,6 +272,11 @@ public:
   DiagnosticIDs();
   ~DiagnosticIDs();
 
+  // Convenience method to construct a new refcounted DiagnosticIDs.
+  static llvm::IntrusiveRefCntPtr<DiagnosticIDs> create() {
+    return llvm::makeIntrusiveRefCnt<DiagnosticIDs>();
+  }
+
   /// Return an ID for a diagnostic with the specified format string and
   /// level.
   ///
diff --git a/clang/include/clang/Basic/OffloadArch.h b/clang/include/clang/Basic/OffloadArch.h
index 4dda3ec..387a684 100644
--- a/clang/include/clang/Basic/OffloadArch.h
+++ b/clang/include/clang/Basic/OffloadArch.h
@@ -45,8 +45,12 @@ enum class OffloadArch {
   SM_100a,
   SM_101,
   SM_101a,
+  SM_103,
+  SM_103a,
   SM_120,
   SM_120a,
+  SM_121,
+  SM_121a,
   GFX600,
   GFX601,
   GFX602,
diff --git a/clang/include/clang/CIR/Dialect/Builder/CIRBaseBuilder.h b/clang/include/clang/CIR/Dialect/Builder/CIRBaseBuilder.h
index b6dd4ee..b26e558 100644
--- a/clang/include/clang/CIR/Dialect/Builder/CIRBaseBuilder.h
+++ b/clang/include/clang/CIR/Dialect/Builder/CIRBaseBuilder.h
@@ -447,6 +447,10 @@ public:
     return create<cir::CmpOp>(loc, getBoolTy(), kind, lhs, rhs);
   }
 
+  mlir::Value createIsNaN(mlir::Location loc, mlir::Value operand) {
+    return createCompare(loc, cir::CmpOpKind::ne, operand, operand);
+  }
+
   mlir::Value createShift(mlir::Location loc, mlir::Value lhs, mlir::Value rhs,
                           bool isShiftLeft) {
     return create<cir::ShiftOp>(loc, lhs.getType(), lhs, rhs, isShiftLeft);
diff --git a/clang/include/clang/CIR/Dialect/IR/CIROps.td b/clang/include/clang/CIR/Dialect/IR/CIROps.td
index 8e16bf8..5ef5b60 100644
--- a/clang/include/clang/CIR/Dialect/IR/CIROps.td
+++ b/clang/include/clang/CIR/Dialect/IR/CIROps.td
@@ -2823,6 +2823,53 @@ def CIR_ComplexSubOp : CIR_Op<"complex.sub", [
   }];
 }
 
+//===----------------------------------------------------------------------===//
+// ComplexMulOp
+//===----------------------------------------------------------------------===//
+
+def CIR_ComplexRangeKind : CIR_I32EnumAttr<
+  "ComplexRangeKind", "complex multiplication and division implementation", [
+    I32EnumAttrCase<"Full", 0, "full">,
+    I32EnumAttrCase<"Improved", 1, "improved">,
+    I32EnumAttrCase<"Promoted", 2, "promoted">,
+    I32EnumAttrCase<"Basic", 3, "basic">,
+]>;
+
+def CIR_ComplexMulOp : CIR_Op<"complex.mul", [
+  Pure, SameOperandsAndResultType
+]> {
+  let summary = "Complex multiplication";
+  let description = [{
+    The `cir.complex.mul` operation takes two complex numbers and returns
+    their product.
+
+    Range is used to select the implementation used when the operation
+    is lowered to the LLVM dialect. For multiplication, 'improved',
+    'promoted', and 'basic' are all handled equivalently, producing the
+    algebraic formula with no special handling for NaN value. If 'full' is
+    used, a runtime-library function is called if one of the intermediate
+    calculations produced a NaN value.
+
+    Example:
+
+    ```mlir
+    %2 = cir.complex.mul %0, %1 range(basic) : !cir.complex<!cir.float>
+    %2 = cir.complex.mul %0, %1 range(full) : !cir.complex<!cir.float>
+    ```
+  }];
+
+  let arguments = (ins
+    CIR_ComplexType:$lhs,
+    CIR_ComplexType:$rhs,
+    CIR_ComplexRangeKind:$range
+  );
+
+  let results = (outs CIR_ComplexType:$result);
+
+  let assemblyFormat = [{
+    $lhs `,` $rhs `range` `(` $range `)` `:` qualified(type($result)) attr-dict
+  }];
+}
 
 //===----------------------------------------------------------------------===//
 // Bit Manipulation Operations
diff --git a/clang/include/clang/Frontend/ASTUnit.h b/clang/include/clang/Frontend/ASTUnit.h
index 1286fe4..7dd9aef 100644
--- a/clang/include/clang/Frontend/ASTUnit.h
+++ b/clang/include/clang/Frontend/ASTUnit.h
@@ -445,6 +445,9 @@ public:
 
   const DiagnosticsEngine &getDiagnostics() const { return *Diagnostics; }
   DiagnosticsEngine &getDiagnostics() { return *Diagnostics; }
+  llvm::IntrusiveRefCntPtr<DiagnosticsEngine> getDiagnosticsPtr() {
+    return Diagnostics;
+  }
 
   const SourceManager &getSourceManager() const { return *SourceMgr; }
   SourceManager &getSourceManager() { return *SourceMgr; }
@@ -918,8 +921,9 @@ public:
                     bool IncludeCodePatterns, bool IncludeBriefComments,
                     CodeCompleteConsumer &Consumer,
                     std::shared_ptr<PCHContainerOperations> PCHContainerOps,
-                    DiagnosticsEngine &Diag, LangOptions &LangOpts,
-                    SourceManager &SourceMgr, FileManager &FileMgr,
+                    llvm::IntrusiveRefCntPtr<DiagnosticsEngine> Diag,
+                    LangOptions &LangOpts, SourceManager &SourceMgr,
+                    FileManager &FileMgr,
                     SmallVectorImpl<StoredDiagnostic> &StoredDiagnostics,
                     SmallVectorImpl<const llvm::MemoryBuffer *> &OwnedBuffers,
                     std::unique_ptr<SyntaxOnlyAction> Act);
diff --git a/clang/include/clang/Frontend/CompilerInstance.h b/clang/include/clang/Frontend/CompilerInstance.h
index 02dd16c..a24decd 100644
--- a/clang/include/clang/Frontend/CompilerInstance.h
+++ b/clang/include/clang/Frontend/CompilerInstance.h
@@ -361,7 +361,7 @@ public:
   }
 
   /// setDiagnostics - Replace the current diagnostics engine.
-  void setDiagnostics(DiagnosticsEngine *Value);
+  void setDiagnostics(llvm::IntrusiveRefCntPtr<DiagnosticsEngine> Value);
 
   DiagnosticConsumer &getDiagnosticClient() const {
     assert(Diagnostics && Diagnostics->getClient() &&
diff --git a/clang/include/clang/Frontend/PrecompiledPreamble.h b/clang/include/clang/Frontend/PrecompiledPreamble.h
index 624df00..565395b 100644
--- a/clang/include/clang/Frontend/PrecompiledPreamble.h
+++ b/clang/include/clang/Frontend/PrecompiledPreamble.h
@@ -84,7 +84,7 @@ public:
   static llvm::ErrorOr<PrecompiledPreamble>
   Build(const CompilerInvocation &Invocation,
         const llvm::MemoryBuffer *MainFileBuffer, PreambleBounds Bounds,
-        DiagnosticsEngine &Diagnostics,
+        IntrusiveRefCntPtr<DiagnosticsEngine> Diagnostics,
         IntrusiveRefCntPtr<llvm::vfs::FileSystem> VFS,
         std::shared_ptr<PCHContainerOperations> PCHContainerOps,
         bool StoreInMemory, StringRef StoragePath,
diff --git a/clang/lib/Basic/Cuda.cpp b/clang/lib/Basic/Cuda.cpp
index 53b36d3..dc81b71 100644
--- a/clang/lib/Basic/Cuda.cpp
+++ b/clang/lib/Basic/Cuda.cpp
@@ -44,6 +44,7 @@ static const CudaVersionMapEntry CudaNameVersionMap[] = {
     CUDA_ENTRY(12, 5),
     CUDA_ENTRY(12, 6),
     CUDA_ENTRY(12, 8),
+    CUDA_ENTRY(12, 9),
     {"", CudaVersion::NEW, llvm::VersionTuple(std::numeric_limits<int>::max())},
     {"unknown", CudaVersion::UNKNOWN, {}} // End of list tombstone.
 };
@@ -119,6 +120,11 @@ CudaVersion MinVersionForOffloadArch(OffloadArch A) {
   case OffloadArch::SM_120:
   case OffloadArch::SM_120a:
     return CudaVersion::CUDA_128;
+  case OffloadArch::SM_103:
+  case OffloadArch::SM_103a:
+  case OffloadArch::SM_121:
+  case OffloadArch::SM_121a:
+    return CudaVersion::CUDA_129;
   default:
     llvm_unreachable("invalid enum");
   }
diff --git a/clang/lib/Basic/OffloadArch.cpp b/clang/lib/Basic/OffloadArch.cpp
index dce9ffa..4348178 100644
--- a/clang/lib/Basic/OffloadArch.cpp
+++ b/clang/lib/Basic/OffloadArch.cpp
@@ -33,8 +33,12 @@ static const OffloadArchToStringMap ArchNames[] = {
     SM(100a),                        // Blackwell
     SM(101),                         // Blackwell
     SM(101a),                        // Blackwell
+    SM(103),                         // Blackwell
+    SM(103a),                        // Blackwell
     SM(120),                         // Blackwell
     SM(120a),                        // Blackwell
+    SM(121),                         // Blackwell
+    SM(121a),                        // Blackwell
     GFX(600),  // gfx600
     GFX(601),  // gfx601
     GFX(602),  // gfx602
diff --git a/clang/lib/Basic/SourceManager.cpp b/clang/lib/Basic/SourceManager.cpp
index 5b8444a..343c26e 100644
--- a/clang/lib/Basic/SourceManager.cpp
+++ b/clang/lib/Basic/SourceManager.cpp
@@ -2379,8 +2379,8 @@ SourceManagerForFile::SourceManagerForFile(StringRef FileName,
   DiagOpts = std::make_unique<DiagnosticOptions>();
   // This is passed to `SM` as reference, so the pointer has to be referenced
   // by `Environment` due to the same reason above.
-  Diagnostics = std::make_unique<DiagnosticsEngine>(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs), *DiagOpts);
+  Diagnostics =
+      std::make_unique<DiagnosticsEngine>(DiagnosticIDs::create(), *DiagOpts);
   SourceMgr = std::make_unique<SourceManager>(*Diagnostics, *FileMgr);
   FileEntryRef FE = llvm::cantFail(FileMgr->getFileRef(FileName));
   FileID ID =
diff --git a/clang/lib/Basic/Targets/NVPTX.cpp b/clang/lib/Basic/Targets/NVPTX.cpp
index 54b39fd..79995cc 100644
--- a/clang/lib/Basic/Targets/NVPTX.cpp
+++ b/clang/lib/Basic/Targets/NVPTX.cpp
@@ -295,10 +295,16 @@ void NVPTXTargetInfo::getTargetDefines(const LangOptions &Opts,
         return "1000";
       case OffloadArch::SM_101:
       case OffloadArch::SM_101a:
-         return "1010";
+        return "1010";
+      case OffloadArch::SM_103:
+      case OffloadArch::SM_103a:
+        return "1030";
       case OffloadArch::SM_120:
       case OffloadArch::SM_120a:
-         return "1200";
+        return "1200";
+      case OffloadArch::SM_121:
+      case OffloadArch::SM_121a:
+        return "1210";
       }
       llvm_unreachable("unhandled OffloadArch");
     }();
@@ -307,7 +313,9 @@ void NVPTXTargetInfo::getTargetDefines(const LangOptions &Opts,
       case OffloadArch::SM_90a:
       case OffloadArch::SM_100a:
       case OffloadArch::SM_101a:
+      case OffloadArch::SM_103a:
       case OffloadArch::SM_120a:
+      case OffloadArch::SM_121a:
         Builder.defineMacro("__CUDA_ARCH_FEAT_SM" + CUDAArchCode.drop_back() + "_ALL", "1");
         break;
       default:
diff --git a/clang/lib/CIR/CodeGen/CIRGenExprComplex.cpp b/clang/lib/CIR/CodeGen/CIRGenExprComplex.cpp
index 18d2860..ea60a95 100644
--- a/clang/lib/CIR/CodeGen/CIRGenExprComplex.cpp
+++ b/clang/lib/CIR/CodeGen/CIRGenExprComplex.cpp
@@ -118,6 +118,7 @@ public:
 
   mlir::Value emitBinAdd(const BinOpInfo &op);
   mlir::Value emitBinSub(const BinOpInfo &op);
+  mlir::Value emitBinMul(const BinOpInfo &op);
 
   QualType getPromotionType(QualType ty, bool isDivOpCode = false) {
     if (auto *complexTy = ty->getAs<ComplexType>()) {
@@ -150,6 +151,7 @@ public:
 
   HANDLEBINOP(Add)
   HANDLEBINOP(Sub)
+  HANDLEBINOP(Mul)
 #undef HANDLEBINOP
 };
 } // namespace
@@ -577,6 +579,7 @@ mlir::Value ComplexExprEmitter::emitPromoted(const Expr *e,
     return emitBin##OP(emitBinOps(bo, promotionTy));
       HANDLE_BINOP(Add)
       HANDLE_BINOP(Sub)
+      HANDLE_BINOP(Mul)
 #undef HANDLE_BINOP
     default:
       break;
@@ -636,6 +639,31 @@ mlir::Value ComplexExprEmitter::emitBinSub(const BinOpInfo &op) {
   return builder.create<cir::ComplexSubOp>(op.loc, op.lhs, op.rhs);
 }
 
+static cir::ComplexRangeKind
+getComplexRangeAttr(LangOptions::ComplexRangeKind range) {
+  switch (range) {
+  case LangOptions::CX_Full:
+    return cir::ComplexRangeKind::Full;
+  case LangOptions::CX_Improved:
+    return cir::ComplexRangeKind::Improved;
+  case LangOptions::CX_Promoted:
+    return cir::ComplexRangeKind::Promoted;
+  case LangOptions::CX_Basic:
+    return cir::ComplexRangeKind::Basic;
+  case LangOptions::CX_None:
+    // The default value for ComplexRangeKind is Full is no option is selected
+    return cir::ComplexRangeKind::Full;
+  }
+}
+
+mlir::Value ComplexExprEmitter::emitBinMul(const BinOpInfo &op) {
+  assert(!cir::MissingFeatures::fastMathFlags());
+  assert(!cir::MissingFeatures::cgFPOptionsRAII());
+  cir::ComplexRangeKind rangeKind =
+      getComplexRangeAttr(op.fpFeatures.getComplexRange());
+  return builder.create<cir::ComplexMulOp>(op.loc, op.lhs, op.rhs, rangeKind);
+}
+
 LValue CIRGenFunction::emitComplexAssignmentLValue(const BinaryOperator *e) {
   assert(e->getOpcode() == BO_Assign && "Expected assign op");
 
diff --git a/clang/lib/CIR/CodeGen/CIRGenFunction.cpp b/clang/lib/CIR/CodeGen/CIRGenFunction.cpp
index c65d025..3e9de17 100644
--- a/clang/lib/CIR/CodeGen/CIRGenFunction.cpp
+++ b/clang/lib/CIR/CodeGen/CIRGenFunction.cpp
@@ -943,6 +943,7 @@ void CIRGenFunction::emitVariablyModifiedType(QualType type) {
     case Type::HLSLInlineSpirv:
     case Type::PredefinedSugar:
       cgm.errorNYI("CIRGenFunction::emitVariablyModifiedType");
+      break;
 
 #define TYPE(Class, Base)
 #define ABSTRACT_TYPE(Class, Base)
diff --git a/clang/lib/CIR/Dialect/Transforms/LoweringPrepare.cpp b/clang/lib/CIR/Dialect/Transforms/LoweringPrepare.cpp
index 670ddaf..66260eb 100644
--- a/clang/lib/CIR/Dialect/Transforms/LoweringPrepare.cpp
+++ b/clang/lib/CIR/Dialect/Transforms/LoweringPrepare.cpp
@@ -15,7 +15,6 @@
 #include "clang/CIR/Dialect/Passes.h"
 #include "clang/CIR/MissingFeatures.h"
 
-#include <iostream>
 #include <memory>
 
 using namespace mlir;
@@ -28,21 +27,47 @@ struct LoweringPreparePass : public LoweringPrepareBase<LoweringPreparePass> {
 
   void runOnOp(mlir::Operation *op);
   void lowerCastOp(cir::CastOp op);
+  void lowerComplexMulOp(cir::ComplexMulOp op);
   void lowerUnaryOp(cir::UnaryOp op);
   void lowerArrayDtor(cir::ArrayDtor op);
   void lowerArrayCtor(cir::ArrayCtor op);
 
+  cir::FuncOp buildRuntimeFunction(
+      mlir::OpBuilder &builder, llvm::StringRef name, mlir::Location loc,
+      cir::FuncType type,
+      cir::GlobalLinkageKind linkage = cir::GlobalLinkageKind::ExternalLinkage);
+
   ///
   /// AST related
   /// -----------
 
   clang::ASTContext *astCtx;
 
+  /// Tracks current module.
+  mlir::ModuleOp mlirModule;
+
   void setASTContext(clang::ASTContext *c) { astCtx = c; }
 };
 
 } // namespace
 
+cir::FuncOp LoweringPreparePass::buildRuntimeFunction(
+    mlir::OpBuilder &builder, llvm::StringRef name, mlir::Location loc,
+    cir::FuncType type, cir::GlobalLinkageKind linkage) {
+  cir::FuncOp f = dyn_cast_or_null<FuncOp>(SymbolTable::lookupNearestSymbolFrom(
+      mlirModule, StringAttr::get(mlirModule->getContext(), name)));
+  if (!f) {
+    f = builder.create<cir::FuncOp>(loc, name, type);
+    f.setLinkageAttr(
+        cir::GlobalLinkageKindAttr::get(builder.getContext(), linkage));
+    mlir::SymbolTable::setSymbolVisibility(
+        f, mlir::SymbolTable::Visibility::Private);
+
+    assert(!cir::MissingFeatures::opFuncExtraAttrs());
+  }
+  return f;
+}
+
 static mlir::Value lowerScalarToComplexCast(mlir::MLIRContext &ctx,
                                             cir::CastOp op) {
   cir::CIRBaseBuilderTy builder(ctx);
@@ -128,6 +153,124 @@ void LoweringPreparePass::lowerCastOp(cir::CastOp op) {
   }
 }
 
+static mlir::Value buildComplexBinOpLibCall(
+    LoweringPreparePass &pass, CIRBaseBuilderTy &builder,
+    llvm::StringRef (*libFuncNameGetter)(llvm::APFloat::Semantics),
+    mlir::Location loc, cir::ComplexType ty, mlir::Value lhsReal,
+    mlir::Value lhsImag, mlir::Value rhsReal, mlir::Value rhsImag) {
+  cir::FPTypeInterface elementTy =
+      mlir::cast<cir::FPTypeInterface>(ty.getElementType());
+
+  llvm::StringRef libFuncName = libFuncNameGetter(
+      llvm::APFloat::SemanticsToEnum(elementTy.getFloatSemantics()));
+  llvm::SmallVector<mlir::Type, 4> libFuncInputTypes(4, elementTy);
+
+  cir::FuncType libFuncTy = cir::FuncType::get(libFuncInputTypes, ty);
+
+  // Insert a declaration for the runtime function to be used in Complex
+  // multiplication and division when needed
+  cir::FuncOp libFunc;
+  {
+    mlir::OpBuilder::InsertionGuard ipGuard{builder};
+    builder.setInsertionPointToStart(pass.mlirModule.getBody());
+    libFunc = pass.buildRuntimeFunction(builder, libFuncName, loc, libFuncTy);
+  }
+
+  cir::CallOp call =
+      builder.createCallOp(loc, libFunc, {lhsReal, lhsImag, rhsReal, rhsImag});
+  return call.getResult();
+}
+
+static llvm::StringRef
+getComplexMulLibCallName(llvm::APFloat::Semantics semantics) {
+  switch (semantics) {
+  case llvm::APFloat::S_IEEEhalf:
+    return "__mulhc3";
+  case llvm::APFloat::S_IEEEsingle:
+    return "__mulsc3";
+  case llvm::APFloat::S_IEEEdouble:
+    return "__muldc3";
+  case llvm::APFloat::S_PPCDoubleDouble:
+    return "__multc3";
+  case llvm::APFloat::S_x87DoubleExtended:
+    return "__mulxc3";
+  case llvm::APFloat::S_IEEEquad:
+    return "__multc3";
+  default:
+    llvm_unreachable("unsupported floating point type");
+  }
+}
+
+static mlir::Value lowerComplexMul(LoweringPreparePass &pass,
+                                   CIRBaseBuilderTy &builder,
+                                   mlir::Location loc, cir::ComplexMulOp op,
+                                   mlir::Value lhsReal, mlir::Value lhsImag,
+                                   mlir::Value rhsReal, mlir::Value rhsImag) {
+  // (a+bi) * (c+di) = (ac-bd) + (ad+bc)i
+  mlir::Value resultRealLhs =
+      builder.createBinop(loc, lhsReal, cir::BinOpKind::Mul, rhsReal);
+  mlir::Value resultRealRhs =
+      builder.createBinop(loc, lhsImag, cir::BinOpKind::Mul, rhsImag);
+  mlir::Value resultImagLhs =
+      builder.createBinop(loc, lhsReal, cir::BinOpKind::Mul, rhsImag);
+  mlir::Value resultImagRhs =
+      builder.createBinop(loc, lhsImag, cir::BinOpKind::Mul, rhsReal);
+  mlir::Value resultReal = builder.createBinop(
+      loc, resultRealLhs, cir::BinOpKind::Sub, resultRealRhs);
+  mlir::Value resultImag = builder.createBinop(
+      loc, resultImagLhs, cir::BinOpKind::Add, resultImagRhs);
+  mlir::Value algebraicResult =
+      builder.createComplexCreate(loc, resultReal, resultImag);
+
+  cir::ComplexType complexTy = op.getType();
+  cir::ComplexRangeKind rangeKind = op.getRange();
+  if (mlir::isa<cir::IntType>(complexTy.getElementType()) ||
+      rangeKind == cir::ComplexRangeKind::Basic ||
+      rangeKind == cir::ComplexRangeKind::Improved ||
+      rangeKind == cir::ComplexRangeKind::Promoted)
+    return algebraicResult;
+
+  assert(!cir::MissingFeatures::fastMathFlags());
+
+  // Check whether the real part and the imaginary part of the result are both
+  // NaN. If so, emit a library call to compute the multiplication instead.
+  // We check a value against NaN by comparing the value against itself.
+  mlir::Value resultRealIsNaN = builder.createIsNaN(loc, resultReal);
+  mlir::Value resultImagIsNaN = builder.createIsNaN(loc, resultImag);
+  mlir::Value resultRealAndImagAreNaN =
+      builder.createLogicalAnd(loc, resultRealIsNaN, resultImagIsNaN);
+
+  return builder
+      .create<cir::TernaryOp>(
+          loc, resultRealAndImagAreNaN,
+          [&](mlir::OpBuilder &, mlir::Location) {
+            mlir::Value libCallResult = buildComplexBinOpLibCall(
+                pass, builder, &getComplexMulLibCallName, loc, complexTy,
+                lhsReal, lhsImag, rhsReal, rhsImag);
+            builder.createYield(loc, libCallResult);
+          },
+          [&](mlir::OpBuilder &, mlir::Location) {
+            builder.createYield(loc, algebraicResult);
+          })
+      .getResult();
+}
+
+void LoweringPreparePass::lowerComplexMulOp(cir::ComplexMulOp op) {
+  cir::CIRBaseBuilderTy builder(getContext());
+  builder.setInsertionPointAfter(op);
+  mlir::Location loc = op.getLoc();
+  mlir::TypedValue<cir::ComplexType> lhs = op.getLhs();
+  mlir::TypedValue<cir::ComplexType> rhs = op.getRhs();
+  mlir::Value lhsReal = builder.createComplexReal(loc, lhs);
+  mlir::Value lhsImag = builder.createComplexImag(loc, lhs);
+  mlir::Value rhsReal = builder.createComplexReal(loc, rhs);
+  mlir::Value rhsImag = builder.createComplexImag(loc, rhs);
+  mlir::Value loweredResult = lowerComplexMul(*this, builder, loc, op, lhsReal,
+                                              lhsImag, rhsReal, rhsImag);
+  op.replaceAllUsesWith(loweredResult);
+  op.erase();
+}
+
 void LoweringPreparePass::lowerUnaryOp(cir::UnaryOp op) {
   mlir::Type ty = op.getType();
   if (!mlir::isa<cir::ComplexType>(ty))
@@ -269,18 +412,22 @@ void LoweringPreparePass::runOnOp(mlir::Operation *op) {
     lowerArrayDtor(arrayDtor);
   else if (auto cast = mlir::dyn_cast<cir::CastOp>(op))
     lowerCastOp(cast);
+  else if (auto complexMul = mlir::dyn_cast<cir::ComplexMulOp>(op))
+    lowerComplexMulOp(complexMul);
   else if (auto unary = mlir::dyn_cast<cir::UnaryOp>(op))
     lowerUnaryOp(unary);
 }
 
 void LoweringPreparePass::runOnOperation() {
   mlir::Operation *op = getOperation();
+  if (isa<::mlir::ModuleOp>(op))
+    mlirModule = cast<::mlir::ModuleOp>(op);
 
   llvm::SmallVector<mlir::Operation *> opsToTransform;
 
   op->walk([&](mlir::Operation *op) {
-    if (mlir::isa<cir::ArrayCtor, cir::ArrayDtor, cir::CastOp, cir::UnaryOp>(
-            op))
+    if (mlir::isa<cir::ArrayCtor, cir::ArrayDtor, cir::CastOp,
+                  cir::ComplexMulOp, cir::UnaryOp>(op))
       opsToTransform.push_back(op);
   });
 
diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
index e25b694..04c9192 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
@@ -2278,8 +2278,12 @@ void CGOpenMPRuntimeGPU::processRequiresDirective(const OMPRequiresDecl *D) {
       case OffloadArch::SM_100a:
       case OffloadArch::SM_101:
       case OffloadArch::SM_101a:
+      case OffloadArch::SM_103:
+      case OffloadArch::SM_103a:
       case OffloadArch::SM_120:
       case OffloadArch::SM_120a:
+      case OffloadArch::SM_121:
+      case OffloadArch::SM_121a:
       case OffloadArch::GFX600:
       case OffloadArch::GFX601:
       case OffloadArch::GFX602:
diff --git a/clang/lib/CrossTU/CrossTranslationUnit.cpp b/clang/lib/CrossTU/CrossTranslationUnit.cpp
index 6d0f042..fb2a79a 100644
--- a/clang/lib/CrossTU/CrossTranslationUnit.cpp
+++ b/clang/lib/CrossTU/CrossTranslationUnit.cpp
@@ -563,9 +563,8 @@ CrossTranslationUnitContext::ASTLoader::loadFromDump(StringRef ASTDumpPath) {
   auto DiagOpts = std::make_shared<DiagnosticOptions>();
   TextDiagnosticPrinter *DiagClient =
       new TextDiagnosticPrinter(llvm::errs(), *DiagOpts);
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-  IntrusiveRefCntPtr<DiagnosticsEngine> Diags(
-      new DiagnosticsEngine(DiagID, *DiagOpts, DiagClient));
+  auto Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
+      DiagnosticIDs::create(), *DiagOpts, DiagClient);
   return ASTUnit::LoadFromASTFile(
       ASTDumpPath, CI.getPCHContainerOperations()->getRawReader(),
       ASTUnit::LoadEverything, DiagOpts, Diags, CI.getFileSystemOpts(),
@@ -607,8 +606,8 @@ CrossTranslationUnitContext::ASTLoader::loadFromSource(
   auto *DiagClient = new ForwardingDiagnosticConsumer{CI.getDiagnosticClient()};
   IntrusiveRefCntPtr<DiagnosticIDs> DiagID{
       CI.getDiagnostics().getDiagnosticIDs()};
-  IntrusiveRefCntPtr<DiagnosticsEngine> Diags(
-      new DiagnosticsEngine{DiagID, *DiagOpts, DiagClient});
+  auto Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(DiagID, *DiagOpts,
+                                                            DiagClient);
 
   return ASTUnit::LoadFromCommandLine(
       CommandLineArgs.begin(), (CommandLineArgs.end()),
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index e15ea50..d0e254f 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -226,17 +226,19 @@ static bool ShouldEnableAutolink(const ArgList &Args, const ToolChain &TC,
 static const char *addDebugCompDirArg(const ArgList &Args,
                                       ArgStringList &CmdArgs,
                                       const llvm::vfs::FileSystem &VFS) {
+  std::string DebugCompDir;
   if (Arg *A = Args.getLastArg(options::OPT_ffile_compilation_dir_EQ,
-                               options::OPT_fdebug_compilation_dir_EQ)) {
-    if (A->getOption().matches(options::OPT_ffile_compilation_dir_EQ))
-      CmdArgs.push_back(Args.MakeArgString(Twine("-fdebug-compilation-dir=") +
-                                           A->getValue()));
+                               options::OPT_fdebug_compilation_dir_EQ))
+    DebugCompDir = A->getValue();
+
+  if (DebugCompDir.empty()) {
+    if (llvm::ErrorOr<std::string> CWD = VFS.getCurrentWorkingDirectory())
+      DebugCompDir = std::move(*CWD);
     else
-      A->render(Args, CmdArgs);
-  } else if (llvm::ErrorOr<std::string> CWD =
-                 VFS.getCurrentWorkingDirectory()) {
-    CmdArgs.push_back(Args.MakeArgString("-fdebug-compilation-dir=" + *CWD));
+      return nullptr;
   }
+  CmdArgs.push_back(
+      Args.MakeArgString("-fdebug-compilation-dir=" + DebugCompDir));
   StringRef Path(CmdArgs.back());
   return Path.substr(Path.find('=') + 1).data();
 }
@@ -525,17 +527,17 @@ static void addPGOAndCoverageFlags(const ToolChain &TC, Compilation &C,
     CmdArgs.push_back("-fcoverage-mcdc");
   }
 
+  StringRef CoverageCompDir;
   if (Arg *A = Args.getLastArg(options::OPT_ffile_compilation_dir_EQ,
-                               options::OPT_fcoverage_compilation_dir_EQ)) {
-    if (A->getOption().matches(options::OPT_ffile_compilation_dir_EQ))
-      CmdArgs.push_back(Args.MakeArgString(
-          Twine("-fcoverage-compilation-dir=") + A->getValue()));
-    else
-      A->render(Args, CmdArgs);
-  } else if (llvm::ErrorOr<std::string> CWD =
-                 D.getVFS().getCurrentWorkingDirectory()) {
-    CmdArgs.push_back(Args.MakeArgString("-fcoverage-compilation-dir=" + *CWD));
-  }
+                               options::OPT_fcoverage_compilation_dir_EQ))
+    CoverageCompDir = A->getValue();
+  if (CoverageCompDir.empty()) {
+    if (auto CWD = D.getVFS().getCurrentWorkingDirectory())
+      CmdArgs.push_back(
+          Args.MakeArgString(Twine("-fcoverage-compilation-dir=") + *CWD));
+  } else
+    CmdArgs.push_back(Args.MakeArgString(Twine("-fcoverage-compilation-dir=") +
+                                         CoverageCompDir));
 
   if (Args.hasArg(options::OPT_fprofile_exclude_files_EQ)) {
     auto *Arg = Args.getLastArg(options::OPT_fprofile_exclude_files_EQ);
diff --git a/clang/lib/Driver/ToolChains/Cuda.cpp b/clang/lib/Driver/ToolChains/Cuda.cpp
index 7d803be..1f0b478 100644
--- a/clang/lib/Driver/ToolChains/Cuda.cpp
+++ b/clang/lib/Driver/ToolChains/Cuda.cpp
@@ -88,6 +88,8 @@ CudaVersion getCudaVersion(uint32_t raw_version) {
     return CudaVersion::CUDA_126;
   if (raw_version < 12090)
     return CudaVersion::CUDA_128;
+  if (raw_version < 13000)
+    return CudaVersion::CUDA_129;
   return CudaVersion::NEW;
 }
 
@@ -683,6 +685,7 @@ void NVPTX::getNVPTXTargetFeatures(const Driver &D, const llvm::Triple &Triple,
   case CudaVersion::CUDA_##CUDA_VER:                                           \
     PtxFeature = "+ptx" #PTX_VER;                                              \
     break;
+    CASE_CUDA_VERSION(129, 88);
     CASE_CUDA_VERSION(128, 87);
     CASE_CUDA_VERSION(126, 85);
     CASE_CUDA_VERSION(125, 85);
diff --git a/clang/lib/Frontend/ASTMerge.cpp b/clang/lib/Frontend/ASTMerge.cpp
index a4ce883..10c10458 100644
--- a/clang/lib/Frontend/ASTMerge.cpp
+++ b/clang/lib/Frontend/ASTMerge.cpp
@@ -41,10 +41,10 @@ void ASTMergeAction::ExecuteAction() {
   auto SharedState = std::make_shared<ASTImporterSharedState>(
       *CI.getASTContext().getTranslationUnitDecl());
   for (unsigned I = 0, N = ASTFiles.size(); I != N; ++I) {
-    IntrusiveRefCntPtr<DiagnosticsEngine> Diags(new DiagnosticsEngine(
+    auto Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
         DiagIDs, CI.getDiagnosticOpts(),
         new ForwardingDiagnosticConsumer(*CI.getDiagnostics().getClient()),
-        /*ShouldOwnClient=*/true));
+        /*ShouldOwnClient=*/true);
     std::unique_ptr<ASTUnit> Unit = ASTUnit::LoadFromASTFile(
         ASTFiles[I], CI.getPCHContainerReader(), ASTUnit::LoadEverything,
         nullptr, Diags, CI.getFileSystemOpts(), CI.getHeaderSearchOpts());
diff --git a/clang/lib/Frontend/ASTUnit.cpp b/clang/lib/Frontend/ASTUnit.cpp
index 09caf85..5711f45 100644
--- a/clang/lib/Frontend/ASTUnit.cpp
+++ b/clang/lib/Frontend/ASTUnit.cpp
@@ -1204,7 +1204,7 @@ bool ASTUnit::Parse(std::shared_ptr<PCHContainerOperations> PCHContainerOps,
 
   // Set up diagnostics, capturing any diagnostics that would
   // otherwise be dropped.
-  Clang->setDiagnostics(&getDiagnostics());
+  Clang->setDiagnostics(getDiagnosticsPtr());
 
   // Create the target instance.
   if (!Clang->createTarget())
@@ -1424,7 +1424,7 @@ ASTUnit::getMainBufferWithPrecompiledPreamble(
       PreambleInvocationIn.getFrontendOpts().SkipFunctionBodies = true;
 
     llvm::ErrorOr<PrecompiledPreamble> NewPreamble = PrecompiledPreamble::Build(
-        PreambleInvocationIn, MainFileBuffer.get(), Bounds, *Diagnostics, VFS,
+        PreambleInvocationIn, MainFileBuffer.get(), Bounds, Diagnostics, VFS,
         PCHContainerOps, StorePreamblesInMemory, PreambleStoragePath,
         Callbacks);
 
@@ -1624,7 +1624,7 @@ ASTUnit *ASTUnit::LoadFromCompilerInvocationAction(
 
   // Set up diagnostics, capturing any diagnostics that would
   // otherwise be dropped.
-  Clang->setDiagnostics(&AST->getDiagnostics());
+  Clang->setDiagnostics(AST->getDiagnosticsPtr());
 
   // Create the target instance.
   if (!Clang->createTarget())
@@ -2209,8 +2209,9 @@ void ASTUnit::CodeComplete(
     bool IncludeCodePatterns, bool IncludeBriefComments,
     CodeCompleteConsumer &Consumer,
     std::shared_ptr<PCHContainerOperations> PCHContainerOps,
-    DiagnosticsEngine &Diag, LangOptions &LangOpts, SourceManager &SourceMgr,
-    FileManager &FileMgr, SmallVectorImpl<StoredDiagnostic> &StoredDiagnostics,
+    llvm::IntrusiveRefCntPtr<DiagnosticsEngine> Diag, LangOptions &LangOpts,
+    SourceManager &SourceMgr, FileManager &FileMgr,
+    SmallVectorImpl<StoredDiagnostic> &StoredDiagnostics,
     SmallVectorImpl<const llvm::MemoryBuffer *> &OwnedBuffers,
     std::unique_ptr<SyntaxOnlyAction> Act) {
   if (!Invocation)
@@ -2259,11 +2260,11 @@ void ASTUnit::CodeComplete(
       std::string(Clang->getFrontendOpts().Inputs[0].getFile());
 
   // Set up diagnostics, capturing any diagnostics produced.
-  Clang->setDiagnostics(&Diag);
+  Clang->setDiagnostics(Diag);
   CaptureDroppedDiagnostics Capture(CaptureDiagsKind::All,
                                     Clang->getDiagnostics(),
                                     &StoredDiagnostics, nullptr);
-  ProcessWarningOptions(Diag, Inv.getDiagnosticOpts(),
+  ProcessWarningOptions(*Diag, Inv.getDiagnosticOpts(),
                         FileMgr.getVirtualFileSystem());
 
   // Create the target instance.
diff --git a/clang/lib/Frontend/ChainedIncludesSource.cpp b/clang/lib/Frontend/ChainedIncludesSource.cpp
index ba7c767..88b1076 100644
--- a/clang/lib/Frontend/ChainedIncludesSource.cpp
+++ b/clang/lib/Frontend/ChainedIncludesSource.cpp
@@ -118,13 +118,12 @@ IntrusiveRefCntPtr<ExternalSemaSource> clang::createChainedIncludesSource(
 
     TextDiagnosticPrinter *DiagClient =
         new TextDiagnosticPrinter(llvm::errs(), CI.getDiagnosticOpts());
-    IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-    IntrusiveRefCntPtr<DiagnosticsEngine> Diags(
-        new DiagnosticsEngine(DiagID, CI.getDiagnosticOpts(), DiagClient));
+    auto Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
+        DiagnosticIDs::create(), CI.getDiagnosticOpts(), DiagClient);
 
     auto Clang = std::make_unique<CompilerInstance>(
         std::move(CInvok), CI.getPCHContainerOperations());
-    Clang->setDiagnostics(Diags.get());
+    Clang->setDiagnostics(Diags);
     Clang->setTarget(TargetInfo::CreateTargetInfo(
         Clang->getDiagnostics(), Clang->getInvocation().getTargetOpts()));
     Clang->createFileManager();
diff --git a/clang/lib/Frontend/CompilerInstance.cpp b/clang/lib/Frontend/CompilerInstance.cpp
index 40fb070..ed6a651 100644
--- a/clang/lib/Frontend/CompilerInstance.cpp
+++ b/clang/lib/Frontend/CompilerInstance.cpp
@@ -87,8 +87,9 @@ bool CompilerInstance::shouldBuildGlobalModuleIndex() const {
          !DisableGeneratingGlobalModuleIndex;
 }
 
-void CompilerInstance::setDiagnostics(DiagnosticsEngine *Value) {
-  Diagnostics = Value;
+void CompilerInstance::setDiagnostics(
+    llvm::IntrusiveRefCntPtr<DiagnosticsEngine> Value) {
+  Diagnostics = std::move(Value);
 }
 
 void CompilerInstance::setVerboseOutputStream(raw_ostream &Value) {
@@ -345,9 +346,8 @@ IntrusiveRefCntPtr<DiagnosticsEngine> CompilerInstance::createDiagnostics(
     llvm::vfs::FileSystem &VFS, DiagnosticOptions &Opts,
     DiagnosticConsumer *Client, bool ShouldOwnClient,
     const CodeGenOptions *CodeGenOpts) {
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-  IntrusiveRefCntPtr<DiagnosticsEngine> Diags(
-      new DiagnosticsEngine(DiagID, Opts));
+  auto Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
+      DiagnosticIDs::create(), Opts);
 
   // Create the diagnostic client for reporting errors or for
   // implementing -verify.
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index ab4384a..9f77e62 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -826,7 +826,7 @@ static bool RoundTrip(ParseFn Parse, GenerateFn Generate,
 
   // Setup a dummy DiagnosticsEngine.
   DiagnosticOptions DummyDiagOpts;
-  DiagnosticsEngine DummyDiags(new DiagnosticIDs(), DummyDiagOpts);
+  DiagnosticsEngine DummyDiags(DiagnosticIDs::create(), DummyDiagOpts);
   DummyDiags.setClient(new TextDiagnosticBuffer());
 
   // Run the first parse on the original arguments with the dummy invocation and
@@ -2685,7 +2685,7 @@ bool clang::ParseDiagnosticArgs(DiagnosticOptions &Opts, ArgList &Args,
   std::optional<DiagnosticsEngine> IgnoringDiags;
   if (!Diags) {
     IgnoringDiagOpts.emplace();
-    IgnoringDiags.emplace(new DiagnosticIDs(), *IgnoringDiagOpts,
+    IgnoringDiags.emplace(DiagnosticIDs::create(), *IgnoringDiagOpts,
                           new IgnoringDiagConsumer());
     Diags = &*IgnoringDiags;
   }
diff --git a/clang/lib/Frontend/FrontendAction.cpp b/clang/lib/Frontend/FrontendAction.cpp
index 1d82fc7..137d225 100644
--- a/clang/lib/Frontend/FrontendAction.cpp
+++ b/clang/lib/Frontend/FrontendAction.cpp
@@ -763,11 +763,11 @@ bool FrontendAction::BeginSourceFile(CompilerInstance &CI,
   // If we're replaying the build of an AST file, import it and set up
   // the initial state from its build.
   if (ReplayASTFile) {
-    IntrusiveRefCntPtr<DiagnosticsEngine> Diags(&CI.getDiagnostics());
+    IntrusiveRefCntPtr<DiagnosticsEngine> Diags = CI.getDiagnosticsPtr();
 
     // The AST unit populates its own diagnostics engine rather than ours.
-    IntrusiveRefCntPtr<DiagnosticsEngine> ASTDiags(new DiagnosticsEngine(
-        Diags->getDiagnosticIDs(), Diags->getDiagnosticOptions()));
+    auto ASTDiags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
+        Diags->getDiagnosticIDs(), Diags->getDiagnosticOptions());
     ASTDiags->setClient(Diags->getClient(), /*OwnsClient*/false);
 
     // FIXME: What if the input is a memory buffer?
@@ -835,7 +835,7 @@ bool FrontendAction::BeginSourceFile(CompilerInstance &CI,
     assert(hasASTFileSupport() &&
            "This action does not have AST file support!");
 
-    IntrusiveRefCntPtr<DiagnosticsEngine> Diags(&CI.getDiagnostics());
+    IntrusiveRefCntPtr<DiagnosticsEngine> Diags = CI.getDiagnosticsPtr();
 
     // FIXME: What if the input is a memory buffer?
     StringRef InputFile = Input.getFile();
diff --git a/clang/lib/Frontend/PrecompiledPreamble.cpp b/clang/lib/Frontend/PrecompiledPreamble.cpp
index 486cd95..7fc1d87 100644
--- a/clang/lib/Frontend/PrecompiledPreamble.cpp
+++ b/clang/lib/Frontend/PrecompiledPreamble.cpp
@@ -412,7 +412,7 @@ PrecompiledPreamble::operator=(PrecompiledPreamble &&) = default;
 llvm::ErrorOr<PrecompiledPreamble> PrecompiledPreamble::Build(
     const CompilerInvocation &Invocation,
     const llvm::MemoryBuffer *MainFileBuffer, PreambleBounds Bounds,
-    DiagnosticsEngine &Diagnostics,
+    IntrusiveRefCntPtr<DiagnosticsEngine> Diagnostics,
     IntrusiveRefCntPtr<llvm::vfs::FileSystem> VFS,
     std::shared_ptr<PCHContainerOperations> PCHContainerOps, bool StoreInMemory,
     StringRef StoragePath, PreambleCallbacks &Callbacks) {
@@ -461,7 +461,7 @@ llvm::ErrorOr<PrecompiledPreamble> PrecompiledPreamble::Build(
   llvm::CrashRecoveryContextCleanupRegistrar<CompilerInstance> CICleanup(
       Clang.get());
 
-  Clang->setDiagnostics(&Diagnostics);
+  Clang->setDiagnostics(Diagnostics);
 
   // Create the target instance.
   if (!Clang->createTarget())
@@ -476,18 +476,18 @@ llvm::ErrorOr<PrecompiledPreamble> PrecompiledPreamble::Build(
   }
 
   // Clear out old caches and data.
-  Diagnostics.Reset();
-  ProcessWarningOptions(Diagnostics, Clang->getDiagnosticOpts(), *VFS);
+  Diagnostics->Reset();
+  ProcessWarningOptions(*Diagnostics, Clang->getDiagnosticOpts(), *VFS);
 
-  VFS =
-      createVFSFromCompilerInvocation(Clang->getInvocation(), Diagnostics, VFS);
+  VFS = createVFSFromCompilerInvocation(Clang->getInvocation(), *Diagnostics,
+                                        VFS);
 
   // Create a file manager object to provide access to and cache the filesystem.
   Clang->setFileManager(new FileManager(Clang->getFileSystemOpts(), VFS));
 
   // Create the source manager.
   Clang->setSourceManager(
-      new SourceManager(Diagnostics, Clang->getFileManager()));
+      new SourceManager(*Diagnostics, Clang->getFileManager()));
 
   auto PreambleDepCollector = std::make_shared<PreambleDependencyCollector>();
   Clang->addDependencyCollector(PreambleDepCollector);
diff --git a/clang/lib/Frontend/SerializedDiagnosticPrinter.cpp b/clang/lib/Frontend/SerializedDiagnosticPrinter.cpp
index 23a1f90..1f5bb47 100644
--- a/clang/lib/Frontend/SerializedDiagnosticPrinter.cpp
+++ b/clang/lib/Frontend/SerializedDiagnosticPrinter.cpp
@@ -753,10 +753,9 @@ DiagnosticsEngine *SDiagsWriter::getMetaDiags() {
   //    to be distinct from the engine the writer was being added to and would
   //    normally not be used.
   if (!State->MetaDiagnostics) {
-    IntrusiveRefCntPtr<DiagnosticIDs> IDs(new DiagnosticIDs());
     auto Client = new TextDiagnosticPrinter(llvm::errs(), State->DiagOpts);
-    State->MetaDiagnostics =
-        std::make_unique<DiagnosticsEngine>(IDs, State->DiagOpts, Client);
+    State->MetaDiagnostics = std::make_unique<DiagnosticsEngine>(
+        DiagnosticIDs::create(), State->DiagOpts, Client);
   }
   return State->MetaDiagnostics.get();
 }
diff --git a/clang/lib/Headers/hlsl/hlsl_intrinsic_helpers.h b/clang/lib/Headers/hlsl/hlsl_intrinsic_helpers.h
index e8ccccb..c877234 100644
--- a/clang/lib/Headers/hlsl/hlsl_intrinsic_helpers.h
+++ b/clang/lib/Headers/hlsl/hlsl_intrinsic_helpers.h
@@ -12,7 +12,7 @@
 namespace hlsl {
 namespace __detail {
 
-constexpr vector<uint, 4> d3d_color_to_ubyte4_impl(vector<float, 4> V) {
+constexpr int4 d3d_color_to_ubyte4_impl(float4 V) {
   // Use the same scaling factor used by FXC, and DXC for DXIL
   // (i.e., 255.001953)
   // https://github.com/microsoft/DirectXShaderCompiler/blob/070d0d5a2beacef9eeb51037a9b04665716fd6f3/lib/HLSL/HLOperationLower.cpp#L666C1-L697C2
diff --git a/clang/lib/Headers/hlsl/hlsl_intrinsics.h b/clang/lib/Headers/hlsl/hlsl_intrinsics.h
index 499a053..d9d87c8 100644
--- a/clang/lib/Headers/hlsl/hlsl_intrinsics.h
+++ b/clang/lib/Headers/hlsl/hlsl_intrinsics.h
@@ -418,7 +418,7 @@ const inline float4 lit(float NDotL, float NDotH, float M) {
 /// This function swizzles and scales components of the \a x parameter. Use this
 /// function to compensate for the lack of UBYTE4 support in some hardware.
 
-constexpr vector<uint, 4> D3DCOLORtoUBYTE4(vector<float, 4> V) {
+constexpr int4 D3DCOLORtoUBYTE4(float4 V) {
   return __detail::d3d_color_to_ubyte4_impl(V);
 }
 
diff --git a/clang/lib/Interpreter/CodeCompletion.cpp b/clang/lib/Interpreter/CodeCompletion.cpp
index dac3888..ecdf489 100644
--- a/clang/lib/Interpreter/CodeCompletion.cpp
+++ b/clang/lib/Interpreter/CodeCompletion.cpp
@@ -380,7 +380,7 @@ void ReplCodeCompleter::codeComplete(CompilerInstance *InterpCI,
   AU->setOwnsRemappedFileBuffers(false);
   AU->CodeComplete(CodeCompletionFileName, 1, Col, RemappedFiles, false, false,
                    false, consumer,
-                   std::make_shared<clang::PCHContainerOperations>(), *diag,
+                   std::make_shared<clang::PCHContainerOperations>(), diag,
                    InterpCI->getLangOpts(), AU->getSourceManager(),
                    AU->getFileManager(), sd, tb, std::move(Act));
 }
diff --git a/clang/lib/Interpreter/Interpreter.cpp b/clang/lib/Interpreter/Interpreter.cpp
index 9b71486..5e5ae81 100644
--- a/clang/lib/Interpreter/Interpreter.cpp
+++ b/clang/lib/Interpreter/Interpreter.cpp
@@ -86,7 +86,6 @@ GetCC1Arguments(DiagnosticsEngine *Diagnostics,
 static llvm::Expected<std::unique_ptr<CompilerInstance>>
 CreateCI(const llvm::opt::ArgStringList &Argv) {
   std::unique_ptr<CompilerInstance> Clang(new CompilerInstance());
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
 
   // Register the support for object-file-wrapped Clang modules.
   // FIXME: Clang should register these container operations automatically.
@@ -98,7 +97,7 @@ CreateCI(const llvm::opt::ArgStringList &Argv) {
   // a well formed diagnostic object.
   DiagnosticOptions DiagOpts;
   TextDiagnosticBuffer *DiagsBuffer = new TextDiagnosticBuffer;
-  DiagnosticsEngine Diags(DiagID, DiagOpts, DiagsBuffer);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts, DiagsBuffer);
   bool Success = CompilerInvocation::CreateFromArgs(
       Clang->getInvocation(), llvm::ArrayRef(Argv.begin(), Argv.size()), Diags);
 
@@ -174,11 +173,10 @@ IncrementalCompilerBuilder::create(std::string TT,
 
   // Buffer diagnostics from argument parsing so that we can output them using a
   // well formed diagnostic object.
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
   std::unique_ptr<DiagnosticOptions> DiagOpts =
       CreateAndPopulateDiagOpts(ClangArgv);
   TextDiagnosticBuffer *DiagsBuffer = new TextDiagnosticBuffer;
-  DiagnosticsEngine Diags(DiagID, *DiagOpts, DiagsBuffer);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), *DiagOpts, DiagsBuffer);
 
   driver::Driver Driver(/*MainBinaryName=*/ClangArgv[0], TT, Diags);
   Driver.setCheckInputsExist(false); // the input comes from mem buffers
diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp
index f896f9f1..682d263 100644
--- a/clang/lib/Serialization/ASTReader.cpp
+++ b/clang/lib/Serialization/ASTReader.cpp
@@ -677,8 +677,7 @@ bool PCHValidator::ReadDiagnosticOptions(DiagnosticOptions &DiagOpts,
                                          bool Complain) {
   DiagnosticsEngine &ExistingDiags = PP.getDiagnostics();
   IntrusiveRefCntPtr<DiagnosticIDs> DiagIDs(ExistingDiags.getDiagnosticIDs());
-  IntrusiveRefCntPtr<DiagnosticsEngine> Diags(
-      new DiagnosticsEngine(DiagIDs, DiagOpts));
+  auto Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(DiagIDs, DiagOpts);
   // This should never fail, because we would have processed these options
   // before writing them to an ASTFile.
   ProcessWarningOptions(*Diags, DiagOpts,
diff --git a/clang/lib/Tooling/CompilationDatabase.cpp b/clang/lib/Tooling/CompilationDatabase.cpp
index d5fc216..860457a 100644
--- a/clang/lib/Tooling/CompilationDatabase.cpp
+++ b/clang/lib/Tooling/CompilationDatabase.cpp
@@ -241,9 +241,8 @@ static bool stripPositionalArgs(std::vector<const char *> Args,
   llvm::raw_string_ostream Output(ErrorMsg);
   TextDiagnosticPrinter DiagnosticPrinter(Output, DiagOpts);
   UnusedInputDiagConsumer DiagClient(DiagnosticPrinter);
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts,
-      &DiagClient, false);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts, &DiagClient,
+                                false);
 
   // The clang executable path isn't required since the jobs the driver builds
   // will not be executed.
diff --git a/clang/lib/Tooling/Core/Replacement.cpp b/clang/lib/Tooling/Core/Replacement.cpp
index a3214de..10bdc22 100644
--- a/clang/lib/Tooling/Core/Replacement.cpp
+++ b/clang/lib/Tooling/Core/Replacement.cpp
@@ -585,8 +585,7 @@ llvm::Expected<std::string> applyAllReplacements(StringRef Code,
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
   FileManager Files(FileSystemOptions(), InMemoryFileSystem);
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs), DiagOpts);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts);
   SourceManager SourceMgr(Diagnostics, Files);
   Rewriter Rewrite(SourceMgr, LangOptions());
   InMemoryFileSystem->addFile(
diff --git a/clang/lib/Tooling/Refactoring.cpp b/clang/lib/Tooling/Refactoring.cpp
index d2b0b37..9b1af237 100644
--- a/clang/lib/Tooling/Refactoring.cpp
+++ b/clang/lib/Tooling/Refactoring.cpp
@@ -39,9 +39,8 @@ int RefactoringTool::runAndSave(FrontendActionFactory *ActionFactory) {
   LangOptions DefaultLangOptions;
   DiagnosticOptions DiagOpts;
   TextDiagnosticPrinter DiagnosticPrinter(llvm::errs(), DiagOpts);
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs()), DiagOpts,
-      &DiagnosticPrinter, false);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts,
+                                &DiagnosticPrinter, false);
   SourceManager Sources(Diagnostics, getFiles());
   Rewriter Rewrite(Sources, DefaultLangOptions);
 
diff --git a/clang/test/CIR/CodeGen/complex-mul-div.cpp b/clang/test/CIR/CodeGen/complex-mul-div.cpp
new file mode 100644
index 0000000..9d71ef7
--- /dev/null
+++ b/clang/test/CIR/CodeGen/complex-mul-div.cpp
@@ -0,0 +1,280 @@
+// complex-range basic
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fclangir -complex-range=basic -Wno-unused-value -fclangir -emit-cir -mmlir --mlir-print-ir-before=cir-canonicalize -o %t.cir %s 2>&1 | FileCheck --check-prefix=CIR-BEFORE-BASIC %s
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=basic -Wno-unused-value -fclangir -emit-cir %s -o %t.cir
+// RUN: FileCheck --input-file=%t.cir %s --check-prefixes=CIR-AFTER-INT,CIR-AFTER-MUL-COMBINED
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=basic -Wno-unused-value -fclangir -emit-llvm %s -o %t-cir.ll
+// RUN: FileCheck --input-file=%t-cir.ll %s --check-prefixes=LLVM-INT,LLVM-MUL-COMBINED
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=basic -Wno-unused-value -emit-llvm %s -o %t.ll
+// RUN: FileCheck --input-file=%t.ll %s --check-prefixes=OGCG-INT,OGCG-MUL-COMBINED
+
+// complex-range improved
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fclangir -complex-range=improved -Wno-unused-value -fclangir -emit-cir -mmlir --mlir-print-ir-before=cir-canonicalize -o %t.cir %s 2>&1 | FileCheck --check-prefix=CIR-BEFORE-IMPROVED %s
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=improved -Wno-unused-value -fclangir -emit-cir %s -o %t.cir
+// RUN: FileCheck --input-file=%t.cir %s --check-prefixes=CIR-AFTER-INT,CIR-AFTER-MUL-COMBINED
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=improved -Wno-unused-value -fclangir -emit-llvm %s -o %t-cir.ll
+// RUN: FileCheck --input-file=%t-cir.ll %s --check-prefixes=LLVM-INT,LLVM-MUL-COMBINED
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=improved -Wno-unused-value -emit-llvm %s -o %t.ll
+// RUN: FileCheck --input-file=%t.ll %s --check-prefixes=OGCG-INT,OGCG-MUL-COMBINED
+
+// complex-range promoted
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fclangir -complex-range=promoted -Wno-unused-value -fclangir -emit-cir -mmlir --mlir-print-ir-before=cir-canonicalize -o %t.cir %s 2>&1 | FileCheck --check-prefix=CIR-BEFORE-PROMOTED %s
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=promoted -Wno-unused-value -fclangir -emit-cir %s -o %t.cir
+// RUN: FileCheck --input-file=%t.cir %s --check-prefixes=CIR-AFTER-INT,CIR-AFTER-MUL-COMBINED
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=promoted -Wno-unused-value -fclangir -emit-llvm %s -o %t-cir.ll
+// RUN: FileCheck --input-file=%t-cir.ll %s --check-prefixes=LLVM-INT,LLVM-MUL-COMBINED
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=promoted -Wno-unused-value -emit-llvm %s -o %t.ll
+// RUN: FileCheck --input-file=%t.ll %s --check-prefixes=OGCG-INT,OGCG-MUL-COMBINED
+
+// complex-range full
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fclangir -complex-range=full -Wno-unused-value -fclangir -emit-cir -mmlir --mlir-print-ir-before=cir-canonicalize -o %t.cir %s 2>&1 | FileCheck --check-prefix=CIR-BEFORE-FULL %s
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=full -Wno-unused-value -fclangir -emit-cir %s -o %t.cir
+// RUN: FileCheck --input-file=%t.cir %s --check-prefixes=CIR-AFTER-FULL,CIR-AFTER-INT
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=full -Wno-unused-value -fclangir -emit-llvm %s -o %t-cir.ll
+// RUN: FileCheck --input-file=%t-cir.ll %s --check-prefixes=LLVM-FULL,LLVM-INT
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu -complex-range=full -Wno-unused-value -emit-llvm %s -o %t.ll
+// RUN: FileCheck --input-file=%t.ll %s --check-prefixes=OGCG-FULL,OGCG-INT
+
+void foo() {
+  float _Complex a;
+  float _Complex b;
+  float _Complex c = a * b;
+}
+
+// CIR-BEFORE-BASIC: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(basic) : !cir.complex<!cir.float>
+
+// CIR-BEFORE-IMPROVED: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(improved) : !cir.complex<!cir.float>
+
+// CIR-BEFORE-PROMOTED: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(promoted) : !cir.complex<!cir.float>
+
+// CIR-AFTER-MUL-COMBINED: %[[A_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["a"]
+// CIR-AFTER-MUL-COMBINED: %[[B_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["b"]
+// CIR-AFTER-MUL-COMBINED: %[[C_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["c", init]
+// CIR-AFTER-MUL-COMBINED: %[[TMP_A:.*]] = cir.load{{.*}} %[[A_ADDR]] : !cir.ptr<!cir.complex<!cir.float>>, !cir.complex<!cir.float>
+// CIR-AFTER-MUL-COMBINED: %[[TMP_B:.*]] = cir.load{{.*}} %[[B_ADDR]] : !cir.ptr<!cir.complex<!cir.float>>, !cir.complex<!cir.float>
+// CIR-AFTER-MUL-COMBINED: %[[A_REAL:.*]] = cir.complex.real %[[TMP_A]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[A_IMAG:.*]] = cir.complex.imag %[[TMP_A]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[B_REAL:.*]] = cir.complex.real %[[TMP_B]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[B_IMAG:.*]] = cir.complex.imag %[[TMP_B]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[MUL_AR_BR:.*]] = cir.binop(mul, %[[A_REAL]], %[[B_REAL]]) : !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[MUL_AI_BI:.*]] = cir.binop(mul, %[[A_IMAG]], %[[B_IMAG]]) : !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[MUL_AR_BI:.*]] = cir.binop(mul, %[[A_REAL]], %[[B_IMAG]]) : !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[MUL_AI_BR:.*]] = cir.binop(mul, %[[A_IMAG]], %[[B_REAL]]) : !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[C_REAL:.*]] = cir.binop(sub, %[[MUL_AR_BR]], %[[MUL_AI_BI]]) : !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[C_IMAG:.*]] = cir.binop(add, %[[MUL_AR_BI]], %[[MUL_AI_BR]]) : !cir.float
+// CIR-AFTER-MUL-COMBINED: %[[RESULT:.*]] = cir.complex.create %[[C_REAL]], %[[C_IMAG]] : !cir.float -> !cir.complex<!cir.float>
+// CIR-AFTER-MUL-COMBINED: cir.store{{.*}} %[[RESULT]], %[[C_ADDR]] : !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>
+
+// LLVM-MUL-COMBINED: %[[A_ADDR:.*]] = alloca { float, float }, i64 1, align 4
+// LLVM-MUL-COMBINED: %[[B_ADDR:.*]] = alloca { float, float }, i64 1, align 4
+// LLVM-MUL-COMBINED: %[[C_ADDR:.*]] = alloca { float, float }, i64 1, align 4
+// LLVM-MUL-COMBINED: %[[TMP_A:.*]] = load { float, float }, ptr %[[A_ADDR]], align 4
+// LLVM-MUL-COMBINED: %[[TMP_B:.*]] = load { float, float }, ptr %[[B_ADDR]], align 4
+// LLVM-MUL-COMBINED: %[[A_REAL:.*]] = extractvalue { float, float } %[[TMP_A]], 0
+// LLVM-MUL-COMBINED: %[[A_IMAG:.*]] = extractvalue { float, float } %[[TMP_A]], 1
+// LLVM-MUL-COMBINED: %[[B_REAL:.*]] = extractvalue { float, float } %[[TMP_B]], 0
+// LLVM-MUL-COMBINED: %[[B_IMAG:.*]] = extractvalue { float, float } %[[TMP_B]], 1
+// LLVM-MUL-COMBINED: %[[MUL_AR_BR:.*]] = fmul float %[[A_REAL]], %[[B_REAL]]
+// LLVM-MUL-COMBINED: %[[MUL_AI_BI:.*]] = fmul float %[[A_IMAG]], %[[B_IMAG]]
+// LLVM-MUL-COMBINED: %[[MUL_AR_BI:.*]] = fmul float %[[A_REAL]], %[[B_IMAG]]
+// LLVM-MUL-COMBINED: %[[MUL_AI_BR:.*]] = fmul float %[[A_IMAG]], %[[B_REAL]]
+// LLVM-MUL-COMBINED: %[[C_REAL:.*]] = fsub float %[[MUL_AR_BR]], %[[MUL_AI_BI]]
+// LLVM-MUL-COMBINED: %[[C_IMAG:.*]] = fadd float %[[MUL_AR_BI]], %[[MUL_AI_BR]]
+// LLVM-MUL-COMBINED: %[[MUL_A_B:.*]] = insertvalue { float, float } {{.*}}, float %[[C_REAL]], 0
+// LLVM-MUL-COMBINED: %[[RESULT:.*]] = insertvalue { float, float } %[[MUL_A_B]], float %[[C_IMAG]], 1
+// LLVM-MUL-COMBINED: store { float, float } %[[RESULT]], ptr %[[C_ADDR]], align 4
+
+// OGCG-MUL-COMBINED: %[[A_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-MUL-COMBINED: %[[B_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-MUL-COMBINED: %[[C_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-MUL-COMBINED: %[[A_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[A_ADDR]], i32 0, i32 0
+// OGCG-MUL-COMBINED: %[[A_REAL:.*]] = load float, ptr %[[A_REAL_PTR]], align 4
+// OGCG-MUL-COMBINED: %[[A_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[A_ADDR]], i32 0, i32 1
+// OGCG-MUL-COMBINED: %[[A_IMAG:.*]] = load float, ptr %[[A_IMAG_PTR]], align 4
+// OGCG-MUL-COMBINED: %[[B_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[B_ADDR]], i32 0, i32 0
+// OGCG-MUL-COMBINED: %[[B_REAL:.*]] = load float, ptr %[[B_REAL_PTR]], align 4
+// OGCG-MUL-COMBINED: %[[B_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[B_ADDR]], i32 0, i32 1
+// OGCG-MUL-COMBINED: %[[B_IMAG:.*]] = load float, ptr %[[B_IMAG_PTR]], align 4
+// OGCG-MUL-COMBINED: %[[MUL_AR_BR:.*]] = fmul float %[[A_REAL]], %[[B_REAL]]
+// OGCG-MUL-COMBINED: %[[MUL_AI_BI:.*]] = fmul float %[[A_IMAG]], %[[B_IMAG]]
+// OGCG-MUL-COMBINED: %[[MUL_AR_BI:.*]] = fmul float %[[A_REAL]], %[[B_IMAG]]
+// OGCG-MUL-COMBINED: %[[MUL_AI_BR:.*]] = fmul float %[[A_IMAG]], %[[B_REAL]]
+// OGCG-MUL-COMBINED: %[[C_REAL:.*]] = fsub float %[[MUL_AR_BR]], %[[MUL_AI_BI]]
+// OGCG-MUL-COMBINED: %[[C_IMAG:.*]] = fadd float %[[MUL_AR_BI]], %[[MUL_AI_BR]]
+// OGCG-MUL-COMBINED: %[[C_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[C_ADDR]], i32 0, i32 0
+// OGCG-MUL-COMBINED: %[[C_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[C_ADDR]], i32 0, i32 1
+// OGCG-MUL-COMBINED: store float %[[C_REAL]], ptr %[[C_REAL_PTR]], align 4
+// OGCG-MUL-COMBINED: store float %[[C_IMAG]], ptr %[[C_IMAG_PTR]], align 4
+
+// CIR-BEFORE-FULL: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(full) : !cir.complex<!cir.float>
+
+// CIR-AFTER-FULL: %[[A_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["a"]
+// CIR-AFTER-FULL: %[[B_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["b"]
+// CIR-AFTER-FULL: %[[C_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["c", init]
+// CIR-AFTER-FULL: %[[TMP_A:.*]] = cir.load{{.*}} %[[A_ADDR]] : !cir.ptr<!cir.complex<!cir.float>>, !cir.complex<!cir.float>
+// CIR-AFTER-FULL: %[[TMP_B:.*]] = cir.load{{.*}} %[[B_ADDR]] : !cir.ptr<!cir.complex<!cir.float>>, !cir.complex<!cir.float>
+// CIR-AFTER-FULL: %[[A_REAL:.*]] = cir.complex.real %[[TMP_A]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-FULL: %[[A_IMAG:.*]] = cir.complex.imag %[[TMP_A]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-FULL: %[[B_REAL:.*]] = cir.complex.real %[[TMP_B]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-FULL: %[[B_IMAG:.*]] = cir.complex.imag %[[TMP_B]] : !cir.complex<!cir.float> -> !cir.float
+// CIR-AFTER-FULL: %[[MUL_AR_BR:.*]] = cir.binop(mul, %[[A_REAL]], %[[B_REAL]]) : !cir.float
+// CIR-AFTER-FULL: %[[MUL_AI_BI:.*]] = cir.binop(mul, %[[A_IMAG]], %[[B_IMAG]]) : !cir.float
+// CIR-AFTER-FULL: %[[MUL_AR_BI:.*]] = cir.binop(mul, %[[A_REAL]], %[[B_IMAG]]) : !cir.float
+// CIR-AFTER-FULL: %[[MUL_AI_BR:.*]] = cir.binop(mul, %[[A_IMAG]], %[[B_REAL]]) : !cir.float
+// CIR-AFTER-FULL: %[[C_REAL:.*]] = cir.binop(sub, %[[MUL_AR_BR]], %[[MUL_AI_BI]]) : !cir.float
+// CIR-AFTER-FULL: %[[C_IMAG:.*]] = cir.binop(add, %[[MUL_AR_BI]], %[[MUL_AI_BR]]) : !cir.float
+// CIR-AFTER-FULL: %[[COMPLEX:.*]] = cir.complex.create %[[C_REAL]], %[[C_IMAG]] : !cir.float -> !cir.complex<!cir.float>
+// CIR-AFTER-FULL: %[[IS_C_REAL_NAN:.*]] = cir.cmp(ne, %[[C_REAL]], %[[C_REAL]]) : !cir.float, !cir.bool
+// CIR-AFTER-FULL: %[[IS_C_IMAG_NAN:.*]] = cir.cmp(ne, %[[C_IMAG]], %[[C_IMAG]]) : !cir.float, !cir.bool
+// CIR-AFTER-FULL: %[[CONST_FALSE:.*]] = cir.const #false
+// CIR-AFTER-FULL: %[[SELECT_CONDITION:.*]] = cir.select if %[[IS_C_REAL_NAN]] then %[[IS_C_IMAG_NAN]] else %[[CONST_FALSE]] : (!cir.bool, !cir.bool, !cir.bool) -> !cir.bool
+// CIR-AFTER-FULL: %[[RESULT:.*]] = cir.ternary(%[[SELECT_CONDITION]], true {
+// CIR-AFTER-FULL:   %[[LIBC_COMPLEX:.*]] = cir.call @__mulsc3(%[[A_REAL]], %[[A_IMAG]], %[[B_REAL]], %[[B_IMAG]]) : (!cir.float, !cir.float, !cir.float, !cir.float) -> !cir.complex<!cir.float>
+// CIR-AFTER-FULL:   cir.yield %[[LIBC_COMPLEX]] : !cir.complex<!cir.float>
+// CIR-AFTER-FULL: }, false {
+// CIR-AFTER-FULL:   cir.yield %[[COMPLEX]] : !cir.complex<!cir.float>
+// CIR-AFTER-FULL: }) : (!cir.bool) -> !cir.complex<!cir.float>
+// CIR-AFTER-FULL: cir.store{{.*}} %[[RESULT]], %[[C_ADDR]] : !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>
+
+// LLVM-FULL: %[[A_ADDR:.*]] = alloca { float, float }, i64 1, align 4
+// LLVM-FULL: %[[B_ADDR:.*]] = alloca { float, float }, i64 1, align 4
+// LLVM-FULL: %[[C_ADDR:.*]] = alloca { float, float }, i64 1, align 4
+// LLVM-FULL: %[[TMP_A:.*]] = load { float, float }, ptr %[[A_ADDR]], align 4
+// LLVM-FULL: %[[TMP_B:.*]] = load { float, float }, ptr %[[B_ADDR]], align 4
+// LLVM-FULL: %[[A_REAL:.*]] = extractvalue { float, float } %[[TMP_A]], 0
+// LLVM-FULL: %[[A_IMAG:.*]] = extractvalue { float, float } %[[TMP_A]], 1
+// LLVM-FULL: %[[B_REAL:.*]] = extractvalue { float, float } %[[TMP_B]], 0
+// LLVM-FULL: %[[B_IMAG:.*]] = extractvalue { float, float } %[[TMP_B]], 1
+// LLVM-FULL: %[[MUL_AR_BR:.*]] = fmul float %[[A_REAL]], %[[B_REAL]]
+// LLVM-FULL: %[[MUL_AI_BI:.*]] = fmul float %[[A_IMAG]], %[[B_IMAG]]
+// LLVM-FULL: %[[MUL_AR_BI:.*]] = fmul float %[[A_REAL]], %[[B_IMAG]]
+// LLVM-FULL: %[[MUL_AI_BR:.*]] = fmul float %[[A_IMAG]], %[[B_REAL]]
+// LLVM-FULL: %[[C_REAL:.*]] = fsub float %[[MUL_AR_BR]], %[[MUL_AI_BI]]
+// LLVM-FULL: %[[C_IMAG:.*]] = fadd float %[[MUL_AR_BI]], %[[MUL_AI_BR]]
+// LLVM-FULL: %[[MUL_A_B:.*]] = insertvalue { float, float } {{.*}}, float %[[C_REAL]], 0
+// LLVM-FULL: %[[COMPLEX:.*]] = insertvalue { float, float } %[[MUL_A_B]], float %[[C_IMAG]], 1
+// LLVM-FULL: %[[IS_C_REAL_NAN:.*]] = fcmp une float %[[C_REAL]], %[[C_REAL]]
+// LLVM-FULL: %[[IS_C_IMAG_NAN:.*]] = fcmp une float %[[C_IMAG]], %[[C_IMAG]]
+// LLVM-FULL: %[[SELECT_CONDITION:.*]] = and i1 %[[IS_C_REAL_NAN]], %[[IS_C_IMAG_NAN]]
+// LLVM-FULL: br i1 %[[SELECT_CONDITION]], label %[[THEN_LABEL:.*]], label %[[ELSE_LABEL:.*]]
+// LLVM-FULL: [[THEN_LABEL]]:
+// LLVM-FULL:  %[[LIBC_COMPLEX:.*]] = call { float, float } @__mulsc3(float %[[A_REAL]], float %[[A_IMAG]], float %[[B_REAL]], float %[[B_IMAG]])
+// LLVM-FULL:  br label %[[PHI_BRANCH:.*]]
+// LLVM-FULL: [[ELSE_LABEL]]:
+// LLVM-FULL:  br label %[[PHI_BRANCH:]]
+// LLVM-FULL: [[PHI_BRANCH:]]:
+// LLVM-FULL:  %[[RESULT:.*]] = phi { float, float } [ %[[COMPLEX]], %[[ELSE_LABEL]] ], [ %[[LIBC_COMPLEX]], %[[THEN_LABEL]] ]
+// LLVM-FULL:  br label %[[END_LABEL:.*]]
+// LLVM-FULL: [[END_LABEL]]:
+// LLVM-FULL:  store { float, float } %[[RESULT]], ptr %[[C_ADDR]], align 4
+
+// OGCG-FULL: %[[A_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-FULL: %[[B_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-FULL: %[[C_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-FULL: %[[COMPLEX_CALL_ADDR:.*]] = alloca { float, float }, align 4
+// OGCG-FULL: %[[A_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[A_ADDR]], i32 0, i32 0
+// OGCG-FULL: %[[A_REAL:.*]] = load float, ptr %[[A_REAL_PTR]], align 4
+// OGCG-FULL: %[[A_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[A_ADDR]], i32 0, i32 1
+// OGCG-FULL: %[[A_IMAG:.*]] = load float, ptr %[[A_IMAG_PTR]], align 4
+// OGCG-FULL: %[[B_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[B_ADDR]], i32 0, i32 0
+// OGCG-FULL: %[[B_REAL:.*]] = load float, ptr %[[B_REAL_PTR]], align 4
+// OGCG-FULL: %[[B_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[B_ADDR]], i32 0, i32 1
+// OGCG-FULL: %[[B_IMAG:.*]] = load float, ptr %[[B_IMAG_PTR]], align 4
+// OGCG-FULL: %[[MUL_AR_BR:.*]] = fmul float %[[A_REAL]], %[[B_REAL]]
+// OGCG-FULL: %[[MUL_AI_BI:.*]] = fmul float %[[A_IMAG]], %[[B_IMAG]]
+// OGCG-FULL: %[[MUL_AR_BI:.*]] = fmul float %[[A_REAL]], %[[B_IMAG]]
+// OGCG-FULL: %[[MUL_AI_BR:.*]] = fmul float %[[A_IMAG]], %[[B_REAL]]
+// OGCG-FULL: %[[C_REAL:.*]] = fsub float %[[MUL_AR_BR]], %[[MUL_AI_BI]]
+// OGCG-FULL: %[[C_IMAG:.*]] = fadd float %[[MUL_AR_BI]], %[[MUL_AI_BR]]
+// OGCG-FULL: %[[IS_C_REAL_NAN:.*]] = fcmp uno float %[[C_REAL]], %[[C_REAL]]
+// OGCG-FULL: br i1 %[[IS_C_REAL_NAN]], label %[[COMPLEX_IS_IMAG_NAN:.*]], label %[[END_LABEL:.*]], !prof !2
+// OGCG-FULL: [[COMPLEX_IS_IMAG_NAN]]:
+// OGCG-FULL:  %[[IS_C_IMAG_NAN:.*]] = fcmp uno float %[[C_IMAG]], %[[C_IMAG]]
+// OGCG-FULL:  br i1 %[[IS_C_IMAG_NAN]], label %[[COMPLEX_LIB_CALL:.*]], label %[[END_LABEL]], !prof !2
+// OGCG-FULL: [[COMPLEX_LIB_CALL]]:
+// OGCG-FULL:  %[[CALL_RESULT:.*]] = call noundef <2 x float> @__mulsc3(float noundef %[[A_REAL]], float noundef %[[A_IMAG]], float noundef %[[B_REAL]], float noundef %[[B_IMAG]])
+// OGCG-FULL:  store <2 x float> %[[CALL_RESULT]], ptr %[[COMPLEX_CALL_ADDR]], align 4
+// OGCG-FULL:  %[[COMPLEX_CALL_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[COMPLEX_CALL_ADDR]], i32 0, i32 0
+// OGCG-FULL:  %[[COMPLEX_CALL_REAL:.*]] = load float, ptr %[[COMPLEX_CALL_REAL_PTR]], align 4
+// OGCG-FULL:  %[[COMPLEX_CALL_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[COMPLEX_CALL_ADDR]], i32 0, i32 1
+// OGCG-FULL:  %[[COMPLEX_CALL_IMAG:.*]] = load float, ptr %[[COMPLEX_CALL_IMAG_PTR]], align 4
+// OGCG-FULL:  br label %[[END_LABEL]]
+// OGCG-FULL: [[END_LABEL]]:
+// OGCG-FULL:  %[[FINAL_REAL:.*]] = phi float [ %[[C_REAL]], %[[ENTRY:.*]] ], [ %[[C_REAL]], %[[COMPLEX_IS_IMAG_NAN]] ], [ %[[COMPLEX_CALL_REAL]], %[[COMPLEX_LIB_CALL]] ]
+// OGCG-FULL:  %[[FINAL_IMAG:.*]] = phi float [ %[[C_IMAG]], %[[ENTRY]] ], [ %[[C_IMAG]], %[[COMPLEX_IS_IMAG_NAN]] ], [ %[[COMPLEX_CALL_IMAG]], %[[COMPLEX_LIB_CALL]] ]
+// OGCG-FULL:  %[[C_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[C_ADDR]], i32 0, i32 0
+// OGCG-FULL:  %[[C_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[C_ADDR]], i32 0, i32 1
+// OGCG-FULL:  store float %[[FINAL_REAL]], ptr %[[C_REAL_PTR]], align 4
+// OGCG-FULL:  store float %[[FINAL_IMAG]], ptr %[[C_IMAG_PTR]], align 4
+
+void foo1() {
+  int _Complex a;
+  int _Complex b;
+  int _Complex c = a * b;
+}
+
+// CIR-BEFORE-BASIC: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(basic) : !cir.complex<!s32i>
+
+// CIR-BEFORE-IMPROVED: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(improved) : !cir.complex<!s32i>
+
+// CIR-BEFORE-PROMOTED: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(promoted) : !cir.complex<!s32i>
+
+// CIR-BEFORE-FULL: %{{.*}} = cir.complex.mul {{.*}}, {{.*}} range(full) : !cir.complex<!s32i>
+
+// CIR-AFTER-INT: %[[A_ADDR:.*]] = cir.alloca !cir.complex<!s32i>, !cir.ptr<!cir.complex<!s32i>>, ["a"]
+// CIR-AFTER-INT: %[[B_ADDR:.*]] = cir.alloca !cir.complex<!s32i>, !cir.ptr<!cir.complex<!s32i>>, ["b"]
+// CIR-AFTER-INT: %[[C_ADDR:.*]] = cir.alloca !cir.complex<!s32i>, !cir.ptr<!cir.complex<!s32i>>, ["c", init]
+// CIR-AFTER-INT: %[[TMP_A:.*]] = cir.load{{.*}} %[[A_ADDR]] : !cir.ptr<!cir.complex<!s32i>>, !cir.complex<!s32i>
+// CIR-AFTER-INT: %[[TMP_B:.*]] = cir.load{{.*}} %[[B_ADDR]] : !cir.ptr<!cir.complex<!s32i>>, !cir.complex<!s32i>
+// CIR-AFTER-INT: %[[A_REAL:.*]] = cir.complex.real %[[TMP_A]] : !cir.complex<!s32i> -> !s32i
+// CIR-AFTER-INT: %[[A_IMAG:.*]] = cir.complex.imag %[[TMP_A]] : !cir.complex<!s32i> -> !s32i
+// CIR-AFTER-INT: %[[B_REAL:.*]] = cir.complex.real %[[TMP_B]] : !cir.complex<!s32i> -> !s32i
+// CIR-AFTER-INT: %[[B_IMAG:.*]] = cir.complex.imag %[[TMP_B]] : !cir.complex<!s32i> -> !s32i
+// CIR-AFTER-INT: %[[MUL_AR_BR:.*]] = cir.binop(mul, %[[A_REAL]], %[[B_REAL]]) : !s32i
+// CIR-AFTER-INT: %[[MUL_AI_BI:.*]] = cir.binop(mul, %[[A_IMAG]], %[[B_IMAG]]) : !s32i
+// CIR-AFTER-INT: %[[MUL_AR_BI:.*]] = cir.binop(mul, %[[A_REAL]], %[[B_IMAG]]) : !s32i
+// CIR-AFTER-INT: %[[MUL_AI_BR:.*]] = cir.binop(mul, %[[A_IMAG]], %[[B_REAL]]) : !s32i
+// CIR-AFTER-INT: %[[C_REAL:.*]] = cir.binop(sub, %[[MUL_AR_BR]], %[[MUL_AI_BI]]) : !s32i
+// CIR-AFTER-INT: %[[C_IMAG:.*]] = cir.binop(add, %[[MUL_AR_BI]], %[[MUL_AI_BR]]) : !s32i
+// CIR-AFTER-INT: %[[RESULT:.*]] = cir.complex.create %[[C_REAL]], %[[C_IMAG]] : !s32i -> !cir.complex<!s32i>
+// CIR-AFTER-INT: cir.store{{.*}} %[[RESULT]], %[[C_ADDR]] : !cir.complex<!s32i>, !cir.ptr<!cir.complex<!s32i>>
+
+// LLVM-INT: %[[A_ADDR:.*]] = alloca { i32, i32 }, i64 1, align 4
+// LLVM-INT: %[[B_ADDR:.*]] = alloca { i32, i32 }, i64 1, align 4
+// LLVM-INT: %[[C_ADDR:.*]] = alloca { i32, i32 }, i64 1, align 4
+// LLVM-INT: %[[TMP_A:.*]] = load { i32, i32 }, ptr %[[A_ADDR]], align 4
+// LLVM-INT: %[[TMP_B:.*]] = load { i32, i32 }, ptr %[[B_ADDR]], align 4
+// LLVM-INT: %[[A_REAL:.*]] = extractvalue { i32, i32 } %[[TMP_A]], 0
+// LLVM-INT: %[[A_IMAG:.*]] = extractvalue { i32, i32 } %[[TMP_A]], 1
+// LLVM-INT: %[[B_REAL:.*]] = extractvalue { i32, i32 } %[[TMP_B]], 0
+// LLVM-INT: %[[B_IMAG:.*]] = extractvalue { i32, i32 } %[[TMP_B]], 1
+// LLVM-INT: %[[MUL_AR_BR:.*]] = mul i32 %[[A_REAL]], %[[B_REAL]]
+// LLVM-INT: %[[MUL_AI_BI:.*]] = mul i32 %[[A_IMAG]], %[[B_IMAG]]
+// LLVM-INT: %[[MUL_AR_BI:.*]] = mul i32 %[[A_REAL]], %[[B_IMAG]]
+// LLVM-INT: %[[MUL_AI_BR:.*]] = mul i32 %[[A_IMAG]], %[[B_REAL]]
+// LLVM-INT: %[[C_REAL:.*]] = sub i32 %[[MUL_AR_BR]], %[[MUL_AI_BI]]
+// LLVM-INT: %[[C_IMAG:.*]] = add i32 %[[MUL_AR_BI]], %[[MUL_AI_BR]]
+// LLVM-INT: %[[MUL_A_B:.*]] = insertvalue { i32, i32 } {{.*}}, i32 %[[C_REAL]], 0
+// LLVM-INT: %[[RESULT:.*]] = insertvalue { i32, i32 } %[[MUL_A_B]], i32 %[[C_IMAG]], 1
+// LLVM-INT: store { i32, i32 } %[[RESULT]], ptr %[[C_ADDR]], align 4
+
+// OGCG-INT: %[[A_ADDR:.*]] = alloca { i32, i32 }, align 4
+// OGCG-INT: %[[B_ADDR:.*]] = alloca { i32, i32 }, align 4
+// OGCG-INT: %[[C_ADDR:.*]] = alloca { i32, i32 }, align 4
+// OGCG-INT: %[[A_REAL_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[A_ADDR]], i32 0, i32 0
+// OGCG-INT: %[[A_REAL:.*]] = load i32, ptr %[[A_REAL_PTR]], align 4
+// OGCG-INT: %[[A_IMAG_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[A_ADDR]], i32 0, i32 1
+// OGCG-INT: %[[A_IMAG:.*]] = load i32, ptr %[[A_IMAG_PTR]], align 4
+// OGCG-INT: %[[B_REAL_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[B_ADDR]], i32 0, i32 0
+// OGCG-INT: %[[B_REAL:.*]] = load i32, ptr %[[B_REAL_PTR]], align 4
+// OGCG-INT: %[[B_IMAG_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[B_ADDR]], i32 0, i32 1
+// OGCG-INT: %[[B_IMAG:.*]] = load i32, ptr %[[B_IMAG_PTR]], align 4
+// OGCG-INT: %[[MUL_AR_BR:.*]] = mul i32 %[[A_REAL]], %[[B_REAL]]
+// OGCG-INT: %[[MUL_AI_BI:.*]] = mul i32 %[[A_IMAG]], %[[B_IMAG]]
+// OGCG-INT: %[[C_REAL:.*]] = sub i32 %[[MUL_AR_BR]], %[[MUL_AI_BI]]
+// OGCG-INT: %[[MUL_AI_BR:.*]] = mul i32 %[[A_IMAG]], %[[B_REAL]]
+// OGCG-INT: %[[MUL_AR_BI:.*]] = mul i32 %[[A_REAL]], %[[B_IMAG]]
+// OGCG-INT: %[[C_IMAG:.*]] = add i32 %[[MUL_AI_BR]], %[[MUL_AR_BI]]
+// OGCG-INT: %[[C_REAL_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[C_ADDR]], i32 0, i32 0
+// OGCG-INT: %[[C_IMAG_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[C_ADDR]], i32 0, i32 1
+// OGCG-INT: store i32 %[[C_REAL]], ptr %[[C_REAL_PTR]], align 4
+// OGCG-INT: store i32 %[[C_IMAG]], ptr %[[C_IMAG_PTR]], align 4
diff --git a/clang/test/CodeGenHLSL/builtins/D3DCOLORtoUBYTE4.hlsl b/clang/test/CodeGenHLSL/builtins/D3DCOLORtoUBYTE4.hlsl
index 990f0aa..3c9e35a 100644
--- a/clang/test/CodeGenHLSL/builtins/D3DCOLORtoUBYTE4.hlsl
+++ b/clang/test/CodeGenHLSL/builtins/D3DCOLORtoUBYTE4.hlsl
@@ -5,8 +5,16 @@
 // CHECK-LABEL: D3DCOLORtoUBYTE4
 int4 test_D3DCOLORtoUBYTE4(float4 p1) {
   // CHECK: %[[SCALED:.*]] = fmul [[FMFLAGS:.*]][[FLOAT_TYPE:<4 x float>]] %{{.*}}, splat (float 0x406FE01000000000)
-  // CHECK: %[[CONVERTED:.*]] = fptoui [[FLOAT_TYPE]] %[[SCALED]] to [[INT_TYPE:<4 x i32>]]
+  // CHECK: %[[CONVERTED:.*]] = fptosi [[FLOAT_TYPE]] %[[SCALED]] to [[INT_TYPE:<4 x i32>]]
   // CHECK: %[[SHUFFLED:.*]] = shufflevector [[INT_TYPE]] %[[CONVERTED]], [[INT_TYPE]] poison, <4 x i32> <i32 2, i32 1, i32 0, i32 3>
   // CHECK: ret [[INT_TYPE]] %[[SHUFFLED]]
   return D3DCOLORtoUBYTE4(p1);
 }
+
+// Note this test confirms issue 150673 is fixed 
+// by confirming the negative does not become a poison
+// CHECK-LABEL: test_constant_inputs
+int4 test_constant_inputs() {
+  // CHECK: ret <4 x i32> <i32 -12877, i32 2833, i32 0, i32 25500>
+  return D3DCOLORtoUBYTE4(float4(0, 11.11, -50.5, 100));
+}
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
index 1c67fc3..77d5673 100644
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
+++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
@@ -282,6 +282,33 @@ void test_cvt_sr_pk_bf16_f32(global bfloat2* out, float a, float b, uint sr)
   *out = __builtin_amdgcn_cvt_sr_pk_bf16_f32(a, b, sr);
 }
 
+// CHECK-LABEL: @test_cvt_sr_pk_f16_f32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4, addrspace(5)
+// CHECK-NEXT:    [[B_ADDR:%.*]] = alloca float, align 4, addrspace(5)
+// CHECK-NEXT:    [[SR_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// CHECK-NEXT:    [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[OUT_ADDR]] to ptr
+// CHECK-NEXT:    [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
+// CHECK-NEXT:    [[B_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[B_ADDR]] to ptr
+// CHECK-NEXT:    [[SR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SR_ADDR]] to ptr
+// CHECK-NEXT:    store ptr addrspace(1) [[OUT:%.*]], ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store float [[A:%.*]], ptr [[A_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    store float [[B:%.*]], ptr [[B_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    store i32 [[SR:%.*]], ptr [[SR_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load float, ptr [[B_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[SR_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = call <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float [[TMP0]], float [[TMP1]], i32 [[TMP2]])
+// CHECK-NEXT:    [[TMP4:%.*]] = load ptr addrspace(1), ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store <2 x half> [[TMP3]], ptr addrspace(1) [[TMP4]], align 4
+// CHECK-NEXT:    ret void
+//
+void test_cvt_sr_pk_f16_f32(global half2* out, float a, float b, uint sr)
+{
+  *out = __builtin_amdgcn_cvt_sr_pk_f16_f32(a, b, sr);
+}
+
 // CHECK-LABEL: @test_cvt_f16_fp8(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
diff --git a/clang/test/Driver/compilation-dir.c b/clang/test/Driver/compilation-dir.c
index dbe801c..70a117b 100644
--- a/clang/test/Driver/compilation-dir.c
+++ b/clang/test/Driver/compilation-dir.c
@@ -8,3 +8,8 @@
 // RUN: %clang -### -integrated-as -ffile-compilation-dir=. -x assembler %s 2>&1 | FileCheck -check-prefixes=CHECK-DEBUG-COMPILATION-DIR %s
 // CHECK-DEBUG-COMPILATION-DIR: "-fdebug-compilation-dir=."
 // CHECK-DEBUG-COMPILATION-DIR-NOT: "-ffile-compilation-dir=."
+
+// RUN: %clang -### -S %s -working-directory %S 2>&1 | FileCheck -check-prefix=CHECK-CWD %s
+// RUN: cd %S
+// RUN: %clang -### -S %s 2>&1 | FileCheck -check-prefix=CHECK-CWD %s
+// CHECK-CWD: -fdebug-compilation-dir={{.*}}Driver
diff --git a/clang/test/Misc/target-invalid-cpu-note/nvptx.c b/clang/test/Misc/target-invalid-cpu-note/nvptx.c
index b5209ff..b90f26e 100644
--- a/clang/test/Misc/target-invalid-cpu-note/nvptx.c
+++ b/clang/test/Misc/target-invalid-cpu-note/nvptx.c
@@ -30,8 +30,12 @@
 // CHECK-SAME: {{^}}, sm_100a
 // CHECK-SAME: {{^}}, sm_101
 // CHECK-SAME: {{^}}, sm_101a
+// CHECK-SAME: {{^}}, sm_103
+// CHECK-SAME: {{^}}, sm_103a
 // CHECK-SAME: {{^}}, sm_120
 // CHECK-SAME: {{^}}, sm_120a
+// CHECK-SAME: {{^}}, sm_121
+// CHECK-SAME: {{^}}, sm_121a
 // CHECK-SAME: {{^}}, gfx600
 // CHECK-SAME: {{^}}, gfx601
 // CHECK-SAME: {{^}}, gfx602
diff --git a/clang/test/SemaHLSL/BuiltIns/D3DCOLORtoUBYTE4-errors.hlsl b/clang/test/SemaHLSL/BuiltIns/D3DCOLORtoUBYTE4-errors.hlsl
index e9ba851..e9bf4c9 100644
--- a/clang/test/SemaHLSL/BuiltIns/D3DCOLORtoUBYTE4-errors.hlsl
+++ b/clang/test/SemaHLSL/BuiltIns/D3DCOLORtoUBYTE4-errors.hlsl
@@ -25,5 +25,5 @@ struct S {
 int4 struct_arg(S v) {
     return D3DCOLORtoUBYTE4(v);
     // expected-error@-1 {{no matching function for call to 'D3DCOLORtoUBYTE4'}}
-    // expected-note@hlsl/hlsl_intrinsics.h:* {{candidate function not viable: no known conversion from 'S' to 'vector<float, 4>' (vector of 4 'float' values) for 1st argument}}
+    // expected-note@hlsl/hlsl_intrinsics.h:* {{candidate function not viable: no known conversion from 'S' to 'float4' (aka 'vector<float, 4>') for 1st argument}}
 }
diff --git a/clang/tools/clang-extdef-mapping/ClangExtDefMapGen.cpp b/clang/tools/clang-extdef-mapping/ClangExtDefMapGen.cpp
index 0b621b8..ddb2944 100644
--- a/clang/tools/clang-extdef-mapping/ClangExtDefMapGen.cpp
+++ b/clang/tools/clang-extdef-mapping/ClangExtDefMapGen.cpp
@@ -134,10 +134,9 @@ GetDiagnosticsEngine(DiagnosticOptions &DiagOpts) {
   TextDiagnosticPrinter *DiagClient =
       new TextDiagnosticPrinter(llvm::errs(), DiagOpts);
   DiagClient->setPrefix("clang-extdef-mappping");
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
 
-  IntrusiveRefCntPtr<DiagnosticsEngine> DiagEngine(
-      new DiagnosticsEngine(DiagID, DiagOpts, DiagClient));
+  auto DiagEngine = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
+      DiagnosticIDs::create(), DiagOpts, DiagClient);
   Diags.swap(DiagEngine);
 
   // Retain this one time so it's not destroyed by ASTUnit::LoadFromASTFile
diff --git a/clang/tools/clang-format/ClangFormat.cpp b/clang/tools/clang-format/ClangFormat.cpp
index afc40e9..5f6502f 100644
--- a/clang/tools/clang-format/ClangFormat.cpp
+++ b/clang/tools/clang-format/ClangFormat.cpp
@@ -241,8 +241,7 @@ static bool fillRanges(MemoryBuffer *Code,
       makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
   FileManager Files(FileSystemOptions(), InMemoryFileSystem);
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs), DiagOpts);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts);
   SourceManager Sources(Diagnostics, Files);
   const auto ID = createInMemoryFile("<irrelevant>", *Code, Sources, Files,
                                      InMemoryFileSystem.get());
@@ -517,9 +516,8 @@ static bool format(StringRef FileName, bool ErrorOnIncompleteFormat = false) {
 
     DiagnosticOptions DiagOpts;
     ClangFormatDiagConsumer IgnoreDiagnostics;
-    DiagnosticsEngine Diagnostics(
-        IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs), DiagOpts,
-        &IgnoreDiagnostics, false);
+    DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts,
+                                  &IgnoreDiagnostics, false);
     SourceManager Sources(Diagnostics, Files);
     FileID ID = createInMemoryFile(AssumedFileName, *Code, Sources, Files,
                                    InMemoryFileSystem.get());
diff --git a/clang/tools/clang-fuzzer/handle-cxx/handle_cxx.cpp b/clang/tools/clang-fuzzer/handle-cxx/handle_cxx.cpp
index 98925c0..8259361 100644
--- a/clang/tools/clang-fuzzer/handle-cxx/handle_cxx.cpp
+++ b/clang/tools/clang-fuzzer/handle-cxx/handle_cxx.cpp
@@ -33,9 +33,8 @@ void clang_fuzzer::HandleCXX(const std::string &S,
       new FileManager(FileSystemOptions()));
   IgnoringDiagConsumer Diags;
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diagnostics(
-      IntrusiveRefCntPtr<clang::DiagnosticIDs>(new DiagnosticIDs()), DiagOpts,
-      &Diags, false);
+  DiagnosticsEngine Diagnostics(DiagnosticIDs::create(), DiagOpts, &Diags,
+                                false);
   std::unique_ptr<clang::CompilerInvocation> Invocation(
       tooling::newInvocation(&Diagnostics, CC1Args, /*BinaryName=*/nullptr));
   std::unique_ptr<llvm::MemoryBuffer> Input =
@@ -49,4 +48,3 @@ void clang_fuzzer::HandleCXX(const std::string &S,
   action->runInvocation(std::move(Invocation), Files.get(), PCHContainerOps,
                         &Diags);
 }
-
diff --git a/clang/tools/clang-installapi/ClangInstallAPI.cpp b/clang/tools/clang-installapi/ClangInstallAPI.cpp
index 509934e..70091fc 100644
--- a/clang/tools/clang-installapi/ClangInstallAPI.cpp
+++ b/clang/tools/clang-installapi/ClangInstallAPI.cpp
@@ -77,8 +77,8 @@ static bool run(ArrayRef<const char *> Args, const char *ProgName) {
       ArrayRef(Args).slice(1), MissingArgIndex, MissingArgCount);
   ParseDiagnosticArgs(DiagOpts, ParsedArgs);
 
-  IntrusiveRefCntPtr<DiagnosticsEngine> Diag = new clang::DiagnosticsEngine(
-      new clang::DiagnosticIDs(), DiagOpts,
+  auto Diag = llvm::makeIntrusiveRefCnt<clang::DiagnosticsEngine>(
+      clang::DiagnosticIDs::create(), DiagOpts,
       new clang::TextDiagnosticPrinter(llvm::errs(), DiagOpts));
 
   // Create file manager for all file operations and holding in-memory generated
diff --git a/clang/tools/clang-repl/CMakeLists.txt b/clang/tools/clang-repl/CMakeLists.txt
index 68d86dd..c3d14ce 100644
--- a/clang/tools/clang-repl/CMakeLists.txt
+++ b/clang/tools/clang-repl/CMakeLists.txt
@@ -19,14 +19,14 @@ if(MSVC)
   set_target_properties(clang-repl PROPERTIES WINDOWS_EXPORT_ALL_SYMBOLS 1)
 
   # RTTI/C++ symbols
-  set(clang_repl_exports ${clang_repl_exports} ??_7type_info@@6B@
-    ?__type_info_root_node@@3U__type_info_node@@A
-    ?nothrow@std@@3Unothrow_t@1@B
+  set(clang_repl_exports ${clang_repl_exports} ??_7type_info@@6B@,DATA
+    ?__type_info_root_node@@3U__type_info_node@@A,DATA
+    ?nothrow@std@@3Unothrow_t@1@B,DATA
   )
 
   # Compiler added symbols for static variables. NOT for VStudio < 2015
-  set(clang_repl_exports ${clang_repl_exports} _Init_thread_abort _Init_thread_epoch
-    _Init_thread_footer _Init_thread_header _tls_index
+  set(clang_repl_exports ${clang_repl_exports} _Init_thread_abort _Init_thread_epoch,DATA
+    _Init_thread_footer _Init_thread_header _tls_index,DATA
   )
 
   if(CMAKE_SIZEOF_VOID_P EQUAL 8)
@@ -50,7 +50,10 @@ if(MSVC)
   endif()
 
   # List to '/EXPORT:sym0 /EXPORT:sym1 /EXPORT:sym2 ...'
-  list(TRANSFORM clang_repl_exports PREPEND "LINKER:/EXPORT:")
+  # The 'SHELL' prefix tells CMake to use a space instead of comma as the
+  # separator between the driver and linker options, which we need since MSVC's
+  # linker uses `,DATA` as a suffix to indicate that data is being exported.
+  list(TRANSFORM clang_repl_exports PREPEND "LINKER:SHELL:/EXPORT:")
 
   set_property(TARGET clang-repl APPEND PROPERTY LINK_OPTIONS ${clang_repl_exports})
 
diff --git a/clang/tools/diagtool/ShowEnabledWarnings.cpp b/clang/tools/diagtool/ShowEnabledWarnings.cpp
index 0d1455d..bea0288 100644
--- a/clang/tools/diagtool/ShowEnabledWarnings.cpp
+++ b/clang/tools/diagtool/ShowEnabledWarnings.cpp
@@ -55,7 +55,6 @@ static char getCharForLevel(DiagnosticsEngine::Level Level) {
 
 static IntrusiveRefCntPtr<DiagnosticsEngine>
 createDiagnostics(unsigned int argc, char **argv) {
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagIDs(new DiagnosticIDs());
   DiagnosticOptions DiagOpts;
 
   // Buffer diagnostics from argument parsing so that we can output them using a
@@ -67,7 +66,8 @@ createDiagnostics(unsigned int argc, char **argv) {
   Args.push_back("diagtool");
   Args.append(argv, argv + argc);
   CreateInvocationOptions CIOpts;
-  CIOpts.Diags = new DiagnosticsEngine(DiagIDs, DiagOpts, DiagsBuffer);
+  CIOpts.Diags = llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(
+      DiagnosticIDs::create(), DiagOpts, DiagsBuffer);
   std::unique_ptr<CompilerInvocation> Invocation =
       createInvocation(Args, CIOpts);
   if (!Invocation)
diff --git a/clang/tools/diagtool/TreeView.cpp b/clang/tools/diagtool/TreeView.cpp
index 7e47c74..13b8015 100644
--- a/clang/tools/diagtool/TreeView.cpp
+++ b/clang/tools/diagtool/TreeView.cpp
@@ -32,7 +32,7 @@ public:
   static bool isIgnored(unsigned DiagID) {
     // FIXME: This feels like a hack.
     static DiagnosticOptions DiagOpts;
-    static clang::DiagnosticsEngine Diags(new DiagnosticIDs, DiagOpts);
+    static clang::DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts);
     return Diags.isIgnored(DiagID, SourceLocation());
   }
 
diff --git a/clang/tools/driver/cc1_main.cpp b/clang/tools/driver/cc1_main.cpp
index 9b1e390..854ab3e 100644
--- a/clang/tools/driver/cc1_main.cpp
+++ b/clang/tools/driver/cc1_main.cpp
@@ -217,7 +217,7 @@ static int PrintEnabledExtensions(const TargetOptions& TargetOpts) {
 int cc1_main(ArrayRef<const char *> Argv, const char *Argv0, void *MainAddr) {
   ensureSufficientStack();
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
 
   // Register the support for object-file-wrapped Clang modules.
   auto PCHOps = std::make_shared<PCHContainerOperations>();
diff --git a/clang/tools/driver/cc1as_main.cpp b/clang/tools/driver/cc1as_main.cpp
index f938e7e..e9243ca 100644
--- a/clang/tools/driver/cc1as_main.cpp
+++ b/clang/tools/driver/cc1as_main.cpp
@@ -662,8 +662,7 @@ int cc1as_main(ArrayRef<const char *> Argv, const char *Argv0, void *MainAddr) {
   TextDiagnosticPrinter *DiagClient =
       new TextDiagnosticPrinter(errs(), DiagOpts);
   DiagClient->setPrefix("clang -cc1as");
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-  DiagnosticsEngine Diags(DiagID, DiagOpts, DiagClient);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts, DiagClient);
 
   // Set an error handler, so that any LLVM backend diagnostics go through our
   // error handler.
diff --git a/clang/tools/driver/cc1gen_reproducer_main.cpp b/clang/tools/driver/cc1gen_reproducer_main.cpp
index 8d7171e..ddff0d0 100644
--- a/clang/tools/driver/cc1gen_reproducer_main.cpp
+++ b/clang/tools/driver/cc1gen_reproducer_main.cpp
@@ -119,8 +119,8 @@ generateReproducerForInvocationArguments(ArrayRef<const char *> Argv,
 
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-  DiagnosticsEngine Diags(DiagID, DiagOpts, new IgnoringDiagConsumer());
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
+                          new IgnoringDiagConsumer());
   auto VFS = llvm::vfs::getRealFileSystem();
   ProcessWarningOptions(Diags, DiagOpts, *VFS, /*ReportDiags=*/false);
   Driver TheDriver(ToolContext.Path, llvm::sys::getDefaultTargetTriple(), Diags,
diff --git a/clang/tools/driver/driver.cpp b/clang/tools/driver/driver.cpp
index 9ed8d4e..e5c3c4e 100644
--- a/clang/tools/driver/driver.cpp
+++ b/clang/tools/driver/driver.cpp
@@ -331,9 +331,7 @@ int clang_main(int Argc, char **Argv, const llvm::ToolContext &ToolContext) {
       new TextDiagnosticPrinter(llvm::errs(), *DiagOpts);
   FixupDiagPrefixExeName(DiagClient, ProgName);
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-
-  DiagnosticsEngine Diags(DiagID, *DiagOpts, DiagClient);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), *DiagOpts, DiagClient);
 
   if (!DiagOpts->DiagnosticSerializationFile.empty()) {
     auto SerializedConsumer =
diff --git a/clang/tools/libclang/CIndexCodeCompletion.cpp b/clang/tools/libclang/CIndexCodeCompletion.cpp
index 8f6729b..adac7c3 100644
--- a/clang/tools/libclang/CIndexCodeCompletion.cpp
+++ b/clang/tools/libclang/CIndexCodeCompletion.cpp
@@ -357,8 +357,8 @@ static std::atomic<unsigned> CodeCompletionResultObjects;
 AllocatedCXCodeCompleteResults::AllocatedCXCodeCompleteResults(
     IntrusiveRefCntPtr<FileManager> FileMgr)
     : CXCodeCompleteResults(),
-      Diag(new DiagnosticsEngine(
-          IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs), DiagOpts)),
+      Diag(llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(DiagnosticIDs::create(),
+                                                        DiagOpts)),
       FileMgr(std::move(FileMgr)),
       SourceMgr(new SourceManager(*Diag, *this->FileMgr)),
       CodeCompletionAllocator(
@@ -763,7 +763,7 @@ clang_codeCompleteAt_Impl(CXTranslationUnit TU, const char *complete_filename,
                     RemappedFiles, (options & CXCodeComplete_IncludeMacros),
                     (options & CXCodeComplete_IncludeCodePatterns),
                     IncludeBriefComments, Capture,
-                    CXXIdx->getPCHContainerOperations(), *Results->Diag,
+                    CXXIdx->getPCHContainerOperations(), Results->Diag,
                     Results->LangOpts, *Results->SourceMgr, *Results->FileMgr,
                     Results->Diagnostics, Results->TemporaryBuffers,
                     /*SyntaxOnlyAction=*/nullptr);
diff --git a/clang/unittests/AST/ASTVectorTest.cpp b/clang/unittests/AST/ASTVectorTest.cpp
index 66003b4..03da549 100644
--- a/clang/unittests/AST/ASTVectorTest.cpp
+++ b/clang/unittests/AST/ASTVectorTest.cpp
@@ -26,14 +26,13 @@ namespace {
 class ASTVectorTest : public ::testing::Test {
 protected:
   ASTVectorTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), Idents(LangOpts, nullptr),
         Ctxt(LangOpts, SourceMgr, Idents, Sels, Builtins, TU_Complete) {}
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/AST/CommentLexer.cpp b/clang/unittests/AST/CommentLexer.cpp
index dc10dae..99f4691 100644
--- a/clang/unittests/AST/CommentLexer.cpp
+++ b/clang/unittests/AST/CommentLexer.cpp
@@ -27,13 +27,12 @@ namespace {
 class CommentLexerTest : public ::testing::Test {
 protected:
   CommentLexerTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), Traits(Allocator, CommentOptions()) {}
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
@@ -2006,4 +2005,3 @@ TEST_F(CommentLexerTest, MultipleComments) {
 
 } // end namespace comments
 } // end namespace clang
-
diff --git a/clang/unittests/AST/CommentParser.cpp b/clang/unittests/AST/CommentParser.cpp
index 67fabe5..3bd2bdb 100644
--- a/clang/unittests/AST/CommentParser.cpp
+++ b/clang/unittests/AST/CommentParser.cpp
@@ -33,13 +33,12 @@ const bool MY_DEBUG = true;
 class CommentParserTest : public ::testing::Test {
 protected:
   CommentParserTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), Traits(Allocator, CommentOptions()) {}
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/AST/CommentTextTest.cpp b/clang/unittests/AST/CommentTextTest.cpp
index 84ec51a..675173c 100644
--- a/clang/unittests/AST/CommentTextTest.cpp
+++ b/clang/unittests/AST/CommentTextTest.cpp
@@ -44,7 +44,7 @@ protected:
     // shouldn't matter.
     RawComment Comment(SourceMgr, CommentRange, EmptyOpts, /*Merged=*/true);
     DiagnosticOptions DiagOpts;
-    DiagnosticsEngine Diags(new DiagnosticIDs, DiagOpts);
+    DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts);
     return Comment.getFormattedText(SourceMgr, Diags);
   }
 };
diff --git a/clang/unittests/Analysis/MacroExpansionContextTest.cpp b/clang/unittests/Analysis/MacroExpansionContextTest.cpp
index af7a8d6..25a76ed 100644
--- a/clang/unittests/Analysis/MacroExpansionContextTest.cpp
+++ b/clang/unittests/Analysis/MacroExpansionContextTest.cpp
@@ -36,8 +36,7 @@ protected:
       : InMemoryFileSystem(
             llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>()),
         FileMgr(FileSystemOptions(), InMemoryFileSystem),
-        DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions()) {
     TargetOpts->Triple = "x86_64-pc-linux-unknown";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -46,7 +45,6 @@ protected:
 
   IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> InMemoryFileSystem;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Analysis/UnsafeBufferUsageTest.cpp b/clang/unittests/Analysis/UnsafeBufferUsageTest.cpp
index 9da2c58..f795918 100644
--- a/clang/unittests/Analysis/UnsafeBufferUsageTest.cpp
+++ b/clang/unittests/Analysis/UnsafeBufferUsageTest.cpp
@@ -12,13 +12,12 @@ namespace {
 class UnsafeBufferUsageTest : public ::testing::Test {
 protected:
   UnsafeBufferUsageTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr) {}
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
@@ -58,4 +57,4 @@ TEST_F(UnsafeBufferUsageTest, FixItHintsConflict) {
 
   Fixes = {H1, H2, H3, MkDummyHint(2, 23) /* overlaps H1, H2, and H3 */};
   EXPECT_TRUE(internal::anyConflict(Fixes, SourceMgr));
-}
\ No newline at end of file
+}
diff --git a/clang/unittests/Basic/DiagnosticTest.cpp b/clang/unittests/Basic/DiagnosticTest.cpp
index b0a034e..4b3af00 100644
--- a/clang/unittests/Basic/DiagnosticTest.cpp
+++ b/clang/unittests/Basic/DiagnosticTest.cpp
@@ -47,7 +47,7 @@ using testing::IsEmpty;
 // Check that DiagnosticErrorTrap works with SuppressAllDiagnostics.
 TEST(DiagnosticTest, suppressAndTrap) {
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(new DiagnosticIDs(), DiagOpts,
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                           new IgnoringDiagConsumer());
   Diags.setSuppressAllDiagnostics(true);
 
@@ -78,7 +78,7 @@ TEST(DiagnosticTest, suppressAndTrap) {
 TEST(DiagnosticTest, fatalsAsError) {
   for (unsigned FatalsAsError = 0; FatalsAsError != 2; ++FatalsAsError) {
     DiagnosticOptions DiagOpts;
-    DiagnosticsEngine Diags(new DiagnosticIDs(), DiagOpts,
+    DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                             new IgnoringDiagConsumer());
     Diags.setFatalsAsError(FatalsAsError);
 
@@ -102,7 +102,7 @@ TEST(DiagnosticTest, fatalsAsError) {
 
 TEST(DiagnosticTest, tooManyErrorsIsAlwaysFatal) {
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(new DiagnosticIDs(), DiagOpts,
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                           new IgnoringDiagConsumer());
   Diags.setFatalsAsError(true);
 
@@ -119,7 +119,7 @@ TEST(DiagnosticTest, tooManyErrorsIsAlwaysFatal) {
 // Check that soft RESET works as intended
 TEST(DiagnosticTest, softReset) {
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(new DiagnosticIDs(), DiagOpts,
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                           new IgnoringDiagConsumer());
 
   unsigned numWarnings = 0U, numErrors = 0U;
@@ -143,7 +143,7 @@ TEST(DiagnosticTest, softReset) {
 
 TEST(DiagnosticTest, diagnosticError) {
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(new DiagnosticIDs(), DiagOpts,
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                           new IgnoringDiagConsumer());
   PartialDiagnostic::DiagStorageAllocator Alloc;
   llvm::Expected<std::pair<int, int>> Value = DiagnosticError::create(
@@ -166,7 +166,7 @@ TEST(DiagnosticTest, diagnosticError) {
 
 TEST(DiagnosticTest, storedDiagEmptyWarning) {
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(new DiagnosticIDs(), DiagOpts);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts);
 
   class CaptureDiagnosticConsumer : public DiagnosticConsumer {
   public:
@@ -197,7 +197,7 @@ protected:
   llvm::IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> FS =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags{new DiagnosticIDs(), DiagOpts};
+  DiagnosticsEngine Diags{DiagnosticIDs::create(), DiagOpts};
 
   llvm::ArrayRef<StoredDiagnostic> diags() {
     return CaptureConsumer.StoredDiags;
diff --git a/clang/unittests/Basic/SarifTest.cpp b/clang/unittests/Basic/SarifTest.cpp
index c6be7ee..089b6cb 100644
--- a/clang/unittests/Basic/SarifTest.cpp
+++ b/clang/unittests/Basic/SarifTest.cpp
@@ -44,13 +44,11 @@ protected:
       : InMemoryFileSystem(
             llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>()),
         FileMgr(FileSystemOptions(), InMemoryFileSystem),
-        DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr) {}
 
   IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> InMemoryFileSystem;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Basic/SourceManagerTest.cpp b/clang/unittests/Basic/SourceManagerTest.cpp
index cbe047b..04b23dd 100644
--- a/clang/unittests/Basic/SourceManagerTest.cpp
+++ b/clang/unittests/Basic/SourceManagerTest.cpp
@@ -40,8 +40,8 @@ namespace {
 class SourceManagerTest : public ::testing::Test {
 protected:
   SourceManagerTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     TargetOpts->Triple = "x86_64-apple-darwin11.1.0";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -49,7 +49,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Driver/DXCModeTest.cpp b/clang/unittests/Driver/DXCModeTest.cpp
index 3a2e9ec..e7d8137 100644
--- a/clang/unittests/Driver/DXCModeTest.cpp
+++ b/clang/unittests/Driver/DXCModeTest.cpp
@@ -55,8 +55,6 @@ static void validateTargetProfile(
 }
 
 TEST(DxcModeTest, TargetProfileValidation) {
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-
   auto InMemoryFileSystem =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
 
@@ -65,7 +63,7 @@ TEST(DxcModeTest, TargetProfileValidation) {
 
   auto *DiagConsumer = new SimpleDiagnosticConsumer;
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts, DiagConsumer);
 
   validateTargetProfile("-Tvs_6_0", "dxilv1.0--shadermodel6.0-vertex",
                         InMemoryFileSystem, Diags);
@@ -105,8 +103,6 @@ TEST(DxcModeTest, TargetProfileValidation) {
 }
 
 TEST(DxcModeTest, ValidatorVersionValidation) {
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
-
   auto InMemoryFileSystem =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
 
@@ -115,7 +111,7 @@ TEST(DxcModeTest, ValidatorVersionValidation) {
 
   auto *DiagConsumer = new SimpleDiagnosticConsumer;
   DiagnosticOptions DiagOpts;
-  DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer);
+  DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts, DiagConsumer);
   Driver TheDriver("/bin/clang", "", Diags, "", InMemoryFileSystem);
   std::unique_ptr<Compilation> C(TheDriver.BuildCompilation(
       {"clang", "--driver-mode=dxc", "-Tlib_6_7", "foo.hlsl"}));
diff --git a/clang/unittests/Driver/SanitizerArgsTest.cpp b/clang/unittests/Driver/SanitizerArgsTest.cpp
index 41884c0..7847947 100644
--- a/clang/unittests/Driver/SanitizerArgsTest.cpp
+++ b/clang/unittests/Driver/SanitizerArgsTest.cpp
@@ -53,7 +53,7 @@ protected:
     assert(!DriverInstance && "Running twice is not allowed");
 
     DiagnosticOptions DiagOpts;
-    DiagnosticsEngine Diags(new DiagnosticIDs, DiagOpts,
+    DiagnosticsEngine Diags(DiagnosticIDs::create(), DiagOpts,
                             new TextDiagnosticPrinter(llvm::errs(), DiagOpts));
     DriverInstance.emplace(ClangBinary, "x86_64-unknown-linux-gnu", Diags,
                            "clang LLVM compiler", prepareFS(ExtraFiles));
diff --git a/clang/unittests/Driver/SimpleDiagnosticConsumer.h b/clang/unittests/Driver/SimpleDiagnosticConsumer.h
index a90524c..7ab409b 100644
--- a/clang/unittests/Driver/SimpleDiagnosticConsumer.h
+++ b/clang/unittests/Driver/SimpleDiagnosticConsumer.h
@@ -42,13 +42,12 @@ struct SimpleDiagnosticConsumer : public clang::DiagnosticConsumer {
 // for testing situations where it will only ever be used for emitting
 // diagnostics, such as being passed to `MultilibSet::select`.
 inline clang::driver::Driver diagnostic_test_driver() {
-  llvm::IntrusiveRefCntPtr<clang::DiagnosticIDs> DiagID(
-      new clang::DiagnosticIDs());
   auto InMemoryFileSystem =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
   auto *DiagConsumer = new SimpleDiagnosticConsumer;
   clang::DiagnosticOptions DiagOpts;
-  clang::DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer);
+  clang::DiagnosticsEngine Diags(clang::DiagnosticIDs::create(), DiagOpts,
+                                 DiagConsumer);
   return clang::driver::Driver("/bin/clang", "", Diags, "", InMemoryFileSystem);
 }
 
diff --git a/clang/unittests/Driver/ToolChainTest.cpp b/clang/unittests/Driver/ToolChainTest.cpp
index 03f2fcf..4fa2729 100644
--- a/clang/unittests/Driver/ToolChainTest.cpp
+++ b/clang/unittests/Driver/ToolChainTest.cpp
@@ -40,7 +40,7 @@ namespace {
 TEST(ToolChainTest, VFSGCCInstallation) {
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   auto InMemoryFileSystem =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
@@ -137,7 +137,7 @@ TEST(ToolChainTest, VFSGCCInstallation) {
 TEST(ToolChainTest, VFSGCCInstallationRelativeDir) {
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto InMemoryFileSystem =
@@ -176,7 +176,7 @@ TEST(ToolChainTest, VFSGCCInstallationRelativeDir) {
 TEST(ToolChainTest, VFSSolarisMultiGCCInstallation) {
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   auto InMemoryFileSystem =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
@@ -340,7 +340,7 @@ MATCHER_P(jobHasArgs, Substr, "") {
 TEST(ToolChainTest, VFSGnuLibcxxPathNoSysroot) {
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   auto InMemoryFileSystem =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
@@ -371,7 +371,7 @@ TEST(ToolChainTest, VFSGnuLibcxxPathNoSysroot) {
 TEST(ToolChainTest, DefaultDriverMode) {
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto InMemoryFileSystem =
@@ -402,7 +402,7 @@ TEST(ToolChainTest, DefaultDriverMode) {
   EXPECT_TRUE(CLDriver.IsCLMode());
 }
 TEST(ToolChainTest, InvalidArgument) {
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
@@ -517,7 +517,7 @@ TEST(ToolChainTest, GetTargetAndMode) {
 TEST(ToolChainTest, CommandOutput) {
   DiagnosticOptions DiagOpts;
 
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto InMemoryFileSystem =
@@ -545,7 +545,7 @@ TEST(ToolChainTest, CommandOutput) {
 
 TEST(ToolChainTest, PostCallback) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto InMemoryFileSystem =
@@ -598,7 +598,7 @@ TEST(ToolChainTest, UEFICallingConventionTest) {
 
 TEST(ToolChainTest, UEFIDefaultDebugFormatTest) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto InMemoryFileSystem =
@@ -640,7 +640,7 @@ struct SimpleDiagnosticConsumer : public DiagnosticConsumer {
 
 TEST(ToolChainTest, ConfigFileSearch) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto FS = llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
@@ -716,7 +716,7 @@ struct FileSystemWithError : public llvm::vfs::FileSystem {
 
 TEST(ToolChainTest, ConfigFileError) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   std::unique_ptr<SimpleDiagnosticConsumer> DiagConsumer(
       new SimpleDiagnosticConsumer());
   DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer.get(), false);
@@ -737,7 +737,7 @@ TEST(ToolChainTest, ConfigFileError) {
 
 TEST(ToolChainTest, BadConfigFile) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   std::unique_ptr<SimpleDiagnosticConsumer> DiagConsumer(
       new SimpleDiagnosticConsumer());
   DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer.get(), false);
@@ -810,7 +810,7 @@ TEST(ToolChainTest, BadConfigFile) {
 
 TEST(ToolChainTest, ConfigInexistentInclude) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   std::unique_ptr<SimpleDiagnosticConsumer> DiagConsumer(
       new SimpleDiagnosticConsumer());
   DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer.get(), false);
@@ -850,7 +850,7 @@ TEST(ToolChainTest, ConfigInexistentInclude) {
 
 TEST(ToolChainTest, ConfigRecursiveInclude) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   std::unique_ptr<SimpleDiagnosticConsumer> DiagConsumer(
       new SimpleDiagnosticConsumer());
   DiagnosticsEngine Diags(DiagID, DiagOpts, DiagConsumer.get(), false);
@@ -895,7 +895,7 @@ TEST(ToolChainTest, ConfigRecursiveInclude) {
 
 TEST(ToolChainTest, NestedConfigFile) {
   DiagnosticOptions DiagOpts;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs());
+  IntrusiveRefCntPtr<DiagnosticIDs> DiagID = DiagnosticIDs::create();
   struct TestDiagnosticConsumer : public DiagnosticConsumer {};
   DiagnosticsEngine Diags(DiagID, DiagOpts, new TestDiagnosticConsumer);
   auto FS = llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
diff --git a/clang/unittests/Frontend/CompilerInstanceTest.cpp b/clang/unittests/Frontend/CompilerInstanceTest.cpp
index 459a386..7c1b653 100644
--- a/clang/unittests/Frontend/CompilerInstanceTest.cpp
+++ b/clang/unittests/Frontend/CompilerInstanceTest.cpp
@@ -71,7 +71,7 @@ TEST(CompilerInstance, DefaultVFSOverlayFromInvocation) {
   // Create a minimal CompilerInstance which should use the VFS we specified
   // in the CompilerInvocation (as we don't explicitly set our own).
   CompilerInstance Instance(std::move(CInvok));
-  Instance.setDiagnostics(Diags.get());
+  Instance.setDiagnostics(Diags);
   Instance.createFileManager();
 
   // Check if the virtual file exists which means that our VFS is used by the
@@ -135,7 +135,7 @@ TEST(CompilerInstance, MultipleInputsCleansFileIDs) {
   ASSERT_TRUE(CInvok) << "could not create compiler invocation";
 
   CompilerInstance Instance(std::move(CInvok));
-  Instance.setDiagnostics(Diags.get());
+  Instance.setDiagnostics(Diags);
   Instance.createFileManager(VFS);
 
   // Run once for `a.cc` and then for `a.h`. This makes sure we get the same
diff --git a/clang/unittests/Frontend/SearchPathTest.cpp b/clang/unittests/Frontend/SearchPathTest.cpp
index 014f482..a8c16fe 100644
--- a/clang/unittests/Frontend/SearchPathTest.cpp
+++ b/clang/unittests/Frontend/SearchPathTest.cpp
@@ -40,7 +40,7 @@ namespace {
 class SearchPathTest : public ::testing::Test {
 protected:
   SearchPathTest()
-      : Diags(new DiagnosticIDs(), DiagOpts, new IgnoringDiagConsumer()),
+      : Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         VFS(llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>()),
         FileMgr(FileSystemOptions(), VFS), SourceMgr(Diags, FileMgr),
         Invocation(std::make_unique<CompilerInvocation>()) {}
diff --git a/clang/unittests/Frontend/TextDiagnosticTest.cpp b/clang/unittests/Frontend/TextDiagnosticTest.cpp
index 8fd8187..622dbc5 100644
--- a/clang/unittests/Frontend/TextDiagnosticTest.cpp
+++ b/clang/unittests/Frontend/TextDiagnosticTest.cpp
@@ -36,9 +36,8 @@ TEST(TextDiagnostic, ShowLine) {
   // Create dummy FileManager and SourceManager.
   FileSystemOptions FSOpts;
   FileManager FileMgr(FSOpts);
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID(new DiagnosticIDs);
   DiagnosticOptions DiagEngineOpts;
-  DiagnosticsEngine DiagEngine(DiagID, DiagEngineOpts,
+  DiagnosticsEngine DiagEngine(DiagnosticIDs::create(), DiagEngineOpts,
                                new IgnoringDiagConsumer());
   SourceManager SrcMgr(DiagEngine, FileMgr);
 
diff --git a/clang/unittests/Lex/HeaderSearchTest.cpp b/clang/unittests/Lex/HeaderSearchTest.cpp
index 45eb855..0213bfe 100644
--- a/clang/unittests/Lex/HeaderSearchTest.cpp
+++ b/clang/unittests/Lex/HeaderSearchTest.cpp
@@ -29,8 +29,8 @@ class HeaderSearchTest : public ::testing::Test {
 protected:
   HeaderSearchTest()
       : VFS(llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>()),
-        FileMgr(FileMgrOpts, VFS), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+        FileMgr(FileMgrOpts, VFS),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions),
         Search(HSOpts, SourceMgr, Diags, LangOpts, Target.get()) {
     TargetOpts->Triple = "x86_64-apple-darwin11.1.0";
@@ -80,7 +80,6 @@ protected:
   IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> VFS;
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Lex/LexerTest.cpp b/clang/unittests/Lex/LexerTest.cpp
index 86df872..56d73ce 100644
--- a/clang/unittests/Lex/LexerTest.cpp
+++ b/clang/unittests/Lex/LexerTest.cpp
@@ -41,8 +41,8 @@ using testing::ElementsAre;
 class LexerTest : public ::testing::Test {
 protected:
   LexerTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     TargetOpts->Triple = "x86_64-apple-darwin11.1.0";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -102,7 +102,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Lex/ModuleDeclStateTest.cpp b/clang/unittests/Lex/ModuleDeclStateTest.cpp
index 6ecba4d..adc6cf1 100644
--- a/clang/unittests/Lex/ModuleDeclStateTest.cpp
+++ b/clang/unittests/Lex/ModuleDeclStateTest.cpp
@@ -54,8 +54,8 @@ public:
 class ModuleDeclStateTest : public ::testing::Test {
 protected:
   ModuleDeclStateTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     TargetOpts->Triple = "x86_64-unknown-linux-gnu";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -93,7 +93,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Lex/PPCallbacksTest.cpp b/clang/unittests/Lex/PPCallbacksTest.cpp
index a186246..990689c 100644
--- a/clang/unittests/Lex/PPCallbacksTest.cpp
+++ b/clang/unittests/Lex/PPCallbacksTest.cpp
@@ -136,7 +136,7 @@ protected:
       : InMemoryFileSystem(
             llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>()),
         FileMgr(FileSystemOptions(), InMemoryFileSystem),
-        DiagID(new DiagnosticIDs()),
+        DiagID(DiagnosticIDs::create()),
         Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions()) {
     TargetOpts->Triple = "x86_64-apple-darwin11.1.0";
diff --git a/clang/unittests/Lex/PPConditionalDirectiveRecordTest.cpp b/clang/unittests/Lex/PPConditionalDirectiveRecordTest.cpp
index 54c1d02..4a88bd4 100644
--- a/clang/unittests/Lex/PPConditionalDirectiveRecordTest.cpp
+++ b/clang/unittests/Lex/PPConditionalDirectiveRecordTest.cpp
@@ -29,8 +29,8 @@ namespace {
 class PPConditionalDirectiveRecordTest : public ::testing::Test {
 protected:
   PPConditionalDirectiveRecordTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     TargetOpts->Triple = "x86_64-apple-darwin11.1.0";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -38,7 +38,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Lex/PPDependencyDirectivesTest.cpp b/clang/unittests/Lex/PPDependencyDirectivesTest.cpp
index dec22a2..15cc283 100644
--- a/clang/unittests/Lex/PPDependencyDirectivesTest.cpp
+++ b/clang/unittests/Lex/PPDependencyDirectivesTest.cpp
@@ -31,8 +31,8 @@ namespace {
 class PPDependencyDirectivesTest : public ::testing::Test {
 protected:
   PPDependencyDirectivesTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     TargetOpts->Triple = "x86_64-apple-macos12";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -40,7 +40,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Lex/PPMemoryAllocationsTest.cpp b/clang/unittests/Lex/PPMemoryAllocationsTest.cpp
index 4d83003..f873774 100644
--- a/clang/unittests/Lex/PPMemoryAllocationsTest.cpp
+++ b/clang/unittests/Lex/PPMemoryAllocationsTest.cpp
@@ -27,8 +27,8 @@ namespace {
 class PPMemoryAllocationsTest : public ::testing::Test {
 protected:
   PPMemoryAllocationsTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Diags(DiagID, DiagOpts, new IgnoringDiagConsumer()),
+      : FileMgr(FileMgrOpts),
+        Diags(DiagnosticIDs::create(), DiagOpts, new IgnoringDiagConsumer()),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     TargetOpts->Triple = "x86_64-apple-darwin11.1.0";
     Target = TargetInfo::CreateTargetInfo(Diags, *TargetOpts);
@@ -36,7 +36,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   DiagnosticsEngine Diags;
   SourceManager SourceMgr;
diff --git a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
index 4d08f8d..44f6b04 100644
--- a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
+++ b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
@@ -71,8 +71,8 @@ public:
 class ParseHLSLRootSignatureTest : public ::testing::Test {
 protected:
   ParseHLSLRootSignatureTest()
-      : FileMgr(FileMgrOpts), DiagID(new DiagnosticIDs()),
-        Consumer(new ExpectedDiagConsumer()), Diags(DiagID, DiagOpts, Consumer),
+      : FileMgr(FileMgrOpts), Consumer(new ExpectedDiagConsumer()),
+        Diags(DiagnosticIDs::create(), DiagOpts, Consumer),
         SourceMgr(Diags, FileMgr), TargetOpts(new TargetOptions) {
     // This is an arbitrarily chosen target triple to create the target info.
     TargetOpts->Triple = "dxil";
@@ -114,7 +114,6 @@ protected:
 
   FileSystemOptions FileMgrOpts;
   FileManager FileMgr;
-  IntrusiveRefCntPtr<DiagnosticIDs> DiagID;
   DiagnosticOptions DiagOpts;
   ExpectedDiagConsumer *Consumer;
   DiagnosticsEngine Diags;
diff --git a/clang/unittests/Sema/SemaNoloadLookupTest.cpp b/clang/unittests/Sema/SemaNoloadLookupTest.cpp
index 5a04f42..e565372 100644
--- a/clang/unittests/Sema/SemaNoloadLookupTest.cpp
+++ b/clang/unittests/Sema/SemaNoloadLookupTest.cpp
@@ -82,7 +82,7 @@ public:
     EXPECT_TRUE(Invocation);
 
     CompilerInstance Instance(std::move(Invocation));
-    Instance.setDiagnostics(Diags.get());
+    Instance.setDiagnostics(Diags);
     Instance.getFrontendOpts().OutputFile = CacheBMIPath;
     GenerateReducedModuleInterfaceAction Action;
     EXPECT_TRUE(Instance.ExecuteAction(Action));
diff --git a/clang/unittests/Serialization/ForceCheckFileInputTest.cpp b/clang/unittests/Serialization/ForceCheckFileInputTest.cpp
index 970eeef..92ff76b 100644
--- a/clang/unittests/Serialization/ForceCheckFileInputTest.cpp
+++ b/clang/unittests/Serialization/ForceCheckFileInputTest.cpp
@@ -87,7 +87,7 @@ export int aa = 43;
     Buf->release();
 
     CompilerInstance Instance(std::move(Invocation));
-    Instance.setDiagnostics(Diags.get());
+    Instance.setDiagnostics(Diags);
 
     Instance.getFrontendOpts().OutputFile = BMIPath;
 
@@ -122,7 +122,7 @@ export int aa = 43;
 
     CompilerInstance Clang(std::move(Invocation));
 
-    Clang.setDiagnostics(Diags.get());
+    Clang.setDiagnostics(Diags);
     FileManager *FM = Clang.createFileManager(CIOpts.VFS);
     Clang.createSourceManager(*FM);
 
diff --git a/clang/unittests/Serialization/LoadSpecLazilyTest.cpp b/clang/unittests/Serialization/LoadSpecLazilyTest.cpp
index 6315474..d7b5549 100644
--- a/clang/unittests/Serialization/LoadSpecLazilyTest.cpp
+++ b/clang/unittests/Serialization/LoadSpecLazilyTest.cpp
@@ -80,7 +80,7 @@ public:
     EXPECT_TRUE(Invocation);
 
     CompilerInstance Instance(std::move(Invocation));
-    Instance.setDiagnostics(Diags.get());
+    Instance.setDiagnostics(Diags);
     Instance.getFrontendOpts().OutputFile = CacheBMIPath;
     // Avoid memory leaks.
     Instance.getFrontendOpts().DisableFree = false;
diff --git a/clang/unittests/Serialization/ModuleCacheTest.cpp b/clang/unittests/Serialization/ModuleCacheTest.cpp
index de6e13a..1f64401 100644
--- a/clang/unittests/Serialization/ModuleCacheTest.cpp
+++ b/clang/unittests/Serialization/ModuleCacheTest.cpp
@@ -121,7 +121,7 @@ TEST_F(ModuleCacheTest, CachedModuleNewPath) {
       createInvocationAndEnableFree(Args, CIOpts);
   ASSERT_TRUE(Invocation);
   CompilerInstance Instance(std::move(Invocation));
-  Instance.setDiagnostics(Diags.get());
+  Instance.setDiagnostics(Diags);
   SyntaxOnlyAction Action;
   ASSERT_TRUE(Instance.ExecuteAction(Action));
   ASSERT_FALSE(Diags->hasErrorOccurred());
@@ -145,7 +145,7 @@ TEST_F(ModuleCacheTest, CachedModuleNewPath) {
   CompilerInstance Instance2(std::move(Invocation2),
                              Instance.getPCHContainerOperations(),
                              &Instance.getModuleCache());
-  Instance2.setDiagnostics(Diags.get());
+  Instance2.setDiagnostics(Diags);
   SyntaxOnlyAction Action2;
   ASSERT_FALSE(Instance2.ExecuteAction(Action2));
   ASSERT_TRUE(Diags->hasErrorOccurred());
@@ -171,7 +171,7 @@ TEST_F(ModuleCacheTest, CachedModuleNewPathAllowErrors) {
       createInvocationAndEnableFree(Args, CIOpts);
   ASSERT_TRUE(Invocation);
   CompilerInstance Instance(std::move(Invocation));
-  Instance.setDiagnostics(Diags.get());
+  Instance.setDiagnostics(Diags);
   SyntaxOnlyAction Action;
   ASSERT_TRUE(Instance.ExecuteAction(Action));
   ASSERT_FALSE(Diags->hasErrorOccurred());
@@ -189,7 +189,7 @@ TEST_F(ModuleCacheTest, CachedModuleNewPathAllowErrors) {
   CompilerInstance Instance2(std::move(Invocation2),
                              Instance.getPCHContainerOperations(),
                              &Instance.getModuleCache());
-  Instance2.setDiagnostics(Diags.get());
+  Instance2.setDiagnostics(Diags);
   SyntaxOnlyAction Action2;
   ASSERT_FALSE(Instance2.ExecuteAction(Action2));
   ASSERT_TRUE(Diags->hasErrorOccurred());
diff --git a/clang/unittests/Serialization/NoCommentsTest.cpp b/clang/unittests/Serialization/NoCommentsTest.cpp
index 05efeef..ed96c7c 100644
--- a/clang/unittests/Serialization/NoCommentsTest.cpp
+++ b/clang/unittests/Serialization/NoCommentsTest.cpp
@@ -99,7 +99,7 @@ void foo() {}
   ASSERT_TRUE(Invocation);
 
   CompilerInstance Instance(std::move(Invocation));
-  Instance.setDiagnostics(Diags.get());
+  Instance.setDiagnostics(Diags);
   Instance.getFrontendOpts().OutputFile = CacheBMIPath;
   GenerateReducedModuleInterfaceAction Action;
   ASSERT_TRUE(Instance.ExecuteAction(Action));
diff --git a/clang/unittests/Serialization/PreambleInNamedModulesTest.cpp b/clang/unittests/Serialization/PreambleInNamedModulesTest.cpp
index c43520f..f9d7736 100644
--- a/clang/unittests/Serialization/PreambleInNamedModulesTest.cpp
+++ b/clang/unittests/Serialization/PreambleInNamedModulesTest.cpp
@@ -101,7 +101,7 @@ export using ::E;
 
   PreambleCallbacks Callbacks;
   llvm::ErrorOr<PrecompiledPreamble> BuiltPreamble = PrecompiledPreamble::Build(
-      *Invocation, Buffer.get(), Bounds, *Diags, VFS,
+      *Invocation, Buffer.get(), Bounds, Diags, VFS,
       std::make_shared<PCHContainerOperations>(),
       /*StoreInMemory=*/false, /*StoragePath=*/TestDir, Callbacks);
 
@@ -112,7 +112,7 @@ export using ::E;
   BuiltPreamble->OverridePreamble(*Invocation, VFS, Buffer.get());
 
   auto Clang = std::make_unique<CompilerInstance>(std::move(Invocation));
-  Clang->setDiagnostics(Diags.get());
+  Clang->setDiagnostics(Diags);
 
   if (auto VFSWithRemapping = createVFSFromCompilerInvocation(
           Clang->getInvocation(), Clang->getDiagnostics(), VFS))
diff --git a/clang/unittests/Serialization/VarDeclConstantInitTest.cpp b/clang/unittests/Serialization/VarDeclConstantInitTest.cpp
index 5b2988e..743f851 100644
--- a/clang/unittests/Serialization/VarDeclConstantInitTest.cpp
+++ b/clang/unittests/Serialization/VarDeclConstantInitTest.cpp
@@ -106,7 +106,7 @@ export namespace Fibonacci
   Invocation->getFrontendOpts().DisableFree = false;
 
   CompilerInstance Instance(std::move(Invocation));
-  Instance.setDiagnostics(Diags.get());
+  Instance.setDiagnostics(Diags);
 
   std::string CacheBMIPath = llvm::Twine(TestDir + "/Cached.pcm").str();
   Instance.getFrontendOpts().OutputFile = CacheBMIPath;
diff --git a/clang/unittests/Tooling/RewriterTestContext.h b/clang/unittests/Tooling/RewriterTestContext.h
index c478b3f..020ef60 100644
--- a/clang/unittests/Tooling/RewriterTestContext.h
+++ b/clang/unittests/Tooling/RewriterTestContext.h
@@ -49,8 +49,7 @@ struct RewriterDiagnosticConsumer : public DiagnosticConsumer {
 class RewriterTestContext {
  public:
    RewriterTestContext()
-       : Diagnostics(IntrusiveRefCntPtr<DiagnosticIDs>(new DiagnosticIDs),
-                     DiagOpts),
+       : Diagnostics(DiagnosticIDs::create(), DiagOpts),
          InMemoryFileSystem(
              llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>()),
          OverlayFileSystem(
diff --git a/clang/unittests/Tooling/Syntax/TokensTest.cpp b/clang/unittests/Tooling/Syntax/TokensTest.cpp
index a643738..e86793f 100644
--- a/clang/unittests/Tooling/Syntax/TokensTest.cpp
+++ b/clang/unittests/Tooling/Syntax/TokensTest.cpp
@@ -133,7 +133,7 @@ public:
     CI->getPreprocessorOpts().addRemappedFile(
         FileName, llvm::MemoryBuffer::getMemBufferCopy(Code).release());
     CompilerInstance Compiler(std::move(CI));
-    Compiler.setDiagnostics(Diags.get());
+    Compiler.setDiagnostics(Diags);
     Compiler.setFileManager(FileMgr.get());
     Compiler.setSourceManager(SourceMgr.get());
 
@@ -250,7 +250,8 @@ public:
   // Data fields.
   DiagnosticOptions DiagOpts;
   llvm::IntrusiveRefCntPtr<DiagnosticsEngine> Diags =
-      new DiagnosticsEngine(new DiagnosticIDs, DiagOpts);
+      llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(DiagnosticIDs::create(),
+                                                   DiagOpts);
   IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> FS =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
   llvm::IntrusiveRefCntPtr<FileManager> FileMgr =
diff --git a/clang/unittests/Tooling/Syntax/TreeTestBase.cpp b/clang/unittests/Tooling/Syntax/TreeTestBase.cpp
index 9f22b1d..4a25863 100644
--- a/clang/unittests/Tooling/Syntax/TreeTestBase.cpp
+++ b/clang/unittests/Tooling/Syntax/TreeTestBase.cpp
@@ -152,7 +152,7 @@ SyntaxTreeTest::buildTree(StringRef Code, const TestClangConfig &ClangConfig) {
   Invocation->getPreprocessorOpts().addRemappedFile(
       FileName, llvm::MemoryBuffer::getMemBufferCopy(Code).release());
   CompilerInstance Compiler(Invocation);
-  Compiler.setDiagnostics(Diags.get());
+  Compiler.setDiagnostics(Diags);
   Compiler.setFileManager(FileMgr.get());
   Compiler.setSourceManager(SourceMgr.get());
 
diff --git a/clang/unittests/Tooling/Syntax/TreeTestBase.h b/clang/unittests/Tooling/Syntax/TreeTestBase.h
index 01ff07b..fce89e2 100644
--- a/clang/unittests/Tooling/Syntax/TreeTestBase.h
+++ b/clang/unittests/Tooling/Syntax/TreeTestBase.h
@@ -42,7 +42,8 @@ protected:
   // Data fields.
   DiagnosticOptions DiagOpts;
   IntrusiveRefCntPtr<DiagnosticsEngine> Diags =
-      new DiagnosticsEngine(new DiagnosticIDs, DiagOpts);
+      llvm::makeIntrusiveRefCnt<DiagnosticsEngine>(DiagnosticIDs::create(),
+                                                   DiagOpts);
   IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> FS =
       llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
   IntrusiveRefCntPtr<FileManager> FileMgr =
diff --git a/flang/lib/Frontend/CompilerInstance.cpp b/flang/lib/Frontend/CompilerInstance.cpp
index 2e0f91f..cd8ddda 100644
--- a/flang/lib/Frontend/CompilerInstance.cpp
+++ b/flang/lib/Frontend/CompilerInstance.cpp
@@ -233,10 +233,8 @@ clang::IntrusiveRefCntPtr<clang::DiagnosticsEngine>
 CompilerInstance::createDiagnostics(clang::DiagnosticOptions &opts,
                                     clang::DiagnosticConsumer *client,
                                     bool shouldOwnClient) {
-  clang::IntrusiveRefCntPtr<clang::DiagnosticIDs> diagID(
-      new clang::DiagnosticIDs());
-  clang::IntrusiveRefCntPtr<clang::DiagnosticsEngine> diags(
-      new clang::DiagnosticsEngine(diagID, opts));
+  auto diags = llvm::makeIntrusiveRefCnt<clang::DiagnosticsEngine>(
+      clang::DiagnosticIDs::create(), opts);
 
   // Create the diagnostic client for reporting errors or for
   // implementing -verify.
diff --git a/flang/test/Parser/OpenMP/declare-target-indirect-tree.f90 b/flang/test/Parser/OpenMP/declare-target-indirect-tree.f90
index df85942..916bd66 100644
--- a/flang/test/Parser/OpenMP/declare-target-indirect-tree.f90
+++ b/flang/test/Parser/OpenMP/declare-target-indirect-tree.f90
@@ -15,7 +15,8 @@ module functions
 contains
   function func1() result(i)
     !$omp declare target enter(func1) indirect(.true.)
-    !CHECK: | | | | | OmpDeclareTargetSpecifier -> OmpDeclareTargetWithClause -> OmpClauseList -> OmpClause -> Enter -> OmpObjectList -> OmpObject -> Designator -> DataRef -> Name = 'func1'
+    !CHECK: | | | | | OmpDeclareTargetSpecifier -> OmpDeclareTargetWithClause -> OmpClauseList -> OmpClause -> Enter -> OmpEnterClause
+    !CHECK-NEXT: | | | | | | OmpObjectList -> OmpObject -> Designator -> DataRef -> Name = 'func1'
     !CHECK-NEXT: | | | | | OmpClause -> Indirect -> OmpIndirectClause -> Scalar -> Logical -> Expr = '.true._4'
     !CHECK-NEXT: | | | | | | LiteralConstant -> LogicalLiteralConstant
     !CHECK-NEXT: | | | | | | | bool = 'true'
@@ -26,7 +27,8 @@ contains
 
   function func2() result(i)
     !$omp declare target enter(func2) indirect
-    !CHECK: | | | | | OmpDeclareTargetSpecifier -> OmpDeclareTargetWithClause -> OmpClauseList -> OmpClause -> Enter -> OmpObjectList -> OmpObject -> Designator -> DataRef -> Name = 'func2'
+    !CHECK: | | | | | OmpDeclareTargetSpecifier -> OmpDeclareTargetWithClause -> OmpClauseList -> OmpClause -> Enter -> OmpEnterClause
+    !CHECK-NEXT: | | | | | | OmpObjectList -> OmpObject -> Designator -> DataRef -> Name = 'func2'
     !CHECK-NEXT: | | | | | OmpClause -> Indirect -> OmpIndirectClause ->
     character(1) :: i
     i = 'b'
diff --git a/flang/tools/flang-driver/driver.cpp b/flang/tools/flang-driver/driver.cpp
index 3a2dffc..8321b16 100644
--- a/flang/tools/flang-driver/driver.cpp
+++ b/flang/tools/flang-driver/driver.cpp
@@ -123,15 +123,14 @@ int main(int argc, const char **argv) {
   // Create DiagnosticsEngine for the compiler driver
   std::unique_ptr<clang::DiagnosticOptions> diagOpts =
       createAndPopulateDiagOpts(args);
-  llvm::IntrusiveRefCntPtr<clang::DiagnosticIDs> diagID(
-      new clang::DiagnosticIDs());
   Fortran::frontend::TextDiagnosticPrinter *diagClient =
       new Fortran::frontend::TextDiagnosticPrinter(llvm::errs(), *diagOpts);
 
   diagClient->setPrefix(
       std::string(llvm::sys::path::stem(getExecutablePath(args[0]))));
 
-  clang::DiagnosticsEngine diags(diagID, *diagOpts, diagClient);
+  clang::DiagnosticsEngine diags(clang::DiagnosticIDs::create(), *diagOpts,
+                                 diagClient);
 
   // Prepare the driver
   clang::driver::Driver theDriver(driverPath,
diff --git a/flang/tools/flang-driver/fc1_main.cpp b/flang/tools/flang-driver/fc1_main.cpp
index f2cd513..d9b103d 100644
--- a/flang/tools/flang-driver/fc1_main.cpp
+++ b/flang/tools/flang-driver/fc1_main.cpp
@@ -65,10 +65,9 @@ int fc1_main(llvm::ArrayRef<const char *> argv, const char *argv0) {
 
   // Create CompilerInvocation - use a dedicated instance of DiagnosticsEngine
   // for parsing the arguments
-  llvm::IntrusiveRefCntPtr<clang::DiagnosticIDs> diagID(
-      new clang::DiagnosticIDs());
   clang::DiagnosticOptions diagOpts;
-  clang::DiagnosticsEngine diags(diagID, diagOpts, diagsBuffer);
+  clang::DiagnosticsEngine diags(clang::DiagnosticIDs::create(), diagOpts,
+                                 diagsBuffer);
   bool success = CompilerInvocation::createFromArgs(flang->getInvocation(),
                                                     argv, diags, argv0);
 
diff --git a/libc/src/__support/math/asin_utils.h b/libc/src/__support/math/asin_utils.h
index e0c9096..efe779c 100644
--- a/libc/src/__support/math/asin_utils.h
+++ b/libc/src/__support/math/asin_utils.h
@@ -45,7 +45,7 @@ static constexpr double ASIN_COEFFS[12] = {
     0x1.2b5993bda1d9bp-6, -0x1.806aff270bf25p-7, 0x1.02614e5ed3936p-5,
 };
 
-LIBC_INLINE static constexpr double asin_eval(double u) {
+LIBC_INLINE double asin_eval(double u) {
   double u2 = u * u;
   double c0 = fputil::multiply_add(u, ASIN_COEFFS[1], ASIN_COEFFS[0]);
   double c1 = fputil::multiply_add(u, ASIN_COEFFS[3], ASIN_COEFFS[2]);
diff --git a/libcxx/include/__config b/libcxx/include/__config
index 549aa06..77a71b6 100644
--- a/libcxx/include/__config
+++ b/libcxx/include/__config
@@ -64,9 +64,21 @@
 
 // HARDENING {
 
-// TODO: Remove in LLVM 21. We're making this an error to catch folks who might not have migrated.
-#  ifdef _LIBCPP_ENABLE_ASSERTIONS
-#    error "_LIBCPP_ENABLE_ASSERTIONS has been removed, please use _LIBCPP_HARDENING_MODE instead"
+// TODO(LLVM 23): Remove this. We're making these an error to catch folks who might not have migrated.
+//       Since hardening went through several changes (many of which impacted user-facing macros),
+//       we're keeping these checks around for a bit longer than usual. Failure to properly configure
+//       hardening results in checks being dropped silently, which is a pretty big deal.
+#  if defined(_LIBCPP_ENABLE_ASSERTIONS)
+#    error "_LIBCPP_ENABLE_ASSERTIONS has been removed, please use _LIBCPP_HARDENING_MODE=<mode> instead (see docs)"
+#  endif
+#  if defined(_LIBCPP_ENABLE_HARDENED_MODE)
+#    error "_LIBCPP_ENABLE_HARDENED_MODE has been removed, please use _LIBCPP_HARDENING_MODE=<mode> instead (see docs)"
+#  endif
+#  if defined(_LIBCPP_ENABLE_SAFE_MODE)
+#    error "_LIBCPP_ENABLE_SAFE_MODE has been removed, please use _LIBCPP_HARDENING_MODE=<mode> instead (see docs)"
+#  endif
+#  if defined(_LIBCPP_ENABLE_DEBUG_MODE)
+#    error "_LIBCPP_ENABLE_DEBUG_MODE has been removed, please use _LIBCPP_HARDENING_MODE=<mode> instead (see docs)"
 #  endif
 
 // The library provides the macro `_LIBCPP_HARDENING_MODE` which can be set to one of the following values:
diff --git a/libcxx/include/__cxx03/__math/logarithms.h b/libcxx/include/__cxx03/__math/logarithms.h
index 2547350..9b9e59a 100644
--- a/libcxx/include/__cxx03/__math/logarithms.h
+++ b/libcxx/include/__cxx03/__math/logarithms.h
@@ -58,7 +58,7 @@ inline _LIBCPP_HIDE_FROM_ABI double log10(_A1 __x) _NOEXCEPT {
 inline _LIBCPP_HIDE_FROM_ABI int ilogb(float __x) _NOEXCEPT { return __builtin_ilogbf(__x); }
 
 template <class = int>
-_LIBCPP_HIDE_FROM_ABI double ilogb(double __x) _NOEXCEPT {
+_LIBCPP_HIDE_FROM_ABI int ilogb(double __x) _NOEXCEPT {
   return __builtin_ilogb(__x);
 }
 
diff --git a/libcxx/include/__math/logarithms.h b/libcxx/include/__math/logarithms.h
index 5f5f943..7343d6a 100644
--- a/libcxx/include/__math/logarithms.h
+++ b/libcxx/include/__math/logarithms.h
@@ -58,7 +58,7 @@ inline _LIBCPP_HIDE_FROM_ABI double log10(_A1 __x) _NOEXCEPT {
 inline _LIBCPP_HIDE_FROM_ABI int ilogb(float __x) _NOEXCEPT { return __builtin_ilogbf(__x); }
 
 template <class = int>
-_LIBCPP_HIDE_FROM_ABI double ilogb(double __x) _NOEXCEPT {
+_LIBCPP_HIDE_FROM_ABI int ilogb(double __x) _NOEXCEPT {
   return __builtin_ilogb(__x);
 }
 
diff --git a/libsycl/.clang-format b/libsycl/.clang-format
new file mode 100644
index 0000000..24550f8
--- /dev/null
+++ b/libsycl/.clang-format
@@ -0,0 +1,4 @@
+BasedOnStyle: LLVM
+
+# Preferred indentions of preprocessor statements.
+IndentPPDirectives: AfterHash
diff --git a/libsycl/.clang-tidy b/libsycl/.clang-tidy
new file mode 100644
index 0000000..cdc7a29
--- /dev/null
+++ b/libsycl/.clang-tidy
@@ -0,0 +1,17 @@
+Checks: >
+  -*,
+  clang-analyzer-*,
+  clang-diagnostic-*,
+  cppcoreguidelines-*,
+  -cppcoreguidelines-pro-bounds-array-to-pointer-decay,
+  -cppcoreguidelines-pro-bounds-constant-array-index,
+  -cppcoreguidelines-pro-bounds-pointer-arithmetic,
+  -cppcoreguidelines-pro-type-member-init,
+  -cppcoreguidelines-pro-type-union-access,
+  google-*,
+  -google-build-using-namespace,
+  -google-explicit-constructor,
+  -google-runtime-references,
+  misc-*,
+  -misc-macro-parentheses,
+  -misc-unused-parameters
diff --git a/libsycl/CMakeLists.txt b/libsycl/CMakeLists.txt
new file mode 100644
index 0000000..fe08a42
--- /dev/null
+++ b/libsycl/CMakeLists.txt
@@ -0,0 +1,126 @@
+#===============================================================================
+# Setup Project
+#===============================================================================
+cmake_minimum_required(VERSION 3.20.0)
+
+set(LLVM_SUBPROJECT_TITLE "libsycl")
+
+set(LIBSYCL_SOURCE_DIR  ${CMAKE_CURRENT_SOURCE_DIR})
+set(LIBSYCL_BINARY_DIR  ${CMAKE_CURRENT_BINARY_DIR})
+
+set(CMAKE_CXX_STANDARD 17)
+set(CMAKE_CXX_STANDARD_REQUIRED YES)
+set(CMAKE_CXX_EXTENSIONS OFF)
+
+#===============================================================================
+# Limitations
+#===============================================================================
+
+if (CMAKE_SYSTEM_NAME STREQUAL Windows AND NOT MSVC)
+# Build with other compilers is not configured, not guaranteed and not tested.
+    message(FATAL_ERROR
+      "When compiling for Windows, libsycl requires a"
+      " version of Microsoft Visual C++ or another compiler"
+      " that uses the Visual C++ cl command-line syntax.")
+endif()
+
+#===============================================================================
+# Setup CMake Options
+#===============================================================================
+
+option(LIBSYCL_ENABLE_WERROR "Treat all warnings as errors in the libsycl project" OFF)
+option(LIBSYCL_ENABLE_PEDANTIC "Compile with pedantic enabled." OFF)
+
+#===============================================================================
+# Configure System
+#===============================================================================
+
+set_property(GLOBAL PROPERTY USE_FOLDERS ON)
+
+set(LIBSYCL_SHARED_OUTPUT_NAME "sycl" CACHE STRING "Output name for the shared libsycl runtime library.")
+
+if(LLVM_ENABLE_PER_TARGET_RUNTIME_DIR AND NOT APPLE)
+  set(LIBSYCL_TARGET_SUBDIR ${LLVM_DEFAULT_TARGET_TRIPLE})
+  if(LIBSYCL_LIBDIR_SUBDIR)
+    string(APPEND LIBSYCL_TARGET_SUBDIR /${LIBSYCL_LIBDIR_SUBDIR})
+  endif()
+  cmake_path(NORMAL_PATH LIBSYCL_TARGET_SUBDIR)
+  set(LIBSYCL_LIBRARY_DIR ${LLVM_LIBRARY_OUTPUT_INTDIR}/${LIBSYCL_TARGET_SUBDIR})
+  set(LIBSYCL_INSTALL_LIBRARY_DIR lib${LLVM_LIBDIR_SUFFIX}/${LIBSYCL_TARGET_SUBDIR} CACHE STRING
+      "Path where built libsycl libraries should be installed.")
+  unset(LIBSYCL_TARGET_SUBDIR)
+else()
+  if(LLVM_LIBRARY_OUTPUT_INTDIR)
+    set(LIBSYCL_LIBRARY_DIR ${LLVM_LIBRARY_OUTPUT_INTDIR})
+  else()
+    set(LIBSYCL_LIBRARY_DIR ${CMAKE_BINARY_DIR}/lib${LIBSYCL_LIBDIR_SUFFIX})
+  endif()
+  set(LIBSYCL_INSTALL_LIBRARY_DIR lib${LIBSYCL_LIBDIR_SUFFIX} CACHE STRING
+      "Path where built libsycl libraries should be installed.")
+endif()
+
+set(LIBSYCL_INCLUDE_DIR include)
+set(LIBSYCL_BUILD_INCLUDE_DIR ${LLVM_BINARY_DIR}/${LIBSYCL_INCLUDE_DIR})
+set(LIBSYCL_SOURCE_INCLUDE_DIR ${CMAKE_CURRENT_SOURCE_DIR}/include)
+
+set(CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${LIBSYCL_LIBRARY_DIR})
+set(CMAKE_LIBRARY_OUTPUT_DIRECTORY ${LIBSYCL_LIBRARY_DIR})
+set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${LIBSYCL_LIBRARY_DIR})
+
+set(LIBSYCL_MAJOR_VERSION 0)
+set(LIBSYCL_MINOR_VERSION 1)
+set(LIBSYCL_PATCH_VERSION 0)
+set(LIBSYCL_VERSION_STRING "${LIBSYCL_MAJOR_VERSION}.${LIBSYCL_MINOR_VERSION}.${LIBSYCL_PATCH_VERSION}")
+set(LIBSYCL_ABI_NAMESPACE "__V${LIBSYCL_MAJOR_VERSION}" CACHE STRING
+    "The inline ABI namespace used by libsycl. It defaults to __Vn where `n` is the current ABI version.")
+if (NOT LIBSYCL_ABI_NAMESPACE MATCHES "__V.*")
+   message(FATAL_ERROR "LIBSYCL_ABI_NAMESPACE must be a reserved identifier, got '${LIBSYCL_ABI_NAMESPACE}'.")
+endif()
+
+#===============================================================================
+# Setup build & install rules
+#===============================================================================
+
+# Generate headers
+configure_file("${LIBSYCL_SOURCE_DIR}/src/version.hpp.in" "${LIBSYCL_BUILD_INCLUDE_DIR}/sycl/__impl/version.hpp")
+
+# Install generated headers.
+install(FILES
+  "${LIBSYCL_BUILD_INCLUDE_DIR}/sycl/__impl/version.hpp"
+  DESTINATION "${LIBSYCL_INCLUDE_DIR}/sycl/__impl"
+  COMPONENT sycl-headers)
+
+# This is a workaround to detect changes (add or modify) in subtree which
+# are not detected by copy_directory command.
+file(GLOB_RECURSE HEADERS_IN_SYCL_DIR CONFIGURE_DEPENDS "${LIBSYCL_SOURCE_INCLUDE_DIR}/sycl/*")
+file(GLOB_RECURSE HEADERS_IN_CL_DIR CONFIGURE_DEPENDS "${LIBSYCL_SOURCE_INCLUDE_DIR}/CL/*")
+
+string(REPLACE "${LIBSYCL_SOURCE_INCLUDE_DIR}" "${LIBSYCL_BUILD_INCLUDE_DIR}"
+  OUT_HEADERS_IN_SYCL_DIR "${HEADERS_IN_SYCL_DIR}")
+string(REPLACE "${LIBSYCL_SOURCE_INCLUDE_DIR}/CL" "${LIBSYCL_BUILD_INCLUDE_DIR}/CL"
+  OUT_HEADERS_IN_CL_DIR "${HEADERS_IN_CL_DIR}")
+
+# Copy SYCL headers from sources to build directory
+add_custom_target(sycl-headers
+  DEPENDS ${OUT_HEADERS_IN_SYCL_DIR}
+          ${OUT_HEADERS_IN_CL_DIR})
+
+add_custom_command(
+  OUTPUT  ${OUT_HEADERS_IN_SYCL_DIR}
+          ${OUT_HEADERS_IN_CL_DIR}
+  DEPENDS ${HEADERS_IN_SYCL_DIR}
+          ${HEADERS_IN_CL_DIR}
+  COMMAND ${CMAKE_COMMAND} -E copy_directory ${LIBSYCL_SOURCE_INCLUDE_DIR}/sycl ${LIBSYCL_BUILD_INCLUDE_DIR}/sycl
+  COMMAND ${CMAKE_COMMAND} -E copy_directory ${LIBSYCL_SOURCE_INCLUDE_DIR}/CL ${LIBSYCL_BUILD_INCLUDE_DIR}/CL
+  COMMENT "Copying SYCL headers...")
+
+install(DIRECTORY "${LIBSYCL_SOURCE_INCLUDE_DIR}/sycl" DESTINATION ${LIBSYCL_INCLUDE_DIR} COMPONENT sycl-headers)
+install(DIRECTORY "${LIBSYCL_SOURCE_INCLUDE_DIR}/CL" DESTINATION ${LIBSYCL_INCLUDE_DIR} COMPONENT sycl-headers)
+
+set(LIBSYCL_RT_LIBS ${LIBSYCL_SHARED_OUTPUT_NAME})
+
+add_subdirectory(src)
+
+add_custom_target(libsycl-runtime-libraries
+  DEPENDS ${LIBSYCL_RT_LIBS}
+)
diff --git a/libsycl/LICENSE.txt b/libsycl/LICENSE.txt
new file mode 100644
index 0000000..5715176
--- /dev/null
+++ b/libsycl/LICENSE.txt
@@ -0,0 +1,278 @@
+==============================================================================
+The LLVM Project is under the Apache License v2.0 with LLVM Exceptions:
+==============================================================================
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+    TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+    1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+    2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+    3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+    4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+    5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+    6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+    7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+    8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+    9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+    END OF TERMS AND CONDITIONS
+
+    APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+    Copyright [yyyy] [name of copyright owner]
+
+    Licensed under the Apache License, Version 2.0 (the "License");
+    you may not use this file except in compliance with the License.
+    You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+    Unless required by applicable law or agreed to in writing, software
+    distributed under the License is distributed on an "AS IS" BASIS,
+    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+    See the License for the specific language governing permissions and
+    limitations under the License.
+
+
+---- LLVM Exceptions to the Apache 2.0 License ----
+
+As an exception, if, as a result of your compiling your source code, portions
+of this Software are embedded into an Object form of such source code, you
+may redistribute such embedded portions in such Object form without complying
+with the conditions of Sections 4(a), 4(b) and 4(d) of the License.
+
+In addition, if you combine or link compiled forms of this Software with
+software that is licensed under the GPLv2 ("Combined Software") and if a
+court of competent jurisdiction determines that the patent provision (Section
+3), the indemnity provision (Section 9) or other Section of the License
+conflicts with the conditions of the GPLv2, you may retroactively and
+prospectively choose to deem waived or otherwise exclude such Section(s) of
+the License, but only in their entirety and only with respect to the Combined
+Software.
+
+==============================================================================
+Software from third parties included in the LLVM Project:
+==============================================================================
+The LLVM Project contains third party software which is under different license
+terms. All such code will be identified clearly using at least one of two
+mechanisms:
+1) It will be in a separate directory tree with its own `LICENSE.txt` or
+   `LICENSE` file at the top containing the specific license and restrictions
+   which apply to that software, or
+2) It will contain specific license and restriction terms at the top of every
+   file.
+
+==============================================================================
+Legacy LLVM License (https://llvm.org/docs/DeveloperPolicy.html#legacy):
+==============================================================================
+University of Illinois/NCSA
+Open Source License
+
+Copyright (c) 2003-2019 University of Illinois at Urbana-Champaign.
+All rights reserved.
+
+Developed by:
+
+    LLVM Team
+
+    University of Illinois at Urbana-Champaign
+
+    http://llvm.org
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal with
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimers.
+
+    * Redistributions in binary form must reproduce the above copyright notice,
+      this list of conditions and the following disclaimers in the
+      documentation and/or other materials provided with the distribution.
+
+    * Neither the names of the LLVM Team, University of Illinois at
+      Urbana-Champaign, nor the names of its contributors may be used to
+      endorse or promote products derived from this Software without specific
+      prior written permission.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
+FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
+CONTRIBUTORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS WITH THE
+SOFTWARE.
diff --git a/libsycl/README.md b/libsycl/README.md
new file mode 100644
index 0000000..9fed3fd
--- /dev/null
+++ b/libsycl/README.md
@@ -0,0 +1,20 @@
+## SYCL runtime library
+
+The libsycl subproject is an implementation of the SYCL runtime library as defined by the
+[SYCL 2020 specification](https://registry.khronos.org/SYCL/specs/sycl-2020/html/sycl-2020.html).
+
+Subproject documentation is available at: [SYCL RT documentation](./docs).
+
+libsycl runtime library and headers require C++17 support or higher.
+
+### How to use libsycl with Clang
+
+TODO
+
+#### How to build
+
+TODO
+
+# License
+
+See [LICENSE](./LICENSE.TXT) for details.
diff --git a/libsycl/docs/index.rst b/libsycl/docs/index.rst
new file mode 100644
index 0000000..78e76e7
--- /dev/null
+++ b/libsycl/docs/index.rst
@@ -0,0 +1,79 @@
+=====================
+SYCL runtime implementation
+=====================
+
+.. contents::
+   :local:
+
+.. _index:
+
+Current Status
+========
+
+The implementation is in the very early stages of upstreaming. The first milestone is to get
+support for a simple SYCL application with device code using Unified Shared Memory:
+
+.. code-block:: c++
+
+   #include <sycl/sycl.hpp>
+   
+   class TestKernel;
+   
+   int main() {
+     sycl::queue q;
+   
+     const size_t dataSize = 32;
+     int *dataPtr = sycl::malloc_shared<int>(32, q);
+     for (int i = 0; i < dataSize; ++i)
+       dataPtr[i] = 0;
+   
+     q.submit([&](sycl::handler &cgh) {
+       cgh.parallel_for<TestKernel>(
+           sycl::range<1>(dataSize),
+           [=](sycl::id<1> idx) { dataPtr[idx] = idx[0]; });
+     });
+     q.wait();
+   
+     bool error = false;
+     for (int i = 0; i < dataSize; ++i)
+       if (dataPtr[i] != i) error = true;
+   
+     free(dataPtr, q);
+   
+     return error;
+   }
+
+This requires at least partial support of the following functionality on the libsycl side:
+  * ``sycl::platform`` class
+  * ``sycl::device`` class
+  * ``sycl::context`` class
+  * ``sycl::queue`` class
+  * ``sycl::handler`` class
+  * ``sycl::id`` and ``sycl::range`` classes
+  * Unified shared memory allocation/deallocation
+  * Program manager, an internal component for retrieving and using device images from the multi-architectural binaries
+
+Build steps
+========
+
+To build LLVM with libsycl runtime enabled the following script can be used.
+
+.. code-block:: console
+
+  #!/bin/sh
+
+  build_llvm=`pwd`/build-llvm
+  installprefix=`pwd`/install
+  llvm=`pwd`
+  mkdir -p $build_llvm
+  mkdir -p $installprefix
+
+  cmake -G Ninja -S $llvm/llvm -B $build_llvm \
+        -DLLVM_ENABLE_PROJECTS="clang;clang-tools-extra" \
+        -DLLVM_INSTALL_UTILS=ON \
+        -DCMAKE_INSTALL_PREFIX=$installprefix \
+        -DLLVM_ENABLE_RUNTIMES="libcxx;libcxxabi;libsycl;libunwind" \
+        -DCMAKE_BUILD_TYPE=Release
+
+  ninja -C $build_llvm install
+  
\ No newline at end of file
diff --git a/libsycl/include/CL/sycl.hpp b/libsycl/include/CL/sycl.hpp
new file mode 100644
index 0000000..7c5de57
--- /dev/null
+++ b/libsycl/include/CL/sycl.hpp
@@ -0,0 +1,30 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file is a SYCL 1.2.1 standard header file. Deprecated.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef _LIBSYCL_CL_SYCL_HPP
+#define _LIBSYCL_CL_SYCL_HPP
+
+#if defined(__clang__)
+#  pragma clang diagnostic push
+#  pragma clang diagnostic ignored "-Wpedantic"
+#  warning "CL/sycl.hpp is deprecated, use sycl/sycl.hpp"
+#  pragma clang diagnostic pop
+#endif
+
+#include <sycl/sycl.hpp>
+
+namespace cl {
+namespace sycl = ::sycl;
+}
+
+#endif // _LIBSYCL_CL_SYCL_HPP
diff --git a/libsycl/include/sycl/__impl/detail/config.hpp b/libsycl/include/sycl/__impl/detail/config.hpp
new file mode 100644
index 0000000..cc90597
--- /dev/null
+++ b/libsycl/include/sycl/__impl/detail/config.hpp
@@ -0,0 +1,59 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file contains the declaration of the macros defining attributes for
+/// exported methods and defining API namespaces.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef _LIBSYCL___IMPL_DETAIL_CONFIG_HPP
+#define _LIBSYCL___IMPL_DETAIL_CONFIG_HPP
+
+#include <sycl/__impl/version.hpp>
+
+#define _LIBSYCL_BEGIN_UNVERSIONED_NAMESPACE_SYCL namespace sycl {
+#define _LIBSYCL_END_UNVERSIONED_NAMESPACE_SYCL }
+
+#define _LIBSYCL_BEGIN_NAMESPACE_SYCL                                          \
+  _LIBSYCL_BEGIN_UNVERSIONED_NAMESPACE_SYCL inline namespace _LIBSYCL_ABI_NAMESPACE {
+#define _LIBSYCL_END_NAMESPACE_SYCL                                            \
+  }                                                                            \
+  _LIBSYCL_END_UNVERSIONED_NAMESPACE_SYCL
+
+#ifndef __SYCL_DEVICE_ONLY__
+
+#  ifndef _LIBSYCL_EXPORT
+#    ifdef _WIN32
+
+#      define _LIBSYCL_DLL_LOCAL
+
+#      ifdef _LIBSYCL_BUILDING_LIBRARY
+#        define _LIBSYCL_EXPORT __declspec(dllexport)
+#      else
+#        define _LIBSYCL_EXPORT __declspec(dllimport)
+#      endif //_LIBSYCL_BUILDING_LIBRARY
+
+#    else // _WIN32
+
+#      define _LIBSYCL_DLL_LOCAL [[__gnu__::__visibility__("hidden")]]
+#      define _LIBSYCL_EXPORT [[__gnu__::__visibility__("default")]]
+
+#    endif // _WIN32
+#  endif   // _LIBSYCL_EXPORT
+
+#else // __SYCL_DEVICE_ONLY__
+
+#  ifndef _LIBSYCL_EXPORT
+#    define _LIBSYCL_EXPORT
+#    define _LIBSYCL_DLL_LOCAL
+#  endif
+
+#endif // __SYCL_DEVICE_ONLY__
+
+#endif // _LIBSYCL___IMPL_DETAIL_CONFIG_HPP
diff --git a/libsycl/include/sycl/__impl/platform.hpp b/libsycl/include/sycl/__impl/platform.hpp
new file mode 100644
index 0000000..bac59ac
--- /dev/null
+++ b/libsycl/include/sycl/__impl/platform.hpp
@@ -0,0 +1,31 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file contains the declaration of the SYCL platform class, which
+/// encapsulates a single platform on which kernel functions may be executed.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef _LIBSYCL___IMPL_PLATFORM_HPP
+#define _LIBSYCL___IMPL_PLATFORM_HPP
+
+#include <sycl/__impl/detail/config.hpp>
+
+_LIBSYCL_BEGIN_NAMESPACE_SYCL
+
+class _LIBSYCL_EXPORT platform {
+public:
+  /// Constructs a SYCL platform which contains the default device.
+  platform();
+
+}; // class platform
+
+_LIBSYCL_END_NAMESPACE_SYCL
+
+#endif // _LIBSYCL___IMPL_PLATFORM_HPP
diff --git a/libsycl/include/sycl/sycl.hpp b/libsycl/include/sycl/sycl.hpp
new file mode 100644
index 0000000..76399eba
--- /dev/null
+++ b/libsycl/include/sycl/sycl.hpp
@@ -0,0 +1,19 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file is a SYCL 2020 standard header file.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef _LIBSYCL_SYCL_HPP
+#define _LIBSYCL_SYCL_HPP
+
+#include <sycl/__impl/platform.hpp>
+
+#endif // _LIBSYCL_SYCL_HPP
diff --git a/libsycl/src/CMakeLists.txt b/libsycl/src/CMakeLists.txt
new file mode 100644
index 0000000..206b856
--- /dev/null
+++ b/libsycl/src/CMakeLists.txt
@@ -0,0 +1,98 @@
+list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../runtimes/cmake/Modules")
+include(WarningFlags)
+
+function(add_sycl_rt_library LIB_TARGET_NAME LIB_OBJ_NAME LIB_OUTPUT_NAME)
+  if (NOT LLVM_ENABLE_PIC)
+    message( FATAL_ERROR "Position-Independent Code generation is required for libsycl shared library" )
+  endif()
+
+  cmake_parse_arguments(ARG "" "" "COMPILE_OPTIONS;SOURCES" ${ARGN})
+
+  add_library(${LIB_OBJ_NAME} OBJECT ${ARG_SOURCES})
+
+  # Common compilation step setup
+  target_compile_definitions(${LIB_OBJ_NAME} PRIVATE
+              $<$<STREQUAL:${CMAKE_SYSTEM_NAME},Windows>:_LIBSYCL_BUILDING_LIBRARY>)
+  cxx_add_warning_flags(${LIB_OBJ_NAME} ${LIBSYCL_ENABLE_WERROR} ${LIBSYCL_ENABLE_PEDANTIC})
+
+  target_include_directories(
+    ${LIB_OBJ_NAME}
+    PRIVATE
+      ${CMAKE_CURRENT_SOURCE_DIR}
+      ${LIBSYCL_BUILD_INCLUDE_DIR}
+  )
+
+  add_library(${LIB_TARGET_NAME} SHARED
+              $<TARGET_OBJECTS:${LIB_OBJ_NAME}>)
+
+  add_dependencies(${LIB_OBJ_NAME}
+    sycl-headers
+  )
+
+  set_target_properties(${LIB_TARGET_NAME} PROPERTIES LINKER_LANGUAGE CXX)
+
+  if (CMAKE_SYSTEM_NAME STREQUAL Windows)
+    # Install stripped PDB
+    set(PDB_FILENAME "${LIB_TARGET_NAME}$<$<CONFIG:Debug>:d>")
+    check_linker_flag(CXX "LINKER:/PDBSTRIPPED:${PDB_FILENAME}.stripped.pdb"
+                          LINKER_SUPPORTS_PDBSTRIPPED)
+    if(LINKER_SUPPORTS_PDBSTRIPPED)
+      target_link_options(${LIB_TARGET_NAME}
+                          PRIVATE "LINKER:/PDBSTRIPPED:${PDB_FILENAME}.stripped.pdb")
+      install(FILES "${CMAKE_CURRENT_BINARY_DIR}/${PDB_FILENAME}.stripped.pdb"
+              DESTINATION ${CMAKE_INSTALL_PREFIX}/bin
+              RENAME "${PDB_FILENAME}.pdb"
+              COMPONENT ${LIB_TARGET_NAME}
+              OPTIONAL)
+    endif()
+
+    target_compile_options(${LIB_OBJ_NAME} PUBLIC /EHsc)
+  else()
+    target_compile_options(${LIB_OBJ_NAME} PUBLIC
+                           -fvisibility=hidden -fvisibility-inlines-hidden)
+
+    if (UNIX AND NOT APPLE)
+      set(linker_script "${CMAKE_CURRENT_SOURCE_DIR}/ld-version-script.txt")
+      target_link_libraries(
+          ${LIB_TARGET_NAME} PRIVATE "-Wl,--version-script=${linker_script}")
+      set_target_properties(${LIB_TARGET_NAME} PROPERTIES LINK_DEPENDS ${linker_script})
+    endif()
+  endif()
+
+  find_package(Threads REQUIRED)
+
+  target_link_libraries(${LIB_TARGET_NAME}
+    PRIVATE
+      ${CMAKE_DL_LIBS}
+      ${CMAKE_THREAD_LIBS_INIT}
+  )
+
+  set_target_properties(${LIB_TARGET_NAME} PROPERTIES
+                        VERSION ${LIBSYCL_VERSION_STRING}
+                        SOVERSION ${LIBSYCL_VERSION_STRING})
+  set_target_properties(${LIB_TARGET_NAME} PROPERTIES OUTPUT_NAME ${LIB_OUTPUT_NAME})
+endfunction(add_sycl_rt_library)
+
+set(LIBSYCL_SOURCES
+    "platform.cpp"
+)
+
+set(LIB_NAME "sycl")
+set(LIB_OUTPUT_NAME "${LIB_NAME}")
+if (CMAKE_SYSTEM_NAME STREQUAL Windows)
+  if (CMAKE_MSVC_RUNTIME_LIBRARY AND (NOT CMAKE_MSVC_RUNTIME_LIBRARY MATCHES "DLL$"))
+      message(FATAL_ERROR "libsycl requires a DLL version of the MSVC CRT.")
+  endif()
+  if ((NOT CMAKE_MSVC_RUNTIME_LIBRARY AND uppercase_CMAKE_BUILD_TYPE STREQUAL "DEBUG")
+      OR (CMAKE_MSVC_RUNTIME_LIBRARY STREQUAL "MultiThreadedDebugDLL"))
+    set(LIB_OUTPUT_NAME "${LIB_OUTPUT_NAME}d")
+  endif()
+endif()
+
+add_sycl_rt_library(${LIB_NAME} sycl_object ${LIB_OUTPUT_NAME}
+                    SOURCES ${LIBSYCL_SOURCES})
+
+install(TARGETS ${LIBSYCL_RT_LIBS}
+  ARCHIVE DESTINATION "lib${LLVM_LIBDIR_SUFFIX}" COMPONENT libsycl
+  LIBRARY DESTINATION "lib${LLVM_LIBDIR_SUFFIX}" COMPONENT libsycl
+  RUNTIME DESTINATION "bin" COMPONENT libsycl)
diff --git a/libsycl/src/ld-version-script.txt b/libsycl/src/ld-version-script.txt
new file mode 100644
index 0000000..a347d20
--- /dev/null
+++ b/libsycl/src/ld-version-script.txt
@@ -0,0 +1,20 @@
+{
+  /* Symbols to be exported are selected based on mangled names rather than   */
+  /* the demangled names provided by the `extern "C++"` matcher because it is */
+  /* easy to express "export everything defined in the sycl namespace" using  */
+  /* the former. Matching demangled names is more complicated in the presence */
+  /* of examples like:                                                        */
+  /*   "vtable for sycl::foo"                  (should be exported)           */
+  /*   "vtable for std::__internal<sycl::foo>" (should not be exported)       */
+
+  global:
+    /* Export everything from sycl namespace */
+    _ZNK4sycl*;  /* function */
+    _ZN4sycl*;   /* function */
+    _ZTIN4sycl*; /* typeinfo */
+    _ZTSN4sycl*; /* typeinfo name */
+    _ZTVN4sycl*; /* vtable */
+
+  local:
+    *;
+};
diff --git a/libsycl/src/platform.cpp b/libsycl/src/platform.cpp
new file mode 100644
index 0000000..b5d6517
--- /dev/null
+++ b/libsycl/src/platform.cpp
@@ -0,0 +1,17 @@
+//===----------------------------------------------------------------------===//
+//
+// 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 <sycl/__impl/platform.hpp>
+
+#include <stdexcept>
+
+_LIBSYCL_BEGIN_NAMESPACE_SYCL
+
+platform::platform() { throw std::runtime_error("Unimplemented"); }
+
+_LIBSYCL_END_NAMESPACE_SYCL
diff --git a/libsycl/src/version.hpp.in b/libsycl/src/version.hpp.in
new file mode 100644
index 0000000..a721576
--- /dev/null
+++ b/libsycl/src/version.hpp.in
@@ -0,0 +1,16 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file contains the declaration of SYCL RT version macro.
+///
+//===----------------------------------------------------------------------===//
+#define _LIBSYCL_MAJOR_VERSION ${LIBSYCL_MAJOR_VERSION}
+#define _LIBSYCL_MINOR_VERSION ${LIBSYCL_MINOR_VERSION}
+#define _LIBSYCL_PATCH_VERSION ${LIBSYCL_PATCH_VERSION}
+#define _LIBSYCL_ABI_NAMESPACE ${LIBSYCL_ABI_NAMESPACE}
diff --git a/lldb/bindings/interface/SBThreadExtensions.i b/lldb/bindings/interface/SBThreadExtensions.i
index 267faad..4ec9f10 100644
--- a/lldb/bindings/interface/SBThreadExtensions.i
+++ b/lldb/bindings/interface/SBThreadExtensions.i
@@ -45,6 +45,9 @@ STRING_EXTENSION_OUTSIDE(SBThread)
                 frames.append(frame)
             return frames
 
+        def get_stop_description(self):
+            return self.GetStopDescription(1024)
+
         def get_stop_reason_data(self):
             return [
                 self.GetStopReasonDataAtIndex(idx)
@@ -69,6 +72,7 @@ STRING_EXTENSION_OUTSIDE(SBThread)
         name = property(GetName, None, doc='''A read only property that returns the name of this thread as a string.''')
         queue = property(GetQueueName, None, doc='''A read only property that returns the dispatch queue name of this thread as a string.''')
         queue_id = property(GetQueueID, None, doc='''A read only property that returns the dispatch queue id of this thread as an integer.''')
+        stop_description = property(get_stop_description, None, doc='''A read only property that returns a string describing the reason this thread stopped.''')
         stop_reason = property(GetStopReason, None, doc='''A read only property that returns an lldb enumeration value (see enumerations that start with "lldb.eStopReason") that represents the reason this thread stopped.''')
         stop_reason_data = property(get_stop_reason_data, None, doc='''A read only property that returns the stop reason data as a list.''')
         is_suspended = property(IsSuspended, None, doc='''A read only property that returns a boolean value that indicates if this thread is suspended.''')
diff --git a/lldb/packages/Python/lldbsuite/support/temp_file.py b/lldb/packages/Python/lldbsuite/support/temp_file.py
new file mode 100644
index 0000000..a21e212
--- /dev/null
+++ b/lldb/packages/Python/lldbsuite/support/temp_file.py
@@ -0,0 +1,23 @@
+"""
+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
+"""
+
+import os
+import tempfile
+
+
+class OnDiskTempFile:
+    def __init__(self, delete=True):
+        self.path = None
+
+    def __enter__(self):
+        fd, path = tempfile.mkstemp()
+        os.close(fd)
+        self.path = path
+        return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        if os.path.exists(self.path):
+            os.remove(self.path)
diff --git a/lldb/packages/Python/lldbsuite/test/builders/builder.py b/lldb/packages/Python/lldbsuite/test/builders/builder.py
index ada6f9f..96c7b39 100644
--- a/lldb/packages/Python/lldbsuite/test/builders/builder.py
+++ b/lldb/packages/Python/lldbsuite/test/builders/builder.py
@@ -26,7 +26,7 @@ class Builder:
 
     def getTriple(self, arch):
         """Returns the triple for the given architecture or None."""
-        return None
+        return configuration.triple
 
     def getExtraMakeArgs(self):
         """
@@ -37,6 +37,9 @@ class Builder:
 
     def getArchCFlags(self, architecture):
         """Returns the ARCH_CFLAGS for the make system."""
+        triple = self.getTriple(architecture)
+        if triple:
+            return ["ARCH_CFLAGS=-target {}".format(triple)]
         return []
 
     def getMake(self, test_subdir, test_name):
diff --git a/lldb/packages/Python/lldbsuite/test/configuration.py b/lldb/packages/Python/lldbsuite/test/configuration.py
index b2d91fd2..5e38109 100644
--- a/lldb/packages/Python/lldbsuite/test/configuration.py
+++ b/lldb/packages/Python/lldbsuite/test/configuration.py
@@ -45,6 +45,9 @@ dsymutil = None
 sdkroot = None
 make_path = None
 
+# Allow specifying a triple for cross compilation.
+triple = None
+
 # The overriden dwarf verison.
 # Don't use this to test the current compiler's
 # DWARF version, as this won't be set if the
@@ -141,6 +144,7 @@ enabled_plugins = []
 # Typical values include Debug, Release, RelWithDebInfo and MinSizeRel
 cmake_build_type = None
 
+
 def shouldSkipBecauseOfCategories(test_categories):
     if use_categories:
         if (
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py b/lldb/packages/Python/lldbsuite/test/decorators.py
index a5f5837..bd10bcc 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -20,6 +20,7 @@ from . import configuration
 from . import test_categories
 from . import lldbtest_config
 from lldbsuite.support import funcutils
+from lldbsuite.support import temp_file
 from lldbsuite.test import lldbplatform
 from lldbsuite.test import lldbplatformutil
 
@@ -94,22 +95,23 @@ def _match_decorator_property(expected, actual):
 
 
 def _compiler_supports(
-    compiler, flag, source="int main() {}", output_file=tempfile.NamedTemporaryFile()
+    compiler, flag, source="int main() {}", output_file=temp_file.OnDiskTempFile()
 ):
     """Test whether the compiler supports the given flag."""
-    if platform.system() == "Darwin":
-        compiler = "xcrun " + compiler
-    try:
-        cmd = "echo '%s' | %s %s -x c -o %s -" % (
-            source,
-            compiler,
-            flag,
-            output_file.name,
-        )
-        subprocess.check_call(cmd, shell=True)
-    except subprocess.CalledProcessError:
-        return False
-    return True
+    with output_file:
+        if platform.system() == "Darwin":
+            compiler = "xcrun " + compiler
+        try:
+            cmd = "echo '%s' | %s %s -x c -o %s -" % (
+                source,
+                compiler,
+                flag,
+                output_file.path,
+            )
+            subprocess.check_call(cmd, shell=True)
+        except subprocess.CalledProcessError:
+            return False
+        return True
 
 
 def expectedFailureIf(condition, bugnumber=None):
@@ -876,19 +878,19 @@ def skipUnlessSupportedTypeAttribute(attr):
 
     def compiler_doesnt_support_struct_attribute():
         compiler_path = lldbplatformutil.getCompiler()
-        f = tempfile.NamedTemporaryFile()
-        cmd = [lldbplatformutil.getCompiler(), "-x", "c++", "-c", "-o", f.name, "-"]
-        p = subprocess.Popen(
-            cmd,
-            stdin=subprocess.PIPE,
-            stdout=subprocess.PIPE,
-            stderr=subprocess.PIPE,
-            universal_newlines=True,
-        )
-        stdout, stderr = p.communicate("struct __attribute__((%s)) Test {};" % attr)
-        if attr in stderr:
-            return "Compiler does not support attribute %s" % (attr)
-        return None
+        with temp_file.OnDiskTempFile() as f:
+            cmd = [lldbplatformutil.getCompiler(), "-x", "c++", "-c", "-o", f.path, "-"]
+            p = subprocess.Popen(
+                cmd,
+                stdin=subprocess.PIPE,
+                stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE,
+                universal_newlines=True,
+            )
+            stdout, stderr = p.communicate("struct __attribute__((%s)) Test {};" % attr)
+            if attr in stderr:
+                return "Compiler does not support attribute %s" % (attr)
+            return None
 
     return skipTestIfFn(compiler_doesnt_support_struct_attribute)
 
@@ -902,21 +904,21 @@ def skipUnlessHasCallSiteInfo(func):
         if not compiler.startswith("clang"):
             return "Test requires clang as compiler"
 
-        f = tempfile.NamedTemporaryFile()
-        cmd = (
-            "echo 'int main() {}' | "
-            "%s -g -glldb -O1 -S -emit-llvm -x c -o %s -" % (compiler_path, f.name)
-        )
-        if os.popen(cmd).close() is not None:
-            return "Compiler can't compile with call site info enabled"
+        with temp_file.OnDiskTempFile() as f:
+            cmd = (
+                "echo 'int main() {}' | "
+                "%s -g -glldb -O1 -S -emit-llvm -x c -o %s -" % (compiler_path, f.path)
+            )
+            if os.popen(cmd).close() is not None:
+                return "Compiler can't compile with call site info enabled"
 
-        with open(f.name, "r") as ir_output_file:
-            buf = ir_output_file.read()
+            with open(f.path, "r") as ir_output_file:
+                buf = ir_output_file.read()
 
-        if "DIFlagAllCallsDescribed" not in buf:
-            return "Compiler did not introduce DIFlagAllCallsDescribed IR flag"
+            if "DIFlagAllCallsDescribed" not in buf:
+                return "Compiler did not introduce DIFlagAllCallsDescribed IR flag"
 
-        return None
+            return None
 
     return skipTestIfFn(is_compiler_clang_with_call_site_info)(func)
 
@@ -957,7 +959,7 @@ def skipUnlessUndefinedBehaviorSanitizer(func):
             )
 
         # We need to write out the object into a named temp file for inspection.
-        outputf = tempfile.NamedTemporaryFile()
+        outputf = temp_file.OnDiskTempFile()
 
         # Try to compile with ubsan turned on.
         if not _compiler_supports(
@@ -969,7 +971,7 @@ def skipUnlessUndefinedBehaviorSanitizer(func):
             return "Compiler cannot compile with -fsanitize=undefined"
 
         # Check that we actually see ubsan instrumentation in the binary.
-        cmd = "nm %s" % outputf.name
+        cmd = "nm %s" % outputf.path
         with os.popen(cmd) as nm_output:
             if "___ubsan_handle_divrem_overflow" not in nm_output.read():
                 return "Division by zero instrumentation is missing"
@@ -1037,40 +1039,37 @@ def skipUnlessAArch64MTELinuxCompiler(func):
 
     def is_toolchain_with_mte():
         compiler_path = lldbplatformutil.getCompiler()
-        f = tempfile.NamedTemporaryFile(delete=False)
-        if lldbplatformutil.getPlatform() == "windows":
-            return "MTE tests are not compatible with 'windows'"
-
-        # Note hostos may be Windows.
-        f.close()
+        with temp_file.OnDiskTempFile() as f:
+            if lldbplatformutil.getPlatform() == "windows":
+                return "MTE tests are not compatible with 'windows'"
+
+            cmd = f"{compiler_path} -x c -o {f.path} -"
+            if (
+                subprocess.run(
+                    cmd, shell=True, input="int main() {}".encode()
+                ).returncode
+                != 0
+            ):
+                # Cannot compile at all, don't skip the test
+                # so that we report the broken compiler normally.
+                return None
 
-        cmd = f"{compiler_path} -x c -o {f.name} -"
-        if (
-            subprocess.run(cmd, shell=True, input="int main() {}".encode()).returncode
-            != 0
-        ):
-            os.remove(f.name)
-            # Cannot compile at all, don't skip the test
-            # so that we report the broken compiler normally.
+            # We need the Linux headers and ACLE MTE intrinsics
+            test_src = """
+                #include <asm/hwcap.h>
+                #include <arm_acle.h>
+                #ifndef HWCAP2_MTE
+                #error
+                #endif
+                int main() {
+                    void* ptr = __arm_mte_create_random_tag((void*)(0), 0);
+                }"""
+            cmd = f"{compiler_path} -march=armv8.5-a+memtag -x c -o {f.path} -"
+            res = subprocess.run(cmd, shell=True, input=test_src.encode())
+            if res.returncode != 0:
+                return "Toolchain does not support MTE"
             return None
 
-        # We need the Linux headers and ACLE MTE intrinsics
-        test_src = """
-            #include <asm/hwcap.h>
-            #include <arm_acle.h>
-            #ifndef HWCAP2_MTE
-            #error
-            #endif
-            int main() {
-                void* ptr = __arm_mte_create_random_tag((void*)(0), 0);
-            }"""
-        cmd = f"{compiler_path} -march=armv8.5-a+memtag -x c -o {f.name} -"
-        res = subprocess.run(cmd, shell=True, input=test_src.encode())
-        os.remove(f.name)
-        if res.returncode != 0:
-            return "Toolchain does not support MTE"
-        return None
-
     return skipTestIfFn(is_toolchain_with_mte)(func)
 
 
diff --git a/lldb/packages/Python/lldbsuite/test/dotest.py b/lldb/packages/Python/lldbsuite/test/dotest.py
index 24236e7..47a3c2e 100644
--- a/lldb/packages/Python/lldbsuite/test/dotest.py
+++ b/lldb/packages/Python/lldbsuite/test/dotest.py
@@ -43,6 +43,7 @@ from . import lldbtest_config
 from . import test_categories
 from . import test_result
 from ..support import seven
+from ..support import temp_file
 
 
 def is_exe(fpath):
@@ -321,8 +322,13 @@ def parseOptionsAndInitTestdirs():
             logging.error("No SDK found with the name %s; aborting...", args.apple_sdk)
             sys.exit(-1)
 
+    if args.triple:
+        configuration.triple = args.triple
+
     if args.arch:
         configuration.arch = args.arch
+    elif args.triple:
+        configuration.arch = args.triple.split("-")[0]
     else:
         configuration.arch = platform_machine
 
@@ -780,8 +786,8 @@ def canRunLibcxxTests():
         return True, "libc++ always present"
 
     if platform == "linux":
-        with tempfile.NamedTemporaryFile() as f:
-            cmd = [configuration.compiler, "-xc++", "-stdlib=libc++", "-o", f.name, "-"]
+        with temp_file.OnDiskTempFile() as f:
+            cmd = [configuration.compiler, "-xc++", "-stdlib=libc++", "-o", f.path, "-"]
             p = subprocess.Popen(
                 cmd,
                 stdin=subprocess.PIPE,
@@ -840,8 +846,8 @@ def canRunMsvcStlTests():
     if platform != "windows":
         return False, f"Don't know how to build with MSVC's STL on {platform}"
 
-    with tempfile.NamedTemporaryFile() as f:
-        cmd = [configuration.compiler, "-xc++", "-o", f.name, "-E", "-"]
+    with temp_file.OnDiskTempFile() as f:
+        cmd = [configuration.compiler, "-xc++", "-o", f.path, "-E", "-"]
         p = subprocess.Popen(
             cmd,
             stdin=subprocess.PIPE,
diff --git a/lldb/packages/Python/lldbsuite/test/dotest_args.py b/lldb/packages/Python/lldbsuite/test/dotest_args.py
index e9c2138..fce9e41 100644
--- a/lldb/packages/Python/lldbsuite/test/dotest_args.py
+++ b/lldb/packages/Python/lldbsuite/test/dotest_args.py
@@ -58,6 +58,14 @@ def create_parser():
             """Specify the path to sysroot. This overrides apple_sdk sysroot."""
         ),
     )
+    group.add_argument(
+        "--triple",
+        metavar="triple",
+        dest="triple",
+        help=textwrap.dedent(
+            """Specify the target triple. Used for cross compilation."""
+        ),
+    )
     if sys.platform == "darwin":
         group.add_argument(
             "--apple-sdk",
diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
index 8521ca5..e72ffd1 100644
--- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
+++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
@@ -149,6 +149,16 @@ else
 endif
 
 #----------------------------------------------------------------------
+# Use LLD when cross compiling on Darwin.
+#----------------------------------------------------------------------
+ifeq "$(HOST_OS)" "Darwin"
+	ifneq (,$(filter $(OS), Android FreeBSD Linux NetBSD Windows_NT))
+		LDFLAGS += -fuse-ld=lld
+	endif
+endif
+
+
+#----------------------------------------------------------------------
 # ARCHFLAG is the flag used to tell the compiler which architecture
 # to compile for. The default is the flag that clang accepts.
 #----------------------------------------------------------------------
diff --git a/lldb/source/Commands/CommandObjectDWIMPrint.cpp b/lldb/source/Commands/CommandObjectDWIMPrint.cpp
index a2c004d..d7589cc 100644
--- a/lldb/source/Commands/CommandObjectDWIMPrint.cpp
+++ b/lldb/source/Commands/CommandObjectDWIMPrint.cpp
@@ -18,11 +18,14 @@
 #include "lldb/Interpreter/OptionGroupValueObjectDisplay.h"
 #include "lldb/Target/StackFrame.h"
 #include "lldb/Utility/ConstString.h"
+#include "lldb/Utility/LLDBLog.h"
+#include "lldb/Utility/Log.h"
 #include "lldb/ValueObject/ValueObject.h"
 #include "lldb/lldb-defines.h"
 #include "lldb/lldb-enumerations.h"
 #include "lldb/lldb-forward.h"
 #include "llvm/ADT/StringRef.h"
+#include "llvm/Support/Error.h"
 
 #include <regex>
 
@@ -132,27 +135,22 @@ void CommandObjectDWIMPrint::DoExecute(StringRef command,
   };
 
   // Dump `valobj` according to whether `po` was requested or not.
-  auto dump_val_object = [&](ValueObject &valobj) {
+  auto dump_val_object = [&](ValueObject &valobj) -> Error {
     if (is_po) {
       StreamString temp_result_stream;
-      if (llvm::Error error = valobj.Dump(temp_result_stream, dump_options)) {
-        result.AppendError(toString(std::move(error)));
-        return;
-      }
+      if (Error err = valobj.Dump(temp_result_stream, dump_options))
+        return err;
       llvm::StringRef output = temp_result_stream.GetString();
       maybe_add_hint(output);
       result.GetOutputStream() << output;
     } else {
-      llvm::Error error =
-        valobj.Dump(result.GetOutputStream(), dump_options);
-      if (error) {
-        result.AppendError(toString(std::move(error)));
-        return;
-      }
+      if (Error err = valobj.Dump(result.GetOutputStream(), dump_options))
+        return err;
     }
     m_interpreter.PrintWarningsIfNecessary(result.GetOutputStream(),
                                            m_cmd_name);
     result.SetStatus(eReturnStatusSuccessFinishResult);
+    return Error::success();
   };
 
   // First, try `expr` as a _limited_ frame variable expression path: only the
@@ -186,8 +184,13 @@ void CommandObjectDWIMPrint::DoExecute(StringRef command,
                                      expr);
       }
 
-      dump_val_object(*valobj_sp);
-      return;
+      Error err = dump_val_object(*valobj_sp);
+      if (!err)
+        return;
+
+      // Dump failed, continue on to expression evaluation.
+      LLDB_LOG_ERROR(GetLog(LLDBLog::Expressions), std::move(err),
+                     "could not print frame variable '{1}': {0}", expr);
     }
   }
 
@@ -196,8 +199,14 @@ void CommandObjectDWIMPrint::DoExecute(StringRef command,
     if (auto *state = target.GetPersistentExpressionStateForLanguage(language))
       if (auto var_sp = state->GetVariable(expr))
         if (auto valobj_sp = var_sp->GetValueObject()) {
-          dump_val_object(*valobj_sp);
-          return;
+          Error err = dump_val_object(*valobj_sp);
+          if (!err)
+            return;
+
+          // Dump failed, continue on to expression evaluation.
+          LLDB_LOG_ERROR(GetLog(LLDBLog::Expressions), std::move(err),
+                         "could not print persistent variable '{1}': {0}",
+                         expr);
         }
 
   // Third, and lastly, try `expr` as a source expression to evaluate.
@@ -248,10 +257,12 @@ void CommandObjectDWIMPrint::DoExecute(StringRef command,
       result.AppendNoteWithFormatv("ran `expression {0}{1}`", flags, expr);
     }
 
-    if (valobj_sp->GetError().GetError() != UserExpression::kNoResult)
-      dump_val_object(*valobj_sp);
-    else
+    if (valobj_sp->GetError().GetError() != UserExpression::kNoResult) {
+      if (Error err = dump_val_object(*valobj_sp))
+        result.SetError(std::move(err));
+    } else {
       result.SetStatus(eReturnStatusSuccessFinishNoResult);
+    }
 
     if (suppress_result)
       if (auto result_var_sp =
diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp b/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp
index 06f3a7e..ff9ed9c 100644
--- a/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp
+++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangExpressionSourceCode.cpp
@@ -260,9 +260,8 @@ TokenVerifier::TokenVerifier(std::string body) {
 
   // Let's build the actual source code Clang needs and setup some utility
   // objects.
-  llvm::IntrusiveRefCntPtr<DiagnosticIDs> diag_ids(new DiagnosticIDs());
   DiagnosticOptions diags_opts;
-  DiagnosticsEngine diags(diag_ids, diags_opts);
+  DiagnosticsEngine diags(DiagnosticIDs::create(), diags_opts);
   clang::SourceManager SM(diags, file_mgr);
   auto buf = llvm::MemoryBuffer::getMemBuffer(body);
 
diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp b/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp
index 2f838b3..d54f072 100644
--- a/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp
+++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangModulesDeclVendor.cpp
@@ -747,7 +747,7 @@ ClangModulesDeclVendor::Create(Target &target) {
 
   // Make sure clang uses the same VFS as LLDB.
   instance->createFileManager(FileSystem::Instance().GetVirtualFileSystem());
-  instance->setDiagnostics(diagnostics_engine.get());
+  instance->setDiagnostics(diagnostics_engine);
 
   std::unique_ptr<clang::FrontendAction> action(new clang::SyntaxOnlyAction);
 
diff --git a/lldb/source/Plugins/Language/ClangCommon/ClangHighlighter.cpp b/lldb/source/Plugins/Language/ClangCommon/ClangHighlighter.cpp
index 8cc5714..9cb5ea3 100644
--- a/lldb/source/Plugins/Language/ClangCommon/ClangHighlighter.cpp
+++ b/lldb/source/Plugins/Language/ClangCommon/ClangHighlighter.cpp
@@ -162,9 +162,8 @@ void ClangHighlighter::Highlight(const HighlightStyle &options,
   // Let's build the actual source code Clang needs and setup some utility
   // objects.
   std::string full_source = previous_lines.str() + line.str();
-  llvm::IntrusiveRefCntPtr<DiagnosticIDs> diag_ids(new DiagnosticIDs());
   DiagnosticOptions diags_opts;
-  DiagnosticsEngine diags(diag_ids, diags_opts);
+  DiagnosticsEngine diags(DiagnosticIDs::create(), diags_opts);
   clang::SourceManager SM(diags, file_mgr);
   auto buf = llvm::MemoryBuffer::getMemBuffer(full_source);
 
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.cpp b/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.cpp
index 4bfbb4d..9762ead 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.cpp
@@ -13,6 +13,7 @@
 
 #include "lldb/Core/Module.h"
 #include "lldb/Symbol/Function.h"
+#include "lldb/lldb-private-enumerations.h"
 #include "llvm/Support/DJB.h"
 
 using namespace lldb;
@@ -275,7 +276,7 @@ void AppleDWARFIndex::GetNamespaces(
 void AppleDWARFIndex::GetFunctions(
     const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
     const CompilerDeclContext &parent_decl_ctx,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
   if (!m_apple_names_up)
     return;
 
@@ -288,15 +289,16 @@ void AppleDWARFIndex::GetFunctions(
       ReportInvalidDIERef(die_ref, name);
       continue;
     }
-    if (!ProcessFunctionDIE(lookup_info, die, parent_decl_ctx, callback))
+    if (ProcessFunctionDIE(lookup_info, die, parent_decl_ctx, callback) ==
+        IterationAction::Stop)
       return;
   }
 }
 
 void AppleDWARFIndex::GetFunctions(
     const RegularExpression &regex,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
-  return GetGlobalVariables(regex, callback);
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
+  return GetGlobalVariables(regex, IterationActionAdaptor(callback));
 }
 
 void AppleDWARFIndex::Dump(Stream &s) {
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.h b/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.h
index 73de75b..c0f0eb6 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/AppleDWARFIndex.h
@@ -61,12 +61,13 @@ public:
                 llvm::function_ref<bool(DWARFDIE die)> callback) override;
   void GetNamespaces(ConstString name,
                      llvm::function_ref<bool(DWARFDIE die)> callback) override;
-  void GetFunctions(const Module::LookupInfo &lookup_info,
-                    SymbolFileDWARF &dwarf,
-                    const CompilerDeclContext &parent_decl_ctx,
-                    llvm::function_ref<bool(DWARFDIE die)> callback) override;
-  void GetFunctions(const RegularExpression &regex,
-                    llvm::function_ref<bool(DWARFDIE die)> callback) override;
+  void GetFunctions(
+      const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
+      const CompilerDeclContext &parent_decl_ctx,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback) override;
+  void GetFunctions(
+      const RegularExpression &regex,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback) override;
 
   void Dump(Stream &s) override;
 
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.cpp b/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.cpp
index 30c890d..a806506 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.cpp
@@ -16,6 +16,7 @@
 #include "lldb/Core/Mangled.h"
 #include "lldb/Core/Module.h"
 #include "lldb/Target/Language.h"
+#include "lldb/lldb-private-enumerations.h"
 
 using namespace lldb_private;
 using namespace lldb;
@@ -23,10 +24,10 @@ using namespace lldb_private::plugin::dwarf;
 
 DWARFIndex::~DWARFIndex() = default;
 
-bool DWARFIndex::ProcessFunctionDIE(
+IterationAction DWARFIndex::ProcessFunctionDIE(
     const Module::LookupInfo &lookup_info, DWARFDIE die,
     const CompilerDeclContext &parent_decl_ctx,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
   llvm::StringRef name = lookup_info.GetLookupName().GetStringRef();
   FunctionNameType name_type_mask = lookup_info.GetNameTypeMask();
 
@@ -43,7 +44,7 @@ bool DWARFIndex::ProcessFunctionDIE(
 
     if (!lookup_info.NameMatchesLookupInfo(name_to_match_against,
                                            lookup_info.GetLanguageType()))
-      return true;
+      return IterationAction::Continue;
   }
 
   // Exit early if we're searching exclusively for methods or selectors and
@@ -51,12 +52,12 @@ bool DWARFIndex::ProcessFunctionDIE(
   uint32_t looking_for_nonmethods =
       name_type_mask & ~(eFunctionNameTypeMethod | eFunctionNameTypeSelector);
   if (!looking_for_nonmethods && parent_decl_ctx.IsValid())
-    return true;
+    return IterationAction::Continue;
 
   // Otherwise, we need to also check that the context matches. If it does not
   // match, we do nothing.
   if (!SymbolFileDWARF::DIEInDeclContext(parent_decl_ctx, die))
-    return true;
+    return IterationAction::Continue;
 
   // In case of a full match, we just insert everything we find.
   if (name_type_mask & eFunctionNameTypeFull && die.GetMangledName() == name)
@@ -79,7 +80,7 @@ bool DWARFIndex::ProcessFunctionDIE(
       return callback(die);
   }
 
-  return true;
+  return IterationAction::Continue;
 }
 
 DWARFIndex::DIERefCallbackImpl::DIERefCallbackImpl(
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.h b/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.h
index 15d8503..3578824 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFIndex.h
@@ -16,6 +16,7 @@
 
 #include "lldb/Core/Module.h"
 #include "lldb/Target/Statistics.h"
+#include "lldb/lldb-private-enumerations.h"
 
 namespace lldb_private::plugin {
 namespace dwarf {
@@ -82,10 +83,10 @@ public:
   virtual void
   GetFunctions(const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
                const CompilerDeclContext &parent_decl_ctx,
-               llvm::function_ref<bool(DWARFDIE die)> callback) = 0;
+               llvm::function_ref<IterationAction(DWARFDIE die)> callback) = 0;
   virtual void
   GetFunctions(const RegularExpression &regex,
-               llvm::function_ref<bool(DWARFDIE die)> callback) = 0;
+               llvm::function_ref<IterationAction(DWARFDIE die)> callback) = 0;
 
   virtual void Dump(Stream &s) = 0;
 
@@ -101,9 +102,10 @@ protected:
   /// the function given by "die" matches search criteria given by
   /// "parent_decl_ctx" and "name_type_mask", it calls the callback with the
   /// given die.
-  bool ProcessFunctionDIE(const Module::LookupInfo &lookup_info, DWARFDIE die,
-                          const CompilerDeclContext &parent_decl_ctx,
-                          llvm::function_ref<bool(DWARFDIE die)> callback);
+  IterationAction ProcessFunctionDIE(
+      const Module::LookupInfo &lookup_info, DWARFDIE die,
+      const CompilerDeclContext &parent_decl_ctx,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback);
 
   class DIERefCallbackImpl {
   public:
@@ -140,6 +142,25 @@ protected:
   bool ProcessNamespaceDieMatchParents(
       const CompilerDeclContext &parent_decl_ctx, DWARFDIE die,
       llvm::function_ref<bool(DWARFDIE die)> callback);
+
+  /// Helper to convert callbacks that return an \c IterationAction
+  /// to a callback that returns a \c bool, where \c true indicates
+  /// we should continue iterating. This will be used to incrementally
+  /// migrate the callbacks to return an \c IterationAction.
+  ///
+  /// FIXME: remove once all callbacks in the DWARFIndex APIs return
+  /// IterationAction.
+  struct IterationActionAdaptor {
+    IterationActionAdaptor(
+        llvm::function_ref<IterationAction(DWARFDIE die)> callback)
+        : m_callback_ref(callback) {}
+
+    bool operator()(DWARFDIE die) {
+      return m_callback_ref(std::move(die)) == IterationAction::Continue;
+    }
+
+    llvm::function_ref<IterationAction(DWARFDIE die)> m_callback_ref;
+  };
 };
 } // namespace dwarf
 } // namespace lldb_private::plugin
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
index ff1a76b..3ae9fcc 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
@@ -14,6 +14,7 @@
 #include "lldb/Core/Module.h"
 #include "lldb/Utility/RegularExpression.h"
 #include "lldb/Utility/Stream.h"
+#include "lldb/lldb-private-enumerations.h"
 #include "llvm/ADT/Sequence.h"
 #include <optional>
 
@@ -607,7 +608,7 @@ void DebugNamesDWARFIndex::GetNamespacesWithParents(
 void DebugNamesDWARFIndex::GetFunctions(
     const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
     const CompilerDeclContext &parent_decl_ctx,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
   ConstString name = lookup_info.GetLookupName();
   std::set<DWARFDebugInfoEntry *> seen;
   for (const DebugNames::Entry &entry :
@@ -617,12 +618,12 @@ void DebugNamesDWARFIndex::GetFunctions(
       continue;
 
     if (DWARFDIE die = GetDIE(entry)) {
-      if (!ProcessFunctionDIE(lookup_info, die, parent_decl_ctx,
-                              [&](DWARFDIE die) {
-                                if (!seen.insert(die.GetDIE()).second)
-                                  return true;
-                                return callback(die);
-                              }))
+      if (ProcessFunctionDIE(lookup_info, die, parent_decl_ctx,
+                             [&](DWARFDIE die) {
+                               if (!seen.insert(die.GetDIE()).second)
+                                 return IterationAction::Continue;
+                               return callback(die);
+                             }) == IterationAction::Stop)
         return;
     }
   }
@@ -632,7 +633,7 @@ void DebugNamesDWARFIndex::GetFunctions(
 
 void DebugNamesDWARFIndex::GetFunctions(
     const RegularExpression &regex,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
   for (const DebugNames::NameIndex &ni: *m_debug_names_up) {
     for (DebugNames::NameTableEntry nte: ni) {
       if (!regex.Execute(nte.getString()))
@@ -645,7 +646,7 @@ void DebugNamesDWARFIndex::GetFunctions(
         if (tag != DW_TAG_subprogram && tag != DW_TAG_inlined_subroutine)
           continue;
 
-        if (!ProcessEntry(*entry_or, callback))
+        if (!ProcessEntry(*entry_or, IterationActionAdaptor(callback)))
           return;
       }
       MaybeLogLookupError(entry_or.takeError(), ni, nte.getString());
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h
index ab6cde1..2105919 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h
@@ -58,12 +58,13 @@ public:
   void GetNamespacesWithParents(
       ConstString name, const CompilerDeclContext &parent_decl_ctx,
       llvm::function_ref<bool(DWARFDIE die)> callback) override;
-  void GetFunctions(const Module::LookupInfo &lookup_info,
-                    SymbolFileDWARF &dwarf,
-                    const CompilerDeclContext &parent_decl_ctx,
-                    llvm::function_ref<bool(DWARFDIE die)> callback) override;
-  void GetFunctions(const RegularExpression &regex,
-                    llvm::function_ref<bool(DWARFDIE die)> callback) override;
+  void GetFunctions(
+      const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
+      const CompilerDeclContext &parent_decl_ctx,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback) override;
+  void GetFunctions(
+      const RegularExpression &regex,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback) override;
 
   void Dump(Stream &s) override;
 
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.cpp b/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.cpp
index c858ce2..f96ac7e 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.cpp
@@ -21,6 +21,7 @@
 #include "lldb/Utility/DataExtractor.h"
 #include "lldb/Utility/Stream.h"
 #include "lldb/Utility/Timer.h"
+#include "lldb/lldb-private-enumerations.h"
 #include "llvm/Support/FormatVariadic.h"
 #include "llvm/Support/ThreadPool.h"
 #include <atomic>
@@ -471,60 +472,62 @@ void ManualDWARFIndex::GetNamespaces(
 void ManualDWARFIndex::GetFunctions(
     const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
     const CompilerDeclContext &parent_decl_ctx,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
   Index();
   ConstString name = lookup_info.GetLookupName();
   FunctionNameType name_type_mask = lookup_info.GetNameTypeMask();
 
   if (name_type_mask & eFunctionNameTypeFull) {
     if (!m_set.function_fullnames.Find(
-            name, DIERefCallback(
-                      [&](DWARFDIE die) {
-                        if (!SymbolFileDWARF::DIEInDeclContext(parent_decl_ctx,
-                                                               die))
-                          return true;
-                        return callback(die);
-                      },
-                      name.GetStringRef())))
+            name, DIERefCallback(IterationActionAdaptor([&](DWARFDIE die) {
+                                   if (!SymbolFileDWARF::DIEInDeclContext(
+                                           parent_decl_ctx, die))
+                                     return IterationAction::Continue;
+                                   return callback(die);
+                                 }),
+                                 name.GetStringRef())))
       return;
   }
   if (name_type_mask & eFunctionNameTypeBase) {
     if (!m_set.function_basenames.Find(
-            name, DIERefCallback(
-                      [&](DWARFDIE die) {
-                        if (!SymbolFileDWARF::DIEInDeclContext(parent_decl_ctx,
-                                                               die))
-                          return true;
-                        return callback(die);
-                      },
-                      name.GetStringRef())))
+            name, DIERefCallback(IterationActionAdaptor([&](DWARFDIE die) {
+                                   if (!SymbolFileDWARF::DIEInDeclContext(
+                                           parent_decl_ctx, die))
+                                     return IterationAction::Continue;
+                                   return callback(die);
+                                 }),
+                                 name.GetStringRef())))
       return;
   }
 
   if (name_type_mask & eFunctionNameTypeMethod && !parent_decl_ctx.IsValid()) {
     if (!m_set.function_methods.Find(
-            name, DIERefCallback(callback, name.GetStringRef())))
+            name, DIERefCallback(IterationActionAdaptor(callback),
+                                 name.GetStringRef())))
       return;
   }
 
   if (name_type_mask & eFunctionNameTypeSelector &&
       !parent_decl_ctx.IsValid()) {
     if (!m_set.function_selectors.Find(
-            name, DIERefCallback(callback, name.GetStringRef())))
+            name, DIERefCallback(IterationActionAdaptor(callback),
+                                 name.GetStringRef())))
       return;
   }
 }
 
 void ManualDWARFIndex::GetFunctions(
     const RegularExpression &regex,
-    llvm::function_ref<bool(DWARFDIE die)> callback) {
+    llvm::function_ref<IterationAction(DWARFDIE die)> callback) {
   Index();
 
-  if (!m_set.function_basenames.Find(regex,
-                                     DIERefCallback(callback, regex.GetText())))
+  if (!m_set.function_basenames.Find(
+          regex,
+          DIERefCallback(IterationActionAdaptor(callback), regex.GetText())))
     return;
-  if (!m_set.function_fullnames.Find(regex,
-                                     DIERefCallback(callback, regex.GetText())))
+  if (!m_set.function_fullnames.Find(
+          regex,
+          DIERefCallback(IterationActionAdaptor(callback), regex.GetText())))
     return;
 }
 
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.h b/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.h
index 04627b0..5685ba4 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/ManualDWARFIndex.h
@@ -50,12 +50,13 @@ public:
                 llvm::function_ref<bool(DWARFDIE die)> callback) override;
   void GetNamespaces(ConstString name,
                      llvm::function_ref<bool(DWARFDIE die)> callback) override;
-  void GetFunctions(const Module::LookupInfo &lookup_info,
-                    SymbolFileDWARF &dwarf,
-                    const CompilerDeclContext &parent_decl_ctx,
-                    llvm::function_ref<bool(DWARFDIE die)> callback) override;
-  void GetFunctions(const RegularExpression &regex,
-                    llvm::function_ref<bool(DWARFDIE die)> callback) override;
+  void GetFunctions(
+      const Module::LookupInfo &lookup_info, SymbolFileDWARF &dwarf,
+      const CompilerDeclContext &parent_decl_ctx,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback) override;
+  void GetFunctions(
+      const RegularExpression &regex,
+      llvm::function_ref<IterationAction(DWARFDIE die)> callback) override;
 
   void Dump(Stream &s) override;
 
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp b/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
index 41ab8d1..2c3f050 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
@@ -74,6 +74,7 @@
 #include "ManualDWARFIndex.h"
 #include "SymbolFileDWARFDebugMap.h"
 #include "SymbolFileDWARFDwo.h"
+#include "lldb/lldb-private-enumerations.h"
 
 #include "llvm/DebugInfo/DWARF/DWARFContext.h"
 #include "llvm/DebugInfo/DWARF/DWARFDebugAbbrev.h"
@@ -2539,7 +2540,7 @@ void SymbolFileDWARF::FindFunctions(const Module::LookupInfo &lookup_info,
   m_index->GetFunctions(lookup_info, *this, parent_decl_ctx, [&](DWARFDIE die) {
     if (resolved_dies.insert(die.GetDIE()).second)
       ResolveFunction(die, include_inlines, sc_list);
-    return true;
+    return IterationAction::Continue;
   });
   // With -gsimple-template-names, a templated type's DW_AT_name will not
   // contain the template parameters. Try again stripping '<' and anything
@@ -2556,7 +2557,7 @@ void SymbolFileDWARF::FindFunctions(const Module::LookupInfo &lookup_info,
                             [&](DWARFDIE die) {
                               if (resolved_dies.insert(die.GetDIE()).second)
                                 ResolveFunction(die, include_inlines, sc_list);
-                              return true;
+                              return IterationAction::Continue;
                             });
     }
   }
@@ -2592,7 +2593,7 @@ void SymbolFileDWARF::FindFunctions(const RegularExpression &regex,
   m_index->GetFunctions(regex, [&](DWARFDIE die) {
     if (resolved_dies.insert(die.GetDIE()).second)
       ResolveFunction(die, include_inlines, sc_list);
-    return true;
+    return IterationAction::Continue;
   });
 }
 
diff --git a/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp b/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp
index 836f9fe..5d61cbb 100644
--- a/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp
+++ b/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp
@@ -665,10 +665,9 @@ void TypeSystemClang::CreateASTContext() {
   m_file_manager_up = std::make_unique<clang::FileManager>(
       file_system_options, FileSystem::Instance().GetVirtualFileSystem());
 
-  llvm::IntrusiveRefCntPtr<DiagnosticIDs> diag_id_sp(new DiagnosticIDs());
   m_diagnostic_options_up = std::make_unique<DiagnosticOptions>();
-  m_diagnostics_engine_up =
-      std::make_unique<DiagnosticsEngine>(diag_id_sp, *m_diagnostic_options_up);
+  m_diagnostics_engine_up = std::make_unique<DiagnosticsEngine>(
+      DiagnosticIDs::create(), *m_diagnostic_options_up);
 
   m_source_manager_up = std::make_unique<clang::SourceManager>(
       *m_diagnostics_engine_up, *m_file_manager_up);
diff --git a/lldb/test/API/commands/watchpoints/step_over_watchpoint/TestStepOverWatchpoint.py b/lldb/test/API/commands/watchpoints/step_over_watchpoint/TestStepOverWatchpoint.py
index 8179d52..67dfbea 100644
--- a/lldb/test/API/commands/watchpoints/step_over_watchpoint/TestStepOverWatchpoint.py
+++ b/lldb/test/API/commands/watchpoints/step_over_watchpoint/TestStepOverWatchpoint.py
@@ -50,11 +50,11 @@ class TestStepOverWatchpoint(TestBase):
             lldb.eStopReasonWatchpoint,
             STOPPED_DUE_TO_WATCHPOINT,
         )
-        self.assertEqual(thread.GetStopDescription(20), "watchpoint 1")
+        self.assertEqual(thread.stop_description, "watchpoint 1")
 
         process.Continue()
         self.assertState(process.GetState(), lldb.eStateStopped, PROCESS_STOPPED)
-        self.assertEqual(thread.GetStopDescription(20), "step over")
+        self.assertEqual(thread.stop_description, "step over")
 
         self.step_inst_for_watchpoint(1)
 
@@ -89,11 +89,11 @@ class TestStepOverWatchpoint(TestBase):
             lldb.eStopReasonWatchpoint,
             STOPPED_DUE_TO_WATCHPOINT,
         )
-        self.assertEqual(thread.GetStopDescription(20), "watchpoint 1")
+        self.assertEqual(thread.stop_description, "watchpoint 1")
 
         process.Continue()
         self.assertState(process.GetState(), lldb.eStateStopped, PROCESS_STOPPED)
-        self.assertEqual(thread.GetStopDescription(20), "step over")
+        self.assertEqual(thread.stop_description, "step over")
 
         self.step_inst_for_watchpoint(1)
 
@@ -106,7 +106,7 @@ class TestStepOverWatchpoint(TestBase):
             if stop_reason == lldb.eStopReasonWatchpoint:
                 self.assertFalse(watchpoint_hit, "Watchpoint already hit.")
                 expected_stop_desc = "watchpoint %d" % wp_id
-                actual_stop_desc = self.thread().GetStopDescription(20)
+                actual_stop_desc = self.thread().stop_description
                 self.assertEqual(
                     actual_stop_desc, expected_stop_desc, "Watchpoint ID didn't match."
                 )
diff --git a/lldb/test/API/commands/watchpoints/watchpoint_count/TestWatchpointCount.py b/lldb/test/API/commands/watchpoints/watchpoint_count/TestWatchpointCount.py
index ff834b5..a0251d4 100644
--- a/lldb/test/API/commands/watchpoints/watchpoint_count/TestWatchpointCount.py
+++ b/lldb/test/API/commands/watchpoints/watchpoint_count/TestWatchpointCount.py
@@ -35,7 +35,7 @@ class TestWatchpointCount(TestBase):
         self.assertStopReason(
             stop_reason, lldb.eStopReasonWatchpoint, "watchpoint for x1 not hit"
         )
-        stop_reason_descr = thread.GetStopDescription(256)
+        stop_reason_descr = thread.stop_description
         self.assertEqual(stop_reason_descr, "watchpoint 1")
 
         process.Continue()
@@ -43,5 +43,5 @@ class TestWatchpointCount(TestBase):
         self.assertStopReason(
             stop_reason, lldb.eStopReasonWatchpoint, "watchpoint for x2 not hit"
         )
-        stop_reason_descr = thread.GetStopDescription(256)
+        stop_reason_descr = thread.stop_description
         self.assertEqual(stop_reason_descr, "watchpoint 2")
diff --git a/lldb/test/API/functionalities/gdb_remote_client/TestGDBRemoteClient.py b/lldb/test/API/functionalities/gdb_remote_client/TestGDBRemoteClient.py
index 12b464d..67c5d7d 100644
--- a/lldb/test/API/functionalities/gdb_remote_client/TestGDBRemoteClient.py
+++ b/lldb/test/API/functionalities/gdb_remote_client/TestGDBRemoteClient.py
@@ -594,7 +594,7 @@ class TestGDBRemoteClient(GDBRemoteTestBase):
         process = self.connect(target)
 
         self.assertEqual(process.threads[0].GetStopReason(), lldb.eStopReasonSignal)
-        self.assertEqual(process.threads[0].GetStopDescription(100), "signal SIGBUS")
+        self.assertEqual(process.threads[0].stop_description, "signal SIGBUS")
 
     def test_signal_lldb_old(self):
         class MyResponder(MockGDBServerResponder):
@@ -620,7 +620,7 @@ class TestGDBRemoteClient(GDBRemoteTestBase):
         process = self.connect(target)
 
         self.assertEqual(process.threads[0].GetStopReason(), lldb.eStopReasonSignal)
-        self.assertEqual(process.threads[0].GetStopDescription(100), "signal SIGUSR1")
+        self.assertEqual(process.threads[0].stop_description, "signal SIGUSR1")
 
     def test_signal_lldb(self):
         class MyResponder(MockGDBServerResponder):
@@ -643,7 +643,7 @@ class TestGDBRemoteClient(GDBRemoteTestBase):
         process = self.connect(target)
 
         self.assertEqual(process.threads[0].GetStopReason(), lldb.eStopReasonSignal)
-        self.assertEqual(process.threads[0].GetStopDescription(100), "signal SIGUSR1")
+        self.assertEqual(process.threads[0].stop_description, "signal SIGUSR1")
 
     def do_siginfo_test(self, platform, target_yaml, raw_data, expected):
         class MyResponder(MockGDBServerResponder):
diff --git a/lldb/test/API/functionalities/plugins/python_os_plugin/stepping_plugin_threads/TestOSPluginStepping.py b/lldb/test/API/functionalities/plugins/python_os_plugin/stepping_plugin_threads/TestOSPluginStepping.py
index 0d06a9d..dc555dd 100644
--- a/lldb/test/API/functionalities/plugins/python_os_plugin/stepping_plugin_threads/TestOSPluginStepping.py
+++ b/lldb/test/API/functionalities/plugins/python_os_plugin/stepping_plugin_threads/TestOSPluginStepping.py
@@ -123,5 +123,5 @@ class TestOSPluginStepping(TestBase):
             os_thread = self.get_os_thread()
             self.assertTrue(os_thread.IsValid(), "The OS thread is back after continue")
             self.assertIn(
-                "step out", os_thread.GetStopDescription(100), "Completed step out plan"
+                "step out", os_thread.stop_description, "Completed step out plan"
             )
diff --git a/lldb/test/API/functionalities/postmortem/minidump-new/TestMiniDumpNew.py b/lldb/test/API/functionalities/postmortem/minidump-new/TestMiniDumpNew.py
index 8776d72..4b7d24e 100644
--- a/lldb/test/API/functionalities/postmortem/minidump-new/TestMiniDumpNew.py
+++ b/lldb/test/API/functionalities/postmortem/minidump-new/TestMiniDumpNew.py
@@ -117,7 +117,7 @@ class MiniDumpNewTestCase(TestBase):
         self.assertEqual(self.process.GetNumThreads(), 1)
         thread = self.process.GetThreadAtIndex(0)
         self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonSignal)
-        stop_description = thread.GetStopDescription(256)
+        stop_description = thread.stop_description
         self.assertIn("SIGSEGV", stop_description)
 
     @skipIfLLVMTargetMissing("X86")
@@ -153,7 +153,7 @@ class MiniDumpNewTestCase(TestBase):
         self.assertEqual(self.process.GetNumThreads(), 1)
         thread = self.process.GetThreadAtIndex(0)
         self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonNone)
-        stop_description = thread.GetStopDescription(256)
+        stop_description = thread.stop_description
         self.assertEqual(stop_description, "")
 
     def test_snapshot_minidump_null_exn_code(self):
@@ -164,7 +164,7 @@ class MiniDumpNewTestCase(TestBase):
         self.assertEqual(self.process.GetNumThreads(), 1)
         thread = self.process.GetThreadAtIndex(0)
         self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonNone)
-        stop_description = thread.GetStopDescription(256)
+        stop_description = thread.stop_description
         self.assertEqual(stop_description, "")
 
     def check_register_unsigned(self, set, name, expected):
@@ -198,7 +198,7 @@ class MiniDumpNewTestCase(TestBase):
         self.assertEqual(self.process.GetNumThreads(), 1)
         thread = self.process.GetThreadAtIndex(0)
         self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonNone)
-        stop_description = thread.GetStopDescription(256)
+        stop_description = thread.stop_description
         self.assertEqual(stop_description, "")
         registers = thread.GetFrameAtIndex(0).GetRegisters()
         # Verify the GPR registers are all correct
@@ -261,7 +261,7 @@ class MiniDumpNewTestCase(TestBase):
         self.assertEqual(self.process.GetNumThreads(), 1)
         thread = self.process.GetThreadAtIndex(0)
         self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonNone)
-        stop_description = thread.GetStopDescription(256)
+        stop_description = thread.stop_description
         self.assertEqual(stop_description, "")
         registers = thread.GetFrameAtIndex(0).GetRegisters()
         # Verify the GPR registers are all correct
@@ -522,7 +522,7 @@ class MiniDumpNewTestCase(TestBase):
         for i in range(2):
             thread = self.process.GetThreadAtIndex(i)
             self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonSignal)
-            stop_description = thread.GetStopDescription(256)
+            stop_description = thread.stop_description
             self.assertIn("SIGSEGV", stop_description)
 
     def test_breakpoint_on_minidump(self):
@@ -539,7 +539,7 @@ class MiniDumpNewTestCase(TestBase):
             process = target.LoadCore(core)
             self.assertTrue(process, VALID_PROCESS)
             thread = process.GetThreadAtIndex(0)
-            stop_reason = thread.GetStopDescription(256)
+            stop_reason = thread.stop_description
             self.assertIn("breakpoint 1.1", stop_reason)
         finally:
             if os.path.isfile(core):
diff --git a/lldb/test/API/functionalities/postmortem/minidump/TestMiniDump.py b/lldb/test/API/functionalities/postmortem/minidump/TestMiniDump.py
index 8fe5d2a..362b219 100644
--- a/lldb/test/API/functionalities/postmortem/minidump/TestMiniDump.py
+++ b/lldb/test/API/functionalities/postmortem/minidump/TestMiniDump.py
@@ -32,7 +32,7 @@ class MiniDumpTestCase(TestBase):
         self.assertEqual(self.process.GetNumThreads(), 1)
         thread = self.process.GetThreadAtIndex(0)
         self.assertStopReason(thread.GetStopReason(), lldb.eStopReasonException)
-        stop_description = thread.GetStopDescription(256)
+        stop_description = thread.stop_description
         self.assertIn("0xc0000005", stop_description)
 
     def test_modules_in_mini_dump(self):
diff --git a/lldb/test/API/functionalities/scripted_process/stack_core_scripted_process.py b/lldb/test/API/functionalities/scripted_process/stack_core_scripted_process.py
index 736bb69..ee5ae32 100644
--- a/lldb/test/API/functionalities/scripted_process/stack_core_scripted_process.py
+++ b/lldb/test/API/functionalities/scripted_process/stack_core_scripted_process.py
@@ -204,9 +204,7 @@ class StackCoreScriptedThread(ScriptedThread):
             if self.is_stopped:
                 if "arm64" in self.scripted_process.arch:
                     stop_reason["type"] = lldb.eStopReasonException
-                    stop_reason["data"][
-                        "desc"
-                    ] = self.corefile_thread.GetStopDescription(100)
+                    stop_reason["data"]["desc"] = self.corefile_thread.stop_description
                 elif self.scripted_process.arch == "x86_64":
                     stop_reason["type"] = lldb.eStopReasonSignal
                     stop_reason["data"]["signal"] = signal.SIGTRAP
diff --git a/lldb/test/API/functionalities/step_scripted/TestStepScripted.py b/lldb/test/API/functionalities/step_scripted/TestStepScripted.py
index 5276369..343236a 100644
--- a/lldb/test/API/functionalities/step_scripted/TestStepScripted.py
+++ b/lldb/test/API/functionalities/step_scripted/TestStepScripted.py
@@ -41,7 +41,7 @@ class StepScriptedTestCase(TestBase):
 
         frame = thread.GetFrameAtIndex(0)
         self.assertEqual("main", frame.GetFunctionName())
-        stop_desc = thread.GetStopDescription(1000)
+        stop_desc = thread.stop_description
         self.assertIn("Stepping out from", stop_desc, "Got right description")
 
     def run_until_branch_instruction(self):
@@ -153,7 +153,7 @@ class StepScriptedTestCase(TestBase):
         self.assertTrue(foo_val.GetValueDidChange(), "Foo changed")
 
         # And we should have a reasonable stop description:
-        desc = thread.GetStopDescription(1000)
+        desc = thread.stop_description
         self.assertIn("Stepped until foo changed", desc, "Got right stop description")
 
     def test_stop_others_from_command(self):
diff --git a/lldb/test/API/functionalities/tsan/multiple/TestTsanMultiple.py b/lldb/test/API/functionalities/tsan/multiple/TestTsanMultiple.py
index 435e180..aa2d1d9 100644
--- a/lldb/test/API/functionalities/tsan/multiple/TestTsanMultiple.py
+++ b/lldb/test/API/functionalities/tsan/multiple/TestTsanMultiple.py
@@ -49,7 +49,7 @@ class TsanMultipleTestCase(TestBase):
             stop_description = (
                 self.dbg.GetSelectedTarget()
                 .process.GetSelectedThread()
-                .GetStopDescription(100)
+                .stop_description
             )
 
             self.assertTrue(
diff --git a/lldb/test/API/macosx/abort_with_payload/TestAbortWithPayload.py b/lldb/test/API/macosx/abort_with_payload/TestAbortWithPayload.py
index c10d958..a850908 100644
--- a/lldb/test/API/macosx/abort_with_payload/TestAbortWithPayload.py
+++ b/lldb/test/API/macosx/abort_with_payload/TestAbortWithPayload.py
@@ -61,7 +61,7 @@ class TestAbortWithPayload(TestBase):
         self.assertEqual(thread, sel_thread, "Selected the original thread")
         # Make sure the stop reason is right:
         self.assertEqual(
-            thread.GetStopDescription(100),
+            thread.stop_description,
             "abort with payload or reason",
             "Description was right",
         )
diff --git a/lldb/test/API/macosx/corefile-exception-reason/TestCorefileExceptionReason.py b/lldb/test/API/macosx/corefile-exception-reason/TestCorefileExceptionReason.py
index ada74a1..e452bb5 100644
--- a/lldb/test/API/macosx/corefile-exception-reason/TestCorefileExceptionReason.py
+++ b/lldb/test/API/macosx/corefile-exception-reason/TestCorefileExceptionReason.py
@@ -44,7 +44,7 @@ class TestCorefileExceptionReason(TestBase):
             self.runCmd("fr v")
 
         self.assertEqual(
-            thread.GetStopDescription(256), "ESR_EC_DABORT_EL0 (fault address: 0x0)"
+            thread.stop_description, "ESR_EC_DABORT_EL0 (fault address: 0x0)"
         )
 
         if self.TraceOn():
diff --git a/lldb/test/API/riscv/break-undecoded/TestBreakpointIllegal.py b/lldb/test/API/riscv/break-undecoded/TestBreakpointIllegal.py
index 41e8901..5b00298 100644
--- a/lldb/test/API/riscv/break-undecoded/TestBreakpointIllegal.py
+++ b/lldb/test/API/riscv/break-undecoded/TestBreakpointIllegal.py
@@ -17,7 +17,7 @@ class TestBreakpointIllegal(TestBase):
         )
         self.runCmd("thread step-inst")
         # we need to step more, as some compilers do not set appropriate debug info.
-        while cur_thread.GetStopDescription(256) == "instruction step into":
+        while cur_thread.stop_description == "instruction step into":
             self.runCmd("thread step-inst")
         # The stop reason of the thread should be illegal opcode.
         self.expect(
@@ -34,7 +34,7 @@ class TestBreakpointIllegal(TestBase):
         )
         self.runCmd("thread step-inst")
         # we need to step more, as some compilers do not set appropriate debug info.
-        while cur_thread.GetStopDescription(256) == "instruction step into":
+        while cur_thread.stop_description == "instruction step into":
             self.runCmd("thread step-inst")
         # The stop reason of the thread should be illegal opcode.
         self.expect(
diff --git a/llvm/CMakeLists.txt b/llvm/CMakeLists.txt
index e874d76..4c70b98 100644
--- a/llvm/CMakeLists.txt
+++ b/llvm/CMakeLists.txt
@@ -149,7 +149,7 @@ endforeach()
 # As we migrate runtimes to using the bootstrapping build, the set of default runtimes
 # should grow as we remove those runtimes from LLVM_ENABLE_PROJECTS above.
 set(LLVM_DEFAULT_RUNTIMES "libcxx;libcxxabi;libunwind")
-set(LLVM_SUPPORTED_RUNTIMES "libc;libunwind;libcxxabi;libcxx;compiler-rt;openmp;llvm-libgcc;offload;flang-rt;libclc")
+set(LLVM_SUPPORTED_RUNTIMES "libc;libunwind;libcxxabi;libcxx;compiler-rt;openmp;llvm-libgcc;offload;flang-rt;libclc;libsycl")
 set(LLVM_ENABLE_RUNTIMES "" CACHE STRING
   "Semicolon-separated list of runtimes to build, or \"all\" (${LLVM_DEFAULT_RUNTIMES}). Supported runtimes are ${LLVM_SUPPORTED_RUNTIMES}.")
 if(LLVM_ENABLE_RUNTIMES STREQUAL "all")
diff --git a/llvm/docs/CommandGuide/lit.rst b/llvm/docs/CommandGuide/lit.rst
index 938b7f9..eb90e95 100644
--- a/llvm/docs/CommandGuide/lit.rst
+++ b/llvm/docs/CommandGuide/lit.rst
@@ -356,6 +356,11 @@ The timing data is stored in the `test_exec_root` in a file named
   primary purpose is to suppress an ``XPASS`` result without modifying a test
   case that uses the ``XFAIL`` directive.
 
+.. option:: --exclude-xfail
+
+  ``XFAIL`` tests won't be run, unless they are listed in the ``--xfail-not``
+  (or ``LIT_XFAIL_NOT``) lists.
+
 .. option:: --num-shards M
 
  Divide the set of selected tests into ``M`` equal-sized subsets or
diff --git a/llvm/include/llvm/CodeGen/MachineBasicBlock.h b/llvm/include/llvm/CodeGen/MachineBasicBlock.h
index 938d71d..9e3d919 100644
--- a/llvm/include/llvm/CodeGen/MachineBasicBlock.h
+++ b/llvm/include/llvm/CodeGen/MachineBasicBlock.h
@@ -323,10 +323,11 @@ public:
   const MachineFunction *getParent() const { return xParent; }
   MachineFunction *getParent() { return xParent; }
 
-  /// Returns true if the original IR terminator is an `indirectbr`. This
-  /// typically corresponds to a `goto` in C, rather than jump tables.
-  bool terminatorIsComputedGoto() const {
-    return back().isIndirectBranch() &&
+  /// Returns true if the original IR terminator is an `indirectbr` with
+  /// successor blocks. This typically corresponds to a `goto` in C, rather than
+  /// jump tables.
+  bool terminatorIsComputedGotoWithSuccessors() const {
+    return back().isIndirectBranch() && !succ_empty() &&
            llvm::all_of(successors(), [](const MachineBasicBlock *Succ) {
              return Succ->isIRBlockAddressTaken();
            });
diff --git a/llvm/include/llvm/CodeGen/MachineInstrBuilder.h b/llvm/include/llvm/CodeGen/MachineInstrBuilder.h
index e63e77a..e705d7d9 100644
--- a/llvm/include/llvm/CodeGen/MachineInstrBuilder.h
+++ b/llvm/include/llvm/CodeGen/MachineInstrBuilder.h
@@ -69,6 +69,32 @@ enum {
 
 } // end namespace RegState
 
+/// Set of metadata that should be preserved when using BuildMI(). This provides
+/// a more convenient way of preserving DebugLoc, PCSections and MMRA.
+class MIMetadata {
+public:
+  MIMetadata() = default;
+  MIMetadata(DebugLoc DL, MDNode *PCSections = nullptr, MDNode *MMRA = nullptr)
+      : DL(std::move(DL)), PCSections(PCSections), MMRA(MMRA) {}
+  MIMetadata(const DILocation *DI, MDNode *PCSections = nullptr,
+             MDNode *MMRA = nullptr)
+      : DL(DI), PCSections(PCSections), MMRA(MMRA) {}
+  explicit MIMetadata(const Instruction &From)
+      : DL(From.getDebugLoc()),
+        PCSections(From.getMetadata(LLVMContext::MD_pcsections)) {}
+  explicit MIMetadata(const MachineInstr &From)
+      : DL(From.getDebugLoc()), PCSections(From.getPCSections()) {}
+
+  const DebugLoc &getDL() const { return DL; }
+  MDNode *getPCSections() const { return PCSections; }
+  MDNode *getMMRAMetadata() const { return MMRA; }
+
+private:
+  DebugLoc DL;
+  MDNode *PCSections = nullptr;
+  MDNode *MMRA = nullptr;
+};
+
 class MachineInstrBuilder {
   MachineFunction *MF = nullptr;
   MachineInstr *MI = nullptr;
@@ -317,15 +343,11 @@ public:
     }
   }
 
-  const MachineInstrBuilder &setPCSections(MDNode *MD) const {
-    if (MD)
-      MI->setPCSections(*MF, MD);
-    return *this;
-  }
-
-  const MachineInstrBuilder &setMMRAMetadata(MDNode *MMRA) const {
-    if (MMRA)
-      MI->setMMRAMetadata(*MF, MMRA);
+  const MachineInstrBuilder &copyMIMetadata(const MIMetadata &MIMD) const {
+    if (MIMD.getPCSections())
+      MI->setPCSections(*MF, MIMD.getPCSections());
+    if (MIMD.getMMRAMetadata())
+      MI->setMMRAMetadata(*MF, MIMD.getMMRAMetadata());
     return *this;
   }
 
@@ -343,38 +365,11 @@ public:
   }
 };
 
-/// Set of metadata that should be preserved when using BuildMI(). This provides
-/// a more convenient way of preserving DebugLoc, PCSections and MMRA.
-class MIMetadata {
-public:
-  MIMetadata() = default;
-  MIMetadata(DebugLoc DL, MDNode *PCSections = nullptr, MDNode *MMRA = nullptr)
-      : DL(std::move(DL)), PCSections(PCSections), MMRA(MMRA) {}
-  MIMetadata(const DILocation *DI, MDNode *PCSections = nullptr,
-             MDNode *MMRA = nullptr)
-      : DL(DI), PCSections(PCSections), MMRA(MMRA) {}
-  explicit MIMetadata(const Instruction &From)
-      : DL(From.getDebugLoc()),
-        PCSections(From.getMetadata(LLVMContext::MD_pcsections)) {}
-  explicit MIMetadata(const MachineInstr &From)
-      : DL(From.getDebugLoc()), PCSections(From.getPCSections()) {}
-
-  const DebugLoc &getDL() const { return DL; }
-  MDNode *getPCSections() const { return PCSections; }
-  MDNode *getMMRAMetadata() const { return MMRA; }
-
-private:
-  DebugLoc DL;
-  MDNode *PCSections = nullptr;
-  MDNode *MMRA = nullptr;
-};
-
 /// Builder interface. Specify how to create the initial instruction itself.
 inline MachineInstrBuilder BuildMI(MachineFunction &MF, const MIMetadata &MIMD,
                                    const MCInstrDesc &MCID) {
   return MachineInstrBuilder(MF, MF.CreateMachineInstr(MCID, MIMD.getDL()))
-      .setPCSections(MIMD.getPCSections())
-      .setMMRAMetadata(MIMD.getMMRAMetadata());
+      .copyMIMetadata(MIMD);
 }
 
 /// This version of the builder sets up the first operand as a
@@ -382,8 +377,7 @@ inline MachineInstrBuilder BuildMI(MachineFunction &MF, const MIMetadata &MIMD,
 inline MachineInstrBuilder BuildMI(MachineFunction &MF, const MIMetadata &MIMD,
                                    const MCInstrDesc &MCID, Register DestReg) {
   return MachineInstrBuilder(MF, MF.CreateMachineInstr(MCID, MIMD.getDL()))
-      .setPCSections(MIMD.getPCSections())
-      .setMMRAMetadata(MIMD.getMMRAMetadata())
+      .copyMIMetadata(MIMD)
       .addReg(DestReg, RegState::Define);
 }
 
@@ -397,10 +391,8 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineFunction &MF = *BB.getParent();
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
-  return MachineInstrBuilder(MF, MI)
-      .setPCSections(MIMD.getPCSections())
-      .setMMRAMetadata(MIMD.getMMRAMetadata())
-      .addReg(DestReg, RegState::Define);
+  return MachineInstrBuilder(MF, MI).copyMIMetadata(MIMD).addReg(
+      DestReg, RegState::Define);
 }
 
 /// This version of the builder inserts the newly-built instruction before
@@ -416,10 +408,8 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineFunction &MF = *BB.getParent();
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
-  return MachineInstrBuilder(MF, MI)
-      .setPCSections(MIMD.getPCSections())
-      .setMMRAMetadata(MIMD.getMMRAMetadata())
-      .addReg(DestReg, RegState::Define);
+  return MachineInstrBuilder(MF, MI).copyMIMetadata(MIMD).addReg(
+      DestReg, RegState::Define);
 }
 
 inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB, MachineInstr &I,
@@ -449,9 +439,7 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineFunction &MF = *BB.getParent();
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
-  return MachineInstrBuilder(MF, MI)
-      .setPCSections(MIMD.getPCSections())
-      .setMMRAMetadata(MIMD.getMMRAMetadata());
+  return MachineInstrBuilder(MF, MI).copyMIMetadata(MIMD);
 }
 
 inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
@@ -461,9 +449,7 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineFunction &MF = *BB.getParent();
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
-  return MachineInstrBuilder(MF, MI)
-      .setPCSections(MIMD.getPCSections())
-      .setMMRAMetadata(MIMD.getMMRAMetadata());
+  return MachineInstrBuilder(MF, MI).copyMIMetadata(MIMD);
 }
 
 inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB, MachineInstr &I,
diff --git a/llvm/include/llvm/CodeGen/SelectionDAGNodes.h b/llvm/include/llvm/CodeGen/SelectionDAGNodes.h
index 8f88811..11ae8cd 100644
--- a/llvm/include/llvm/CodeGen/SelectionDAGNodes.h
+++ b/llvm/include/llvm/CodeGen/SelectionDAGNodes.h
@@ -182,7 +182,7 @@ public:
     return SDValue(Node, R);
   }
 
-  /// Return true if this node is an operand of N.
+  /// Return true if the referenced return value is an operand of N.
   LLVM_ABI bool isOperandOf(const SDNode *N) const;
 
   /// Return the ValueType of the referenced return value.
diff --git a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
index 4a50558..e57f9b3 100644
--- a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
+++ b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
@@ -593,6 +593,10 @@ def int_amdgcn_tanh : DefaultAttrsIntrinsic<
   [llvm_anyfloat_ty], [LLVMMatchType<0>], [IntrNoMem, IntrSpeculatable]
 >;
 
+def int_amdgcn_cvt_sr_pk_f16_f32 : DefaultAttrsIntrinsic<
+  [llvm_v2f16_ty], [llvm_float_ty, llvm_float_ty, llvm_i32_ty], [IntrNoMem, IntrSpeculatable]
+>, ClangBuiltin<"__builtin_amdgcn_cvt_sr_pk_f16_f32">;
+
 def int_amdgcn_cvt_sr_pk_bf16_f32 : DefaultAttrsIntrinsic<
   [llvm_v2bf16_ty], [llvm_float_ty, llvm_float_ty, llvm_i32_ty], [IntrNoMem, IntrSpeculatable]
 >, ClangBuiltin<"__builtin_amdgcn_cvt_sr_pk_bf16_f32">;
diff --git a/llvm/lib/Analysis/VectorUtils.cpp b/llvm/lib/Analysis/VectorUtils.cpp
index 425ea31..b3b4c37 100644
--- a/llvm/lib/Analysis/VectorUtils.cpp
+++ b/llvm/lib/Analysis/VectorUtils.cpp
@@ -81,7 +81,6 @@ bool llvm::isTriviallyVectorizable(Intrinsic::ID ID) {
   case Intrinsic::exp:
   case Intrinsic::exp10:
   case Intrinsic::exp2:
-  case Intrinsic::ldexp:
   case Intrinsic::log:
   case Intrinsic::log10:
   case Intrinsic::log2:
@@ -109,8 +108,6 @@ bool llvm::isTriviallyVectorizable(Intrinsic::ID ID) {
   case Intrinsic::canonicalize:
   case Intrinsic::fptosi_sat:
   case Intrinsic::fptoui_sat:
-  case Intrinsic::lround:
-  case Intrinsic::llround:
   case Intrinsic::lrint:
   case Intrinsic::llrint:
   case Intrinsic::ucmp:
@@ -192,8 +189,6 @@ bool llvm::isVectorIntrinsicWithOverloadTypeAtArg(
   switch (ID) {
   case Intrinsic::fptosi_sat:
   case Intrinsic::fptoui_sat:
-  case Intrinsic::lround:
-  case Intrinsic::llround:
   case Intrinsic::lrint:
   case Intrinsic::llrint:
   case Intrinsic::vp_lrint:
@@ -208,7 +203,6 @@ bool llvm::isVectorIntrinsicWithOverloadTypeAtArg(
   case Intrinsic::vp_is_fpclass:
     return OpdIdx == 0;
   case Intrinsic::powi:
-  case Intrinsic::ldexp:
     return OpdIdx == -1 || OpdIdx == 1;
   default:
     return OpdIdx == -1;
diff --git a/llvm/lib/CGData/StableFunctionMapRecord.cpp b/llvm/lib/CGData/StableFunctionMapRecord.cpp
index 4e4fcef..423e068 100644
--- a/llvm/lib/CGData/StableFunctionMapRecord.cpp
+++ b/llvm/lib/CGData/StableFunctionMapRecord.cpp
@@ -160,14 +160,18 @@ void StableFunctionMapRecord::deserialize(const unsigned char *&Ptr,
   for (unsigned I = 0; I < NumFuncs; ++I) {
     auto Hash =
         endian::readNext<stable_hash, endianness::little, unaligned>(Ptr);
-    auto FunctionNameId =
+    [[maybe_unused]] auto FunctionNameId =
         endian::readNext<uint32_t, endianness::little, unaligned>(Ptr);
-    assert(FunctionMap->getNameForId(FunctionNameId) &&
-           "FunctionNameId out of range");
-    auto ModuleNameId =
+    [[maybe_unused]] auto ModuleNameId =
         endian::readNext<uint32_t, endianness::little, unaligned>(Ptr);
-    assert(FunctionMap->getNameForId(ModuleNameId) &&
-           "ModuleNameId out of range");
+    // Only validate IDs if we've read the names
+    if (ReadStableFunctionMapNames) {
+      assert(FunctionMap->getNameForId(FunctionNameId) &&
+             "FunctionNameId out of range");
+      assert(FunctionMap->getNameForId(ModuleNameId) &&
+             "ModuleNameId out of range");
+    }
+
     auto InstCount =
         endian::readNext<uint32_t, endianness::little, unaligned>(Ptr);
 
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
index 02d1100..f41b6eb 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
@@ -12782,7 +12782,7 @@ bool SDNode::areOnlyUsersOf(ArrayRef<const SDNode *> Nodes, const SDNode *N) {
   return Seen;
 }
 
-/// isOperand - Return true if this node is an operand of N.
+/// Return true if the referenced return value is an operand of N.
 bool SDValue::isOperandOf(const SDNode *N) const {
   return is_contained(N->op_values(), *this);
 }
diff --git a/llvm/lib/CodeGen/TailDuplicator.cpp b/llvm/lib/CodeGen/TailDuplicator.cpp
index a88c57f..5d720fb 100644
--- a/llvm/lib/CodeGen/TailDuplicator.cpp
+++ b/llvm/lib/CodeGen/TailDuplicator.cpp
@@ -604,12 +604,21 @@ bool TailDuplicator::shouldTailDuplicate(bool IsSimple,
   bool HasComputedGoto = false;
   if (!TailBB.empty()) {
     HasIndirectbr = TailBB.back().isIndirectBranch();
-    HasComputedGoto = TailBB.terminatorIsComputedGoto();
+    HasComputedGoto = TailBB.terminatorIsComputedGotoWithSuccessors();
   }
 
   if (HasIndirectbr && PreRegAlloc)
     MaxDuplicateCount = TailDupIndirectBranchSize;
 
+  // Allow higher limits when the block has computed-gotos and running after
+  // register allocation. NB. This basically unfactors computed gotos that were
+  // factored early on in the compilation process to speed up edge based data
+  // flow. If we do not unfactor them again, it can seriously pessimize code
+  // with many computed jumps in the source code, such as interpreters.
+  // Therefore we do not restrict the computed gotos.
+  if (HasComputedGoto && !PreRegAlloc)
+    MaxDuplicateCount = std::max(MaxDuplicateCount, 10u);
+
   // Check the instructions in the block to determine whether tail-duplication
   // is invalid or unlikely to be profitable.
   unsigned InstrCount = 0;
@@ -663,12 +672,7 @@ bool TailDuplicator::shouldTailDuplicate(bool IsSimple,
   // Duplicating a BB which has both multiple predecessors and successors will
   // may cause huge amount of PHI nodes. If we want to remove this limitation,
   // we have to address https://github.com/llvm/llvm-project/issues/78578.
-  // NB. This basically unfactors computed gotos that were factored early on in
-  // the compilation process to speed up edge based data flow. If we do not
-  // unfactor them again, it can seriously pessimize code with many computed
-  // jumps in the source code, such as interpreters. Therefore we do not
-  // restrict the computed gotos.
-  if (!HasComputedGoto && TailBB.pred_size() > TailDupPredSize &&
+  if (PreRegAlloc && TailBB.pred_size() > TailDupPredSize &&
       TailBB.succ_size() > TailDupSuccSize) {
     // If TailBB or any of its successors contains a phi, we may have to add a
     // large number of additional phis with additional incoming values.
diff --git a/llvm/lib/CodeGen/TargetInstrInfo.cpp b/llvm/lib/CodeGen/TargetInstrInfo.cpp
index 18d6bbc..705e046e 100644
--- a/llvm/lib/CodeGen/TargetInstrInfo.cpp
+++ b/llvm/lib/CodeGen/TargetInstrInfo.cpp
@@ -1406,7 +1406,7 @@ void TargetInstrInfo::reassociateOps(
                               const MCInstrDesc &MCID, Register DestReg) {
     return MachineInstrBuilder(
                MF, MF.CreateMachineInstr(MCID, MIMD.getDL(), /*NoImpl=*/true))
-        .setPCSections(MIMD.getPCSections())
+        .copyMIMetadata(MIMD)
         .addReg(DestReg, RegState::Define);
   };
 
diff --git a/llvm/lib/Support/BLAKE3/CMakeLists.txt b/llvm/lib/Support/BLAKE3/CMakeLists.txt
index eae2b02..90311ae 100644
--- a/llvm/lib/Support/BLAKE3/CMakeLists.txt
+++ b/llvm/lib/Support/BLAKE3/CMakeLists.txt
@@ -26,7 +26,8 @@ endmacro()
 if (CAN_USE_ASSEMBLER)
   if (MSVC)
     check_symbol_exists(_M_X64 "" IS_X64)
-    if (IS_X64)
+    check_symbol_exists(_M_ARM64EC "" IS_ARM64EC)
+    if (IS_X64 AND NOT IS_ARM64EC)
       enable_language(ASM_MASM)
       set(LLVM_BLAKE3_ASM_FILES
         blake3_sse2_x86-64_windows_msvc.asm
diff --git a/llvm/lib/Support/Windows/Threading.inc b/llvm/lib/Support/Windows/Threading.inc
index 8dd7c88..b11f216ad 100644
--- a/llvm/lib/Support/Windows/Threading.inc
+++ b/llvm/lib/Support/Windows/Threading.inc
@@ -136,6 +136,7 @@ HMODULE loadSystemModuleSecure(LPCWSTR lpModuleName) {
 } // namespace llvm::sys::windows
 
 SetThreadPriorityResult llvm::set_thread_priority(ThreadPriority Priority) {
+#ifdef THREAD_POWER_THROTTLING_CURRENT_VERSION
   HMODULE kernelM = llvm::sys::windows::loadSystemModuleSecure(L"kernel32.dll");
   if (kernelM) {
     // SetThreadInformation is only available on Windows 8 and later. Since we
@@ -166,6 +167,7 @@ SetThreadPriorityResult llvm::set_thread_priority(ThreadPriority Priority) {
                                : 0);
     }
   }
+#endif
 
   // https://docs.microsoft.com/en-us/windows/desktop/api/processthreadsapi/nf-processthreadsapi-setthreadpriority
   // Begin background processing mode. The system lowers the resource scheduling
diff --git a/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
index 6bca2fe..c3b14ab 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
@@ -4574,6 +4574,7 @@ AMDGPURegisterBankInfo::getInstrMapping(const MachineInstr &MI) const {
     case Intrinsic::amdgcn_cvt_pknorm_u16:
     case Intrinsic::amdgcn_cvt_pk_i16:
     case Intrinsic::amdgcn_cvt_pk_u16:
+    case Intrinsic::amdgcn_cvt_sr_pk_f16_f32:
     case Intrinsic::amdgcn_cvt_sr_pk_bf16_f32:
     case Intrinsic::amdgcn_cvt_pk_f16_fp8:
     case Intrinsic::amdgcn_cvt_pk_f16_bf8:
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.td b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
index efcc88e..c5931fc 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
@@ -2928,6 +2928,7 @@ def VOP_V32F32_V6I32_F32 : VOPProfile <[v32f32, v6i32, f32, untyped]>;
 def VOP_V32F16_V6I32_F32 : VOPProfile <[v32f16, v6i32, f32, untyped]>;
 def VOP_V32BF16_V6I32_F32 : VOPProfile <[v32bf16, v6i32, f32, untyped]>;
 def VOP_V2BF16_F32_F32_I32 : VOPProfile <[v2bf16, f32, f32, i32]>;
+def VOP_V2F16_F32_F32_I32 : VOPProfile <[v2f16, f32, f32, i32]>;
 def VOP_V6I32_V32F16_F32 : VOPProfile<[v6i32, v32f16, f32, untyped]>;
 def VOP_V6I32_V32BF16_F32 : VOPProfile<[v6i32, v32bf16, f32, untyped]>;
 def VOP_V6I32_V16F32_V16F32_F32 : VOPProfile<[v6i32, v16f32, v16f32, f32]>;
diff --git a/llvm/lib/Target/AMDGPU/VOP3Instructions.td b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
index 2d3caec..cfc5fe5 100644
--- a/llvm/lib/Target/AMDGPU/VOP3Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
@@ -1650,6 +1650,8 @@ def VOP3_CVT_SR_F8_F16_Fake16_Profile : VOP3_Profile_Fake16<VOP3_CVT_SR_F8_F16_P
 
 let SubtargetPredicate = isGFX1250Plus in {
   let ReadsModeReg = 0 in {
+    defm V_CVT_SR_PK_F16_F32 : VOP3Inst<"v_cvt_sr_pk_f16_f32", VOP3_Profile<VOP_V2F16_F32_F32_I32>, int_amdgcn_cvt_sr_pk_f16_f32>;
+
     // These instructions have non-standard use of op_sel. They are using bits 2 and 3 of opsel
     // to select a byte in the vdst. Bits 0 and 1 are unused.
     let Constraints = "$vdst = $vdst_in", DisableEncoding = "$vdst_in" in {
@@ -2066,6 +2068,8 @@ defm V_ASHR_PK_I8_I32                : VOP3Only_Realtriple_gfx1250<0x290>;
 defm V_ASHR_PK_U8_I32                : VOP3Only_Realtriple_gfx1250<0x291>;
 defm V_CVT_PK_BF16_F32               : VOP3Only_Realtriple_gfx1250<0x36d>;
 defm V_CVT_SR_PK_BF16_F32            : VOP3Only_Realtriple_gfx1250<0x36e>;
+defm V_CVT_PK_F16_F32                : VOP3Only_Realtriple_gfx1250<0x36f>;
+defm V_CVT_SR_PK_F16_F32             : VOP3Only_Realtriple_gfx1250<0x370>;
 defm V_CVT_PK_FP8_F16_gfx1250        : VOP3Only_Realtriple_t16_and_fake16_gfx1250<0x372, "v_cvt_pk_fp8_f16">;
 defm V_CVT_PK_BF8_F16_gfx1250        : VOP3Only_Realtriple_t16_and_fake16_gfx1250<0x373, "v_cvt_pk_bf8_f16">;
 defm V_CVT_SR_FP8_F16                : VOP3Only_Realtriple_t16_and_fake16_gfx1250<0x374>;
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index 61dbd06..0d5eb86 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -2627,18 +2627,17 @@ void RISCVTTIImpl::getUnrollingPreferences(
   if (L->getNumBlocks() > 4)
     return;
 
-  // Don't unroll vectorized loops, including the remainder loop
-  if (getBooleanLoopAttribute(L, "llvm.loop.isvectorized"))
-    return;
-
   // Scan the loop: don't unroll loops with calls as this could prevent
-  // inlining.
+  // inlining. Don't unroll auto-vectorized loops either, though do allow
+  // unrolling of the scalar remainder.
+  bool IsVectorized = getBooleanLoopAttribute(L, "llvm.loop.isvectorized");
   InstructionCost Cost = 0;
   for (auto *BB : L->getBlocks()) {
     for (auto &I : *BB) {
-      // Initial setting - Don't unroll loops containing vectorized
-      // instructions.
-      if (I.getType()->isVectorTy())
+      // Both auto-vectorized loops and the scalar remainder have the
+      // isvectorized attribute, so differentiate between them by the presence
+      // of vector instructions.
+      if (IsVectorized && I.getType()->isVectorTy())
         return;
 
       if (isa<CallInst>(I) || isa<InvokeInst>(I)) {
diff --git a/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp b/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
index e30d723..fb0a47d 100644
--- a/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
+++ b/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
@@ -9044,7 +9044,7 @@ static unsigned detectEvenOddMultiplyOperand(const SelectionDAG &DAG,
         if (unsigned(ShuffleMask[Elt]) != 2 * Elt)
           CanUseEven = false;
         if (unsigned(ShuffleMask[Elt]) != 2 * Elt + 1)
-          CanUseEven = true;
+          CanUseOdd = false;
       }
       Op = Op.getOperand(0);
       if (CanUseEven)
diff --git a/llvm/lib/TargetParser/Host.cpp b/llvm/lib/TargetParser/Host.cpp
index 78bd5b4..7e09d30 100644
--- a/llvm/lib/TargetParser/Host.cpp
+++ b/llvm/lib/TargetParser/Host.cpp
@@ -587,8 +587,9 @@ StringRef sys::detail::getHostCPUNameForBPF() {
 #endif
 }
 
-#if defined(__i386__) || defined(_M_IX86) || defined(__x86_64__) ||            \
-    defined(_M_X64)
+#if (defined(__i386__) || defined(_M_IX86) || defined(__x86_64__) ||           \
+     defined(_M_X64)) &&                                                       \
+    !defined(_M_ARM64EC)
 
 /// getX86CpuIDAndInfo - Execute the specified cpuid and return the 4 values in
 /// the specified arguments.  If we can't run cpuid on the host, return true.
@@ -1853,8 +1854,9 @@ VendorSignatures getVendorSignature(unsigned *MaxLeaf) {
 } // namespace llvm
 #endif
 
-#if defined(__i386__) || defined(_M_IX86) || \
-    defined(__x86_64__) || defined(_M_X64)
+#if (defined(__i386__) || defined(_M_IX86) || defined(__x86_64__) ||           \
+     defined(_M_X64)) &&                                                       \
+    !defined(_M_ARM64EC)
 StringMap<bool> sys::getHostCPUFeatures() {
   unsigned EAX = 0, EBX = 0, ECX = 0, EDX = 0;
   unsigned MaxLevel;
@@ -2147,7 +2149,8 @@ StringMap<bool> sys::getHostCPUFeatures() {
 
   return Features;
 }
-#elif defined(_WIN32) && (defined(__aarch64__) || defined(_M_ARM64))
+#elif defined(_WIN32) && (defined(__aarch64__) || defined(_M_ARM64) ||         \
+                          defined(__arm64ec__) || defined(_M_ARM64EC))
 StringMap<bool> sys::getHostCPUFeatures() {
   StringMap<bool> Features;
 
diff --git a/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp b/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
index 7af5ba4..40a7f80 100644
--- a/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
+++ b/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
@@ -458,29 +458,19 @@ static bool foldSqrt(CallInst *Call, LibFunc Func, TargetTransformInfo &TTI,
 // Check if this array of constants represents a cttz table.
 // Iterate over the elements from \p Table by trying to find/match all
 // the numbers from 0 to \p InputBits that should represent cttz results.
-static bool isCTTZTable(const ConstantDataArray &Table, uint64_t Mul,
-                        uint64_t Shift, uint64_t InputBits) {
-  unsigned Length = Table.getNumElements();
-  if (Length < InputBits || Length > InputBits * 2)
-    return false;
-
-  APInt Mask = APInt::getBitsSetFrom(InputBits, Shift);
-  unsigned Matched = 0;
-
-  for (unsigned i = 0; i < Length; i++) {
-    uint64_t Element = Table.getElementAsInteger(i);
-    if (Element >= InputBits)
-      continue;
-
-    // Check if \p Element matches a concrete answer. It could fail for some
-    // elements that are never accessed, so we keep iterating over each element
-    // from the table. The number of matched elements should be equal to the
-    // number of potential right answers which is \p InputBits actually.
-    if ((((Mul << Element) & Mask.getZExtValue()) >> Shift) == i)
-      Matched++;
+static bool isCTTZTable(Constant *Table, const APInt &Mul, const APInt &Shift,
+                        const APInt &AndMask, Type *AccessTy,
+                        unsigned InputBits, const APInt &GEPIdxFactor,
+                        const DataLayout &DL) {
+  for (unsigned Idx = 0; Idx < InputBits; Idx++) {
+    APInt Index = (APInt(InputBits, 1).shl(Idx) * Mul).lshr(Shift) & AndMask;
+    ConstantInt *C = dyn_cast_or_null<ConstantInt>(
+        ConstantFoldLoadFromConst(Table, AccessTy, Index * GEPIdxFactor, DL));
+    if (!C || C->getValue() != Idx)
+      return false;
   }
 
-  return Matched == InputBits;
+  return true;
 }
 
 // Try to recognize table-based ctz implementation.
@@ -495,6 +485,11 @@ static bool isCTTZTable(const ConstantDataArray &Table, uint64_t Mul,
 // this can be lowered to `cttz` instruction.
 // There is also a special case when the element is 0.
 //
+// The (x & -x) sets the lowest non-zero bit to 1. The multiply is a de-bruijn
+// sequence that contains each pattern of bits in it. The shift extracts
+// the top bits after the multiply, and that index into the table should
+// represent the number of trailing zeros in the original number.
+//
 // Here are some examples or LLVM IR for a 64-bit target:
 //
 // CASE 1:
@@ -536,8 +531,8 @@ static bool isCTTZTable(const ConstantDataArray &Table, uint64_t Mul,
 //     i64 %shr
 // %0 = load i8, i8* %arrayidx, align 1, !tbaa !8
 //
-// All this can be lowered to @llvm.cttz.i32/64 intrinsic.
-static bool tryToRecognizeTableBasedCttz(Instruction &I) {
+// All these can be lowered to @llvm.cttz.i32/64 intrinsics.
+static bool tryToRecognizeTableBasedCttz(Instruction &I, const DataLayout &DL) {
   LoadInst *LI = dyn_cast<LoadInst>(&I);
   if (!LI)
     return false;
@@ -547,53 +542,47 @@ static bool tryToRecognizeTableBasedCttz(Instruction &I) {
     return false;
 
   GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(LI->getPointerOperand());
-  if (!GEP || !GEP->hasNoUnsignedSignedWrap() || GEP->getNumIndices() != 2)
-    return false;
-
-  if (!GEP->getSourceElementType()->isArrayTy())
-    return false;
-
-  uint64_t ArraySize = GEP->getSourceElementType()->getArrayNumElements();
-  if (ArraySize != 32 && ArraySize != 64)
+  if (!GEP || !GEP->hasNoUnsignedSignedWrap())
     return false;
 
   GlobalVariable *GVTable = dyn_cast<GlobalVariable>(GEP->getPointerOperand());
   if (!GVTable || !GVTable->hasInitializer() || !GVTable->isConstant())
     return false;
 
-  ConstantDataArray *ConstData =
-      dyn_cast<ConstantDataArray>(GVTable->getInitializer());
-  if (!ConstData)
-    return false;
-
-  if (!match(GEP->idx_begin()->get(), m_ZeroInt()))
+  unsigned BW = DL.getIndexTypeSizeInBits(GEP->getType());
+  APInt ModOffset(BW, 0);
+  SmallMapVector<Value *, APInt, 4> VarOffsets;
+  if (!GEP->collectOffset(DL, BW, VarOffsets, ModOffset) ||
+      VarOffsets.size() != 1 || ModOffset != 0)
     return false;
+  auto [GepIdx, GEPScale] = VarOffsets.front();
 
-  Value *Idx2 = std::next(GEP->idx_begin())->get();
   Value *X1;
-  uint64_t MulConst, ShiftConst;
-  // FIXME: 64-bit targets have `i64` type for the GEP index, so this match will
-  // probably fail for other (e.g. 32-bit) targets.
-  if (!match(Idx2, m_ZExtOrSelf(
-                       m_LShr(m_Mul(m_c_And(m_Neg(m_Value(X1)), m_Deferred(X1)),
-                                    m_ConstantInt(MulConst)),
-                              m_ConstantInt(ShiftConst)))))
+  const APInt *MulConst, *ShiftConst, *AndCst = nullptr;
+  // Check that the gep variable index is ((x & -x) * MulConst) >> ShiftConst.
+  // This might be extended to the pointer index type, and if the gep index type
+  // has been replaced with an i8 then a new And (and different ShiftConst) will
+  // be present.
+  auto MatchInner = m_LShr(
+      m_Mul(m_c_And(m_Neg(m_Value(X1)), m_Deferred(X1)), m_APInt(MulConst)),
+      m_APInt(ShiftConst));
+  if (!match(GepIdx, m_CastOrSelf(MatchInner)) &&
+      !match(GepIdx, m_CastOrSelf(m_And(MatchInner, m_APInt(AndCst)))))
     return false;
 
   unsigned InputBits = X1->getType()->getScalarSizeInBits();
-  if (InputBits != 32 && InputBits != 64)
-    return false;
-
-  // Shift should extract top 5..7 bits.
-  if (InputBits - Log2_32(InputBits) != ShiftConst &&
-      InputBits - Log2_32(InputBits) - 1 != ShiftConst)
+  if (InputBits != 16 && InputBits != 32 && InputBits != 64 && InputBits != 128)
     return false;
 
-  if (!isCTTZTable(*ConstData, MulConst, ShiftConst, InputBits))
+  if (!GEPScale.isIntN(InputBits) ||
+      !isCTTZTable(GVTable->getInitializer(), *MulConst, *ShiftConst,
+                   AndCst ? *AndCst : APInt::getAllOnes(InputBits), AccessType,
+                   InputBits, GEPScale.zextOrTrunc(InputBits), DL))
     return false;
 
-  auto ZeroTableElem = ConstData->getElementAsInteger(0);
-  bool DefinedForZero = ZeroTableElem == InputBits;
+  ConstantInt *ZeroTableElem = cast<ConstantInt>(
+      ConstantFoldLoadFromConst(GVTable->getInitializer(), AccessType, DL));
+  bool DefinedForZero = ZeroTableElem->getZExtValue() == InputBits;
 
   IRBuilder<> B(LI);
   ConstantInt *BoolConst = B.getInt1(!DefinedForZero);
@@ -607,8 +596,7 @@ static bool tryToRecognizeTableBasedCttz(Instruction &I) {
     // If the value in elem 0 isn't the same as InputBits, we still want to
     // produce the value from the table.
     auto Cmp = B.CreateICmpEQ(X1, ConstantInt::get(XType, 0));
-    auto Select =
-        B.CreateSelect(Cmp, ConstantInt::get(XType, ZeroTableElem), Cttz);
+    auto Select = B.CreateSelect(Cmp, B.CreateZExt(ZeroTableElem, XType), Cttz);
 
     // NOTE: If the table[0] is 0, but the cttz(0) is defined by the Target
     // it should be handled as: `cttz(x) & (typeSize - 1)`.
@@ -1477,7 +1465,7 @@ static bool foldUnusualPatterns(Function &F, DominatorTree &DT,
       MadeChange |= foldGuardedFunnelShift(I, DT);
       MadeChange |= tryToRecognizePopCount(I);
       MadeChange |= tryToFPToSat(I, TTI);
-      MadeChange |= tryToRecognizeTableBasedCttz(I);
+      MadeChange |= tryToRecognizeTableBasedCttz(I, DL);
       MadeChange |= foldConsecutiveLoads(I, DL, TTI, AA, DT);
       MadeChange |= foldPatternedLoads(I, DL);
       MadeChange |= foldICmpOrChain(I, DL, TTI, AA, DT);
diff --git a/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp b/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
index 4e5a8d1..bcb90d6 100644
--- a/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
+++ b/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
@@ -160,6 +160,16 @@ static cl::opt<bool> ClGenerateTagsWithCalls(
 static cl::opt<bool> ClGlobals("hwasan-globals", cl::desc("Instrument globals"),
                                cl::Hidden, cl::init(false));
 
+static cl::opt<bool> ClAllGlobals(
+    "hwasan-all-globals",
+    cl::desc(
+        "Instrument globals, even those within user-defined sections. Warning: "
+        "This may break existing code which walks globals via linker-generated "
+        "symbols, expects certain globals to be contiguous with each other, or "
+        "makes other assumptions which are invalidated by HWASan "
+        "instrumentation."),
+    cl::Hidden, cl::init(false));
+
 static cl::opt<int> ClMatchAllTag(
     "hwasan-match-all-tag",
     cl::desc("don't report bad accesses via pointers with this tag"),
@@ -681,11 +691,11 @@ void HWAddressSanitizer::initializeModule() {
       !CompileKernel && !UsePageAliases && optOr(ClGlobals, NewRuntime);
 
   if (!CompileKernel) {
-    createHwasanCtorComdat();
-
     if (InstrumentGlobals)
       instrumentGlobals();
 
+    createHwasanCtorComdat();
+
     bool InstrumentPersonalityFunctions =
         optOr(ClInstrumentPersonalityFunctions, NewRuntime);
     if (InstrumentPersonalityFunctions)
@@ -1772,11 +1782,17 @@ void HWAddressSanitizer::instrumentGlobals() {
     if (GV.hasCommonLinkage())
       continue;
 
-    // Globals with custom sections may be used in __start_/__stop_ enumeration,
-    // which would be broken both by adding tags and potentially by the extra
-    // padding/alignment that we insert.
-    if (GV.hasSection())
-      continue;
+    if (ClAllGlobals) {
+      // Avoid instrumenting intrinsic global variables.
+      if (GV.getSection() == "llvm.metadata")
+        continue;
+    } else {
+      // Globals with custom sections may be used in __start_/__stop_
+      // enumeration, which would be broken both by adding tags and potentially
+      // by the extra padding/alignment that we insert.
+      if (GV.hasSection())
+        continue;
+    }
 
     Globals.push_back(&GV);
   }
diff --git a/llvm/test/CodeGen/AArch64/late-taildup-computed-goto.ll b/llvm/test/CodeGen/AArch64/late-taildup-computed-goto.ll
index c4a027c..381904f 100644
--- a/llvm/test/CodeGen/AArch64/late-taildup-computed-goto.ll
+++ b/llvm/test/CodeGen/AArch64/late-taildup-computed-goto.ll
@@ -25,77 +25,58 @@ define void @test_interp(ptr %frame, ptr %dst) {
 ; CHECK-NEXT:    adrp x21, _opcode.targets@PAGE
 ; CHECK-NEXT:  Lloh1:
 ; CHECK-NEXT:    add x21, x21, _opcode.targets@PAGEOFF
-; CHECK-NEXT:    mov x22, xzr
+; CHECK-NEXT:    mov x24, xzr
 ; CHECK-NEXT:    add x8, x21, xzr, lsl #3
 ; CHECK-NEXT:    mov x19, x1
 ; CHECK-NEXT:    mov x20, x0
-; CHECK-NEXT:    add x23, x22, #1
+; CHECK-NEXT:    mov x23, xzr
+; CHECK-NEXT:    mov w22, #1 ; =0x1
+; CHECK-NEXT:    add x24, x24, #1
 ; CHECK-NEXT:    br x8
 ; CHECK-NEXT:  Ltmp0: ; Block address taken
 ; CHECK-NEXT:  LBB0_1: ; %loop.header
 ; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    add x8, x21, x23, lsl #3
+; CHECK-NEXT:    add x8, x21, x24, lsl #3
 ; CHECK-NEXT:    mov x20, xzr
-; CHECK-NEXT:    mov x22, xzr
-; CHECK-NEXT:    add x23, x23, #1
+; CHECK-NEXT:    mov x23, xzr
+; CHECK-NEXT:    add x24, x24, #1
 ; CHECK-NEXT:    br x8
 ; CHECK-NEXT:  Ltmp1: ; Block address taken
 ; CHECK-NEXT:  LBB0_2: ; %op1.bb
-; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    str xzr, [x19]
-; CHECK-NEXT:    mov w8, #1 ; =0x1
+; CHECK-NEXT:  Ltmp2: ; Block address taken
+; CHECK-NEXT:  LBB0_3: ; %op6.bb
+; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    ldr x0, [x20, #-8]!
-; CHECK-NEXT:    ldr x9, [x0, #8]
-; CHECK-NEXT:    str x8, [x0]
-; CHECK-NEXT:    ldr x8, [x9, #48]
+; CHECK-NEXT:    ldr x8, [x0, #8]
+; CHECK-NEXT:    str x22, [x0]
+; CHECK-NEXT:    ldr x8, [x8, #48]
 ; CHECK-NEXT:    blr x8
-; CHECK-NEXT:    add x8, x21, x23, lsl #3
-; CHECK-NEXT:    add x23, x23, #1
+; CHECK-NEXT:    add x8, x21, x24, lsl #3
+; CHECK-NEXT:    add x24, x24, #1
 ; CHECK-NEXT:    br x8
-; CHECK-NEXT:  Ltmp2: ; Block address taken
-; CHECK-NEXT:  LBB0_3: ; %op2.bb
+; CHECK-NEXT:  Ltmp3: ; Block address taken
+; CHECK-NEXT:  LBB0_4: ; %op2.bb
 ; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    add x8, x21, x23, lsl #3
+; CHECK-NEXT:    add x8, x21, x24, lsl #3
 ; CHECK-NEXT:    mov x20, xzr
-; CHECK-NEXT:    add x23, x23, #1
-; CHECK-NEXT:    str x22, [x19]
-; CHECK-NEXT:    mov x22, xzr
+; CHECK-NEXT:    str x23, [x19]
+; CHECK-NEXT:    mov x23, xzr
+; CHECK-NEXT:    add x24, x24, #1
 ; CHECK-NEXT:    br x8
-; CHECK-NEXT:  Ltmp3: ; Block address taken
-; CHECK-NEXT:  LBB0_4: ; %op4.bb
-; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    str x22, [x19]
-; CHECK-NEXT:    add x10, x21, x23, lsl #3
-; CHECK-NEXT:    add x23, x23, #1
-; CHECK-NEXT:    ldur x8, [x22, #12]
-; CHECK-NEXT:    ldur x9, [x20, #-8]
-; CHECK-NEXT:    add x22, x22, #20
-; CHECK-NEXT:    stp x8, x9, [x20, #-8]
-; CHECK-NEXT:    add x20, x20, #8
-; CHECK-NEXT:    br x10
 ; CHECK-NEXT:  Ltmp4: ; Block address taken
-; CHECK-NEXT:  LBB0_5: ; %op5.bb
+; CHECK-NEXT:  LBB0_5: ; %op4.bb
+; CHECK-NEXT:  Ltmp5: ; Block address taken
+; CHECK-NEXT:  LBB0_6: ; %op5.bb
 ; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    str x22, [x19]
-; CHECK-NEXT:    add x10, x21, x23, lsl #3
-; CHECK-NEXT:    add x23, x23, #1
-; CHECK-NEXT:    ldur x8, [x22, #12]
+; CHECK-NEXT:    str x23, [x19]
+; CHECK-NEXT:    ldur x8, [x23, #12]
 ; CHECK-NEXT:    ldur x9, [x20, #-8]
-; CHECK-NEXT:    add x22, x22, #20
+; CHECK-NEXT:    add x23, x23, #20
 ; CHECK-NEXT:    stp x8, x9, [x20, #-8]
+; CHECK-NEXT:    add x8, x21, x24, lsl #3
 ; CHECK-NEXT:    add x20, x20, #8
-; CHECK-NEXT:    br x10
-; CHECK-NEXT:  Ltmp5: ; Block address taken
-; CHECK-NEXT:  LBB0_6: ; %op6.bb
-; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    ldr x0, [x20, #-8]!
-; CHECK-NEXT:    mov w8, #1 ; =0x1
-; CHECK-NEXT:    ldr x9, [x0, #8]
-; CHECK-NEXT:    str x8, [x0]
-; CHECK-NEXT:    ldr x8, [x9, #48]
-; CHECK-NEXT:    blr x8
-; CHECK-NEXT:    add x8, x21, x23, lsl #3
-; CHECK-NEXT:    add x23, x23, #1
+; CHECK-NEXT:    add x24, x24, #1
 ; CHECK-NEXT:    br x8
 ; CHECK-NEXT:    .loh AdrpAdd Lloh0, Lloh1
 entry:
diff --git a/llvm/test/CodeGen/AMDGPU/fptrunc.f16.ll b/llvm/test/CodeGen/AMDGPU/fptrunc.f16.ll
index 1d33c26..c52fb61 100644
--- a/llvm/test/CodeGen/AMDGPU/fptrunc.f16.ll
+++ b/llvm/test/CodeGen/AMDGPU/fptrunc.f16.ll
@@ -11,6 +11,10 @@
 ; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1100 -global-isel=0 -mattr=-flat-for-global,-real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX11-SDAG-FAKE16 %s
 ; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1100 -global-isel=1 -mattr=-flat-for-global,+real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX11-GISEL-TRUE16 %s
 ; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1100 -global-isel=1 -mattr=-flat-for-global,-real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX11-GISEL-FAKE16 %s
+;  TODO: FIXME-TRUE16 llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1250 -global-isel=0 -mattr=-flat-for-global,+real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX1250-SDAG-TRUE16 %s
+; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1250 -global-isel=0 -mattr=-flat-for-global,-real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX1250-SDAG-FAKE16 %s
+; TODO: FIXME-TRUE16  llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1250 -global-isel=1 -mattr=-flat-for-global,+real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX1250-GISEL-TRUE16 %s
+; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=gfx1250 -global-isel=1 -mattr=-flat-for-global,-real-true16 -denormal-fp-math=preserve-sign < %s | FileCheck -enable-var-scope -check-prefixes=GFX1250-GISEL-FAKE16 %s
 
 define amdgpu_kernel void @fptrunc_f32_to_f16(
 ; SI-SDAG-LABEL: fptrunc_f32_to_f16:
@@ -192,6 +196,39 @@ define amdgpu_kernel void @fptrunc_f32_to_f16(
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_f32_to_f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_f32_to_f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_3)
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -381,6 +418,39 @@ define amdgpu_kernel void @fptrunc_f32_to_f16_afn(ptr addrspace(1) %r,
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_f32_to_f16_afn:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_f32_to_f16_afn:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_3)
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %a) {
 entry:
   %a.val = load float, ptr addrspace(1) %a
@@ -1089,6 +1159,130 @@ define amdgpu_kernel void @fptrunc_f64_to_f16(
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_f64_to_f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b64 v[0:1], off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s2, v1
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s3, s2, 0x1ff
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s5, s2, 8
+; GFX1250-SDAG-FAKE16-NEXT:    v_or_b32_e32 v0, s3, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_bfe_u32 s3, s2, 0xb0014
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s5, s5, 0xffe
+; GFX1250-SDAG-FAKE16-NEXT:    s_sub_co_i32 s4, 0x3f1, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cmp_ne_u32_e32 vcc_lo, 0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    v_med3_i32 v1, s4, 0, 13
+; GFX1250-SDAG-FAKE16-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc_lo
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s8, v1
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s4, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s4, s5, s4
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s5, s4, 0x1000
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s9, s5, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshl_b32 s8, s9, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lg_u32 s8, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s5, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_addk_co_i32 s3, 0xfc10
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s5, s9, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshl_b32 s8, s3, 12
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s8, s4, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lt_i32 s3, 1
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s5, s5, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s8, s5, 7
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_gt_i32 s8, 5
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s9, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_eq_u32 s8, 3
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s8, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s5, s5, 2
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s8, s8, s9
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_add_co_i32 s5, s5, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lt_i32 s3, 31
+; GFX1250-SDAG-FAKE16-NEXT:    s_movk_i32 s8, 0x7e00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s5, s5, 0x7c00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lg_u32 s4, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s4, s8, 0x7c00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_eq_u32 s3, 0x40f
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s3, s4, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s2, s2, 16
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s2, s2, 0x8000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s2, s2, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_f64_to_f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b64 s[2:3], s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s6, s3, 0x1ff
+; GFX1250-GISEL-FAKE16-NEXT:    s_bfe_u32 s4, s3, 0xb0014
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s5, s3, 8
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s2, s6, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_addk_co_i32 s4, 0xfc10
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s5, s5, 0xffe
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s2, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s2, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s2, s5, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s2, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s5, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_sub_co_i32 s6, 1, s4
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s8, s2, 0x1000
+; GFX1250-GISEL-FAKE16-NEXT:    s_max_i32 s6, s6, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s7, s4, 12
+; GFX1250-GISEL-FAKE16-NEXT:    s_min_i32 s6, s6, 13
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s5, s5, 9
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s9, s8, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s2, s2, s7
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s6, s9, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s5, s5, 0x7c00
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s6, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s6, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s6, s9, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lt_i32 s4, 1
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s2, s6, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s6, s2, 7
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s2, s2, 2
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_eq_u32 s6, 3
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s7, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_gt_i32 s6, 5
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s6, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s6, s7, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_add_co_i32 s2, s2, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_gt_i32 s4, 30
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s2, 0x7c00, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_eq_u32 s4, 0x40f
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s2, s5, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s3, s3, 16
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s3, s3, 0x8000
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s2, s3, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -1294,6 +1488,40 @@ define amdgpu_kernel void @fptrunc_f64_to_f16_afn(
 ; GFX11-GISEL-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_f64_to_f16_afn:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b64 v[0:1], off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f32_f64_e32 v0, v[0:1]
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_f64_to_f16_afn:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b64 s[2:3], s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_f32_f64_e32 v0, s[2:3]
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -1511,6 +1739,38 @@ define amdgpu_kernel void @fptrunc_v2f32_to_v2f16(
 ; GFX11-GISEL-FAKE16-NEXT:    v_pack_b32_f16 v0, v0, v1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_v2f32_to_v2f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b64 v[0:1], off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_pk_f16_f32 v0, v0, v1
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b32 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_v2f32_to_v2f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b64 s[2:3], s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b64_e32 v[0:1], s[2:3]
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_pk_f16_f32 v0, v0, v1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -2756,6 +3016,225 @@ define amdgpu_kernel void @fptrunc_v2f64_to_v2f16(
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_v2f64_to_v2f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b128 v[0:3], off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s2, v3
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s3, s2, 0x1ff
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s5, s2, 8
+; GFX1250-SDAG-FAKE16-NEXT:    v_or_b32_e32 v2, s3, v2
+; GFX1250-SDAG-FAKE16-NEXT:    s_bfe_u32 s3, s2, 0xb0014
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s5, s5, 0xffe
+; GFX1250-SDAG-FAKE16-NEXT:    s_sub_co_i32 s4, 0x3f1, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cmp_ne_u32_e32 vcc_lo, 0, v2
+; GFX1250-SDAG-FAKE16-NEXT:    v_med3_i32 v3, s4, 0, 13
+; GFX1250-SDAG-FAKE16-NEXT:    v_cndmask_b32_e64 v2, 0, 1, vcc_lo
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s8, v3
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s4, v2
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s4, s5, s4
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s5, s4, 0x1000
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s9, s5, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshl_b32 s8, s9, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lg_u32 s8, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s5, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_addk_co_i32 s3, 0xfc10
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s5, s9, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshl_b32 s8, s3, 12
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s8, s4, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lt_i32 s3, 1
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s5, s5, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s8, s5, 7
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_gt_i32 s8, 5
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s9, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_eq_u32 s8, 3
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s8, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s5, s5, 2
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s8, s8, s9
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_add_co_i32 s5, s5, s8
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lt_i32 s3, 31
+; GFX1250-SDAG-FAKE16-NEXT:    s_movk_i32 s8, 0x7e00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s5, s5, 0x7c00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lg_u32 s4, 0
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s4, v1
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s9, s8, 0x7c00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_eq_u32 s3, 0x40f
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s3, s9, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s5, s4, 0x1ff
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s10, s4, 8
+; GFX1250-SDAG-FAKE16-NEXT:    v_or_b32_e32 v0, s5, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_bfe_u32 s5, s4, 0xb0014
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s10, s10, 0xffe
+; GFX1250-SDAG-FAKE16-NEXT:    s_sub_co_i32 s9, 0x3f1, s5
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s2, s2, 16
+; GFX1250-SDAG-FAKE16-NEXT:    v_cmp_ne_u32_e32 vcc_lo, 0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    v_med3_i32 v1, s9, 0, 13
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s2, s2, 0x8000
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s2, s2, s3
+; GFX1250-SDAG-FAKE16-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc_lo
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s11, v1
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_readfirstlane_b32 s9, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s9, s10, s9
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s10, s9, 0x1000
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s12, s10, s11
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshl_b32 s11, s12, s11
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lg_u32 s11, s10
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s3, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_addk_co_i32 s5, 0xfc10
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s3, s12, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshl_b32 s10, s5, 12
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s10, s9, s10
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lt_i32 s5, 1
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s3, s3, s10
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s10, s3, 7
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_gt_i32 s10, 5
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s11, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_eq_u32 s10, 3
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s10, 1, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s3, s3, 2
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s10, s10, s11
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_add_co_i32 s3, s3, s10
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lt_i32 s5, 31
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s3, s3, 0x7c00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_lg_u32 s9, 0
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s8, s8, 0x7c00
+; GFX1250-SDAG-FAKE16-NEXT:    s_cmp_eq_u32 s5, 0x40f
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_cselect_b32 s3, s8, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_lshr_b32 s4, s4, 16
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    s_and_b32 s4, s4, 0x8000
+; GFX1250-SDAG-FAKE16-NEXT:    s_or_b32 s3, s4, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    s_pack_ll_b32_b16 s2, s3, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b32 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_v2f64_to_v2f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[4:7], s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s8, s5, 0x1ff
+; GFX1250-GISEL-FAKE16-NEXT:    s_bfe_u32 s2, s5, 0xb0014
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s3, s5, 8
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s4, s8, s4
+; GFX1250-GISEL-FAKE16-NEXT:    s_addk_co_i32 s2, 0xfc10
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s3, s3, 0xffe
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s4, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s4, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s3, s3, s4
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s3, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s4, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_sub_co_i32 s8, 1, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s10, s3, 0x1000
+; GFX1250-GISEL-FAKE16-NEXT:    s_max_i32 s8, s8, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s9, s2, 12
+; GFX1250-GISEL-FAKE16-NEXT:    s_min_i32 s8, s8, 13
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s4, s4, 9
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s11, s10, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s3, s3, s9
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s8, s11, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s4, s4, 0x7c00
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s8, s10
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s8, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s8, s11, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lt_i32 s2, 1
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s3, s8, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s8, s3, 7
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s3, s3, 2
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_eq_u32 s8, 3
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s9, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_gt_i32 s8, 5
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s8, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s8, s9, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_add_co_i32 s3, s3, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_gt_i32 s2, 30
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s3, 0x7c00, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_eq_u32 s2, 0x40f
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s2, s4, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s3, s5, 16
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s8, s7, 0x1ff
+; GFX1250-GISEL-FAKE16-NEXT:    s_bfe_u32 s4, s7, 0xb0014
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s5, s7, 8
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s3, s3, 0x8000
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s6, s8, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_addk_co_i32 s4, 0xfc10
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s5, s5, 0xffe
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s2, s3, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s6, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s3, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s3, s5, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s3, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s5, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_sub_co_i32 s6, 1, s4
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s9, s3, 0x1000
+; GFX1250-GISEL-FAKE16-NEXT:    s_max_i32 s6, s6, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s8, s4, 12
+; GFX1250-GISEL-FAKE16-NEXT:    s_min_i32 s6, s6, 13
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s5, s5, 9
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s10, s9, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s3, s3, s8
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshl_b32 s6, s10, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s5, s5, 0x7c00
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lg_u32 s6, s9
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s6, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s6, s10, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_lt_i32 s4, 1
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s3, s6, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s6, s3, 7
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s3, s3, 2
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_eq_u32 s6, 3
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s8, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_gt_i32 s6, 5
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s6, 1, 0
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s6, s8, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_add_co_i32 s3, s3, s6
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_gt_i32 s4, 30
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s3, 0x7c00, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_cmp_eq_u32 s4, 0x40f
+; GFX1250-GISEL-FAKE16-NEXT:    s_cselect_b32 s3, s5, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_lshr_b32 s4, s7, 16
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s4, s4, 0x8000
+; GFX1250-GISEL-FAKE16-NEXT:    s_or_b32 s3, s4, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_pack_ll_b32_b16 s2, s2, s3
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -3001,6 +3480,46 @@ define amdgpu_kernel void @fptrunc_v2f64_to_v2f16_afn(
 ; GFX11-GISEL-FAKE16-NEXT:    v_pack_b32_f16 v0, v0, v1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_v2f64_to_v2f16_afn:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b128 v[0:3], off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f32_f64_e32 v2, v[2:3]
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f32_f64_e32 v0, v[0:1]
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_pk_f16_f32 v0, v0, v2
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b32 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_v2f64_to_v2f16_afn:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[4:7], s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_f32_f64_e32 v0, s[4:5]
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_f32_f64_e32 v1, s[6:7]
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-GISEL-FAKE16-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-GISEL-FAKE16-NEXT:    v_pack_b32_f16 v0, v0, v1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -3190,6 +3709,42 @@ define amdgpu_kernel void @fneg_fptrunc_f32_to_f16(
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fneg_fptrunc_f32_to_f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_xor_b32_e32 v0, 0x80000000, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fneg_fptrunc_f32_to_f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_xor_b32 s2, s2, 0x80000000
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_3)
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -3380,6 +3935,42 @@ define amdgpu_kernel void @fabs_fptrunc_f32_to_f16(
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fabs_fptrunc_f32_to_f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_and_b32_e32 v0, 0x7fffffff, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fabs_fptrunc_f32_to_f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_bitset0_b32 s2, 31
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_3)
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) {
 entry:
@@ -3570,6 +4161,42 @@ define amdgpu_kernel void @fneg_fabs_fptrunc_f32_to_f16(
 ; GFX11-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fneg_fabs_fptrunc_f32_to_f16:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_or_b32_e32 v0, 0x80000000, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b16 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fneg_fabs_fptrunc_f32_to_f16:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_bitset1_b32 s2, 31
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_3)
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b16 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) #0 {
 entry:
@@ -3769,6 +4396,42 @@ define amdgpu_kernel void @fptrunc_f32_to_f16_zext_i32(
 ; GFX11-GISEL-FAKE16-NEXT:    v_and_b32_e32 v0, 0xffff, v0
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_f32_to_f16_zext_i32:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b32 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_f32_to_f16_zext_i32:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_3) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s2, 0xffff, s2
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) #0 {
 entry:
@@ -3968,6 +4631,45 @@ define amdgpu_kernel void @fptrunc_fabs_f32_to_f16_zext_i32(
 ; GFX11-GISEL-FAKE16-NEXT:    v_and_b32_e32 v0, 0xffff, v0
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_fabs_f32_to_f16_zext_i32:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_and_b32_e32 v0, 0x7fffffff, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b32 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_fabs_f32_to_f16_zext_i32:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_bitset0_b32 s2, 31
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_3)
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_and_b32 s2, 0xffff, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) #0 {
 entry:
@@ -4176,6 +4878,42 @@ define amdgpu_kernel void @fptrunc_f32_to_f16_sext_i32(
 ; GFX11-GISEL-FAKE16-NEXT:    v_bfe_i32 v0, v0, 0, 16
 ; GFX11-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], 0
 ; GFX11-GISEL-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-SDAG-FAKE16-LABEL: fptrunc_f32_to_f16_sext_i32:
+; GFX1250-SDAG-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-SDAG-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s6, -1
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s7, 0x31016000
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s10, s6
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s11, s7
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s8, s2
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s9, s3
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s4, s0
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_load_b32 v0, off, s[8:11], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_mov_b32 s5, s1
+; GFX1250-SDAG-FAKE16-NEXT:    s_wait_loadcnt 0x0
+; GFX1250-SDAG-FAKE16-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GFX1250-SDAG-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-SDAG-FAKE16-NEXT:    v_bfe_i32 v0, v0, 0, 16
+; GFX1250-SDAG-FAKE16-NEXT:    buffer_store_b32 v0, off, s[4:7], null
+; GFX1250-SDAG-FAKE16-NEXT:    s_endpgm
+;
+; GFX1250-GISEL-FAKE16-LABEL: fptrunc_f32_to_f16_sext_i32:
+; GFX1250-GISEL-FAKE16:       ; %bb.0: ; %entry
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_load_b32 s2, s[2:3], 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_xcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s3, 0x31016000
+; GFX1250-GISEL-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; GFX1250-GISEL-FAKE16-NEXT:    s_cvt_f16_f32 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_delay_alu instid0(SALU_CYCLE_3) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; GFX1250-GISEL-FAKE16-NEXT:    s_sext_i32_i16 s2, s2
+; GFX1250-GISEL-FAKE16-NEXT:    v_mov_b32_e32 v0, s2
+; GFX1250-GISEL-FAKE16-NEXT:    s_mov_b32 s2, -1
+; GFX1250-GISEL-FAKE16-NEXT:    buffer_store_b32 v0, off, s[0:3], null
+; GFX1250-GISEL-FAKE16-NEXT:    s_endpgm
     ptr addrspace(1) %r,
     ptr addrspace(1) %a) #0 {
 entry:
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.pk.f16.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.pk.f16.ll
new file mode 100644
index 0000000..2179800
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.pk.f16.ll
@@ -0,0 +1,64 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1250 < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1250 < %s | FileCheck -check-prefix=GCN %s
+
+declare <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float, float, i32) #0
+
+define amdgpu_ps float @cvt_sr_pk_f16_f32_vvv(float %src0, float %src1, i32 %src2) #1 {
+; GCN-LABEL: cvt_sr_pk_f16_f32_vvv:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    v_cvt_sr_pk_f16_f32 v0, v0, v1, v2
+; GCN-NEXT:    ; return to shader part epilog
+  %cvt = call <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float %src0, float %src1, i32 %src2) #0
+  %ret = bitcast <2 x half> %cvt to float
+  ret float %ret
+}
+
+define amdgpu_ps float @cvt_sr_pk_f16_f32_sss(float inreg %src0, float inreg %src1, i32 inreg %src2) #1 {
+; GCN-LABEL: cvt_sr_pk_f16_f32_sss:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    v_mov_b32_e32 v0, s2
+; GCN-NEXT:    s_delay_alu instid0(VALU_DEP_1)
+; GCN-NEXT:    v_cvt_sr_pk_f16_f32 v0, s0, s1, v0
+; GCN-NEXT:    ; return to shader part epilog
+  %cvt = call <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float %src0, float %src1, i32 %src2) #0
+  %ret = bitcast <2 x half> %cvt to float
+  ret float %ret
+}
+
+define amdgpu_ps float @cvt_sr_pk_f16_f32_vvi(float %src0, float %src1) #1 {
+; GCN-LABEL: cvt_sr_pk_f16_f32_vvi:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    v_cvt_sr_pk_f16_f32 v0, v0, v1, 0x10002
+; GCN-NEXT:    ; return to shader part epilog
+  %cvt = call <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float %src0, float %src1, i32 65538) #0
+  %ret = bitcast <2 x half> %cvt to float
+  ret float %ret
+}
+
+define amdgpu_ps float @cvt_sr_pk_f16_f32_vvi_mods(float %src0, float %src1) #1 {
+; GCN-LABEL: cvt_sr_pk_f16_f32_vvi_mods:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    v_cvt_sr_pk_f16_f32 v0, -v0, |v1|, 1
+; GCN-NEXT:    ; return to shader part epilog
+  %s0 = fneg float %src0
+  %s1 = call float @llvm.fabs.f32(float %src1) #0
+  %cvt = call <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float %s0, float %s1, i32 1) #0
+  %ret = bitcast <2 x half> %cvt to float
+  ret float %ret
+}
+
+define amdgpu_ps float @cvt_sr_pk_f16_f32_ssi(float inreg %src0, float inreg %src1) #1 {
+; GCN-LABEL: cvt_sr_pk_f16_f32_ssi:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    v_cvt_sr_pk_f16_f32 v0, s0, s1, 1
+; GCN-NEXT:    ; return to shader part epilog
+  %cvt = call <2 x half> @llvm.amdgcn.cvt.sr.pk.f16.f32(float %src0, float %src1, i32 1) #0
+  %ret = bitcast <2 x half> %cvt to float
+  ret float %ret
+}
+
+declare float @llvm.fabs.f32(float) #0
+
+attributes #0 = { nounwind readnone }
+attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/SPIRV/transcoding/spirv-target-types.ll b/llvm/test/CodeGen/SPIRV/transcoding/spirv-target-types.ll
new file mode 100644
index 0000000..8b5efe7
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/transcoding/spirv-target-types.ll
@@ -0,0 +1,104 @@
+; RUN: llc -O0 -mtriple=spirv32-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv32-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK-DAG: OpCapability Float16
+; CHECK-DAG: OpCapability ImageBasic
+; CHECK-DAG: OpCapability ImageReadWrite
+; CHECK-DAG: OpCapability Pipes
+; CHECK-DAG: OpCapability DeviceEnqueue
+
+; CHECK-DAG: %[[#VOID:]] = OpTypeVoid
+; CHECK-DAG: %[[#INT:]] = OpTypeInt 32 0
+; CHECK-DAG: %[[#HALF:]] = OpTypeFloat 16
+; CHECK-DAG: %[[#FLOAT:]] = OpTypeFloat 32
+; CHECK-DAG: %[[#PIPE_RD:]] = OpTypePipe ReadOnly
+; CHECK-DAG: %[[#PIPE_WR:]] = OpTypePipe WriteOnly
+; CHECK-DAG: %[[#IMG1D_RD:]] = OpTypeImage %[[#VOID]] 1D 0 0 0 0 Unknown ReadOnly
+; CHECK-DAG: %[[#IMG2D_RD:]] = OpTypeImage %[[#INT]] 2D 0 0 0 0
+; CHECK-DAG: %[[#IMG3D_RD:]] = OpTypeImage %[[#INT]] 3D 0 0 0 0
+; CHECK-DAG: %[[#IMG2DA_RD:]] = OpTypeImage %[[#HALF]] 2D 0 1 0 0
+; CHECK-DAG: %[[#IMG2DD_RD:]] = OpTypeImage %[[#FLOAT]] Buffer 0 0 0
+; CHECK-DAG: %[[#IMG1D_WR:]] = OpTypeImage %[[#VOID]] 1D 0 0 0 0 Unknown WriteOnly
+; CHECK-DAG: %[[#IMG2D_RW:]] = OpTypeImage %[[#VOID]] 2D 0 0 0 0 Unknown ReadWrite
+; CHECK-DAG: %[[#IMG1DB_RD:]] = OpTypeImage %[[#FLOAT]] 2D 1 0 0 0
+
+; CHECK-DAG: %[[#DEVEVENT:]] = OpTypeDeviceEvent
+; CHECK-DAG: %[[#EVENT:]] = OpTypeEvent
+; CHECK-DAG: %[[#QUEUE:]] = OpTypeQueue
+; CHECK-DAG: %[[#RESID:]] = OpTypeReserveId
+; CHECK-DAG: %[[#SAMP:]] = OpTypeSampler
+; CHECK-DAG: %[[#SAMPIMG:]] = OpTypeSampledImage %[[#IMG1DB_RD]]
+
+; CHECK-DAG: %[[#]] = OpFunction %[[#VOID]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#PIPE_RD]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#PIPE_WR]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG1D_RD]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG2D_RD]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG3D_RD]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG2DA_RD]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG2DD_RD]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG1D_WR]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#IMG2D_RW]]
+
+define spir_kernel void @foo(
+  target("spirv.Pipe", 0) %a,
+  target("spirv.Pipe", 1) %b,
+  target("spirv.Image", void, 0, 0, 0, 0, 0, 0, 0) %c1,
+  target("spirv.Image", i32, 1, 0, 0, 0, 0, 0, 0) %d1,
+  target("spirv.Image", i32, 2, 0, 0, 0, 0, 0, 0) %e1,
+  target("spirv.Image", half, 1, 0, 1, 0, 0, 0, 0) %f1,
+  target("spirv.Image", float, 5, 0, 0, 0, 0, 0, 0) %g1,
+  target("spirv.Image", void, 0, 0, 0, 0, 0, 0, 1) %c2,
+  target("spirv.Image", void, 1, 0, 0, 0, 0, 0, 2) %d3) #0 !kernel_arg_addr_space !1 !kernel_arg_access_qual !2 !kernel_arg_type !3 !kernel_arg_base_type !4 !kernel_arg_type_qual !5 {
+entry:
+  ret void
+}
+
+; CHECK-DAG: %[[#]] = OpFunction
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#DEVEVENT]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#EVENT]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#QUEUE]]
+; CHECK-DAG: %[[#]] = OpFunctionParameter %[[#RESID]]
+
+; CHECK-DAG: %[[#IMARG:]] = OpFunctionParameter %[[#IMG1DB_RD]]
+; CHECK-DAG: %[[#SAMARG:]] = OpFunctionParameter %[[#SAMP]]
+; CHECK-DAG: %[[#SAMPIMVAR:]] = OpSampledImage %[[#SAMPIMG]] %[[#IMARG]] %[[#SAMARG]]
+; CHECK-DAG: %[[#]] = OpImageSampleExplicitLod %[[#]] %[[#SAMPIMVAR]]
+
+define spir_func void @bar(
+  target("spirv.DeviceEvent") %a,
+  target("spirv.Event") %b,
+  target("spirv.Queue") %c,
+  target("spirv.ReserveId") %d) {
+  ret void
+}
+
+define spir_func void @test_sampler(target("spirv.Image", float, 1, 1, 0, 0, 0, 0, 0) %srcimg.coerce,
+                                    target("spirv.Sampler") %s.coerce) {
+  %1 = tail call spir_func target("spirv.SampledImage", float, 1, 1, 0, 0, 0, 0, 0) @_Z20__spirv_SampledImagePU3AS1K34__spirv_Image__float_1_1_0_0_0_0_0PU3AS1K15__spirv_Sampler(target("spirv.Image", float, 1, 1, 0, 0, 0, 0, 0) %srcimg.coerce, target("spirv.Sampler") %s.coerce) #1
+  %2 = tail call spir_func <4 x float> @_Z38__spirv_ImageSampleExplicitLod_Rfloat4PU3AS120__spirv_SampledImageDv4_iif(target("spirv.SampledImage", float, 1, 1, 0, 0, 0, 0, 0) %1, <4 x i32> zeroinitializer, i32 2, float 1.000000e+00) #1
+  ret void
+}
+
+declare spir_func target("spirv.SampledImage", float, 1, 1, 0, 0, 0, 0, 0) @_Z20__spirv_SampledImagePU3AS1K34__spirv_Image__float_1_1_0_0_0_0_0PU3AS1K15__spirv_Sampler(target("spirv.Image", float, 1, 1, 0, 0, 0, 0, 0), target("spirv.Sampler"))
+
+declare spir_func <4 x float> @_Z38__spirv_ImageSampleExplicitLod_Rfloat4PU3AS120__spirv_SampledImageDv4_iif(target("spirv.SampledImage", float, 1, 1, 0, 0, 0, 0, 0), <4 x i32>, i32, float)
+
+attributes #0 = { nounwind readnone "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-realign-stack" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!opencl.enable.FP_CONTRACT = !{}
+!opencl.spir.version = !{!6}
+!opencl.ocl.version = !{!7}
+!opencl.used.extensions = !{!8}
+!opencl.used.optional.core.features = !{!9}
+!opencl.compiler.options = !{!8}
+
+!1 = !{i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1}
+!2 = !{!"read_only", !"write_only", !"read_only", !"read_only", !"read_only", !"read_only", !"read_only", !"write_only", !"read_write"}
+!3 = !{!"int", !"int", !"image1d_t", !"image2d_t", !"image3d_t", !"image2d_array_t", !"image1d_buffer_t", !"image1d_t", !"image2d_t"}
+!4 = !{!"int", !"int", !"image1d_t", !"image2d_t", !"image3d_t", !"image2d_array_t", !"image1d_buffer_t", !"image1d_t", !"image2d_t"}
+!5 = !{!"pipe", !"pipe", !"", !"", !"", !"", !"", !"", !""}
+!6 = !{i32 1, i32 2}
+!7 = !{i32 2, i32 0}
+!8 = !{!"cl_khr_fp16"}
+!9 = !{!"cl_images"}
diff --git a/llvm/test/CodeGen/SPIRV/transcoding/unused-sret-opaque-ptr.ll b/llvm/test/CodeGen/SPIRV/transcoding/unused-sret-opaque-ptr.ll
new file mode 100644
index 0000000..63b2604
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/transcoding/unused-sret-opaque-ptr.ll
@@ -0,0 +1,19 @@
+; RUN: llc -O0 -mtriple=spirv32-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv32-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK-DAG: OpName %[[#Fun:]] "_Z3booi"
+; CHECK-DAG: OpDecorate %[[#Param:]] FuncParamAttr Sret
+; CHECK-DAG: %[[#PtrTy:]] = OpTypePointer Function %[[#StructTy:]]
+; CHECK-DAG: %[[#StructTy]] = OpTypeStruct 
+; CHECK: %[[#Fun]] = OpFunction %[[#]] 
+; CHECK: %[[#Param]] = OpFunctionParameter %[[#PtrTy]] 
+
+%struct.Example = type { }
+
+define spir_func i32 @foo() {
+  %1 = alloca %struct.Example, align 8
+  call void @_Z3booi(ptr sret(%struct.Example) align 8 %1, i32 noundef 42)
+  ret i32 0
+}
+
+declare void @_Z3booi(ptr sret(%struct.Example) align 8, i32 noundef)
diff --git a/llvm/test/CodeGen/SystemZ/vec-mul-07.ll b/llvm/test/CodeGen/SystemZ/vec-mul-07.ll
index 73c7a8d..5835616 100644
--- a/llvm/test/CodeGen/SystemZ/vec-mul-07.ll
+++ b/llvm/test/CodeGen/SystemZ/vec-mul-07.ll
@@ -3,6 +3,23 @@
 ;
 ; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z13 | FileCheck %s
 
+; Test a v16i8 -> v8i16 unsigned widening multiplication
+; which is not folded into an even/odd widening operation.
+define <8 x i16> @f1_not(<16 x i8> %val1, <16 x i8> %val2) {
+; CHECK-LABEL: f1_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vuplhb %v0, %v24
+; CHECK-NEXT:    vuplhb %v1, %v26
+; CHECK-NEXT:    vmlhw %v24, %v0, %v1
+; CHECK-NEXT:    br %r14
+  %shuf1 = shufflevector <16 x i8> %val1, <16 x i8> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %zext1 = zext <8 x i8> %shuf1 to <8 x i16>
+  %shuf2 = shufflevector <16 x i8> %val2, <16 x i8> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %zext2 = zext <8 x i8> %shuf2 to <8 x i16>
+  %ret = mul <8 x i16> %zext1, %zext2
+  ret <8 x i16> %ret
+}
+
 ; Test a v16i8 (even) -> v8i16 unsigned widening multiplication.
 define <8 x i16> @f1(<16 x i8> %val1, <16 x i8> %val2) {
 ; CHECK-LABEL: f1:
@@ -31,6 +48,23 @@ define <8 x i16> @f2(<16 x i8> %val1, <16 x i8> %val2) {
   ret <8 x i16> %ret
 }
 
+; Test a v16i8 -> v8i16 signed widening multiplication
+; which is not folded into an even/odd widening operation.
+define <8 x i16> @f3_not(<16 x i8> %val1, <16 x i8> %val2) {
+; CHECK-LABEL: f3_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vuphb %v0, %v26
+; CHECK-NEXT:    vuphb %v1, %v24
+; CHECK-NEXT:    vmlhw %v24, %v1, %v0
+; CHECK-NEXT:    br %r14
+  %shuf1 = shufflevector <16 x i8> %val1, <16 x i8> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %sext1 = sext <8 x i8> %shuf1 to <8 x i16>
+  %shuf2 = shufflevector <16 x i8> %val2, <16 x i8> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %sext2 = sext <8 x i8> %shuf2 to <8 x i16>
+  %ret = mul <8 x i16> %sext1, %sext2
+  ret <8 x i16> %ret
+}
+
 ; Test a v16i8 (even) -> v8i16 signed widening multiplication.
 define <8 x i16> @f3(<16 x i8> %val1, <16 x i8> %val2) {
 ; CHECK-LABEL: f3:
@@ -59,6 +93,23 @@ define <8 x i16> @f4(<16 x i8> %val1, <16 x i8> %val2) {
   ret <8 x i16> %ret
 }
 
+; Test a v8i16 -> v4i32 unsigned widening multiplication
+; which is not folded into an even/odd widening operation.
+define <4 x i32> @f5_not(<8 x i16> %val1, <8 x i16> %val2) {
+; CHECK-LABEL: f5_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vuplhh %v0, %v24
+; CHECK-NEXT:    vuplhh %v1, %v26
+; CHECK-NEXT:    vmlf %v24, %v0, %v1
+; CHECK-NEXT:    br %r14
+  %shuf1 = shufflevector <8 x i16> %val1, <8 x i16> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %zext1 = zext <4 x i16> %shuf1 to <4 x i32>
+  %shuf2 = shufflevector <8 x i16> %val2, <8 x i16> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %zext2 = zext <4 x i16> %shuf2 to <4 x i32>
+  %ret = mul <4 x i32> %zext1, %zext2
+  ret <4 x i32> %ret
+}
+
 ; Test a v8i16 (even) -> v4i32 unsigned widening multiplication.
 define <4 x i32> @f5(<8 x i16> %val1, <8 x i16> %val2) {
 ; CHECK-LABEL: f5:
@@ -87,6 +138,23 @@ define <4 x i32> @f6(<8 x i16> %val1, <8 x i16> %val2) {
   ret <4 x i32> %ret
 }
 
+; Test a v8i16 -> v4i32 signed widening multiplication
+; which is not folded into an even/odd widening operation.
+define <4 x i32> @f7_not(<8 x i16> %val1, <8 x i16> %val2) {
+; CHECK-LABEL: f7_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vuphh %v0, %v26
+; CHECK-NEXT:    vuphh %v1, %v24
+; CHECK-NEXT:    vmlf %v24, %v1, %v0
+; CHECK-NEXT:    br %r14
+  %shuf1 = shufflevector <8 x i16> %val1, <8 x i16> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %sext1 = sext <4 x i16> %shuf1 to <4 x i32>
+  %shuf2 = shufflevector <8 x i16> %val2, <8 x i16> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %sext2 = sext <4 x i16> %shuf2 to <4 x i32>
+  %ret = mul <4 x i32> %sext1, %sext2
+  ret <4 x i32> %ret
+}
+
 ; Test a v8i16 (even) -> v4i32 signed widening multiplication.
 define <4 x i32> @f7(<8 x i16> %val1, <8 x i16> %val2) {
 ; CHECK-LABEL: f7:
@@ -115,6 +183,29 @@ define <4 x i32> @f8(<8 x i16> %val1, <8 x i16> %val2) {
   ret <4 x i32> %ret
 }
 
+; Test a v4i32 -> v2i64 unsigned widening multiplication
+; which is not folded into an even/odd widening operation.
+define <2 x i64> @f9_not(<4 x i32> %val1, <4 x i32> %val2) {
+; CHECK-LABEL: f9_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vuplhf %v0, %v24
+; CHECK-NEXT:    vuplhf %v1, %v26
+; CHECK-NEXT:    vlgvg %r0, %v1, 1
+; CHECK-NEXT:    vlgvg %r1, %v0, 1
+; CHECK-NEXT:    msgr %r1, %r0
+; CHECK-NEXT:    vlgvg %r0, %v1, 0
+; CHECK-NEXT:    vlgvg %r2, %v0, 0
+; CHECK-NEXT:    msgr %r2, %r0
+; CHECK-NEXT:    vlvgp %v24, %r2, %r1
+; CHECK-NEXT:    br %r14
+  %shuf1 = shufflevector <4 x i32> %val1, <4 x i32> poison, <2 x i32> <i32 0, i32 1>
+  %zext1 = zext <2 x i32> %shuf1 to <2 x i64>
+  %shuf2 = shufflevector <4 x i32> %val2, <4 x i32> poison, <2 x i32> <i32 0, i32 1>
+  %zext2 = zext <2 x i32> %shuf2 to <2 x i64>
+  %ret = mul <2 x i64> %zext1, %zext2
+  ret <2 x i64> %ret
+}
+
 ; Test a v4i32 (even) -> v2i64 unsigned widening multiplication.
 define <2 x i64> @f9(<4 x i32> %val1, <4 x i32> %val2) {
 ; CHECK-LABEL: f9:
@@ -143,6 +234,29 @@ define <2 x i64> @f10(<4 x i32> %val1, <4 x i32> %val2) {
   ret <2 x i64> %ret
 }
 
+; Test a v4i32 -> v2i64 signed widening multiplication
+; which is not folded into an even/odd widening operation.
+define <2 x i64> @f11_not(<4 x i32> %val1, <4 x i32> %val2) {
+; CHECK-LABEL: f11_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vuphf %v0, %v24
+; CHECK-NEXT:    vuphf %v1, %v26
+; CHECK-NEXT:    vlgvg %r0, %v1, 1
+; CHECK-NEXT:    vlgvg %r1, %v0, 1
+; CHECK-NEXT:    msgr %r1, %r0
+; CHECK-NEXT:    vlgvg %r0, %v1, 0
+; CHECK-NEXT:    vlgvg %r2, %v0, 0
+; CHECK-NEXT:    msgr %r2, %r0
+; CHECK-NEXT:    vlvgp %v24, %r2, %r1
+; CHECK-NEXT:    br %r14
+  %shuf1 = shufflevector <4 x i32> %val1, <4 x i32> poison, <2 x i32> <i32 0, i32 1>
+  %sext1 = sext <2 x i32> %shuf1 to <2 x i64>
+  %shuf2 = shufflevector <4 x i32> %val2, <4 x i32> poison, <2 x i32> <i32 0, i32 1>
+  %sext2 = sext <2 x i32> %shuf2 to <2 x i64>
+  %ret = mul <2 x i64> %sext1, %sext2
+  ret <2 x i64> %ret
+}
+
 ; Test a v4i32 (even) -> v2i64 signed widening multiplication.
 define <2 x i64> @f11(<4 x i32> %val1, <4 x i32> %val2) {
 ; CHECK-LABEL: f11:
diff --git a/llvm/test/CodeGen/X86/tail-dup-computed-goto.mir b/llvm/test/CodeGen/X86/early-tail-dup-computed-goto.mir
index 17de405..0f28964 100644
--- a/llvm/test/CodeGen/X86/tail-dup-computed-goto.mir
+++ b/llvm/test/CodeGen/X86/early-tail-dup-computed-goto.mir
@@ -1,6 +1,8 @@
 # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
 # RUN: llc -mtriple=x86_64-unknown-linux-gnu -run-pass=early-tailduplication -tail-dup-pred-size=1 -tail-dup-succ-size=1 %s -o - | FileCheck %s
-# Check that only the computed goto is not be restrict by tail-dup-pred-size and tail-dup-succ-size.
+#
+# Check that only the computed goto and others are restricted by tail-dup-pred-size and tail-dup-succ-size.
+#
 --- |
   @computed_goto.dispatch = constant [5 x ptr] [ptr null, ptr blockaddress(@computed_goto, %bb1), ptr blockaddress(@computed_goto, %bb2), ptr blockaddress(@computed_goto, %bb3), ptr blockaddress(@computed_goto, %bb4)]
   declare i64 @f0()
@@ -30,54 +32,54 @@ tracksRegLiveness: true
 body:             |
   ; CHECK-LABEL: name: computed_goto
   ; CHECK: bb.0:
-  ; CHECK-NEXT:   successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+  ; CHECK-NEXT:   successors: %bb.5(0x80000000)
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   ADJCALLSTACKDOWN64 0, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
   ; CHECK-NEXT:   CALL64pcrel32 target-flags(x86-plt) @f0, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
   ; CHECK-NEXT:   ADJCALLSTACKUP64 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
-  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:gr64_nosp = COPY $rax
-  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:gr64_nosp = COPY [[COPY]]
-  ; CHECK-NEXT:   JMP64m $noreg, 8, [[COPY]], @computed_goto.dispatch, $noreg
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:gr64 = COPY $rax
+  ; CHECK-NEXT:   JMP_1 %bb.5
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.1.bb1 (ir-block-address-taken %ir-block.bb1):
-  ; CHECK-NEXT:   successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+  ; CHECK-NEXT:   successors: %bb.5(0x80000000)
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   ADJCALLSTACKDOWN64 0, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
   ; CHECK-NEXT:   CALL64pcrel32 target-flags(x86-plt) @f1, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
   ; CHECK-NEXT:   ADJCALLSTACKUP64 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
-  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:gr64_nosp = COPY $rax
-  ; CHECK-NEXT:   [[COPY3:%[0-9]+]]:gr64_nosp = COPY [[COPY2]]
-  ; CHECK-NEXT:   JMP64m $noreg, 8, [[COPY2]], @computed_goto.dispatch, $noreg
+  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:gr64 = COPY $rax
+  ; CHECK-NEXT:   JMP_1 %bb.5
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.2.bb2 (ir-block-address-taken %ir-block.bb2):
-  ; CHECK-NEXT:   successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+  ; CHECK-NEXT:   successors: %bb.5(0x80000000)
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   ADJCALLSTACKDOWN64 0, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
   ; CHECK-NEXT:   CALL64pcrel32 target-flags(x86-plt) @f2, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
   ; CHECK-NEXT:   ADJCALLSTACKUP64 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
-  ; CHECK-NEXT:   [[COPY4:%[0-9]+]]:gr64_nosp = COPY $rax
-  ; CHECK-NEXT:   [[COPY5:%[0-9]+]]:gr64_nosp = COPY [[COPY4]]
-  ; CHECK-NEXT:   JMP64m $noreg, 8, [[COPY4]], @computed_goto.dispatch, $noreg
+  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:gr64 = COPY $rax
+  ; CHECK-NEXT:   JMP_1 %bb.5
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.3.bb3 (ir-block-address-taken %ir-block.bb3):
-  ; CHECK-NEXT:   successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+  ; CHECK-NEXT:   successors: %bb.5(0x80000000)
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   ADJCALLSTACKDOWN64 0, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
   ; CHECK-NEXT:   CALL64pcrel32 target-flags(x86-plt) @f3, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
   ; CHECK-NEXT:   ADJCALLSTACKUP64 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
-  ; CHECK-NEXT:   [[COPY6:%[0-9]+]]:gr64_nosp = COPY $rax
-  ; CHECK-NEXT:   [[COPY7:%[0-9]+]]:gr64_nosp = COPY [[COPY6]]
-  ; CHECK-NEXT:   JMP64m $noreg, 8, [[COPY6]], @computed_goto.dispatch, $noreg
+  ; CHECK-NEXT:   [[COPY3:%[0-9]+]]:gr64 = COPY $rax
+  ; CHECK-NEXT:   JMP_1 %bb.5
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.4.bb4 (ir-block-address-taken %ir-block.bb4):
-  ; CHECK-NEXT:   successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+  ; CHECK-NEXT:   successors: %bb.5(0x80000000)
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   ADJCALLSTACKDOWN64 0, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
   ; CHECK-NEXT:   CALL64pcrel32 target-flags(x86-plt) @f4, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
   ; CHECK-NEXT:   ADJCALLSTACKUP64 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
-  ; CHECK-NEXT:   [[COPY8:%[0-9]+]]:gr64_nosp = COPY $rax
-  ; CHECK-NEXT:   [[COPY9:%[0-9]+]]:gr64_nosp = COPY [[COPY8]]
-  ; CHECK-NEXT:   JMP64m $noreg, 8, [[COPY8]], @computed_goto.dispatch, $noreg
+  ; CHECK-NEXT:   [[COPY4:%[0-9]+]]:gr64 = COPY $rax
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.5:
+  ; CHECK-NEXT:   successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:gr64_nosp = PHI [[COPY]], %bb.0, [[COPY4]], %bb.4, [[COPY3]], %bb.3, [[COPY2]], %bb.2, [[COPY1]], %bb.1
+  ; CHECK-NEXT:   JMP64m $noreg, 8, [[PHI]], @computed_goto.dispatch, $noreg
   bb.0:
     ADJCALLSTACKDOWN64 0, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
     CALL64pcrel32 target-flags(x86-plt) @f0, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
diff --git a/llvm/test/Instrumentation/HWAddressSanitizer/X86/globals.ll b/llvm/test/Instrumentation/HWAddressSanitizer/X86/globals.ll
index ca1153a..bdada7d 100644
--- a/llvm/test/Instrumentation/HWAddressSanitizer/X86/globals.ll
+++ b/llvm/test/Instrumentation/HWAddressSanitizer/X86/globals.ll
@@ -1,12 +1,5 @@
 ; RUN: opt < %s -S -passes=hwasan -mtriple=x86_64-unknown-linux-gnu | FileCheck %s
 
-; CHECK: @__start_hwasan_globals = external hidden constant [0 x i8]
-; CHECK: @__stop_hwasan_globals = external hidden constant [0 x i8]
-
-; CHECK: @hwasan.note = private constant { i32, i32, i32, [8 x i8], i32, i32 } { i32 8, i32 8, i32 3, [8 x i8] c"LLVM\00\00\00\00", i32 trunc (i64 sub (i64 ptrtoint (ptr @__start_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @__stop_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32) }, section ".note.hwasan.globals", comdat($hwasan.module_ctor), align 4
-
-; CHECK: @hwasan.dummy.global = private constant [0 x i8] zeroinitializer, section "hwasan_globals", comdat($hwasan.module_ctor), !associated [[NOTE:![0-9]+]]
-
 ; CHECK: @four.hwasan = private global { i32, [12 x i8] } { i32 1, [12 x i8] c"\00\00\00\00\00\00\00\00\00\00\00\10" }, align 16
 ; CHECK: @four.hwasan.descriptor = private constant { i32, i32 } { i32 trunc (i64 sub (i64 ptrtoint (ptr @four.hwasan to i64), i64 ptrtoint (ptr @four.hwasan.descriptor to i64)) to i32), i32 268435460 }, section "hwasan_globals", !associated [[FOUR:![0-9]+]]
 
@@ -17,14 +10,21 @@
 ; CHECK: @huge.hwasan.descriptor = private constant { i32, i32 } { i32 trunc (i64 sub (i64 ptrtoint (ptr @huge.hwasan to i64), i64 ptrtoint (ptr @huge.hwasan.descriptor to i64)) to i32), i32 318767088 }, section "hwasan_globals", !associated [[HUGE:![0-9]+]]
 ; CHECK: @huge.hwasan.descriptor.1 = private constant { i32, i32 } { i32 trunc (i64 add (i64 sub (i64 ptrtoint (ptr @huge.hwasan to i64), i64 ptrtoint (ptr @huge.hwasan.descriptor.1 to i64)), i64 16777200) to i32), i32 301989920 }, section "hwasan_globals", !associated [[HUGE]]
 
+; CHECK: @__start_hwasan_globals = external hidden constant [0 x i8]
+; CHECK: @__stop_hwasan_globals = external hidden constant [0 x i8]
+
+; CHECK: @hwasan.note = private constant { i32, i32, i32, [8 x i8], i32, i32 } { i32 8, i32 8, i32 3, [8 x i8] c"LLVM\00\00\00\00", i32 trunc (i64 sub (i64 ptrtoint (ptr @__start_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @__stop_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32) }, section ".note.hwasan.globals", comdat($hwasan.module_ctor), align 4
+
+; CHECK: @hwasan.dummy.global = private constant [0 x i8] zeroinitializer, section "hwasan_globals", comdat($hwasan.module_ctor), !associated [[NOTE:![0-9]+]]
+
 ; CHECK: @four = alias i32, inttoptr (i64 add (i64 ptrtoint (ptr @four.hwasan to i64), i64 2305843009213693952) to ptr)
 ; CHECK: @sixteen = alias [16 x i8], inttoptr (i64 add (i64 ptrtoint (ptr @sixteen.hwasan to i64), i64 2449958197289549824) to ptr)
 ; CHECK: @huge = alias [16777232 x i8], inttoptr (i64 add (i64 ptrtoint (ptr @huge.hwasan to i64), i64 2594073385365405696) to ptr)
 
-; CHECK: [[NOTE]] = !{ptr @hwasan.note}
 ; CHECK: [[FOUR]] = !{ptr @four.hwasan}
 ; CHECK: [[SIXTEEN]] = !{ptr @sixteen.hwasan}
 ; CHECK: [[HUGE]] = !{ptr @huge.hwasan}
+; CHECK: [[NOTE]] = !{ptr @hwasan.note}
 
 source_filename = "foo"
 
diff --git a/llvm/test/Instrumentation/HWAddressSanitizer/globals.ll b/llvm/test/Instrumentation/HWAddressSanitizer/globals.ll
index f5ae1c0..4c28523 100644
--- a/llvm/test/Instrumentation/HWAddressSanitizer/globals.ll
+++ b/llvm/test/Instrumentation/HWAddressSanitizer/globals.ll
@@ -1,16 +1,11 @@
-; RUN: opt < %s -S -passes=hwasan -mtriple=aarch64--linux-android29 | FileCheck --check-prefixes=CHECK,CHECK29 %s
-; RUN: opt < %s -S -passes=hwasan -mtriple=aarch64--linux-android30 | FileCheck --check-prefixes=CHECK,CHECK30 %s
+; RUN: opt < %s -S -passes=hwasan -mtriple=aarch64--linux-android29 | FileCheck --check-prefixes=CHECK,CHECK29,NOALLGLOBALS %s
+; RUN: opt < %s -S -passes=hwasan -mtriple=aarch64--linux-android30 | FileCheck --check-prefixes=CHECK,CHECK30,NOALLGLOBALS %s
+; RUN: opt < %s -S -passes=hwasan -mtriple=riscv64-unknown-elf -hwasan-globals=1 -hwasan-all-globals=1 | FileCheck --check-prefixes=CHECK,CHECK30,ALLGLOBALS %s
 
 ; CHECK29: @four = global
 
 ; CHECK: @specialcaselisted = global i16 2, no_sanitize_hwaddress
-; CHECK: @insection = global i16 2, section "custom"
-; CHECK: @__start_hwasan_globals = external hidden constant [0 x i8]
-; CHECK: @__stop_hwasan_globals = external hidden constant [0 x i8]
-
-; CHECK: @hwasan.note = private constant { i32, i32, i32, [8 x i8], i32, i32 } { i32 8, i32 8, i32 3, [8 x i8] c"LLVM\00\00\00\00", i32 trunc (i64 sub (i64 ptrtoint (ptr @__start_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @__stop_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32) }, section ".note.hwasan.globals", comdat($hwasan.module_ctor), align 4
-
-; CHECK: @hwasan.dummy.global = private constant [0 x i8] zeroinitializer, section "hwasan_globals", comdat($hwasan.module_ctor), !associated [[NOTE:![0-9]+]]
+; NOALLGLOBALS: @insection = global i16 2, section "custom"
 
 ; CHECK30: @four.hwasan = private global { i32, [12 x i8] } { i32 1, [12 x i8] c"\00\00\00\00\00\00\00\00\00\00\00\AC" }, align 16
 ; CHECK30: @four.hwasan.descriptor = private constant { i32, i32 } { i32 trunc (i64 sub (i64 ptrtoint (ptr @four.hwasan to i64), i64 ptrtoint (ptr @four.hwasan.descriptor to i64)) to i32), i32 -1409286140 }, section "hwasan_globals", !associated [[FOUR:![0-9]+]]
@@ -22,14 +17,23 @@
 ; CHECK30: @huge.hwasan.descriptor = private constant { i32, i32 } { i32 trunc (i64 sub (i64 ptrtoint (ptr @huge.hwasan to i64), i64 ptrtoint (ptr @huge.hwasan.descriptor to i64)) to i32), i32 -1358954512 }, section "hwasan_globals", !associated [[HUGE:![0-9]+]]
 ; CHECK30: @huge.hwasan.descriptor.1 = private constant { i32, i32 } { i32 trunc (i64 add (i64 sub (i64 ptrtoint (ptr @huge.hwasan to i64), i64 ptrtoint (ptr @huge.hwasan.descriptor.1 to i64)), i64 16777200) to i32), i32 -1375731680 }, section "hwasan_globals", !associated [[HUGE]]
 
+; ALLGLOBALS: @insection.hwasan = private global { i16, [14 x i8] } { i16 2, [14 x i8] c"\00\00\00\00\00\00\00\00\00\00\00\00\00\AF" }, section "custom", align 16
+
+; CHECK: @__start_hwasan_globals = external hidden constant [0 x i8]
+; CHECK: @__stop_hwasan_globals = external hidden constant [0 x i8]
+
+; CHECK: @hwasan.note = private constant { i32, i32, i32, [8 x i8], i32, i32 } { i32 8, i32 8, i32 3, [8 x i8] c"LLVM\00\00\00\00", i32 trunc (i64 sub (i64 ptrtoint (ptr @__start_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @__stop_hwasan_globals to i64), i64 ptrtoint (ptr @hwasan.note to i64)) to i32) }, section ".note.hwasan.globals", comdat($hwasan.module_ctor), align 4
+
+; CHECK: @hwasan.dummy.global = private constant [0 x i8] zeroinitializer, section "hwasan_globals", comdat($hwasan.module_ctor), !associated [[NOTE:![0-9]+]]
+
 ; CHECK30: @four = alias i32, inttoptr (i64 add (i64 ptrtoint (ptr @four.hwasan to i64), i64 -6052837899185946624) to ptr)
 ; CHECK30: @sixteen = alias [16 x i8], inttoptr (i64 add (i64 ptrtoint (ptr @sixteen.hwasan to i64), i64 -5980780305148018688) to ptr)
 ; CHECK30: @huge = alias [16777232 x i8], inttoptr (i64 add (i64 ptrtoint (ptr @huge.hwasan to i64), i64 -5908722711110090752) to ptr)
 
-; CHECK: [[NOTE]] = !{ptr @hwasan.note}
 ; CHECK30: [[FOUR]] = !{ptr @four.hwasan}
 ; CHECK30: [[SIXTEEN]] = !{ptr @sixteen.hwasan}
 ; CHECK30: [[HUGE]] = !{ptr @huge.hwasan}
+; CHECK: [[NOTE]] = !{ptr @hwasan.note}
 
 source_filename = "foo"
 
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3-fake16.s
index 81fc477..70090ac 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3-fake16.s
@@ -467,6 +467,96 @@ v_cvt_pk_fp8_f16 v1, 0x3118
 v_cvt_pk_fp8_f16 v1, 0.15915494
 // GFX1250: v_cvt_pk_fp8_f16 v1, 0x3118             ; encoding: [0x01,0x00,0x72,0xd7,0xff,0x00,0x00,0x00,0x18,0x31,0x00,0x00]
 
+v_cvt_pk_f16_f32 v5, v1, v2
+// GFX1250: v_cvt_pk_f16_f32 v5, v1, v2             ; encoding: [0x05,0x00,0x6f,0xd7,0x01,0x05,0x02,0x00]
+
+v_cvt_pk_f16_f32 v5, v255, v255
+// GFX1250: v_cvt_pk_f16_f32 v5, v255, v255         ; encoding: [0x05,0x00,0x6f,0xd7,0xff,0xff,0x03,0x00]
+
+v_cvt_pk_f16_f32 v5, s1, s2
+// GFX1250: v_cvt_pk_f16_f32 v5, s1, s2             ; encoding: [0x05,0x00,0x6f,0xd7,0x01,0x04,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, s105, s105
+// GFX1250: v_cvt_pk_f16_f32 v5, s105, s105         ; encoding: [0x05,0x00,0x6f,0xd7,0x69,0xd2,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, vcc_lo, ttmp15
+// GFX1250: v_cvt_pk_f16_f32 v5, vcc_lo, ttmp15     ; encoding: [0x05,0x00,0x6f,0xd7,0x6a,0xf6,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, vcc_hi, 0xaf123456
+// GFX1250: v_cvt_pk_f16_f32 v5, vcc_hi, 0xaf123456 ; encoding: [0x05,0x00,0x6f,0xd7,0x6b,0xfe,0x01,0x00,0x56,0x34,0x12,0xaf]
+
+v_cvt_pk_f16_f32 v5, ttmp15, src_scc
+// GFX1250: v_cvt_pk_f16_f32 v5, ttmp15, src_scc    ; encoding: [0x05,0x00,0x6f,0xd7,0x7b,0xfa,0x01,0x00]
+
+v_cvt_pk_f16_f32 v5, m0, 0.5
+// GFX1250: v_cvt_pk_f16_f32 v5, m0, 0.5            ; encoding: [0x05,0x00,0x6f,0xd7,0x7d,0xe0,0x01,0x00]
+
+v_cvt_pk_f16_f32 v5, exec_lo, -1
+// GFX1250: v_cvt_pk_f16_f32 v5, exec_lo, -1        ; encoding: [0x05,0x00,0x6f,0xd7,0x7e,0x82,0x01,0x00]
+
+v_cvt_pk_f16_f32 v5, exec_hi, null
+// GFX1250: v_cvt_pk_f16_f32 v5, exec_hi, null      ; encoding: [0x05,0x00,0x6f,0xd7,0x7f,0xf8,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, null, exec_lo
+// GFX1250: v_cvt_pk_f16_f32 v5, null, exec_lo      ; encoding: [0x05,0x00,0x6f,0xd7,0x7c,0xfc,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, -1, exec_hi
+// GFX1250: v_cvt_pk_f16_f32 v5, -1, exec_hi        ; encoding: [0x05,0x00,0x6f,0xd7,0xc1,0xfe,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, 0.5, m0 mul:2
+// GFX1250: v_cvt_pk_f16_f32 v5, 0.5, m0 mul:2      ; encoding: [0x05,0x00,0x6f,0xd7,0xf0,0xfa,0x00,0x08]
+
+v_cvt_pk_f16_f32 v5, src_scc, vcc_lo mul:4
+// GFX1250: v_cvt_pk_f16_f32 v5, src_scc, vcc_lo mul:4 ; encoding: [0x05,0x00,0x6f,0xd7,0xfd,0xd4,0x00,0x10]
+
+v_cvt_pk_f16_f32 v255, -|0xaf123456|, vcc_hi clamp div:2
+// GFX1250: v_cvt_pk_f16_f32 v255, -|0xaf123456|, vcc_hi clamp div:2 ; encoding: [0xff,0x81,0x6f,0xd7,0xff,0xd6,0x00,0x38,0x56,0x34,0x12,0xaf]
+
+v_cvt_sr_pk_f16_f32 v5, v1, v2, s3
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, v1, v2, s3      ; encoding: [0x05,0x00,0x70,0xd7,0x01,0x05,0x0e,0x00]
+
+v_cvt_sr_pk_f16_f32 v5, v255, s2, s105
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, v255, s2, s105  ; encoding: [0x05,0x00,0x70,0xd7,0xff,0x05,0xa4,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, s1, v255, exec_hi
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, s1, v255, exec_hi ; encoding: [0x05,0x00,0x70,0xd7,0x01,0xfe,0xff,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, s105, s105, exec_lo
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, s105, s105, exec_lo ; encoding: [0x05,0x00,0x70,0xd7,0x69,0xd2,0xf8,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, vcc_lo, ttmp15, v3
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, vcc_lo, ttmp15, v3 ; encoding: [0x05,0x00,0x70,0xd7,0x6a,0xf6,0x0c,0x04]
+
+v_cvt_sr_pk_f16_f32 v5, vcc_hi, 0xaf123456, v255
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, vcc_hi, 0xaf123456, v255 ; encoding: [0x05,0x00,0x70,0xd7,0x6b,0xfe,0xfd,0x07,0x56,0x34,0x12,0xaf]
+
+v_cvt_sr_pk_f16_f32 v5, -|ttmp15|, -|src_scc|, ttmp15
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -|ttmp15|, -|src_scc|, ttmp15 ; encoding: [0x05,0x03,0x70,0xd7,0x7b,0xfa,0xed,0x61]
+
+v_cvt_sr_pk_f16_f32 v5, m0, 0.5, m0
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, m0, 0.5, m0     ; encoding: [0x05,0x00,0x70,0xd7,0x7d,0xe0,0xf5,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, |exec_lo|, -1, vcc_hi
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, |exec_lo|, -1, vcc_hi ; encoding: [0x05,0x01,0x70,0xd7,0x7e,0x82,0xad,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, -|exec_hi|, null, vcc_lo
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -|exec_hi|, null, vcc_lo ; encoding: [0x05,0x01,0x70,0xd7,0x7f,0xf8,0xa8,0x21]
+
+v_cvt_sr_pk_f16_f32 v5, null, exec_lo, 0xaf123456
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, null, exec_lo, 0xaf123456 ; encoding: [0x05,0x00,0x70,0xd7,0x7c,0xfc,0xfc,0x03,0x56,0x34,0x12,0xaf]
+
+v_cvt_sr_pk_f16_f32 v5, -1, -|exec_hi|, src_scc
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -1, -|exec_hi|, src_scc ; encoding: [0x05,0x02,0x70,0xd7,0xc1,0xfe,0xf4,0x43]
+
+v_cvt_sr_pk_f16_f32 v5, 0.5, -m0, 0.5 mul:2
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, 0.5, -m0, 0.5 mul:2 ; encoding: [0x05,0x00,0x70,0xd7,0xf0,0xfa,0xc0,0x4b]
+
+v_cvt_sr_pk_f16_f32 v5, -src_scc, |vcc_lo|, -1 mul:4
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -src_scc, |vcc_lo|, -1 mul:4 ; encoding: [0x05,0x02,0x70,0xd7,0xfd,0xd4,0x04,0x33]
+
+v_cvt_sr_pk_f16_f32 v255, -|0xaf123456|, -|vcc_hi|, null clamp div:2
+// GFX1250: v_cvt_sr_pk_f16_f32 v255, -|0xaf123456|, -|vcc_hi|, null clamp div:2 ; encoding: [0xff,0x83,0x70,0xd7,0xff,0xd6,0xf0,0x79,0x56,0x34,0x12,0xaf]
+
 v_cvt_sr_bf8_f16 v1, v2, v3
 // GFX1250: v_cvt_sr_bf8_f16 v1, v2, v3             ; encoding: [0x01,0x00,0x75,0xd7,0x02,0x07,0x02,0x00]
 
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3.s
index 209951d..8e6f238 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3.s
@@ -467,6 +467,96 @@ v_cvt_pk_fp8_f16 v1.l, 0x3118
 v_cvt_pk_fp8_f16 v1.l, 0.15915494
 // GFX1250: v_cvt_pk_fp8_f16 v1.l, 0x3118           ; encoding: [0x01,0x00,0x72,0xd7,0xff,0x00,0x00,0x00,0x18,0x31,0x00,0x00]
 
+v_cvt_pk_f16_f32 v5, v1, v2
+// GFX1250: v_cvt_pk_f16_f32 v5, v1, v2             ; encoding: [0x05,0x00,0x6f,0xd7,0x01,0x05,0x02,0x00]
+
+v_cvt_pk_f16_f32 v5, v255, v255
+// GFX1250: v_cvt_pk_f16_f32 v5, v255, v255         ; encoding: [0x05,0x00,0x6f,0xd7,0xff,0xff,0x03,0x00]
+
+v_cvt_pk_f16_f32 v5, s1, s2
+// GFX1250: v_cvt_pk_f16_f32 v5, s1, s2             ; encoding: [0x05,0x00,0x6f,0xd7,0x01,0x04,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, s105, s105
+// GFX1250: v_cvt_pk_f16_f32 v5, s105, s105         ; encoding: [0x05,0x00,0x6f,0xd7,0x69,0xd2,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, vcc_lo, ttmp15
+// GFX1250: v_cvt_pk_f16_f32 v5, vcc_lo, ttmp15     ; encoding: [0x05,0x00,0x6f,0xd7,0x6a,0xf6,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, vcc_hi, 0xaf123456
+// GFX1250: v_cvt_pk_f16_f32 v5, vcc_hi, 0xaf123456 ; encoding: [0x05,0x00,0x6f,0xd7,0x6b,0xfe,0x01,0x00,0x56,0x34,0x12,0xaf]
+
+v_cvt_pk_f16_f32 v5, ttmp15, src_scc
+// GFX1250: v_cvt_pk_f16_f32 v5, ttmp15, src_scc    ; encoding: [0x05,0x00,0x6f,0xd7,0x7b,0xfa,0x01,0x00]
+
+v_cvt_pk_f16_f32 v5, m0, 0.5
+// GFX1250: v_cvt_pk_f16_f32 v5, m0, 0.5            ; encoding: [0x05,0x00,0x6f,0xd7,0x7d,0xe0,0x01,0x00]
+
+v_cvt_pk_f16_f32 v5, exec_lo, -1
+// GFX1250: v_cvt_pk_f16_f32 v5, exec_lo, -1        ; encoding: [0x05,0x00,0x6f,0xd7,0x7e,0x82,0x01,0x00]
+
+v_cvt_pk_f16_f32 v5, exec_hi, null
+// GFX1250: v_cvt_pk_f16_f32 v5, exec_hi, null      ; encoding: [0x05,0x00,0x6f,0xd7,0x7f,0xf8,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, null, exec_lo
+// GFX1250: v_cvt_pk_f16_f32 v5, null, exec_lo      ; encoding: [0x05,0x00,0x6f,0xd7,0x7c,0xfc,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, -1, exec_hi
+// GFX1250: v_cvt_pk_f16_f32 v5, -1, exec_hi        ; encoding: [0x05,0x00,0x6f,0xd7,0xc1,0xfe,0x00,0x00]
+
+v_cvt_pk_f16_f32 v5, 0.5, m0 mul:2
+// GFX1250: v_cvt_pk_f16_f32 v5, 0.5, m0 mul:2      ; encoding: [0x05,0x00,0x6f,0xd7,0xf0,0xfa,0x00,0x08]
+
+v_cvt_pk_f16_f32 v5, src_scc, vcc_lo mul:4
+// GFX1250: v_cvt_pk_f16_f32 v5, src_scc, vcc_lo mul:4 ; encoding: [0x05,0x00,0x6f,0xd7,0xfd,0xd4,0x00,0x10]
+
+v_cvt_pk_f16_f32 v255, -|0xaf123456|, vcc_hi clamp div:2
+// GFX1250: v_cvt_pk_f16_f32 v255, -|0xaf123456|, vcc_hi clamp div:2 ; encoding: [0xff,0x81,0x6f,0xd7,0xff,0xd6,0x00,0x38,0x56,0x34,0x12,0xaf]
+
+v_cvt_sr_pk_f16_f32 v5, v1, v2, s3
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, v1, v2, s3      ; encoding: [0x05,0x00,0x70,0xd7,0x01,0x05,0x0e,0x00]
+
+v_cvt_sr_pk_f16_f32 v5, v255, s2, s105
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, v255, s2, s105  ; encoding: [0x05,0x00,0x70,0xd7,0xff,0x05,0xa4,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, s1, v255, exec_hi
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, s1, v255, exec_hi ; encoding: [0x05,0x00,0x70,0xd7,0x01,0xfe,0xff,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, s105, s105, exec_lo
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, s105, s105, exec_lo ; encoding: [0x05,0x00,0x70,0xd7,0x69,0xd2,0xf8,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, vcc_lo, ttmp15, v3
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, vcc_lo, ttmp15, v3 ; encoding: [0x05,0x00,0x70,0xd7,0x6a,0xf6,0x0c,0x04]
+
+v_cvt_sr_pk_f16_f32 v5, vcc_hi, 0xaf123456, v255
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, vcc_hi, 0xaf123456, v255 ; encoding: [0x05,0x00,0x70,0xd7,0x6b,0xfe,0xfd,0x07,0x56,0x34,0x12,0xaf]
+
+v_cvt_sr_pk_f16_f32 v5, -|ttmp15|, -|src_scc|, ttmp15
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -|ttmp15|, -|src_scc|, ttmp15 ; encoding: [0x05,0x03,0x70,0xd7,0x7b,0xfa,0xed,0x61]
+
+v_cvt_sr_pk_f16_f32 v5, m0, 0.5, m0
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, m0, 0.5, m0     ; encoding: [0x05,0x00,0x70,0xd7,0x7d,0xe0,0xf5,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, |exec_lo|, -1, vcc_hi
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, |exec_lo|, -1, vcc_hi ; encoding: [0x05,0x01,0x70,0xd7,0x7e,0x82,0xad,0x01]
+
+v_cvt_sr_pk_f16_f32 v5, -|exec_hi|, null, vcc_lo
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -|exec_hi|, null, vcc_lo ; encoding: [0x05,0x01,0x70,0xd7,0x7f,0xf8,0xa8,0x21]
+
+v_cvt_sr_pk_f16_f32 v5, null, exec_lo, 0xaf123456
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, null, exec_lo, 0xaf123456 ; encoding: [0x05,0x00,0x70,0xd7,0x7c,0xfc,0xfc,0x03,0x56,0x34,0x12,0xaf]
+
+v_cvt_sr_pk_f16_f32 v5, -1, -|exec_hi|, src_scc
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -1, -|exec_hi|, src_scc ; encoding: [0x05,0x02,0x70,0xd7,0xc1,0xfe,0xf4,0x43]
+
+v_cvt_sr_pk_f16_f32 v5, 0.5, -m0, 0.5 mul:2
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, 0.5, -m0, 0.5 mul:2 ; encoding: [0x05,0x00,0x70,0xd7,0xf0,0xfa,0xc0,0x4b]
+
+v_cvt_sr_pk_f16_f32 v5, -src_scc, |vcc_lo|, -1 mul:4
+// GFX1250: v_cvt_sr_pk_f16_f32 v5, -src_scc, |vcc_lo|, -1 mul:4 ; encoding: [0x05,0x02,0x70,0xd7,0xfd,0xd4,0x04,0x33]
+
+v_cvt_sr_pk_f16_f32 v255, -|0xaf123456|, -|vcc_hi|, null clamp div:2
+// GFX1250: v_cvt_sr_pk_f16_f32 v255, -|0xaf123456|, -|vcc_hi|, null clamp div:2 ; encoding: [0xff,0x83,0x70,0xd7,0xff,0xd6,0xf0,0x79,0x56,0x34,0x12,0xaf]
+
 v_cvt_sr_bf8_f16 v1, v2.l, v3
 // GFX1250: v_cvt_sr_bf8_f16 v1, v2.l, v3           ; encoding: [0x01,0x00,0x75,0xd7,0x02,0x07,0x02,0x00]
 
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16-fake16.s
index c406890..6430469 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16-fake16.s
@@ -346,6 +346,110 @@ v_cvt_pk_fp8_f16_e64_dpp v1, v2 op_sel:[0,1] row_share:0 row_mask:0x5 bank_mask:
 // GFX1250: v_cvt_pk_fp8_f16_e64_dpp v1, v2 op_sel:[0,1] row_share:0 row_mask:0x5 bank_mask:0x3 fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xfa,0x00,0x00,0x00,0x02,0x50,0x05,0x53]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
 
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[0,1,2,3]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0xe4,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_mirror
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x40,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_half_mirror
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x41,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x01,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:15
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x0f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x11,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:15
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x21,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:15
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x2f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x08,0x01,0x5f,0x01,0x01]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 fi:0
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x10,0x01,0x60,0x09,0x13]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xff,0x81,0x6f,0xd7,0xfa,0xfe,0x03,0x38,0xff,0x6f,0x05,0x30]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x1b,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[0,1,2,3]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0xe4,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 row_mirror
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x40,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 row_half_mirror
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xfe,0x07,0x01,0x41,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 row_shl:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xa6,0x01,0x01,0x01,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi row_shl:15
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xae,0x01,0x01,0x0f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo row_shr:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xaa,0x01,0x01,0x11,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi row_ror:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x02,0x70,0xd7,0xfa,0x04,0xfe,0x41,0x01,0x21,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null row_share:0 row_mask:0xf bank_mask:0xf
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x03,0x70,0xd7,0xfa,0x04,0xf2,0x61,0x01,0x50,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0x05,0x01,0x70,0xd7,0xfa,0x04,0x06,0x2b,0x01,0x5f,0x01,0x01]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 fi:0
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0x05,0x02,0x70,0xd7,0xfa,0x04,0x16,0x52,0x01,0x60,0x09,0x13]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xff,0x83,0x70,0xd7,0xfa,0xfe,0xf7,0x7b,0xff,0x6f,0x05,0x30]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
 v_cvt_sr_bf8_f16 v1, v2, v3 quad_perm:[0,1,2,3] fi:1
 // GFX1250: v_cvt_sr_bf8_f16_e64_dpp v1, v2, v3 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf fi:1 ; encoding: [0x01,0x00,0x75,0xd7,0xfa,0x06,0x02,0x00,0x02,0xe4,0x04,0xff]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16.s
index 741d1a1..1d6cf07 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp16.s
@@ -346,6 +346,110 @@ v_cvt_pk_fp8_f16_e64_dpp v1.h, v2 row_share:0 row_mask:0x5 bank_mask:0x3 fi:1
 // GFX1250: v_cvt_pk_fp8_f16_e64_dpp v1.h, v2 op_sel:[0,1] row_share:0 row_mask:0x5 bank_mask:0x3 fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xfa,0x00,0x00,0x00,0x02,0x50,0x05,0x53]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
 
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[0,1,2,3]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0xe4,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_mirror
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x40,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_half_mirror
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x41,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x01,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:15
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x0f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x11,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:15
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x21,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:15
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x2f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x08,0x01,0x5f,0x01,0x01]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 fi:0
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x10,0x01,0x60,0x09,0x13]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xff,0x81,0x6f,0xd7,0xfa,0xfe,0x03,0x38,0xff,0x6f,0x05,0x30]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x1b,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[0,1,2,3]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0xe4,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 row_mirror
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x40,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 row_half_mirror
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xfe,0x07,0x01,0x41,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 row_shl:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xa6,0x01,0x01,0x01,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi row_shl:15
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xae,0x01,0x01,0x0f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo row_shr:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xaa,0x01,0x01,0x11,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi row_ror:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x02,0x70,0xd7,0xfa,0x04,0xfe,0x41,0x01,0x21,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null row_share:0 row_mask:0xf bank_mask:0xf
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x03,0x70,0xd7,0xfa,0x04,0xf2,0x61,0x01,0x50,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0x05,0x01,0x70,0xd7,0xfa,0x04,0x06,0x2b,0x01,0x5f,0x01,0x01]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 fi:0
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0x05,0x02,0x70,0xd7,0xfa,0x04,0x16,0x52,0x01,0x60,0x09,0x13]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xff,0x83,0x70,0xd7,0xfa,0xfe,0xf7,0x7b,0xff,0x6f,0x05,0x30]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
 v_cvt_sr_bf8_f16 v1, v2.l, v3 quad_perm:[0,1,2,3] fi:1
 // GFX1250: v_cvt_sr_bf8_f16_e64_dpp v1, v2.l, v3 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf fi:1 ; encoding: [0x01,0x00,0x75,0xd7,0xfa,0x06,0x02,0x00,0x02,0xe4,0x04,0xff]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8-fake16.s
index 8a3e7ad..a7aef85 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8-fake16.s
@@ -242,6 +242,62 @@ v_cvt_pk_fp8_f16_e64_dpp v1, v2 op_sel:[0,1] dpp8:[7,6,5,4,3,2,1,0] fi:1
 // GFX1250: v_cvt_pk_fp8_f16_e64_dpp v1, v2 op_sel:[0,1] dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xea,0x00,0x00,0x00,0x02,0x77,0x39,0x05]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
 
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x08,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x05,0x00,0x6f,0xd7,0xea,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 dpp8:[0,0,0,0,0,0,0,0] fi:0
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x81,0x6f,0xd7,0xe9,0xfe,0x03,0x38,0xff,0x00,0x00,0x00]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xfe,0x07,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xa6,0x01,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xae,0x01,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xaa,0x01,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x02,0x70,0xd7,0xe9,0x04,0xfe,0x41,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x03,0x70,0xd7,0xe9,0x04,0xf2,0x61,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x01,0x70,0xd7,0xe9,0x04,0x06,0x2b,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x05,0x02,0x70,0xd7,0xea,0x04,0x16,0x52,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 dpp8:[0,0,0,0,0,0,0,0] fi:0
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x83,0x70,0xd7,0xe9,0xfe,0xf7,0x7b,0xff,0x00,0x00,0x00]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
 v_cvt_sr_bf8_f16 v1, v2, v3 dpp8:[1,2,3,4,5,6,7,0] fi:1
 // GFX1250: v_cvt_sr_bf8_f16_e64_dpp v1, v2, v3 dpp8:[1,2,3,4,5,6,7,0] fi:1 ; encoding: [0x01,0x00,0x75,0xd7,0xea,0x06,0x02,0x00,0x02,0xd1,0x58,0x1f]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8.s
index f79b5c5..3556f06 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_dpp8.s
@@ -242,6 +242,62 @@ v_cvt_pk_fp8_f16_e64_dpp v1.h, v2 dpp8:[7,6,5,4,3,2,1,0] fi:1
 // GFX1250: v_cvt_pk_fp8_f16_e64_dpp v1.h, v2 op_sel:[0,1] dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xea,0x00,0x00,0x00,0x02,0x77,0x39,0x05]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
 
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x08,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x05,0x00,0x6f,0xd7,0xea,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 dpp8:[0,0,0,0,0,0,0,0] fi:0
+// GFX1250: v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x81,0x6f,0xd7,0xe9,0xfe,0x03,0x38,0xff,0x00,0x00,0x00]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xfe,0x07,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xa6,0x01,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xae,0x01,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xaa,0x01,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x02,0x70,0xd7,0xe9,0x04,0xfe,0x41,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x03,0x70,0xd7,0xe9,0x04,0xf2,0x61,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x01,0x70,0xd7,0xe9,0x04,0x06,0x2b,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x05,0x02,0x70,0xd7,0xea,0x04,0x16,0x52,0x01,0x77,0x39,0x05]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 dpp8:[0,0,0,0,0,0,0,0] fi:0
+// GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x83,0x70,0xd7,0xe9,0xfe,0xf7,0x7b,0xff,0x00,0x00,0x00]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
 v_cvt_sr_bf8_f16 v1, v2.l, v3 dpp8:[1,2,3,4,5,6,7,0] fi:1
 // GFX1250: v_cvt_sr_bf8_f16_e64_dpp v1, v2.l, v3 dpp8:[1,2,3,4,5,6,7,0] fi:1 ; encoding: [0x01,0x00,0x75,0xd7,0xea,0x06,0x02,0x00,0x02,0xd1,0x58,0x1f]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3.txt
index 10ffc2c..fc76fc3 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3.txt
@@ -486,6 +486,96 @@
 # GFX1250-REAL16: v_cvt_pk_fp8_f16 v1.l, 0x3118           ; encoding: [0x01,0x00,0x72,0xd7,0xff,0x00,0x00,0x00,0x18,0x31,0x00,0x00]
 # GFX1250-FAKE16: v_cvt_pk_fp8_f16 v1, 0x3118             ; encoding: [0x01,0x00,0x72,0xd7,0xff,0x00,0x00,0x00,0x18,0x31,0x00,0x00]
 
+0xff,0x81,0x6f,0xd7,0xff,0xd6,0x00,0x38,0x56,0x34,0x12,0xaf
+# GFX1250: v_cvt_pk_f16_f32 v255, -|0xaf123456|, vcc_hi clamp div:2 ; encoding: [0xff,0x81,0x6f,0xd7,0xff,0xd6,0x00,0x38,0x56,0x34,0x12,0xaf]
+
+0x05,0x00,0x6f,0xd7,0xc1,0xfe,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, -1, exec_hi        ; encoding: [0x05,0x00,0x6f,0xd7,0xc1,0xfe,0x00,0x00]
+
+0x05,0x00,0x6f,0xd7,0xf0,0xfa,0x00,0x08
+# GFX1250: v_cvt_pk_f16_f32 v5, 0.5, m0 mul:2      ; encoding: [0x05,0x00,0x6f,0xd7,0xf0,0xfa,0x00,0x08]
+
+0x05,0x00,0x6f,0xd7,0x7f,0xf8,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, exec_hi, null      ; encoding: [0x05,0x00,0x6f,0xd7,0x7f,0xf8,0x00,0x00]
+
+0x05,0x00,0x6f,0xd7,0x7e,0x82,0x01,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, exec_lo, -1        ; encoding: [0x05,0x00,0x6f,0xd7,0x7e,0x82,0x01,0x00]
+
+0x05,0x00,0x6f,0xd7,0x7d,0xe0,0x01,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, m0, 0.5            ; encoding: [0x05,0x00,0x6f,0xd7,0x7d,0xe0,0x01,0x00]
+
+0x05,0x00,0x6f,0xd7,0x7c,0xfc,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, null, exec_lo      ; encoding: [0x05,0x00,0x6f,0xd7,0x7c,0xfc,0x00,0x00]
+
+0x05,0x00,0x6f,0xd7,0x01,0x04,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, s1, s2             ; encoding: [0x05,0x00,0x6f,0xd7,0x01,0x04,0x00,0x00]
+
+0x05,0x00,0x6f,0xd7,0x69,0xd2,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, s105, s105         ; encoding: [0x05,0x00,0x6f,0xd7,0x69,0xd2,0x00,0x00]
+
+0x05,0x00,0x6f,0xd7,0xfd,0xd4,0x00,0x10
+# GFX1250: v_cvt_pk_f16_f32 v5, src_scc, vcc_lo mul:4 ; encoding: [0x05,0x00,0x6f,0xd7,0xfd,0xd4,0x00,0x10]
+
+0x05,0x00,0x6f,0xd7,0x7b,0xfa,0x01,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, ttmp15, src_scc    ; encoding: [0x05,0x00,0x6f,0xd7,0x7b,0xfa,0x01,0x00]
+
+0x05,0x00,0x6f,0xd7,0x01,0x05,0x02,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, v1, v2             ; encoding: [0x05,0x00,0x6f,0xd7,0x01,0x05,0x02,0x00]
+
+0x05,0x00,0x6f,0xd7,0xff,0xff,0x03,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, v255, v255         ; encoding: [0x05,0x00,0x6f,0xd7,0xff,0xff,0x03,0x00]
+
+0x05,0x00,0x6f,0xd7,0x6b,0xfe,0x01,0x00,0x56,0x34,0x12,0xaf
+# GFX1250: v_cvt_pk_f16_f32 v5, vcc_hi, 0xaf123456 ; encoding: [0x05,0x00,0x6f,0xd7,0x6b,0xfe,0x01,0x00,0x56,0x34,0x12,0xaf]
+
+0x05,0x00,0x6f,0xd7,0x6a,0xf6,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32 v5, vcc_lo, ttmp15     ; encoding: [0x05,0x00,0x6f,0xd7,0x6a,0xf6,0x00,0x00]
+
+0xff,0x83,0x70,0xd7,0xff,0xd6,0xf0,0x79,0x56,0x34,0x12,0xaf
+# GFX1250: v_cvt_sr_pk_f16_f32 v255, -|0xaf123456|, -|vcc_hi|, null clamp div:2 ; encoding: [0xff,0x83,0x70,0xd7,0xff,0xd6,0xf0,0x79,0x56,0x34,0x12,0xaf]
+
+0x05,0x02,0x70,0xd7,0xc1,0xfe,0xf4,0x43
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, -1, -|exec_hi|, src_scc ; encoding: [0x05,0x02,0x70,0xd7,0xc1,0xfe,0xf4,0x43]
+
+0x05,0x02,0x70,0xd7,0xfd,0xd4,0x04,0x33
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, -src_scc, |vcc_lo|, -1 mul:4 ; encoding: [0x05,0x02,0x70,0xd7,0xfd,0xd4,0x04,0x33]
+
+0x05,0x01,0x70,0xd7,0x7f,0xf8,0xa8,0x21
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, -|exec_hi|, null, vcc_lo ; encoding: [0x05,0x01,0x70,0xd7,0x7f,0xf8,0xa8,0x21]
+
+0x05,0x03,0x70,0xd7,0x7b,0xfa,0xed,0x61
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, -|ttmp15|, -|src_scc|, ttmp15 ; encoding: [0x05,0x03,0x70,0xd7,0x7b,0xfa,0xed,0x61]
+
+0x05,0x00,0x70,0xd7,0xf0,0xfa,0xc0,0x4b
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, 0.5, -m0, 0.5 mul:2 ; encoding: [0x05,0x00,0x70,0xd7,0xf0,0xfa,0xc0,0x4b]
+
+0x05,0x00,0x70,0xd7,0x7d,0xe0,0xf5,0x01
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, m0, 0.5, m0     ; encoding: [0x05,0x00,0x70,0xd7,0x7d,0xe0,0xf5,0x01]
+
+0x05,0x00,0x70,0xd7,0x7c,0xfc,0xfc,0x03,0x56,0x34,0x12,0xaf
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, null, exec_lo, 0xaf123456 ; encoding: [0x05,0x00,0x70,0xd7,0x7c,0xfc,0xfc,0x03,0x56,0x34,0x12,0xaf]
+
+0x05,0x00,0x70,0xd7,0x01,0xfe,0xff,0x01
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, s1, v255, exec_hi ; encoding: [0x05,0x00,0x70,0xd7,0x01,0xfe,0xff,0x01]
+
+0x05,0x00,0x70,0xd7,0x69,0xd2,0xf8,0x01
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, s105, s105, exec_lo ; encoding: [0x05,0x00,0x70,0xd7,0x69,0xd2,0xf8,0x01]
+
+0x05,0x00,0x70,0xd7,0x01,0x05,0x0e,0x00
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, v1, v2, s3      ; encoding: [0x05,0x00,0x70,0xd7,0x01,0x05,0x0e,0x00]
+
+0x05,0x00,0x70,0xd7,0xff,0x05,0xa4,0x01
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, v255, s2, s105  ; encoding: [0x05,0x00,0x70,0xd7,0xff,0x05,0xa4,0x01]
+
+0x05,0x00,0x70,0xd7,0x6b,0xfe,0xfd,0x07,0x56,0x34,0x12,0xaf
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, vcc_hi, 0xaf123456, v255 ; encoding: [0x05,0x00,0x70,0xd7,0x6b,0xfe,0xfd,0x07,0x56,0x34,0x12,0xaf]
+
+0x05,0x00,0x70,0xd7,0x6a,0xf6,0x0c,0x04
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, vcc_lo, ttmp15, v3 ; encoding: [0x05,0x00,0x70,0xd7,0x6a,0xf6,0x0c,0x04]
+
+0x05,0x01,0x70,0xd7,0x7e,0x82,0xad,0x01
+# GFX1250: v_cvt_sr_pk_f16_f32 v5, |exec_lo|, -1, vcc_hi ; encoding: [0x05,0x01,0x70,0xd7,0x7e,0x82,0xad,0x01]
+
 0x01,0x00,0x75,0xd7,0x02,0x07,0x02,0x20
 # GFX1250-REAL16: v_cvt_sr_bf8_f16 v1, -v2.l, v3          ; encoding: [0x01,0x00,0x75,0xd7,0x02,0x07,0x02,0x20]
 # GFX1250-FAKE16: v_cvt_sr_bf8_f16 v1, -v2, v3            ; encoding: [0x01,0x00,0x75,0xd7,0x02,0x07,0x02,0x20]
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp16.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp16.txt
index e138425..a12d4e8 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp16.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp16.txt
@@ -286,6 +286,84 @@
 # GFX1250-REAL16: v_cvt_pk_fp8_f16_e64_dpp v1.h, v2 op_sel:[0,1] row_share:0 row_mask:0x5 bank_mask:0x3 fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xfa,0x00,0x00,0x00,0x02,0x50,0x05,0x53]
 # GFX1250-FAKE16: v_cvt_pk_fp8_f16_e64_dpp v1, v2 op_sel:[0,1] row_share:0 row_mask:0x5 bank_mask:0x3 fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xfa,0x00,0x00,0x00,0x02,0x50,0x05,0x53]
 
+0xff,0x81,0x6f,0xd7,0xfa,0xfe,0x03,0x38,0xff,0x6f,0x05,0x30
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xff,0x81,0x6f,0xd7,0xfa,0xfe,0x03,0x38,0xff,0x6f,0x05,0x30]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x08,0x01,0x5f,0x01,0x01
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x08,0x01,0x5f,0x01,0x01]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x10,0x01,0x60,0x09,0x13
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x10,0x01,0x60,0x09,0x13]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0xe4,0x00,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0xe4,0x00,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x41,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x41,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x40,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x40,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x21,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x21,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x2f,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_ror:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x2f,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x50,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x01,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x01,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x0f,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x0f,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x11,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x11,0x01,0xff]
+
+0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1f,0x01,0xff
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 row_shr:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x6f,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1f,0x01,0xff]
+
+0xff,0x83,0x70,0xd7,0xfa,0xfe,0xf7,0x7b,0xff,0x6f,0x05,0x30
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xff,0x83,0x70,0xd7,0xfa,0xfe,0xf7,0x7b,0xff,0x6f,0x05,0x30]
+
+0x05,0x03,0x70,0xd7,0xfa,0x04,0xf2,0x61,0x01,0x50,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x03,0x70,0xd7,0xfa,0x04,0xf2,0x61,0x01,0x50,0x01,0xff]
+
+0x05,0x01,0x70,0xd7,0xfa,0x04,0x06,0x2b,0x01,0x5f,0x01,0x01
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0x05,0x01,0x70,0xd7,0xfa,0x04,0x06,0x2b,0x01,0x5f,0x01,0x01]
+
+0x05,0x02,0x70,0xd7,0xfa,0x04,0x16,0x52,0x01,0x60,0x09,0x13
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0x05,0x02,0x70,0xd7,0xfa,0x04,0x16,0x52,0x01,0x60,0x09,0x13]
+
+0x05,0x02,0x70,0xd7,0xfa,0x04,0xfe,0x41,0x01,0x21,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x02,0x70,0xd7,0xfa,0x04,0xfe,0x41,0x01,0x21,0x01,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0xa6,0x01,0x01,0x01,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xa6,0x01,0x01,0x01,0x01,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0xfe,0x07,0x01,0x41,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xfe,0x07,0x01,0x41,0x01,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0xe4,0x00,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0xe4,0x00,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x1b,0x00,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x1b,0x00,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x40,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0x0e,0x04,0x01,0x40,0x01,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0xae,0x01,0x01,0x0f,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xae,0x01,0x01,0x0f,0x01,0xff]
+
+0x05,0x00,0x70,0xd7,0xfa,0x04,0xaa,0x01,0x01,0x11,0x01,0xff
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x70,0xd7,0xfa,0x04,0xaa,0x01,0x01,0x11,0x01,0xff]
+
 0x01,0x20,0x75,0xd7,0xfa,0x06,0x02,0x00,0x02,0xe4,0x00,0xff
 # GFX1250-REAL16: v_cvt_sr_bf8_f16_e64_dpp v1, v2.l, v3 byte_sel:1 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x20,0x75,0xd7,0xfa,0x06,0x02,0x00,0x02,0xe4,0x00,0xff]
 # GFX1250-FAKE16: v_cvt_sr_bf8_f16_e64_dpp v1, v2, v3 byte_sel:1 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x20,0x75,0xd7,0xfa,0x06,0x02,0x00,0x02,0xe4,0x00,0xff]
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp8.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp8.txt
index c6bde22..796d362 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp8.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_dpp8.txt
@@ -204,6 +204,48 @@
 # GFX1250-REAL16: v_cvt_pk_fp8_f16_e64_dpp v1.h, v2 op_sel:[0,1] dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xea,0x00,0x00,0x00,0x02,0x77,0x39,0x05]
 # GFX1250-FAKE16: v_cvt_pk_fp8_f16_e64_dpp v1, v2 op_sel:[0,1] dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x01,0x40,0x72,0xd7,0xea,0x00,0x00,0x00,0x02,0x77,0x39,0x05]
 
+0xff,0x81,0x6f,0xd7,0xe9,0xfe,0x03,0x38,0xff,0x00,0x00,0x00
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v255, -|v255|, v255 clamp div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x81,0x6f,0xd7,0xe9,0xfe,0x03,0x38,0xff,0x00,0x00,0x00]
+
+0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x08,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6f,0xd7,0xe9,0x04,0x02,0x08,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x6f,0xd7,0xea,0x04,0x02,0x10,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_pk_f16_f32_e64_dpp v5, v1, v2 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x05,0x00,0x6f,0xd7,0xea,0x04,0x02,0x10,0x01,0x77,0x39,0x05]
+
+0xff,0x83,0x70,0xd7,0xe9,0xfe,0xf7,0x7b,0xff,0x00,0x00,0x00
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v255, -|v255|, -|v255|, src_scc clamp div:2 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x83,0x70,0xd7,0xe9,0xfe,0xf7,0x7b,0xff,0x00,0x00,0x00]
+
+0x05,0x03,0x70,0xd7,0xe9,0x04,0xf2,0x61,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, -|v2|, null dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x03,0x70,0xd7,0xe9,0x04,0xf2,0x61,0x01,0x77,0x39,0x05]
+
+0x05,0x01,0x70,0xd7,0xe9,0x04,0x06,0x2b,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, -|v1|, v2, -1 mul:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x01,0x70,0xd7,0xe9,0x04,0x06,0x2b,0x01,0x77,0x39,0x05]
+
+0x05,0x02,0x70,0xd7,0xea,0x04,0x16,0x52,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, 5 mul:4 dpp8:[7,6,5,4,3,2,1,0] fi:1 ; encoding: [0x05,0x02,0x70,0xd7,0xea,0x04,0x16,0x52,0x01,0x77,0x39,0x05]
+
+0x05,0x02,0x70,0xd7,0xe9,0x04,0xfe,0x41,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, -|v2|, exec_hi dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x02,0x70,0xd7,0xe9,0x04,0xfe,0x41,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x70,0xd7,0xe9,0x04,0xa6,0x01,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, s105 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xa6,0x01,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x70,0xd7,0xe9,0x04,0xfe,0x07,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v255 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xfe,0x07,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x70,0xd7,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0x0e,0x04,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x70,0xd7,0xe9,0x04,0xae,0x01,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_hi dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xae,0x01,0x01,0x77,0x39,0x05]
+
+0x05,0x00,0x70,0xd7,0xe9,0x04,0xaa,0x01,0x01,0x77,0x39,0x05
+# GFX1250: v_cvt_sr_pk_f16_f32_e64_dpp v5, v1, v2, vcc_lo dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x70,0xd7,0xe9,0x04,0xaa,0x01,0x01,0x77,0x39,0x05]
+
 0x01,0x00,0x75,0xd7,0xea,0x06,0x02,0x00,0x02,0xd1,0x58,0x1f
 # GFX1250-REAL16: v_cvt_sr_bf8_f16_e64_dpp v1, v2.l, v3 dpp8:[1,2,3,4,5,6,7,0] fi:1 ; encoding: [0x01,0x00,0x75,0xd7,0xea,0x06,0x02,0x00,0x02,0xd1,0x58,0x1f]
 # GFX1250-FAKE16: v_cvt_sr_bf8_f16_e64_dpp v1, v2, v3 dpp8:[1,2,3,4,5,6,7,0] fi:1 ; encoding: [0x01,0x00,0x75,0xd7,0xea,0x06,0x02,0x00,0x02,0xd1,0x58,0x1f]
diff --git a/llvm/test/ThinLTO/AArch64/cgdata-merge-read.ll b/llvm/test/ThinLTO/AArch64/cgdata-merge-read.ll
index da756e7..9eb9bda 100644
--- a/llvm/test/ThinLTO/AArch64/cgdata-merge-read.ll
+++ b/llvm/test/ThinLTO/AArch64/cgdata-merge-read.ll
@@ -30,6 +30,20 @@
 ; RUN: llvm-objdump -d %tout-read.1 | FileCheck %s --check-prefix=THUNK1
 ; RUN: llvm-objdump -d %tout-read.2 | FileCheck %s --check-prefix=THUNK2
 
+; It runs the same if we use -indexed-codegen-data-read-function-map-names=false.
+; RUN: llvm-lto2 run -enable-global-merge-func=true \
+; RUN:    -indexed-codegen-data-read-function-map-names=false \
+; RUN:    -codegen-data-use-path=%tout.cgdata \
+; RUN:    %t-foo.bc %t-goo.bc -o %tout-read \
+; RUN:    -r %t-foo.bc,_f1,px \
+; RUN:    -r %t-goo.bc,_f2,px \
+; RUN:    -r %t-foo.bc,_g,l -r %t-foo.bc,_g1,l -r %t-foo.bc,_g2,l \
+; RUN:    -r %t-goo.bc,_g,l -r %t-goo.bc,_g1,l -r %t-goo.bc,_g2,l
+; RUN: llvm-nm %tout-read.1 | FileCheck %s --check-prefix=READ1
+; RUN: llvm-nm %tout-read.2 | FileCheck %s --check-prefix=READ2
+; RUN: llvm-objdump -d %tout-read.1 | FileCheck %s --check-prefix=THUNK1
+; RUN: llvm-objdump -d %tout-read.2 | FileCheck %s --check-prefix=THUNK2
+
 ; READ1: _f1.Tgm
 ; READ2: _f2.Tgm
 
diff --git a/llvm/test/Transforms/AggressiveInstCombine/lower-table-based-cttz-basics.ll b/llvm/test/Transforms/AggressiveInstCombine/lower-table-based-cttz-basics.ll
index 4d57199..bb3001e 100644
--- a/llvm/test/Transforms/AggressiveInstCombine/lower-table-based-cttz-basics.ll
+++ b/llvm/test/Transforms/AggressiveInstCombine/lower-table-based-cttz-basics.ll
@@ -190,6 +190,39 @@ return:                                           ; preds = %entry, %if.end
   ret i32 %retval.0
 }
 
+define i32 @ctz3_with_i8gep(i32 %x) {
+; CHECK-LABEL: @ctz3_with_i8gep(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[X:%.*]], 0
+; CHECK-NEXT:    br i1 [[CMP]], label [[RETURN:%.*]], label [[IF_END:%.*]]
+; CHECK:       if.end:
+; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.cttz.i32(i32 [[X]], i1 true)
+; CHECK-NEXT:    br label [[RETURN]]
+; CHECK:       return:
+; CHECK-NEXT:    [[RETVAL_0:%.*]] = phi i32 [ [[TMP2]], [[IF_END]] ], [ 32, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    ret i32 [[RETVAL_0]]
+;
+entry:
+  %cmp = icmp eq i32 %x, 0
+  br i1 %cmp, label %return, label %if.end
+
+if.end:                                           ; preds = %entry
+  %sub = sub i32 0, %x
+  %and = and i32 %x, %sub
+  %mul = mul i32 %and, 81224991
+  %0 = lshr i32 %mul, 25
+  %1 = and i32 %0, 124
+  %arrayidx.idx = zext nneg i32 %1 to i64
+  %arrayidx = getelementptr inbounds nuw i8, ptr @ctz3.table, i64 %arrayidx.idx
+  %2 = load i32, ptr %arrayidx, align 4
+  br label %return
+
+return:                                           ; preds = %if.end, %entry
+  %retval.0 = phi i32 [ %2, %if.end ], [ 32, %entry ]
+  ret i32 %retval.0
+}
+
+
 @table = internal unnamed_addr constant [64 x i32] [i32 0, i32 1, i32 12, i32 2, i32 13, i32 22, i32 17, i32 3, i32 14, i32 33, i32 23, i32 36, i32 18, i32 58, i32 28, i32 4, i32 62, i32 15, i32 34, i32 26, i32 24, i32 48, i32 50, i32 37, i32 19, i32 55, i32 59, i32 52, i32 29, i32 44, i32 39, i32 5, i32 63, i32 11, i32 21, i32 16, i32 32, i32 35, i32 57, i32 27, i32 61, i32 25, i32 47, i32 49, i32 54, i32 51, i32 43, i32 38, i32 10, i32 20, i32 31, i32 56, i32 60, i32 46, i32 53, i32 42, i32 9, i32 30, i32 45, i32 41, i32 8, i32 40, i32 7, i32 6], align 4
 
 define i32 @ctz4(i64 %b) {
@@ -276,3 +309,192 @@ entry:
   %0 = load i32, ptr %arrayidx, align 4
   ret i32 %0
 }
+
+;; This has a wrong table size but is otherwise fine.
+@ctz9.table = internal unnamed_addr constant [128 x i8] c"\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09", align 1
+define i32 @ctz9(i32 %x) {
+; CHECK-LABEL: @ctz9(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[X]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i32 0, i32 [[TMP0]]
+; CHECK-NEXT:    [[TMP3:%.*]] = trunc i32 [[TMP2]] to i8
+; CHECK-NEXT:    [[CONV:%.*]] = zext i8 [[TMP3]] to i32
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
+entry:
+  %sub = sub i32 0, %x
+  %and = and i32 %sub, %x
+  %mul = mul i32 %and, 125613361
+  %shr = lshr i32 %mul, 27
+  %idxprom = zext i32 %shr to i64
+  %arrayidx = getelementptr inbounds [128 x i8], ptr @ctz9.table, i64 0, i64 %idxprom
+  %0 = load i8, ptr %arrayidx, align 1
+  %conv = zext i8 %0 to i32
+  ret i32 %conv
+}
+
+define i32 @ctz1_with_i8_gep(i32 %x) {
+; CHECK-LABEL: @ctz1_with_i8_gep(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[X]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i32 0, i32 [[TMP0]]
+; CHECK-NEXT:    [[TMP3:%.*]] = trunc i32 [[TMP2]] to i8
+; CHECK-NEXT:    [[CONV:%.*]] = zext i8 [[TMP3]] to i32
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
+entry:
+  %sub = sub i32 0, %x
+  %and = and i32 %sub, %x
+  %mul = mul i32 %and, 125613361
+  %shr = lshr i32 %mul, 27
+  %idxprom = zext i32 %shr to i64
+  %arrayidx = getelementptr inbounds i8, ptr @ctz7.table, i64 %idxprom
+  %0 = load i8, ptr %arrayidx, align 1
+  %conv = zext i8 %0 to i32
+  ret i32 %conv
+}
+
+; This is the same a ctz2 (i16 table) with an i8 gep making the indices invalid
+define i32 @ctz2_with_i8_gep(i32 %x) {
+; CHECK-LABEL: @ctz2_with_i8_gep(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SUB:%.*]] = sub i32 0, [[X:%.*]]
+; CHECK-NEXT:    [[AND:%.*]] = and i32 [[SUB]], [[X]]
+; CHECK-NEXT:    [[MUL:%.*]] = mul i32 [[AND]], 72416175
+; CHECK-NEXT:    [[SHR:%.*]] = lshr i32 [[MUL]], 26
+; CHECK-NEXT:    [[IDXPROM:%.*]] = zext i32 [[SHR]] to i64
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [64 x i8], ptr @ctz2.table, i64 0, i64 [[IDXPROM]]
+; CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr [[ARRAYIDX]], align 1
+; CHECK-NEXT:    [[CONV:%.*]] = sext i16 [[TMP0]] to i32
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
+entry:
+  %sub = sub i32 0, %x
+  %and = and i32 %sub, %x
+  %mul = mul i32 %and, 72416175
+  %shr = lshr i32 %mul, 26
+  %idxprom = zext i32 %shr to i64
+  %arrayidx = getelementptr inbounds [64 x i8], ptr @ctz2.table, i64 0, i64 %idxprom
+  %0 = load i16, ptr %arrayidx, align 1
+  %conv = sext i16 %0 to i32
+  ret i32 %conv
+}
+
+; This is the same a ctz2_with_i8_gep but with the gep index multiplied by 2.
+define i32 @ctz2_with_i8_gep_fixed(i32 %x) {
+; CHECK-LABEL: @ctz2_with_i8_gep_fixed(
+; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
+; CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
+; CHECK-NEXT:    [[CONV:%.*]] = sext i16 [[TMP2]] to i32
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
+  %sub = sub i32 0, %x
+  %and = and i32 %x, %sub
+  %mul = mul i32 %and, 72416175
+  %shr = lshr i32 %mul, 25
+  %shr2 = and i32 %shr, 126
+  %1 = zext nneg i32 %shr2 to i64
+  %arrayidx = getelementptr inbounds nuw i8, ptr @ctz2.table, i64 %1
+  %2 = load i16, ptr %arrayidx, align 2
+  %conv = sext i16 %2 to i32
+  ret i32 %conv
+}
+
+; This is a i16 input with the debruijn table stored in a single i128.
+@tablei128 = internal unnamed_addr constant i128 16018378897745984667142067713738932480, align 16
+define i32 @cttz_i16_via_i128(i16 noundef %x) {
+; CHECK-LABEL: @cttz_i16_via_i128(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true)
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i16 [[X]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP3]], i16 0, i16 [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc i16 [[TMP2]] to i8
+; CHECK-NEXT:    [[CONV6:%.*]] = zext i8 [[TMP1]] to i32
+; CHECK-NEXT:    ret i32 [[CONV6]]
+;
+entry:
+  %sub = sub i16 0, %x
+  %and = and i16 %x, %sub
+  %mul = mul i16 %and, 2479
+  %0 = lshr i16 %mul, 12
+  %idxprom = zext nneg i16 %0 to i64
+  %arrayidx = getelementptr inbounds nuw i8, ptr @tablei128, i64 %idxprom
+  %1 = load i8, ptr %arrayidx, align 1
+  %conv6 = zext i8 %1 to i32
+  ret i32 %conv6
+}
+
+; Same as above but the table is a little off
+@tablei128b = internal unnamed_addr constant i128 16018378897745984667142068813250560256, align 16
+define i32 @cttz_i16_via_i128_incorrecttable(i16 noundef %x) {
+; CHECK-LABEL: @cttz_i16_via_i128_incorrecttable(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SUB:%.*]] = sub i16 0, [[X:%.*]]
+; CHECK-NEXT:    [[AND:%.*]] = and i16 [[X]], [[SUB]]
+; CHECK-NEXT:    [[MUL:%.*]] = mul i16 [[AND]], 2479
+; CHECK-NEXT:    [[TMP0:%.*]] = lshr i16 [[MUL]], 12
+; CHECK-NEXT:    [[IDXPROM:%.*]] = zext nneg i16 [[TMP0]] to i64
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds nuw i8, ptr @tablei128b, i64 [[IDXPROM]]
+; CHECK-NEXT:    [[TMP3:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
+; CHECK-NEXT:    [[CONV6:%.*]] = zext i8 [[TMP3]] to i32
+; CHECK-NEXT:    ret i32 [[CONV6]]
+;
+entry:
+  %sub = sub i16 0, %x
+  %and = and i16 %x, %sub
+  %mul = mul i16 %and, 2479
+  %0 = lshr i16 %mul, 12
+  %idxprom = zext nneg i16 %0 to i64
+  %arrayidx = getelementptr inbounds nuw i8, ptr @tablei128b, i64 %idxprom
+  %1 = load i8, ptr %arrayidx, align 1
+  %conv6 = zext i8 %1 to i32
+  ret i32 %conv6
+}
+
+; Same as ctz1 but the table and load is very large
+@ctz7i128.table = internal unnamed_addr constant [32 x i128] [i128 0, i128 1, i128 28, i128 2, i128 29, i128 14, i128 24, i128 3, i128 30, i128 22, i128 20, i128 15, i128 25, i128 17, i128 4, i128 8, i128 31, i128 27, i128 13, i128 23, i128 21, i128 19, i128 16, i128 7, i128 26, i128 12, i128 18, i128 6, i128 11, i128 5, i128 10, i128 9], align 16
+define i128 @ctz1_i128(i32 %x) {
+; CHECK-LABEL: @ctz1_i128(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[X]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i32 0, i32 [[TMP0]]
+; CHECK-NEXT:    [[TMP3:%.*]] = zext i32 [[TMP2]] to i128
+; CHECK-NEXT:    ret i128 [[TMP3]]
+;
+entry:
+  %sub = sub i32 0, %x
+  %and = and i32 %sub, %x
+  %mul = mul i32 %and, 125613361
+  %shr = lshr i32 %mul, 27
+  %idxprom = zext i32 %shr to i64
+  %arrayidx = getelementptr inbounds [32 x i128], ptr @ctz7i128.table, i64 0, i64 %idxprom
+  %l = load i128, ptr %arrayidx, align 1
+  ret i128 %l
+}
+
+; This is roughly the same as ctz1 but using i128.
+@table.i128 = internal unnamed_addr constant [128 x i8] c"\00\01e\02tf<\03|ug^R=!\04}yvWoh_5ZSE>0\22\14\05~rzPwmX.pkiI`K6\1Ab[TBMF?'81*#\1C\15\0E\06\7Fds;{]Q xVn4YD/\13qOl-jHJ\19aAL&7)\1B\0Dc:\\\1FU3C\12N,G\18@%(\0C9\1E2\11+\17$\0B\1D\10\16\0A\0F\09\08\07", align 1
+define i32 @src(i128 noundef %x) {
+; CHECK-LABEL: @src(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP3:%.*]] = call i128 @llvm.cttz.i128(i128 [[X:%.*]], i1 true)
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i128 [[X]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i128 0, i128 [[TMP3]]
+; CHECK-NEXT:    [[TMP0:%.*]] = trunc i128 [[TMP2]] to i8
+; CHECK-NEXT:    [[CONV:%.*]] = zext i8 [[TMP0]] to i32
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
+entry:
+  %sub = sub i128 0, %x
+  %and = and i128 %x, %sub
+  %mul = mul i128 %and, 2647824804797170443043024478319300753
+  %shr = lshr i128 %mul, 121
+  %idxprom = trunc i128 %shr to i64
+  %arrayidx = getelementptr inbounds nuw i8, ptr @table.i128, i64 %idxprom
+  %0 = load i8, ptr %arrayidx, align 1
+  %conv = zext i8 %0 to i32
+  ret i32 %conv
+}
diff --git a/llvm/test/Transforms/AggressiveInstCombine/negative-lower-table-based-cttz.ll b/llvm/test/Transforms/AggressiveInstCombine/negative-lower-table-based-cttz.ll
index 714acd7..90836db8 100644
--- a/llvm/test/Transforms/AggressiveInstCombine/negative-lower-table-based-cttz.ll
+++ b/llvm/test/Transforms/AggressiveInstCombine/negative-lower-table-based-cttz.ll
@@ -66,7 +66,7 @@ entry:
 
 ;; This is a negative test with a wrong table size and constants.
 
-@ctz3.table = internal unnamed_addr constant [128 x i8] c"\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09", align 1
+@ctz3.table = internal unnamed_addr constant [128 x i8] c"\01\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09", align 1
 
 define i32 @ctz5(i32 %x) {
 entry:
diff --git a/llvm/test/Transforms/LoopUnroll/RISCV/vector.ll b/llvm/test/Transforms/LoopUnroll/RISCV/vector.ll
new file mode 100644
index 0000000..811d055
--- /dev/null
+++ b/llvm/test/Transforms/LoopUnroll/RISCV/vector.ll
@@ -0,0 +1,603 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -p loop-unroll -mtriple riscv64 -mattr=+v,+f -S %s | FileCheck %s --check-prefixes=COMMON,CHECK
+; RUN: opt -p loop-unroll -mtriple=riscv64 -mcpu=sifive-s76 -S %s | FileCheck %s --check-prefixes=COMMON,SIFIVE
+
+define void @reverse(ptr %dst, ptr %src, i64 %len) {
+; CHECK-LABEL: define void @reverse(
+; CHECK-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], i64 [[LEN:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  [[ENTRY:.*]]:
+; CHECK-NEXT:    br label %[[FOR_BODY:.*]]
+; CHECK:       [[FOR_BODY]]:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[IV_NEXT:%.*]], %[[FOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = sub nsw i64 [[LEN]], [[IV]]
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
+; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store <4 x float> [[TMP1]], ptr [[ARRAYIDX2]], align 16
+; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[LEN]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label %[[EXIT:.*]], label %[[FOR_BODY]]
+; CHECK:       [[EXIT]]:
+; CHECK-NEXT:    ret void
+;
+; SIFIVE-LABEL: define void @reverse(
+; SIFIVE-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], i64 [[LEN:%.*]]) #[[ATTR0:[0-9]+]] {
+; SIFIVE-NEXT:  [[ENTRY:.*]]:
+; SIFIVE-NEXT:    [[TMP2:%.*]] = add i64 [[LEN]], -1
+; SIFIVE-NEXT:    [[XTRAITER:%.*]] = and i64 [[LEN]], 7
+; SIFIVE-NEXT:    [[TMP3:%.*]] = icmp ult i64 [[TMP2]], 7
+; SIFIVE-NEXT:    br i1 [[TMP3]], label %[[EXIT_UNR_LCSSA:.*]], label %[[ENTRY_NEW:.*]]
+; SIFIVE:       [[ENTRY_NEW]]:
+; SIFIVE-NEXT:    [[UNROLL_ITER:%.*]] = sub i64 [[LEN]], [[XTRAITER]]
+; SIFIVE-NEXT:    br label %[[FOR_BODY:.*]]
+; SIFIVE:       [[FOR_BODY]]:
+; SIFIVE-NEXT:    [[IV:%.*]] = phi i64 [ 0, %[[ENTRY_NEW]] ], [ [[IV_NEXT_7:%.*]], %[[FOR_BODY]] ]
+; SIFIVE-NEXT:    [[NITER:%.*]] = phi i64 [ 0, %[[ENTRY_NEW]] ], [ [[NITER_NEXT_7:%.*]], %[[FOR_BODY]] ]
+; SIFIVE-NEXT:    [[TMP0:%.*]] = sub nsw i64 [[LEN]], [[IV]]
+; SIFIVE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP0]]
+; SIFIVE-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP1]], ptr [[ARRAYIDX2]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT:%.*]] = add nuw nsw i64 [[IV]], 1
+; SIFIVE-NEXT:    [[TMP4:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT]]
+; SIFIVE-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP4]]
+; SIFIVE-NEXT:    [[TMP5:%.*]] = load <4 x float>, ptr [[ARRAYIDX_1]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP5]], ptr [[ARRAYIDX2_1]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_1:%.*]] = add nuw nsw i64 [[IV]], 2
+; SIFIVE-NEXT:    [[TMP6:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_1]]
+; SIFIVE-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP6]]
+; SIFIVE-NEXT:    [[TMP7:%.*]] = load <4 x float>, ptr [[ARRAYIDX_2]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_1]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP7]], ptr [[ARRAYIDX2_2]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_2:%.*]] = add nuw nsw i64 [[IV]], 3
+; SIFIVE-NEXT:    [[TMP8:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_2]]
+; SIFIVE-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP8]]
+; SIFIVE-NEXT:    [[TMP9:%.*]] = load <4 x float>, ptr [[ARRAYIDX_3]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_2]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP9]], ptr [[ARRAYIDX2_3]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_3:%.*]] = add nuw nsw i64 [[IV]], 4
+; SIFIVE-NEXT:    [[TMP10:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_3]]
+; SIFIVE-NEXT:    [[ARRAYIDX_4:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP10]]
+; SIFIVE-NEXT:    [[TMP11:%.*]] = load <4 x float>, ptr [[ARRAYIDX_4]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_4:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_3]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP11]], ptr [[ARRAYIDX2_4]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_4:%.*]] = add nuw nsw i64 [[IV]], 5
+; SIFIVE-NEXT:    [[TMP12:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_4]]
+; SIFIVE-NEXT:    [[ARRAYIDX_5:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP12]]
+; SIFIVE-NEXT:    [[TMP13:%.*]] = load <4 x float>, ptr [[ARRAYIDX_5]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_5:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_4]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP13]], ptr [[ARRAYIDX2_5]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_5:%.*]] = add nuw nsw i64 [[IV]], 6
+; SIFIVE-NEXT:    [[TMP14:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_5]]
+; SIFIVE-NEXT:    [[ARRAYIDX_6:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP14]]
+; SIFIVE-NEXT:    [[TMP15:%.*]] = load <4 x float>, ptr [[ARRAYIDX_6]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_6:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_5]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP15]], ptr [[ARRAYIDX2_6]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_6:%.*]] = add nuw nsw i64 [[IV]], 7
+; SIFIVE-NEXT:    [[TMP16:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_6]]
+; SIFIVE-NEXT:    [[ARRAYIDX_7:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP16]]
+; SIFIVE-NEXT:    [[TMP17:%.*]] = load <4 x float>, ptr [[ARRAYIDX_7]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_7:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_6]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP17]], ptr [[ARRAYIDX2_7]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_7]] = add nuw nsw i64 [[IV]], 8
+; SIFIVE-NEXT:    [[NITER_NEXT_7]] = add i64 [[NITER]], 8
+; SIFIVE-NEXT:    [[NITER_NCMP_7:%.*]] = icmp eq i64 [[NITER_NEXT_7]], [[UNROLL_ITER]]
+; SIFIVE-NEXT:    br i1 [[NITER_NCMP_7]], label %[[EXIT_UNR_LCSSA_LOOPEXIT:.*]], label %[[FOR_BODY]]
+; SIFIVE:       [[EXIT_UNR_LCSSA_LOOPEXIT]]:
+; SIFIVE-NEXT:    [[IV_UNR_PH:%.*]] = phi i64 [ [[IV_NEXT_7]], %[[FOR_BODY]] ]
+; SIFIVE-NEXT:    br label %[[EXIT_UNR_LCSSA]]
+; SIFIVE:       [[EXIT_UNR_LCSSA]]:
+; SIFIVE-NEXT:    [[IV_UNR:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[IV_UNR_PH]], %[[EXIT_UNR_LCSSA_LOOPEXIT]] ]
+; SIFIVE-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; SIFIVE-NEXT:    br i1 [[LCMP_MOD]], label %[[FOR_BODY_EPIL_PREHEADER:.*]], label %[[EXIT:.*]]
+; SIFIVE:       [[FOR_BODY_EPIL_PREHEADER]]:
+; SIFIVE-NEXT:    br label %[[FOR_BODY_EPIL:.*]]
+; SIFIVE:       [[FOR_BODY_EPIL]]:
+; SIFIVE-NEXT:    [[TMP18:%.*]] = sub nsw i64 [[LEN]], [[IV_UNR]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP18]]
+; SIFIVE-NEXT:    [[TMP19:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_UNR]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP19]], ptr [[ARRAYIDX2_EPIL]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_EPIL:%.*]] = add nuw nsw i64 [[IV_UNR]], 1
+; SIFIVE-NEXT:    [[EPIL_ITER_CMP:%.*]] = icmp ne i64 1, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[EPIL_ITER_CMP]], label %[[FOR_BODY_EPIL_1:.*]], label %[[EXIT_EPILOG_LCSSA:.*]]
+; SIFIVE:       [[FOR_BODY_EPIL_1]]:
+; SIFIVE-NEXT:    [[TMP20:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_EPIL]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL_1:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP20]]
+; SIFIVE-NEXT:    [[TMP21:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL_1]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL_1:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_EPIL]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP21]], ptr [[ARRAYIDX2_EPIL_1]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_EPIL_1:%.*]] = add nuw nsw i64 [[IV_UNR]], 2
+; SIFIVE-NEXT:    [[EPIL_ITER_CMP_1:%.*]] = icmp ne i64 2, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[EPIL_ITER_CMP_1]], label %[[FOR_BODY_EPIL_2:.*]], label %[[EXIT_EPILOG_LCSSA]]
+; SIFIVE:       [[FOR_BODY_EPIL_2]]:
+; SIFIVE-NEXT:    [[TMP22:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_EPIL_1]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL_2:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP22]]
+; SIFIVE-NEXT:    [[TMP23:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL_2]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL_2:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_EPIL_1]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP23]], ptr [[ARRAYIDX2_EPIL_2]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_EPIL_2:%.*]] = add nuw nsw i64 [[IV_UNR]], 3
+; SIFIVE-NEXT:    [[EPIL_ITER_CMP_2:%.*]] = icmp ne i64 3, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[EPIL_ITER_CMP_2]], label %[[FOR_BODY_EPIL_3:.*]], label %[[EXIT_EPILOG_LCSSA]]
+; SIFIVE:       [[FOR_BODY_EPIL_3]]:
+; SIFIVE-NEXT:    [[TMP24:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_EPIL_2]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL_3:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP24]]
+; SIFIVE-NEXT:    [[TMP25:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL_3]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL_3:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_EPIL_2]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP25]], ptr [[ARRAYIDX2_EPIL_3]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_EPIL_3:%.*]] = add nuw nsw i64 [[IV_UNR]], 4
+; SIFIVE-NEXT:    [[EPIL_ITER_CMP_3:%.*]] = icmp ne i64 4, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[EPIL_ITER_CMP_3]], label %[[FOR_BODY_EPIL_4:.*]], label %[[EXIT_EPILOG_LCSSA]]
+; SIFIVE:       [[FOR_BODY_EPIL_4]]:
+; SIFIVE-NEXT:    [[TMP26:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_EPIL_3]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL_4:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP26]]
+; SIFIVE-NEXT:    [[TMP27:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL_4]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL_4:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_EPIL_3]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP27]], ptr [[ARRAYIDX2_EPIL_4]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_EPIL_4:%.*]] = add nuw nsw i64 [[IV_UNR]], 5
+; SIFIVE-NEXT:    [[EPIL_ITER_CMP_4:%.*]] = icmp ne i64 5, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[EPIL_ITER_CMP_4]], label %[[FOR_BODY_EPIL_5:.*]], label %[[EXIT_EPILOG_LCSSA]]
+; SIFIVE:       [[FOR_BODY_EPIL_5]]:
+; SIFIVE-NEXT:    [[TMP28:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_EPIL_4]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL_5:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP28]]
+; SIFIVE-NEXT:    [[TMP29:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL_5]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL_5:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_EPIL_4]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP29]], ptr [[ARRAYIDX2_EPIL_5]], align 16
+; SIFIVE-NEXT:    [[IV_NEXT_EPIL_5:%.*]] = add nuw nsw i64 [[IV_UNR]], 6
+; SIFIVE-NEXT:    [[EPIL_ITER_CMP_5:%.*]] = icmp ne i64 6, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[EPIL_ITER_CMP_5]], label %[[FOR_BODY_EPIL_6:.*]], label %[[EXIT_EPILOG_LCSSA]]
+; SIFIVE:       [[FOR_BODY_EPIL_6]]:
+; SIFIVE-NEXT:    [[TMP30:%.*]] = sub nsw i64 [[LEN]], [[IV_NEXT_EPIL_5]]
+; SIFIVE-NEXT:    [[ARRAYIDX_EPIL_6:%.*]] = getelementptr inbounds <4 x float>, ptr [[SRC]], i64 [[TMP30]]
+; SIFIVE-NEXT:    [[TMP31:%.*]] = load <4 x float>, ptr [[ARRAYIDX_EPIL_6]], align 16
+; SIFIVE-NEXT:    [[ARRAYIDX2_EPIL_6:%.*]] = getelementptr inbounds nuw <4 x float>, ptr [[DST]], i64 [[IV_NEXT_EPIL_5]]
+; SIFIVE-NEXT:    store <4 x float> [[TMP31]], ptr [[ARRAYIDX2_EPIL_6]], align 16
+; SIFIVE-NEXT:    br label %[[EXIT_EPILOG_LCSSA]]
+; SIFIVE:       [[EXIT_EPILOG_LCSSA]]:
+; SIFIVE-NEXT:    br label %[[EXIT]]
+; SIFIVE:       [[EXIT]]:
+; SIFIVE-NEXT:    ret void
+;
+entry:                               ; preds = %entry
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %for.body
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %1 = sub nsw i64 %len, %iv
+  %arrayidx = getelementptr inbounds <4 x float>, ptr %src, i64 %1
+  %2 = load <4 x float>, ptr %arrayidx, align 16
+  %arrayidx2 = getelementptr inbounds nuw <4 x float>, ptr %dst, i64 %iv
+  store <4 x float> %2, ptr %arrayidx2, align 16
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond.not = icmp eq i64 %iv.next, %len
+  br i1 %exitcond.not, label %exit, label %for.body
+
+exit:                                 ; preds = %for.body, %entry
+  ret void
+}
+
+
+define void @saxpy_tripcount8_full_unroll(ptr %dst, ptr %src, float %a) {
+; COMMON-LABEL: define void @saxpy_tripcount8_full_unroll(
+; COMMON-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], float [[A:%.*]]) #[[ATTR0:[0-9]+]] {
+; COMMON-NEXT:  [[ENTRY:.*:]]
+; COMMON-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[A]], i64 0
+; COMMON-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
+; COMMON-NEXT:    br label %[[VECTOR_BODY:.*]]
+; COMMON:       [[VECTOR_BODY]]:
+; COMMON-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[SRC]], align 4
+; COMMON-NEXT:    [[WIDE_LOAD12:%.*]] = load <4 x float>, ptr [[DST]], align 4
+; COMMON-NEXT:    [[TMP0:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD]], <4 x float> [[WIDE_LOAD12]])
+; COMMON-NEXT:    store <4 x float> [[TMP0]], ptr [[DST]], align 4
+; COMMON-NEXT:    [[TMP1:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 4
+; COMMON-NEXT:    [[WIDE_LOAD_1:%.*]] = load <4 x float>, ptr [[TMP1]], align 4
+; COMMON-NEXT:    [[TMP2:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 4
+; COMMON-NEXT:    [[WIDE_LOAD12_1:%.*]] = load <4 x float>, ptr [[TMP2]], align 4
+; COMMON-NEXT:    [[TMP3:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD_1]], <4 x float> [[WIDE_LOAD12_1]])
+; COMMON-NEXT:    store <4 x float> [[TMP3]], ptr [[TMP2]], align 4
+; COMMON-NEXT:    ret void
+;
+entry:
+  %broadcast.splatinsert = insertelement <4 x float> poison, float %a, i64 0
+  %broadcast.splat = shufflevector <4 x float> %broadcast.splatinsert, <4 x float> poison, <4 x i32> zeroinitializer
+  br label %vector.body
+
+vector.body:                                      ; preds = %vector.body, %entry
+  %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
+  %0 = getelementptr inbounds nuw float, ptr %src, i64 %index
+  %wide.load = load <4 x float>, ptr %0, align 4
+  %1 = getelementptr inbounds nuw float, ptr %dst, i64 %index
+  %wide.load12 = load <4 x float>, ptr %1, align 4
+  %2 = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> %broadcast.splat, <4 x float> %wide.load, <4 x float> %wide.load12)
+  store <4 x float> %2, ptr %1, align 4
+  %index.next = add nuw i64 %index, 4
+  %3 = icmp eq i64 %index.next, 8
+  br i1 %3, label %exit, label %vector.body
+
+exit:                                 ; preds = %vector.body
+  ret void
+}
+
+
+define void @saxpy_tripcount1K_av0(ptr %dst, ptr %src, float %a) {
+; CHECK-LABEL: define void @saxpy_tripcount1K_av0(
+; CHECK-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], float [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  [[ENTRY:.*]]:
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[A]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
+; CHECK:       [[VECTOR_BODY]]:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 [[INDEX]]
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[TMP0]], align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 [[INDEX]]
+; CHECK-NEXT:    [[WIDE_LOAD12:%.*]] = load <4 x float>, ptr [[TMP1]], align 4
+; CHECK-NEXT:    [[TMP2:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD]], <4 x float> [[WIDE_LOAD12]])
+; CHECK-NEXT:    store <4 x float> [[TMP2]], ptr [[TMP1]], align 4
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
+; CHECK-NEXT:    br i1 [[TMP3]], label %[[EXIT:.*]], label %[[VECTOR_BODY]]
+; CHECK:       [[EXIT]]:
+; CHECK-NEXT:    ret void
+;
+; SIFIVE-LABEL: define void @saxpy_tripcount1K_av0(
+; SIFIVE-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], float [[A:%.*]]) #[[ATTR0]] {
+; SIFIVE-NEXT:  [[ENTRY:.*]]:
+; SIFIVE-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[A]], i64 0
+; SIFIVE-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
+; SIFIVE-NEXT:    br label %[[VECTOR_BODY:.*]]
+; SIFIVE:       [[VECTOR_BODY]]:
+; SIFIVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; SIFIVE-NEXT:    [[TMP0:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 [[INDEX]]
+; SIFIVE-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[TMP0]], align 4
+; SIFIVE-NEXT:    [[TMP1:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 [[INDEX]]
+; SIFIVE-NEXT:    [[WIDE_LOAD12:%.*]] = load <4 x float>, ptr [[TMP1]], align 4
+; SIFIVE-NEXT:    [[TMP2:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD]], <4 x float> [[WIDE_LOAD12]])
+; SIFIVE-NEXT:    store <4 x float> [[TMP2]], ptr [[TMP1]], align 4
+; SIFIVE-NEXT:    [[INDEX_NEXT1:%.*]] = add nuw nsw i64 [[INDEX]], 4
+; SIFIVE-NEXT:    [[TMP12:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 [[INDEX_NEXT1]]
+; SIFIVE-NEXT:    [[WIDE_LOAD_1:%.*]] = load <4 x float>, ptr [[TMP12]], align 4
+; SIFIVE-NEXT:    [[TMP4:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 [[INDEX_NEXT1]]
+; SIFIVE-NEXT:    [[WIDE_LOAD12_1:%.*]] = load <4 x float>, ptr [[TMP4]], align 4
+; SIFIVE-NEXT:    [[TMP5:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD_1]], <4 x float> [[WIDE_LOAD12_1]])
+; SIFIVE-NEXT:    store <4 x float> [[TMP5]], ptr [[TMP4]], align 4
+; SIFIVE-NEXT:    [[INDEX_NEXT_1:%.*]] = add nuw nsw i64 [[INDEX]], 8
+; SIFIVE-NEXT:    [[TMP6:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 [[INDEX_NEXT_1]]
+; SIFIVE-NEXT:    [[WIDE_LOAD_2:%.*]] = load <4 x float>, ptr [[TMP6]], align 4
+; SIFIVE-NEXT:    [[TMP7:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 [[INDEX_NEXT_1]]
+; SIFIVE-NEXT:    [[WIDE_LOAD12_2:%.*]] = load <4 x float>, ptr [[TMP7]], align 4
+; SIFIVE-NEXT:    [[TMP8:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD_2]], <4 x float> [[WIDE_LOAD12_2]])
+; SIFIVE-NEXT:    store <4 x float> [[TMP8]], ptr [[TMP7]], align 4
+; SIFIVE-NEXT:    [[INDEX_NEXT_2:%.*]] = add nuw nsw i64 [[INDEX]], 12
+; SIFIVE-NEXT:    [[TMP9:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 [[INDEX_NEXT_2]]
+; SIFIVE-NEXT:    [[WIDE_LOAD_3:%.*]] = load <4 x float>, ptr [[TMP9]], align 4
+; SIFIVE-NEXT:    [[TMP10:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 [[INDEX_NEXT_2]]
+; SIFIVE-NEXT:    [[WIDE_LOAD12_3:%.*]] = load <4 x float>, ptr [[TMP10]], align 4
+; SIFIVE-NEXT:    [[TMP11:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD_3]], <4 x float> [[WIDE_LOAD12_3]])
+; SIFIVE-NEXT:    store <4 x float> [[TMP11]], ptr [[TMP10]], align 4
+; SIFIVE-NEXT:    [[INDEX_NEXT]] = add nuw nsw i64 [[INDEX]], 16
+; SIFIVE-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
+; SIFIVE-NEXT:    br i1 [[TMP3]], label %[[EXIT:.*]], label %[[VECTOR_BODY]]
+; SIFIVE:       [[EXIT]]:
+; SIFIVE-NEXT:    ret void
+;
+entry:
+  %broadcast.splatinsert = insertelement <4 x float> poison, float %a, i64 0
+  %broadcast.splat = shufflevector <4 x float> %broadcast.splatinsert, <4 x float> poison, <4 x i32> zeroinitializer
+  br label %vector.body
+
+vector.body:                                      ; preds = %vector.body, %entry
+  %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
+  %0 = getelementptr inbounds nuw float, ptr %src, i64 %index
+  %wide.load = load <4 x float>, ptr %0, align 4
+  %1 = getelementptr inbounds nuw float, ptr %dst, i64 %index
+  %wide.load12 = load <4 x float>, ptr %1, align 4
+  %2 = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> %broadcast.splat, <4 x float> %wide.load, <4 x float> %wide.load12)
+  store <4 x float> %2, ptr %1, align 4
+  %index.next = add nuw i64 %index, 4
+  %3 = icmp eq i64 %index.next, 1024
+  br i1 %3, label %exit, label %vector.body
+
+exit:                                             ; preds = %vector.body
+  ret void
+}
+
+
+define void @saxpy_tripcount1K_av1(ptr %dst, ptr %src, float %a) {
+; COMMON-LABEL: define void @saxpy_tripcount1K_av1(
+; COMMON-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], float [[A:%.*]]) #[[ATTR0]] {
+; COMMON-NEXT:  [[ENTRY:.*]]:
+; COMMON-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[A]], i64 0
+; COMMON-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
+; COMMON-NEXT:    br label %[[VECTOR_BODY:.*]]
+; COMMON:       [[VECTOR_BODY]]:
+; COMMON-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; COMMON-NEXT:    [[TMP0:%.*]] = getelementptr inbounds nuw float, ptr [[SRC]], i64 [[INDEX]]
+; COMMON-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[TMP0]], align 4
+; COMMON-NEXT:    [[TMP1:%.*]] = getelementptr inbounds nuw float, ptr [[DST]], i64 [[INDEX]]
+; COMMON-NEXT:    [[WIDE_LOAD12:%.*]] = load <4 x float>, ptr [[TMP1]], align 4
+; COMMON-NEXT:    [[TMP2:%.*]] = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[WIDE_LOAD]], <4 x float> [[WIDE_LOAD12]])
+; COMMON-NEXT:    store <4 x float> [[TMP2]], ptr [[TMP1]], align 4
+; COMMON-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; COMMON-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
+; COMMON-NEXT:    br i1 [[TMP3]], label %[[EXIT:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; COMMON:       [[EXIT]]:
+; COMMON-NEXT:    ret void
+;
+entry:
+  %broadcast.splatinsert = insertelement <4 x float> poison, float %a, i64 0
+  %broadcast.splat = shufflevector <4 x float> %broadcast.splatinsert, <4 x float> poison, <4 x i32> zeroinitializer
+  br label %vector.body
+
+vector.body:                                      ; preds = %vector.body, %entry
+  %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
+  %0 = getelementptr inbounds nuw float, ptr %src, i64 %index
+  %wide.load = load <4 x float>, ptr %0, align 4
+  %1 = getelementptr inbounds nuw float, ptr %dst, i64 %index
+  %wide.load12 = load <4 x float>, ptr %1, align 4
+  %2 = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> %broadcast.splat, <4 x float> %wide.load, <4 x float> %wide.load12)
+  store <4 x float> %2, ptr %1, align 4
+  %index.next = add nuw i64 %index, 4
+  %3 = icmp eq i64 %index.next, 1024
+  br i1 %3, label %exit, label %vector.body, !llvm.loop !0
+
+exit:                                 ; preds = %vector.body
+  ret void
+}
+!0 = !{!0, !1}
+!1 = !{!"llvm.loop.isvectorized", i32 1}
+
+; On SiFive we should runtime unroll the scalar epilogue loop, but not the
+; vector loop.
+define void @scalar_epilogue(ptr %p, i8 %splat.scalar, i64 %n) {
+; CHECK-LABEL: define void @scalar_epilogue(
+; CHECK-SAME: ptr [[P:%.*]], i8 [[SPLAT_SCALAR:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  [[ENTRY:.*]]:
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], 32
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_REMAINDER_PREHEADER:.*]], label %[[VECTOR_PH:.*]]
+; CHECK:       [[VECTOR_PH]]:
+; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[N]], -32
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i8> poison, i8 [[SPLAT_SCALAR]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i8> [[BROADCAST_SPLATINSERT]], <16 x i8> poison, <16 x i32> zeroinitializer
+; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
+; CHECK:       [[VECTOR_BODY]]:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[IV_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[GEP_P_IV:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[IV]]
+; CHECK-NEXT:    [[GEP_P_IV_16:%.*]] = getelementptr inbounds nuw i8, ptr [[GEP_P_IV]], i64 16
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[GEP_P_IV]], align 1
+; CHECK-NEXT:    [[WIDE_LOAD_2:%.*]] = load <16 x i8>, ptr [[GEP_P_IV_16]], align 1
+; CHECK-NEXT:    [[ADD_BROADCAST:%.*]] = add <16 x i8> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[ADD_BROADCAST_2:%.*]] = add <16 x i8> [[WIDE_LOAD_2]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    store <16 x i8> [[ADD_BROADCAST]], ptr [[GEP_P_IV]], align 1
+; CHECK-NEXT:    store <16 x i8> [[ADD_BROADCAST_2]], ptr [[GEP_P_IV_16]], align 1
+; CHECK-NEXT:    [[IV_NEXT]] = add nuw i64 [[IV]], 32
+; CHECK-NEXT:    [[EXIT_COND:%.*]] = icmp eq i64 [[IV_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[EXIT_COND]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP2:![0-9]+]]
+; CHECK:       [[MIDDLE_BLOCK]]:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_REMAINDER_PREHEADER]]
+; CHECK:       [[SCALAR_REMAINDER_PREHEADER]]:
+; CHECK-NEXT:    [[IV_SCALAR_LOOP_PH:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[N_VEC]], %[[MIDDLE_BLOCK]] ]
+; CHECK-NEXT:    br label %[[SCALAR_REMAINDER:.*]]
+; CHECK:       [[SCALAR_REMAINDER]]:
+; CHECK-NEXT:    [[IV_SCALAR_LOOP:%.*]] = phi i64 [ [[INC:%.*]], %[[SCALAR_REMAINDER]] ], [ [[IV_SCALAR_LOOP_PH]], %[[SCALAR_REMAINDER_PREHEADER]] ]
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[IV_SCALAR_LOOP]]
+; CHECK-NEXT:    [[SCALAR_LOAD:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
+; CHECK-NEXT:    [[ADD:%.*]] = add i8 [[SCALAR_LOAD]], [[SPLAT_SCALAR]]
+; CHECK-NEXT:    store i8 [[ADD]], ptr [[ARRAYIDX]], align 1
+; CHECK-NEXT:    [[INC]] = add nuw i64 [[IV_SCALAR_LOOP]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label %[[EXIT_LOOPEXIT:.*]], label %[[SCALAR_REMAINDER]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       [[EXIT_LOOPEXIT]]:
+; CHECK-NEXT:    br label %[[EXIT]]
+; CHECK:       [[EXIT]]:
+; CHECK-NEXT:    ret void
+;
+; SIFIVE-LABEL: define void @scalar_epilogue(
+; SIFIVE-SAME: ptr [[P:%.*]], i8 [[SPLAT_SCALAR:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; SIFIVE-NEXT:  [[ENTRY:.*]]:
+; SIFIVE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], 32
+; SIFIVE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_REMAINDER_PREHEADER:.*]], label %[[VECTOR_PH:.*]]
+; SIFIVE:       [[VECTOR_PH]]:
+; SIFIVE-NEXT:    [[N_VEC:%.*]] = and i64 [[N]], -32
+; SIFIVE-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i8> poison, i8 [[SPLAT_SCALAR]], i64 0
+; SIFIVE-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i8> [[BROADCAST_SPLATINSERT]], <16 x i8> poison, <16 x i32> zeroinitializer
+; SIFIVE-NEXT:    br label %[[VECTOR_BODY:.*]]
+; SIFIVE:       [[VECTOR_BODY]]:
+; SIFIVE-NEXT:    [[IV:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[IV_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; SIFIVE-NEXT:    [[GEP_P_IV:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[IV]]
+; SIFIVE-NEXT:    [[GEP_P_IV_16:%.*]] = getelementptr inbounds nuw i8, ptr [[GEP_P_IV]], i64 16
+; SIFIVE-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[GEP_P_IV]], align 1
+; SIFIVE-NEXT:    [[WIDE_LOAD_2:%.*]] = load <16 x i8>, ptr [[GEP_P_IV_16]], align 1
+; SIFIVE-NEXT:    [[ADD_BROADCAST:%.*]] = add <16 x i8> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
+; SIFIVE-NEXT:    [[ADD_BROADCAST_2:%.*]] = add <16 x i8> [[WIDE_LOAD_2]], [[BROADCAST_SPLAT]]
+; SIFIVE-NEXT:    store <16 x i8> [[ADD_BROADCAST]], ptr [[GEP_P_IV]], align 1
+; SIFIVE-NEXT:    store <16 x i8> [[ADD_BROADCAST_2]], ptr [[GEP_P_IV_16]], align 1
+; SIFIVE-NEXT:    [[IV_NEXT]] = add nuw i64 [[IV]], 32
+; SIFIVE-NEXT:    [[EXIT_COND:%.*]] = icmp eq i64 [[IV_NEXT]], [[N_VEC]]
+; SIFIVE-NEXT:    br i1 [[EXIT_COND]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP2:![0-9]+]]
+; SIFIVE:       [[MIDDLE_BLOCK]]:
+; SIFIVE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
+; SIFIVE-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_REMAINDER_PREHEADER]]
+; SIFIVE:       [[SCALAR_REMAINDER_PREHEADER]]:
+; SIFIVE-NEXT:    [[IV_SCALAR_LOOP:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[N_VEC]], %[[MIDDLE_BLOCK]] ]
+; SIFIVE-NEXT:    [[TMP0:%.*]] = sub i64 [[N]], [[IV_SCALAR_LOOP]]
+; SIFIVE-NEXT:    [[TMP1:%.*]] = add i64 [[N]], -1
+; SIFIVE-NEXT:    [[TMP2:%.*]] = sub i64 [[TMP1]], [[IV_SCALAR_LOOP]]
+; SIFIVE-NEXT:    [[XTRAITER:%.*]] = and i64 [[TMP0]], 7
+; SIFIVE-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; SIFIVE-NEXT:    br i1 [[LCMP_MOD]], label %[[SCALAR_REMAINDER_PROL_PREHEADER:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT:.*]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_PREHEADER]]:
+; SIFIVE-NEXT:    br label %[[SCALAR_REMAINDER_PROL:.*]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL]]:
+; SIFIVE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[IV_SCALAR_LOOP]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
+; SIFIVE-NEXT:    [[ADD:%.*]] = add i8 [[SCALAR_LOAD]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD]], ptr [[ARRAYIDX]], align 1
+; SIFIVE-NEXT:    [[INC:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 1
+; SIFIVE-NEXT:    [[PROL_ITER_CMP:%.*]] = icmp ne i64 1, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[PROL_ITER_CMP]], label %[[SCALAR_REMAINDER_PROL_1:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA:.*]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_1]]:
+; SIFIVE-NEXT:    [[ARRAYIDX_PROL_1:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_PROL_1:%.*]] = load i8, ptr [[ARRAYIDX_PROL_1]], align 1
+; SIFIVE-NEXT:    [[ADD_PROL_1:%.*]] = add i8 [[SCALAR_LOAD_PROL_1]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_PROL_1]], ptr [[ARRAYIDX_PROL_1]], align 1
+; SIFIVE-NEXT:    [[INC_PROL_1:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 2
+; SIFIVE-NEXT:    [[PROL_ITER_CMP_1:%.*]] = icmp ne i64 2, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[PROL_ITER_CMP_1]], label %[[SCALAR_REMAINDER_PROL_2:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_2]]:
+; SIFIVE-NEXT:    [[ARRAYIDX_PROL_2:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_PROL_1]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_PROL_2:%.*]] = load i8, ptr [[ARRAYIDX_PROL_2]], align 1
+; SIFIVE-NEXT:    [[ADD_PROL_2:%.*]] = add i8 [[SCALAR_LOAD_PROL_2]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_PROL_2]], ptr [[ARRAYIDX_PROL_2]], align 1
+; SIFIVE-NEXT:    [[INC_PROL_2:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 3
+; SIFIVE-NEXT:    [[PROL_ITER_CMP_2:%.*]] = icmp ne i64 3, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[PROL_ITER_CMP_2]], label %[[SCALAR_REMAINDER_PROL_3:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_3]]:
+; SIFIVE-NEXT:    [[ARRAYIDX_PROL_3:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_PROL_2]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_PROL_3:%.*]] = load i8, ptr [[ARRAYIDX_PROL_3]], align 1
+; SIFIVE-NEXT:    [[ADD_PROL_3:%.*]] = add i8 [[SCALAR_LOAD_PROL_3]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_PROL_3]], ptr [[ARRAYIDX_PROL_3]], align 1
+; SIFIVE-NEXT:    [[INC_PROL_3:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 4
+; SIFIVE-NEXT:    [[PROL_ITER_CMP_3:%.*]] = icmp ne i64 4, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[PROL_ITER_CMP_3]], label %[[SCALAR_REMAINDER_PROL_4:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_4]]:
+; SIFIVE-NEXT:    [[ARRAYIDX_PROL_4:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_PROL_3]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_PROL_4:%.*]] = load i8, ptr [[ARRAYIDX_PROL_4]], align 1
+; SIFIVE-NEXT:    [[ADD_PROL_4:%.*]] = add i8 [[SCALAR_LOAD_PROL_4]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_PROL_4]], ptr [[ARRAYIDX_PROL_4]], align 1
+; SIFIVE-NEXT:    [[INC_PROL_4:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 5
+; SIFIVE-NEXT:    [[PROL_ITER_CMP_4:%.*]] = icmp ne i64 5, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[PROL_ITER_CMP_4]], label %[[SCALAR_REMAINDER_PROL_5:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_5]]:
+; SIFIVE-NEXT:    [[ARRAYIDX_PROL_5:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_PROL_4]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_PROL_5:%.*]] = load i8, ptr [[ARRAYIDX_PROL_5]], align 1
+; SIFIVE-NEXT:    [[ADD_PROL_5:%.*]] = add i8 [[SCALAR_LOAD_PROL_5]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_PROL_5]], ptr [[ARRAYIDX_PROL_5]], align 1
+; SIFIVE-NEXT:    [[INC_PROL_5:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 6
+; SIFIVE-NEXT:    [[PROL_ITER_CMP_5:%.*]] = icmp ne i64 6, [[XTRAITER]]
+; SIFIVE-NEXT:    br i1 [[PROL_ITER_CMP_5]], label %[[SCALAR_REMAINDER_PROL_6:.*]], label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_6]]:
+; SIFIVE-NEXT:    [[ARRAYIDX_PROL_6:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_PROL_5]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_PROL_6:%.*]] = load i8, ptr [[ARRAYIDX_PROL_6]], align 1
+; SIFIVE-NEXT:    [[ADD_PROL_6:%.*]] = add i8 [[SCALAR_LOAD_PROL_6]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_PROL_6]], ptr [[ARRAYIDX_PROL_6]], align 1
+; SIFIVE-NEXT:    [[INC_PROL_6:%.*]] = add nuw i64 [[IV_SCALAR_LOOP]], 7
+; SIFIVE-NEXT:    br label %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]]:
+; SIFIVE-NEXT:    [[IV_SCALAR_LOOP_UNR_PH:%.*]] = phi i64 [ [[INC]], %[[SCALAR_REMAINDER_PROL]] ], [ [[INC_PROL_1]], %[[SCALAR_REMAINDER_PROL_1]] ], [ [[INC_PROL_2]], %[[SCALAR_REMAINDER_PROL_2]] ], [ [[INC_PROL_3]], %[[SCALAR_REMAINDER_PROL_3]] ], [ [[INC_PROL_4]], %[[SCALAR_REMAINDER_PROL_4]] ], [ [[INC_PROL_5]], %[[SCALAR_REMAINDER_PROL_5]] ], [ [[INC_PROL_6]], %[[SCALAR_REMAINDER_PROL_6]] ]
+; SIFIVE-NEXT:    br label %[[SCALAR_REMAINDER_PROL_LOOPEXIT]]
+; SIFIVE:       [[SCALAR_REMAINDER_PROL_LOOPEXIT]]:
+; SIFIVE-NEXT:    [[IV_SCALAR_LOOP_UNR:%.*]] = phi i64 [ [[IV_SCALAR_LOOP]], %[[SCALAR_REMAINDER_PREHEADER]] ], [ [[IV_SCALAR_LOOP_UNR_PH]], %[[SCALAR_REMAINDER_PROL_LOOPEXIT_UNR_LCSSA]] ]
+; SIFIVE-NEXT:    [[TMP3:%.*]] = icmp ult i64 [[TMP2]], 7
+; SIFIVE-NEXT:    br i1 [[TMP3]], label %[[EXIT_LOOPEXIT:.*]], label %[[SCALAR_REMAINDER_PREHEADER_NEW:.*]]
+; SIFIVE:       [[SCALAR_REMAINDER_PREHEADER_NEW]]:
+; SIFIVE-NEXT:    br label %[[SCALAR_REMAINDER:.*]]
+; SIFIVE:       [[SCALAR_REMAINDER]]:
+; SIFIVE-NEXT:    [[IV_SCALAR_LOOP1:%.*]] = phi i64 [ [[IV_SCALAR_LOOP_UNR]], %[[SCALAR_REMAINDER_PREHEADER_NEW]] ], [ [[INC_7:%.*]], %[[SCALAR_REMAINDER]] ]
+; SIFIVE-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[IV_SCALAR_LOOP1]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD1:%.*]] = load i8, ptr [[ARRAYIDX1]], align 1
+; SIFIVE-NEXT:    [[ADD1:%.*]] = add i8 [[SCALAR_LOAD1]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD1]], ptr [[ARRAYIDX1]], align 1
+; SIFIVE-NEXT:    [[INC1:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 1
+; SIFIVE-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC1]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_1:%.*]] = load i8, ptr [[ARRAYIDX_1]], align 1
+; SIFIVE-NEXT:    [[ADD_1:%.*]] = add i8 [[SCALAR_LOAD_1]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_1]], ptr [[ARRAYIDX_1]], align 1
+; SIFIVE-NEXT:    [[INC_1:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 2
+; SIFIVE-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_1]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_2:%.*]] = load i8, ptr [[ARRAYIDX_2]], align 1
+; SIFIVE-NEXT:    [[ADD_2:%.*]] = add i8 [[SCALAR_LOAD_2]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_2]], ptr [[ARRAYIDX_2]], align 1
+; SIFIVE-NEXT:    [[INC_2:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 3
+; SIFIVE-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_2]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_3:%.*]] = load i8, ptr [[ARRAYIDX_3]], align 1
+; SIFIVE-NEXT:    [[ADD_3:%.*]] = add i8 [[SCALAR_LOAD_3]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_3]], ptr [[ARRAYIDX_3]], align 1
+; SIFIVE-NEXT:    [[INC_3:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 4
+; SIFIVE-NEXT:    [[ARRAYIDX_4:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_3]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_4:%.*]] = load i8, ptr [[ARRAYIDX_4]], align 1
+; SIFIVE-NEXT:    [[ADD_4:%.*]] = add i8 [[SCALAR_LOAD_4]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_4]], ptr [[ARRAYIDX_4]], align 1
+; SIFIVE-NEXT:    [[INC_4:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 5
+; SIFIVE-NEXT:    [[ARRAYIDX_5:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_4]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_5:%.*]] = load i8, ptr [[ARRAYIDX_5]], align 1
+; SIFIVE-NEXT:    [[ADD_5:%.*]] = add i8 [[SCALAR_LOAD_5]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_5]], ptr [[ARRAYIDX_5]], align 1
+; SIFIVE-NEXT:    [[INC_5:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 6
+; SIFIVE-NEXT:    [[ARRAYIDX_6:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_5]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_6:%.*]] = load i8, ptr [[ARRAYIDX_6]], align 1
+; SIFIVE-NEXT:    [[ADD_6:%.*]] = add i8 [[SCALAR_LOAD_6]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_6]], ptr [[ARRAYIDX_6]], align 1
+; SIFIVE-NEXT:    [[INC_6:%.*]] = add nuw i64 [[IV_SCALAR_LOOP1]], 7
+; SIFIVE-NEXT:    [[ARRAYIDX_7:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 [[INC_6]]
+; SIFIVE-NEXT:    [[SCALAR_LOAD_7:%.*]] = load i8, ptr [[ARRAYIDX_7]], align 1
+; SIFIVE-NEXT:    [[ADD_7:%.*]] = add i8 [[SCALAR_LOAD_7]], [[SPLAT_SCALAR]]
+; SIFIVE-NEXT:    store i8 [[ADD_7]], ptr [[ARRAYIDX_7]], align 1
+; SIFIVE-NEXT:    [[INC_7]] = add nuw i64 [[IV_SCALAR_LOOP1]], 8
+; SIFIVE-NEXT:    [[EXITCOND_NOT_7:%.*]] = icmp eq i64 [[INC_7]], [[N]]
+; SIFIVE-NEXT:    br i1 [[EXITCOND_NOT_7]], label %[[EXIT_LOOPEXIT_UNR_LCSSA:.*]], label %[[SCALAR_REMAINDER]], !llvm.loop [[LOOP3:![0-9]+]]
+; SIFIVE:       [[EXIT_LOOPEXIT_UNR_LCSSA]]:
+; SIFIVE-NEXT:    br label %[[EXIT_LOOPEXIT]]
+; SIFIVE:       [[EXIT_LOOPEXIT]]:
+; SIFIVE-NEXT:    br label %[[EXIT]]
+; SIFIVE:       [[EXIT]]:
+; SIFIVE-NEXT:    ret void
+;
+entry:
+  %min.iters.check = icmp ult i64 %n, 32
+  br i1 %min.iters.check, label %scalar.remainder, label %vector.ph
+
+vector.ph:
+  %n.vec = and i64 %n, -32
+  %broadcast.splatinsert = insertelement <16 x i8> poison, i8 %splat.scalar, i64 0
+  %broadcast.splat = shufflevector <16 x i8> %broadcast.splatinsert, <16 x i8> poison, <16 x i32> zeroinitializer
+  br label %vector.body
+
+vector.body:
+  %iv = phi i64 [ 0, %vector.ph ], [ %iv.next, %vector.body ]
+  %gep.p.iv = getelementptr inbounds nuw i8, ptr %p, i64 %iv
+  %gep.p.iv.16 = getelementptr inbounds nuw i8, ptr %gep.p.iv, i64 16
+  %wide.load = load <16 x i8>, ptr %gep.p.iv, align 1
+  %wide.load.2 = load <16 x i8>, ptr %gep.p.iv.16, align 1
+  %add.broadcast = add <16 x i8> %wide.load, %broadcast.splat
+  %add.broadcast.2 = add <16 x i8> %wide.load.2, %broadcast.splat
+  store <16 x i8> %add.broadcast, ptr %gep.p.iv, align 1
+  store <16 x i8> %add.broadcast.2, ptr %gep.p.iv.16, align 1
+  %iv.next = add nuw i64 %iv, 32
+  %exit.cond = icmp eq i64 %iv.next, %n.vec
+  br i1 %exit.cond, label %middle.block, label %vector.body, !llvm.loop !2
+
+middle.block:
+  %cmp.n = icmp eq i64 %n, %n.vec
+  br i1 %cmp.n, label %exit, label %scalar.remainder
+
+scalar.remainder:
+  %iv.scalar.loop = phi i64 [ %inc, %scalar.remainder ], [ %n.vec, %middle.block ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds nuw i8, ptr %p, i64 %iv.scalar.loop
+  %scalar.load = load i8, ptr %arrayidx, align 1
+  %add = add i8 %scalar.load, %splat.scalar
+  store i8 %add, ptr %arrayidx, align 1
+  %inc = add nuw i64 %iv.scalar.loop, 1
+  %exitcond.not = icmp eq i64 %inc, %n
+  br i1 %exitcond.not, label %exit, label %scalar.remainder, !llvm.loop !3
+
+exit:
+  ret void
+}
+
+!2 = distinct !{!2, !1}
+!3 = distinct !{!3, !1}
+
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; CHECK: [[LOOP2]] = distinct !{[[LOOP2]], [[META1]]}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]]}
+;.
+; SIFIVE: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]]}
+; SIFIVE: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; SIFIVE: [[LOOP2]] = distinct !{[[LOOP2]], [[META1]]}
+; SIFIVE: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/intrinsic.ll b/llvm/test/Transforms/LoopVectorize/intrinsic.ll
index 10d83a4..9c910d7 100644
--- a/llvm/test/Transforms/LoopVectorize/intrinsic.ll
+++ b/llvm/test/Transforms/LoopVectorize/intrinsic.ll
@@ -324,56 +324,6 @@ for.end:                                          ; preds = %for.body, %entry
 
 declare double @llvm.exp2.f64(double)
 
-define void @ldexp_f32i32(i32 %n, ptr %y, ptr %x, i32 %exp) {
-; CHECK-LABEL: @ldexp_f32i32(
-; CHECK: llvm.ldexp.v4f32.v4i32
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float, ptr %y, i32 %iv
-  %0 = load float, ptr %arrayidx, align 4
-  %call = tail call float @llvm.ldexp.f32.i32(float %0, i32 %exp)
-  %arrayidx2 = getelementptr inbounds float, ptr %x, i32 %iv
-  store float %call, ptr %arrayidx2, align 4
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare float @llvm.ldexp.f32.i32(float, i32)
-
-define void @ldexp_f64i32(i32 %n, ptr %y, ptr %x, i32 %exp) {
-; CHECK-LABEL: @ldexp_f64i32(
-; CHECK: llvm.ldexp.v4f64.v4i32
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double, ptr %y, i32 %iv
-  %0 = load double, ptr %arrayidx, align 8
-  %call = tail call double @llvm.ldexp.f64.i32(double %0, i32 %exp)
-  %arrayidx2 = getelementptr inbounds double, ptr %x, i32 %iv
-  store double %call, ptr %arrayidx2, align 8
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare double @llvm.ldexp.f64.i32(double, i32)
-
 define void @log_f32(i32 %n, ptr %y, ptr %x) {
 ; CHECK-LABEL: @log_f32(
 ; CHECK: llvm.log.v4f32
@@ -1026,157 +976,6 @@ for.end:                                          ; preds = %for.body, %entry
 
 declare double @llvm.roundeven.f64(double)
 
-
-define void @lround_i32f32(i32 %n, ptr %y, ptr %x) {
-; CHECK-LABEL: @lround_i32f32(
-; CHECK: llvm.lround.v4i32.v4f32
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float, ptr %y, i32 %iv
-  %0 = load float, ptr %arrayidx, align 4
-  %call = tail call i32 @llvm.lround.i32.f32(float %0)
-  %arrayidx2 = getelementptr inbounds i32, ptr %x, i32 %iv
-  store i32 %call, ptr %arrayidx2, align 4
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare i32 @llvm.lround.i32.f32(float)
-
-define void @lround_i32f64(i32 %n, ptr %y, ptr %x) {
-; CHECK-LABEL: @lround_i32f64(
-; CHECK: llvm.lround.v4i32.v4f64
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double, ptr %y, i32 %iv
-  %0 = load double, ptr %arrayidx, align 8
-  %call = tail call i32 @llvm.lround.i32.f64(double %0)
-  %arrayidx2 = getelementptr inbounds i32, ptr %x, i32 %iv
-  store i32 %call, ptr %arrayidx2, align 8
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare i32 @llvm.lround.i32.f64(double)
-
-define void @lround_i64f32(i32 %n, ptr %y, ptr %x) {
-; CHECK-LABEL: @lround_i64f32(
-; CHECK: llvm.lround.v4i64.v4f32
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float, ptr %y, i32 %iv
-  %0 = load float, ptr %arrayidx, align 4
-  %call = tail call i64 @llvm.lround.i64.f32(float %0)
-  %arrayidx2 = getelementptr inbounds i64, ptr %x, i32 %iv
-  store i64 %call, ptr %arrayidx2, align 4
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare i64 @llvm.lround.i64.f32(float)
-
-define void @lround_i64f64(i32 %n, ptr %y, ptr %x) {
-; CHECK-LABEL: @lround_i64f64(
-; CHECK: llvm.lround.v4i64.v4f64
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double, ptr %y, i32 %iv
-  %0 = load double, ptr %arrayidx, align 8
-  %call = tail call i64 @llvm.lround.i64.f64(double %0)
-  %arrayidx2 = getelementptr inbounds i64, ptr %x, i32 %iv
-  store i64 %call, ptr %arrayidx2, align 8
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare i64 @llvm.lround.i64.f64(double)
-
-define void @llround_i64f32(i32 %n, ptr %y, ptr %x) {
-; CHECK-LABEL: @llround_i64f32(
-; CHECK: llvm.llround.v4i64.v4f32
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float, ptr %y, i32 %iv
-  %0 = load float, ptr %arrayidx, align 4
-  %call = tail call i64 @llvm.llround.i64.f32(float %0)
-  %arrayidx2 = getelementptr inbounds i64, ptr %x, i32 %iv
-  store i64 %call, ptr %arrayidx2, align 4
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare i64 @llvm.llround.i64.f32(float)
-
-define void @llround_i64f64(i32 %n, ptr %y, ptr %x) {
-; CHECK-LABEL: @llround_i64f64(
-; CHECK: llvm.llround.v4i64.v4f64
-; CHECK: ret void
-;
-entry:
-  br label %for.body
-
-for.body:                                         ; preds = %entry, %for.body
-  %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double, ptr %y, i32 %iv
-  %0 = load double, ptr %arrayidx, align 8
-  %call = tail call i64 @llvm.llround.i64.f64(double %0)
-  %arrayidx2 = getelementptr inbounds i64, ptr %x, i32 %iv
-  store i64 %call, ptr %arrayidx2, align 8
-  %iv.next = add i32 %iv, 1
-  %exitcond = icmp eq i32 %iv.next, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
-declare i64 @llvm.llround.i64.f64(double)
-
 define void @fma_f32(i32 %n, ptr %y, ptr %x, ptr %z, ptr %w) {
 ; CHECK-LABEL: @fma_f32(
 ; CHECK: llvm.fma.v4f32
diff --git a/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll b/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll
index 19fbc1f..4455016 100644
--- a/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll
+++ b/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll
@@ -1,3 +1,6 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -O3 -S < %s | FileCheck %s
+
 ;; This tests lowering of the implementations of table-based ctz
 ;; algorithm to the llvm.cttz instruction in the -O3 case.
 
@@ -13,13 +16,17 @@
 ;; }
 ;; Compiled as: clang -O3 test.c -S -emit-llvm -Xclang -disable-llvm-optzns
 
-; RUN: opt -O3 -S < %s | FileCheck %s
-
-; CHECK: call range(i32 0, 33) i32 @llvm.cttz.i32
-
 @ctz1.table = internal constant [32 x i8] c"\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09", align 16
 
-define i32 @ctz1(i32 noundef %x) {
+define i32 @ctz(i32 noundef %x) {
+; CHECK-LABEL: define range(i32 0, 32) i32 @ctz(
+; CHECK-SAME: i32 noundef [[X:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  [[ENTRY:.*:]]
+; CHECK-NEXT:    [[TMP0:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 true)
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[X]], 0
+; CHECK-NEXT:    [[CONV:%.*]] = select i1 [[TMP1]], i32 0, i32 [[TMP0]]
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
 entry:
   %x.addr = alloca i32, align 4
   store i32 %x, ptr %x.addr, align 4
@@ -35,3 +42,28 @@ entry:
   %conv = sext i8 %2 to i32
   ret i32 %conv
 }
+
+define i32 @ctz_nonarraygep(i32 noundef %x) {
+; CHECK-LABEL: define range(i32 0, 32) i32 @ctz_nonarraygep(
+; CHECK-SAME: i32 noundef [[X:%.*]]) local_unnamed_addr #[[ATTR0]] {
+; CHECK-NEXT:  [[ENTRY:.*:]]
+; CHECK-NEXT:    [[TMP0:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 true)
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[X]], 0
+; CHECK-NEXT:    [[CONV:%.*]] = select i1 [[TMP1]], i32 0, i32 [[TMP0]]
+; CHECK-NEXT:    ret i32 [[CONV]]
+;
+entry:
+  %x.addr = alloca i32, align 4
+  store i32 %x, ptr %x.addr, align 4
+  %0 = load i32, ptr %x.addr, align 4
+  %1 = load i32, ptr %x.addr, align 4
+  %sub = sub i32 0, %1
+  %and = and i32 %0, %sub
+  %mul = mul i32 %and, 125613361
+  %shr = lshr i32 %mul, 27
+  %idxprom = zext i32 %shr to i64
+  %arrayidx = getelementptr inbounds i8, ptr @ctz1.table, i64 %idxprom
+  %2 = load i8, ptr %arrayidx, align 1
+  %conv = sext i8 %2 to i32
+  ret i32 %conv
+}
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/exp.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/exp.ll
deleted file mode 100644
index 301e5da..0000000
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/exp.ll
+++ /dev/null
@@ -1,279 +0,0 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt -S -passes=slp-vectorizer -mtriple=aarch64 < %s | FileCheck %s
-
-target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
-
-define void @ldexp_f32i32(ptr %x, ptr %y, i32 %exp) {
-; CHECK-LABEL: @ldexp_f32i32(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load float, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load float, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load float, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load float, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call float @llvm.ldexp.f32.i32(float [[L0]], i32 [[EXP:%.*]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call float @llvm.ldexp.f32.i32(float [[L2]], i32 [[EXP]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call float @llvm.ldexp.f32.i32(float [[L4]], i32 [[EXP]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call float @llvm.ldexp.f32.i32(float [[L6]], i32 [[EXP]])
-; CHECK-NEXT:    store float [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 1
-; CHECK-NEXT:    store float [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 2
-; CHECK-NEXT:    store float [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 3
-; CHECK-NEXT:    store float [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load float, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds float, ptr %x, i64 1
-  %l2 = load float, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds float, ptr %x, i64 2
-  %l4 = load float, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds float, ptr %x, i64 3
-  %l6 = load float, ptr %arrayidx.3, align 4
-  %l1 = tail call float @llvm.ldexp.f32.i32(float %l0, i32 %exp)
-  %l3 = tail call float @llvm.ldexp.f32.i32(float %l2, i32 %exp)
-  %l5 = tail call float @llvm.ldexp.f32.i32(float %l4, i32 %exp)
-  %l7 = tail call float @llvm.ldexp.f32.i32(float %l6, i32 %exp)
-  store float %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds float, ptr %y, i64 1
-  store float %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds float, ptr %y, i64 2
-  store float %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds float, ptr %y, i64 3
-  store float %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @ldexp_f64i32(ptr %x, ptr %y, i32 %exp) {
-; CHECK-LABEL: @ldexp_f64i32(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load double, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds double, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load double, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds double, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load double, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds double, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load double, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call double @llvm.ldexp.f64.i32(double [[L0]], i32 [[EXP:%.*]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call double @llvm.ldexp.f64.i32(double [[L2]], i32 [[EXP]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call double @llvm.ldexp.f64.i32(double [[L4]], i32 [[EXP]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call double @llvm.ldexp.f64.i32(double [[L6]], i32 [[EXP]])
-; CHECK-NEXT:    store double [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 1
-; CHECK-NEXT:    store double [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 2
-; CHECK-NEXT:    store double [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 3
-; CHECK-NEXT:    store double [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load double, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds double, ptr %x, i64 1
-  %l2 = load double, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds double, ptr %x, i64 2
-  %l4 = load double, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds double, ptr %x, i64 3
-  %l6 = load double, ptr %arrayidx.3, align 4
-  %l1 = tail call double @llvm.ldexp.f64.i32(double %l0, i32 %exp)
-  %l3 = tail call double @llvm.ldexp.f64.i32(double %l2, i32 %exp)
-  %l5 = tail call double @llvm.ldexp.f64.i32(double %l4, i32 %exp)
-  %l7 = tail call double @llvm.ldexp.f64.i32(double %l6, i32 %exp)
-  store double %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds double, ptr %y, i64 1
-  store double %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds double, ptr %y, i64 2
-  store double %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds double, ptr %y, i64 3
-  store double %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @ldexp_f32i64(ptr %x, ptr %y, i64 %exp) {
-; CHECK-LABEL: @ldexp_f32i64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load float, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load float, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load float, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load float, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call float @llvm.ldexp.f32.i64(float [[L0]], i64 [[EXP:%.*]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call float @llvm.ldexp.f32.i64(float [[L2]], i64 [[EXP]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call float @llvm.ldexp.f32.i64(float [[L4]], i64 [[EXP]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call float @llvm.ldexp.f32.i64(float [[L6]], i64 [[EXP]])
-; CHECK-NEXT:    store float [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 1
-; CHECK-NEXT:    store float [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 2
-; CHECK-NEXT:    store float [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 3
-; CHECK-NEXT:    store float [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load float, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds float, ptr %x, i64 1
-  %l2 = load float, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds float, ptr %x, i64 2
-  %l4 = load float, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds float, ptr %x, i64 3
-  %l6 = load float, ptr %arrayidx.3, align 4
-  %l1 = tail call float @llvm.ldexp.f32.i64(float %l0, i64 %exp)
-  %l3 = tail call float @llvm.ldexp.f32.i64(float %l2, i64 %exp)
-  %l5 = tail call float @llvm.ldexp.f32.i64(float %l4, i64 %exp)
-  %l7 = tail call float @llvm.ldexp.f32.i64(float %l6, i64 %exp)
-  store float %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds float, ptr %y, i64 1
-  store float %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds float, ptr %y, i64 2
-  store float %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds float, ptr %y, i64 3
-  store float %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @ldexp_f64i64(ptr %x, ptr %y, i64 %exp) {
-; CHECK-LABEL: @ldexp_f64i64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load double, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds double, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load double, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds double, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load double, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds double, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load double, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call double @llvm.ldexp.f64.i64(double [[L0]], i64 [[EXP:%.*]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call double @llvm.ldexp.f64.i64(double [[L2]], i64 [[EXP]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call double @llvm.ldexp.f64.i64(double [[L4]], i64 [[EXP]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call double @llvm.ldexp.f64.i64(double [[L6]], i64 [[EXP]])
-; CHECK-NEXT:    store double [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 1
-; CHECK-NEXT:    store double [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 2
-; CHECK-NEXT:    store double [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 3
-; CHECK-NEXT:    store double [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load double, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds double, ptr %x, i64 1
-  %l2 = load double, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds double, ptr %x, i64 2
-  %l4 = load double, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds double, ptr %x, i64 3
-  %l6 = load double, ptr %arrayidx.3, align 4
-  %l1 = tail call double @llvm.ldexp.f64.i64(double %l0, i64 %exp)
-  %l3 = tail call double @llvm.ldexp.f64.i64(double %l2, i64 %exp)
-  %l5 = tail call double @llvm.ldexp.f64.i64(double %l4, i64 %exp)
-  %l7 = tail call double @llvm.ldexp.f64.i64(double %l6, i64 %exp)
-  store double %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds double, ptr %y, i64 1
-  store double %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds double, ptr %y, i64 2
-  store double %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds double, ptr %y, i64 3
-  store double %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @ldexp_f32i32_i64(ptr %x, ptr %y, i32 %exp32, i64 %exp64) {
-; CHECK-LABEL: @ldexp_f32i32_i64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load float, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load float, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load float, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load float, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call float @llvm.ldexp.f32.i32(float [[L0]], i32 [[EXP32:%.*]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call float @llvm.ldexp.f32.i32(float [[L2]], i32 [[EXP32]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call float @llvm.ldexp.f32.i64(float [[L4]], i64 [[EXP64:%.*]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call float @llvm.ldexp.f32.i64(float [[L6]], i64 [[EXP64]])
-; CHECK-NEXT:    store float [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 1
-; CHECK-NEXT:    store float [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 2
-; CHECK-NEXT:    store float [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds float, ptr [[Y]], i64 3
-; CHECK-NEXT:    store float [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load float, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds float, ptr %x, i64 1
-  %l2 = load float, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds float, ptr %x, i64 2
-  %l4 = load float, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds float, ptr %x, i64 3
-  %l6 = load float, ptr %arrayidx.3, align 4
-  %l1 = tail call float @llvm.ldexp.f32.i32(float %l0, i32 %exp32)
-  %l3 = tail call float @llvm.ldexp.f32.i32(float %l2, i32 %exp32)
-  %l5 = tail call float @llvm.ldexp.f32.i64(float %l4, i64 %exp64)
-  %l7 = tail call float @llvm.ldexp.f32.i64(float %l6, i64 %exp64)
-  store float %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds float, ptr %y, i64 1
-  store float %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds float, ptr %y, i64 2
-  store float %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds float, ptr %y, i64 3
-  store float %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @ldexp_f64_i32_i64(ptr %x, ptr %y, i32 %exp32, i64 %exp64) {
-; CHECK-LABEL: @ldexp_f64_i32_i64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load double, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds double, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load double, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds double, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load double, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds double, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load double, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call double @llvm.ldexp.f64.i32(double [[L0]], i32 [[EXP32:%.*]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call double @llvm.ldexp.f64.i32(double [[L2]], i32 [[EXP32]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call double @llvm.ldexp.f64.i64(double [[L4]], i64 [[EXP64:%.*]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call double @llvm.ldexp.f64.i64(double [[L6]], i64 [[EXP64]])
-; CHECK-NEXT:    store double [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 1
-; CHECK-NEXT:    store double [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 2
-; CHECK-NEXT:    store double [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds double, ptr [[Y]], i64 3
-; CHECK-NEXT:    store double [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load double, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds double, ptr %x, i64 1
-  %l2 = load double, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds double, ptr %x, i64 2
-  %l4 = load double, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds double, ptr %x, i64 3
-  %l6 = load double, ptr %arrayidx.3, align 4
-  %l1 = tail call double @llvm.ldexp.f64.i32(double %l0, i32 %exp32)
-  %l3 = tail call double @llvm.ldexp.f64.i32(double %l2, i32 %exp32)
-  %l5 = tail call double @llvm.ldexp.f64.i64(double %l4, i64 %exp64)
-  %l7 = tail call double @llvm.ldexp.f64.i64(double %l6, i64 %exp64)
-  store double %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds double, ptr %y, i64 1
-  store double %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds double, ptr %y, i64 2
-  store double %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds double, ptr %y, i64 3
-  store double %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-declare float @llvm.ldexp.f32.i32(float, i32)
-declare double @llvm.ldexp.f64.i32(double, i32)
-declare float @llvm.ldexp.f32.i64(float, i64)
-declare double @llvm.ldexp.f64.i64(double, i64)
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/fround.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/fround.ll
deleted file mode 100644
index 07a3fe7..0000000
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/fround.ll
+++ /dev/null
@@ -1,280 +0,0 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt -S -passes=slp-vectorizer -mtriple=aarch64 < %s | FileCheck %s
-
-target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
-
-define void @lround_i32f32(ptr %x, ptr %y, i32 %n) {
-; CHECK-LABEL: @lround_i32f32(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load float, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load float, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load float, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load float, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call i32 @llvm.lround.i32.f32(float [[L0]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call i32 @llvm.lround.i32.f32(float [[L2]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call i32 @llvm.lround.i32.f32(float [[L4]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call i32 @llvm.lround.i32.f32(float [[L6]])
-; CHECK-NEXT:    store i32 [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i32, ptr [[Y]], i64 1
-; CHECK-NEXT:    store i32 [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i32, ptr [[Y]], i64 2
-; CHECK-NEXT:    store i32 [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i32, ptr [[Y]], i64 3
-; CHECK-NEXT:    store i32 [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load float, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds float, ptr %x, i64 1
-  %l2 = load float, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds float, ptr %x, i64 2
-  %l4 = load float, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds float, ptr %x, i64 3
-  %l6 = load float, ptr %arrayidx.3, align 4
-  %l1 = tail call i32 @llvm.lround.i32.f32(float %l0)
-  %l3 = tail call i32 @llvm.lround.i32.f32(float %l2)
-  %l5 = tail call i32 @llvm.lround.i32.f32(float %l4)
-  %l7 = tail call i32 @llvm.lround.i32.f32(float %l6)
-  store i32 %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds i32, ptr %y, i64 1
-  store i32 %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds i32, ptr %y, i64 2
-  store i32 %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds i32, ptr %y, i64 3
-  store i32 %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @lround_i32f64(ptr %x, ptr %y, i32 %n) {
-; CHECK-LABEL: @lround_i32f64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load double, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds double, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load double, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds double, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load double, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds double, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load double, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call i32 @llvm.lround.i32.f64(double [[L0]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call i32 @llvm.lround.i32.f64(double [[L2]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call i32 @llvm.lround.i32.f64(double [[L4]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call i32 @llvm.lround.i32.f64(double [[L6]])
-; CHECK-NEXT:    store i32 [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i32, ptr [[Y]], i64 1
-; CHECK-NEXT:    store i32 [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i32, ptr [[Y]], i64 2
-; CHECK-NEXT:    store i32 [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i32, ptr [[Y]], i64 3
-; CHECK-NEXT:    store i32 [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load double, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds double, ptr %x, i64 1
-  %l2 = load double, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds double, ptr %x, i64 2
-  %l4 = load double, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds double, ptr %x, i64 3
-  %l6 = load double, ptr %arrayidx.3, align 4
-  %l1 = tail call i32 @llvm.lround.i32.f64(double %l0)
-  %l3 = tail call i32 @llvm.lround.i32.f64(double %l2)
-  %l5 = tail call i32 @llvm.lround.i32.f64(double %l4)
-  %l7 = tail call i32 @llvm.lround.i32.f64(double %l6)
-  store i32 %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds i32, ptr %y, i64 1
-  store i32 %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds i32, ptr %y, i64 2
-  store i32 %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds i32, ptr %y, i64 3
-  store i32 %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @lround_i64f32(ptr %x, ptr %y, i64 %n) {
-; CHECK-LABEL: @lround_i64f32(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load float, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load float, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load float, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load float, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call i64 @llvm.lround.i64.f32(float [[L0]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call i64 @llvm.lround.i64.f32(float [[L2]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call i64 @llvm.lround.i64.f32(float [[L4]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call i64 @llvm.lround.i64.f32(float [[L6]])
-; CHECK-NEXT:    store i64 [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 1
-; CHECK-NEXT:    store i64 [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 2
-; CHECK-NEXT:    store i64 [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 3
-; CHECK-NEXT:    store i64 [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load float, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds float, ptr %x, i64 1
-  %l2 = load float, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds float, ptr %x, i64 2
-  %l4 = load float, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds float, ptr %x, i64 3
-  %l6 = load float, ptr %arrayidx.3, align 4
-  %l1 = tail call i64 @llvm.lround.i64.f32(float %l0)
-  %l3 = tail call i64 @llvm.lround.i64.f32(float %l2)
-  %l5 = tail call i64 @llvm.lround.i64.f32(float %l4)
-  %l7 = tail call i64 @llvm.lround.i64.f32(float %l6)
-  store i64 %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds i64, ptr %y, i64 1
-  store i64 %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds i64, ptr %y, i64 2
-  store i64 %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds i64, ptr %y, i64 3
-  store i64 %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @lround_i64f64(ptr %x, ptr %y, i64 %n) {
-; CHECK-LABEL: @lround_i64f64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load double, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds double, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load double, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds double, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load double, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds double, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load double, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call i64 @llvm.lround.i64.f64(double [[L0]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call i64 @llvm.lround.i64.f64(double [[L2]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call i64 @llvm.lround.i64.f64(double [[L4]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call i64 @llvm.lround.i64.f64(double [[L6]])
-; CHECK-NEXT:    store i64 [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 1
-; CHECK-NEXT:    store i64 [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 2
-; CHECK-NEXT:    store i64 [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 3
-; CHECK-NEXT:    store i64 [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load double, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds double, ptr %x, i64 1
-  %l2 = load double, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds double, ptr %x, i64 2
-  %l4 = load double, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds double, ptr %x, i64 3
-  %l6 = load double, ptr %arrayidx.3, align 4
-  %l1 = tail call i64 @llvm.lround.i64.f64(double %l0)
-  %l3 = tail call i64 @llvm.lround.i64.f64(double %l2)
-  %l5 = tail call i64 @llvm.lround.i64.f64(double %l4)
-  %l7 = tail call i64 @llvm.lround.i64.f64(double %l6)
-  store i64 %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds i64, ptr %y, i64 1
-  store i64 %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds i64, ptr %y, i64 2
-  store i64 %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds i64, ptr %y, i64 3
-  store i64 %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @llround_i64f32(ptr %x, ptr %y, i64 %n) {
-; CHECK-LABEL: @llround_i64f32(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load float, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load float, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load float, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load float, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call i64 @llvm.llround.i64.f32(float [[L0]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call i64 @llvm.llround.i64.f32(float [[L2]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call i64 @llvm.llround.i64.f32(float [[L4]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call i64 @llvm.llround.i64.f32(float [[L6]])
-; CHECK-NEXT:    store i64 [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 1
-; CHECK-NEXT:    store i64 [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 2
-; CHECK-NEXT:    store i64 [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 3
-; CHECK-NEXT:    store i64 [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load float, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds float, ptr %x, i64 1
-  %l2 = load float, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds float, ptr %x, i64 2
-  %l4 = load float, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds float, ptr %x, i64 3
-  %l6 = load float, ptr %arrayidx.3, align 4
-  %l1 = tail call i64 @llvm.llround.i64.f32(float %l0)
-  %l3 = tail call i64 @llvm.llround.i64.f32(float %l2)
-  %l5 = tail call i64 @llvm.llround.i64.f32(float %l4)
-  %l7 = tail call i64 @llvm.llround.i64.f32(float %l6)
-  store i64 %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds i64, ptr %y, i64 1
-  store i64 %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds i64, ptr %y, i64 2
-  store i64 %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds i64, ptr %y, i64 3
-  store i64 %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-define void @llround_i64f64(ptr %x, ptr %y, i64 %n) {
-; CHECK-LABEL: @llround_i64f64(
-; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[L0:%.*]] = load double, ptr [[X:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_1:%.*]] = getelementptr inbounds double, ptr [[X]], i64 1
-; CHECK-NEXT:    [[L2:%.*]] = load double, ptr [[ARRAYIDX_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds double, ptr [[X]], i64 2
-; CHECK-NEXT:    [[L4:%.*]] = load double, ptr [[ARRAYIDX_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds double, ptr [[X]], i64 3
-; CHECK-NEXT:    [[L6:%.*]] = load double, ptr [[ARRAYIDX_3]], align 4
-; CHECK-NEXT:    [[L1:%.*]] = tail call i64 @llvm.llround.i64.f64(double [[L0]])
-; CHECK-NEXT:    [[L3:%.*]] = tail call i64 @llvm.llround.i64.f64(double [[L2]])
-; CHECK-NEXT:    [[L5:%.*]] = tail call i64 @llvm.llround.i64.f64(double [[L4]])
-; CHECK-NEXT:    [[L7:%.*]] = tail call i64 @llvm.llround.i64.f64(double [[L6]])
-; CHECK-NEXT:    store i64 [[L1]], ptr [[Y:%.*]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 1
-; CHECK-NEXT:    store i64 [[L3]], ptr [[ARRAYIDX2_1]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 2
-; CHECK-NEXT:    store i64 [[L5]], ptr [[ARRAYIDX2_2]], align 4
-; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i64, ptr [[Y]], i64 3
-; CHECK-NEXT:    store i64 [[L7]], ptr [[ARRAYIDX2_3]], align 4
-; CHECK-NEXT:    ret void
-;
-entry:
-  %l0 = load double, ptr %x, align 4
-  %arrayidx.1 = getelementptr inbounds double, ptr %x, i64 1
-  %l2 = load double, ptr %arrayidx.1, align 4
-  %arrayidx.2 = getelementptr inbounds double, ptr %x, i64 2
-  %l4 = load double, ptr %arrayidx.2, align 4
-  %arrayidx.3 = getelementptr inbounds double, ptr %x, i64 3
-  %l6 = load double, ptr %arrayidx.3, align 4
-  %l1 = tail call i64 @llvm.llround.i64.f64(double %l0)
-  %l3 = tail call i64 @llvm.llround.i64.f64(double %l2)
-  %l5 = tail call i64 @llvm.llround.i64.f64(double %l4)
-  %l7 = tail call i64 @llvm.llround.i64.f64(double %l6)
-  store i64 %l1, ptr %y, align 4
-  %arrayidx2.1 = getelementptr inbounds i64, ptr %y, i64 1
-  store i64 %l3, ptr %arrayidx2.1, align 4
-  %arrayidx2.2 = getelementptr inbounds i64, ptr %y, i64 2
-  store i64 %l5, ptr %arrayidx2.2, align 4
-  %arrayidx2.3 = getelementptr inbounds i64, ptr %y, i64 3
-  store i64 %l7, ptr %arrayidx2.3, align 4
-  ret void
-}
-
-declare i32 @llvm.lround.i32.f32(float)
-declare i64 @llvm.lround.i64.f32(float)
-declare i64 @llvm.lround.i64.f64(double)
-declare i64 @llvm.llround.i64.f32(float)
-declare i64 @llvm.llround.i64.f64(double)
diff --git a/llvm/test/Transforms/Scalarizer/intrinsics.ll b/llvm/test/Transforms/Scalarizer/intrinsics.ll
index 070c765..cee44ef 100644
--- a/llvm/test/Transforms/Scalarizer/intrinsics.ll
+++ b/llvm/test/Transforms/Scalarizer/intrinsics.ll
@@ -8,7 +8,6 @@ declare <2 x float> @llvm.sqrt.v2f32(<2 x float>)
 declare <2 x float> @llvm.minnum.v2f32(<2 x float>, <2 x float>)
 declare <2 x float> @llvm.minimum.v2f32(<2 x float>, <2 x float>)
 declare <2 x float> @llvm.maximum.v2f32(<2 x float>, <2 x float>)
-declare <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float>, <2 x i32>)
 
 ; Ternary fp
 declare <2 x float> @llvm.fma.v2f32(<2 x float>, <2 x float>, <2 x float>)
@@ -33,8 +32,6 @@ declare <2 x i32> @llvm.fptoui.sat.v2i32.v2f32(<2 x float>)
 ; Unary fp operand, int return type
 declare <2 x i32> @llvm.lrint.v2i32.v2f32(<2 x float>)
 declare <2 x i32> @llvm.llrint.v2i32.v2f32(<2 x float>)
-declare <2 x i32> @llvm.lround.v2i32.v2f32(<2 x float>)
-declare <2 x i32> @llvm.llround.v2i32.v2f32(<2 x float>)
 
 ; Bool return type, overloaded on fp operand type
 declare <2 x i1> @llvm.is.fpclass(<2 x float>, i32)
@@ -162,22 +159,6 @@ define <2 x float> @scalarize_powi_v2f32(<2 x float> %x, i32 %y) #0 {
   ret <2 x float> %powi
 }
 
-define <2 x float> @scalarize_ldexp_v2f32(<2 x float> %x, <2 x i32> %y) #0 {
-; CHECK-LABEL: @scalarize_ldexp_v2f32(
-; CHECK-NEXT:    [[X_I0:%.*]] = extractelement <2 x float> [[X:%.*]], i64 0
-; CHECK-NEXT:    [[Y:%.*]] = extractelement <2 x i32> [[Y1:%.*]], i64 0
-; CHECK-NEXT:    [[POWI_I0:%.*]] = call float @llvm.ldexp.f32.i32(float [[X_I0]], i32 [[Y]])
-; CHECK-NEXT:    [[X_I1:%.*]] = extractelement <2 x float> [[X]], i64 1
-; CHECK-NEXT:    [[Y_I1:%.*]] = extractelement <2 x i32> [[Y1]], i64 1
-; CHECK-NEXT:    [[POWI_I1:%.*]] = call float @llvm.ldexp.f32.i32(float [[X_I1]], i32 [[Y_I1]])
-; CHECK-NEXT:    [[POWI_UPTO0:%.*]] = insertelement <2 x float> poison, float [[POWI_I0]], i64 0
-; CHECK-NEXT:    [[POWI:%.*]] = insertelement <2 x float> [[POWI_UPTO0]], float [[POWI_I1]], i64 1
-; CHECK-NEXT:    ret <2 x float> [[POWI]]
-;
-  %powi = call <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float> %x, <2 x i32> %y)
-  ret <2 x float> %powi
-}
-
 define <2 x i32> @scalarize_smul_fix_sat_v2i32(<2 x i32> %x) #0 {
 ; CHECK-LABEL: @scalarize_smul_fix_sat_v2i32(
 ; CHECK-NEXT:    [[X_I0:%.*]] = extractelement <2 x i32> [[X:%.*]], i64 0
@@ -262,34 +243,6 @@ define <2 x i32> @scalarize_llrint(<2 x float> %x) #0 {
   ret <2 x i32> %rnd
 }
 
-define <2 x i32> @scalarize_lround(<2 x float> %x) #0 {
-; CHECK-LABEL: @scalarize_lround(
-; CHECK-NEXT:    [[X_I0:%.*]] = extractelement <2 x float> [[X:%.*]], i64 0
-; CHECK-NEXT:    [[RND_I0:%.*]] = call i32 @llvm.lround.i32.f32(float [[X_I0]])
-; CHECK-NEXT:    [[X_I1:%.*]] = extractelement <2 x float> [[X]], i64 1
-; CHECK-NEXT:    [[RND_I1:%.*]] = call i32 @llvm.lround.i32.f32(float [[X_I1]])
-; CHECK-NEXT:    [[RND_UPTO0:%.*]] = insertelement <2 x i32> poison, i32 [[RND_I0]], i64 0
-; CHECK-NEXT:    [[RND:%.*]] = insertelement <2 x i32> [[RND_UPTO0]], i32 [[RND_I1]], i64 1
-; CHECK-NEXT:    ret <2 x i32> [[RND]]
-;
-  %rnd = call <2 x i32> @llvm.lround.v2i32.v2f32(<2 x float> %x)
-  ret <2 x i32> %rnd
-}
-
-define <2 x i32> @scalarize_llround(<2 x float> %x) #0 {
-; CHECK-LABEL: @scalarize_llround(
-; CHECK-NEXT:    [[X_I0:%.*]] = extractelement <2 x float> [[X:%.*]], i64 0
-; CHECK-NEXT:    [[RND_I0:%.*]] = call i32 @llvm.llround.i32.f32(float [[X_I0]])
-; CHECK-NEXT:    [[X_I1:%.*]] = extractelement <2 x float> [[X]], i64 1
-; CHECK-NEXT:    [[RND_I1:%.*]] = call i32 @llvm.llround.i32.f32(float [[X_I1]])
-; CHECK-NEXT:    [[RND_UPTO0:%.*]] = insertelement <2 x i32> poison, i32 [[RND_I0]], i64 0
-; CHECK-NEXT:    [[RND:%.*]] = insertelement <2 x i32> [[RND_UPTO0]], i32 [[RND_I1]], i64 1
-; CHECK-NEXT:    ret <2 x i32> [[RND]]
-;
-  %rnd = call <2 x i32> @llvm.llround.v2i32.v2f32(<2 x float> %x)
-  ret <2 x i32> %rnd
-}
-
 define <2 x i1> @scalarize_is_fpclass(<2 x float> %x) #0 {
 ; CHECK-LABEL: @scalarize_is_fpclass(
 ; CHECK-NEXT:    [[X_I0:%.*]] = extractelement <2 x float> [[X:%.*]], i64 0
diff --git a/llvm/test/tools/llvm-profdata/memprof-padding-histogram.test b/llvm/test/tools/llvm-profdata/memprof-padding-histogram.test
index 79521f3..2d0346e 100644
--- a/llvm/test/tools/llvm-profdata/memprof-padding-histogram.test
+++ b/llvm/test/tools/llvm-profdata/memprof-padding-histogram.test
@@ -21,79 +21,79 @@ CHECK-NEXT:     Offset: 0x{{[[:xdigit:]]+}}
 CHECK-NEXT:   -
 
 CHECK:   Records:
-CHEC-NEXT    FunctionGUID: {{[0-9]+}}
-CHEC-NEXT    AllocSites:
-CHEC-NEXT    -
-CHEC-NEXT      Callstack:
-CHEC-NEXT      -
-CHEC-NEXT        Function: {{[0-9]+}}
-CHEC-NEXT        SymbolName: main
-CHEC-NEXT        LineOffset: 3
-CHEC-NEXT        Column: 10
-CHEC-NEXT        Inline: 0
-CHEC-NEXT      MemInfoBlock:
-CHEC-NEXT        AllocCount: 1
-CHEC-NEXT        TotalAccessCount: 5
-CHEC-NEXT        MinAccessCount: 5
-CHEC-NEXT        MaxAccessCount: 5
-CHEC-NEXT        TotalSize: 24
-CHEC-NEXT        MinSize: 24
-CHEC-NEXT        MaxSize: 24
-CHEC-NEXT        AllocTimestamp: {{[0-9]+}}
-CHEC-NEXT        DeallocTimestamp: {{[0-9]+}}
-CHEC-NEXT        TotalLifetime: 0
-CHEC-NEXT        MinLifetime: 0
-CHEC-NEXT        MaxLifetime: 0
-CHEC-NEXT        AllocCpuId: 11
-CHEC-NEXT        DeallocCpuId: 11
-CHEC-NEXT        NumMigratedCpu: 0
-CHEC-NEXT        NumLifetimeOverlaps: 0
-CHEC-NEXT        NumSameAllocCpu: 0
-CHEC-NEXT        NumSameDeallocCpu: 0
-CHEC-NEXT        DataTypeId: 0
-CHEC-NEXT        TotalAccessDensity: 20
-CHEC-NEXT        MinAccessDensity: 20
-CHEC-NEXT        MaxAccessDensity: 20
-CHEC-NEXT        TotalLifetimeAccessDensity: 20000
-CHEC-NEXT        MinLifetimeAccessDensity: 20000
-CHEC-NEXT        MaxLifetimeAccessDensity: 20000
-CHEC-NEXT        AccessHistogramSize: 3
-CHEC-NEXT        AccessHistogram: {{[0-9]+}}
-CHEC-NEXT        AccessHistogramValues: -2 -1 -2
-CHEC-NEXT    -
-CHEC-NEXT      Callstack:
-CHEC-NEXT      -
-CHEC-NEXT        Function: {{[0-9]+}}
-CHEC-NEXT        SymbolName: main
-CHEC-NEXT        LineOffset: 10
-CHEC-NEXT        Column: 10
-CHEC-NEXT        Inline: 0
-CHEC-NEXT      MemInfoBlock:
-CHEC-NEXT        AllocCount: 1
-CHEC-NEXT        TotalAccessCount: 4
-CHEC-NEXT        MinAccessCount: 4
-CHEC-NEXT        MaxAccessCount: 4
-CHEC-NEXT        TotalSize: 48
-CHEC-NEXT        MinSize: 48
-CHEC-NEXT        MaxSize: 48
-CHEC-NEXT        AllocTimestamp: {{[0-9]+}}
-CHEC-NEXT        DeallocTimestamp: {{[0-9]+}}
-CHEC-NEXT        TotalLifetime: 0
-CHEC-NEXT        MinLifetime: 0
-CHEC-NEXT        MaxLifetime: 0
-CHEC-NEXT        AllocCpuId: 11
-CHEC-NEXT        DeallocCpuId: 11
-CHEC-NEXT        NumMigratedCpu: 0
-CHEC-NEXT        NumLifetimeOverlaps: 0
-CHEC-NEXT        NumSameAllocCpu: 0
-CHEC-NEXT        NumSameDeallocCpu: 0
-CHEC-NEXT        DataTypeId: 0
-CHEC-NEXT        TotalAccessDensity: 8
-CHEC-NEXT        MinAccessDensity: 8
-CHEC-NEXT        MaxAccessDensity: 8
-CHEC-NEXT        TotalLifetimeAccessDensity: 8000
-CHEC-NEXT        MinLifetimeAccessDensity: 8000
-CHEC-NEXT        MaxLifetimeAccessDensity: 8000
-CHEC-NEXT        AccessHistogramSize: 6
-CHEC-NEXT        AccessHistogram: {{[0-9]+}}
-CHEC-NEXT        AccessHistogramValues: -2 -0 -0 -0 -1 -1
+CHECK-NEXT    FunctionGUID: {{[0-9]+}}
+CHECK-NEXT    AllocSites:
+CHECK-NEXT    -
+CHECK-NEXT      Callstack:
+CHECK-NEXT      -
+CHECK-NEXT        Function: {{[0-9]+}}
+CHECK-NEXT        SymbolName: main
+CHECK-NEXT        LineOffset: 3
+CHECK-NEXT        Column: 10
+CHECK-NEXT        Inline: 0
+CHECK-NEXT      MemInfoBlock:
+CHECK-NEXT        AllocCount: 1
+CHECK-NEXT        TotalAccessCount: 5
+CHECK-NEXT        MinAccessCount: 5
+CHECK-NEXT        MaxAccessCount: 5
+CHECK-NEXT        TotalSize: 24
+CHECK-NEXT        MinSize: 24
+CHECK-NEXT        MaxSize: 24
+CHECK-NEXT        AllocTimestamp: {{[0-9]+}}
+CHECK-NEXT        DeallocTimestamp: {{[0-9]+}}
+CHECK-NEXT        TotalLifetime: 0
+CHECK-NEXT        MinLifetime: 0
+CHECK-NEXT        MaxLifetime: 0
+CHECK-NEXT        AllocCpuId: 11
+CHECK-NEXT        DeallocCpuId: 11
+CHECK-NEXT        NumMigratedCpu: 0
+CHECK-NEXT        NumLifetimeOverlaps: 0
+CHECK-NEXT        NumSameAllocCpu: 0
+CHECK-NEXT        NumSameDeallocCpu: 0
+CHECK-NEXT        DataTypeId: 0
+CHECK-NEXT        TotalAccessDensity: 20
+CHECK-NEXT        MinAccessDensity: 20
+CHECK-NEXT        MaxAccessDensity: 20
+CHECK-NEXT        TotalLifetimeAccessDensity: 20000
+CHECK-NEXT        MinLifetimeAccessDensity: 20000
+CHECK-NEXT        MaxLifetimeAccessDensity: 20000
+CHECK-NEXT        AccessHistogramSize: 3
+CHECK-NEXT        AccessHistogram: {{[0-9]+}}
+CHECK-NEXT        AccessHistogramValues: -2 -1 -2
+CHECK-NEXT    -
+CHECK-NEXT      Callstack:
+CHECK-NEXT      -
+CHECK-NEXT        Function: {{[0-9]+}}
+CHECK-NEXT        SymbolName: main
+CHECK-NEXT        LineOffset: 10
+CHECK-NEXT        Column: 10
+CHECK-NEXT        Inline: 0
+CHECK-NEXT      MemInfoBlock:
+CHECK-NEXT        AllocCount: 1
+CHECK-NEXT        TotalAccessCount: 4
+CHECK-NEXT        MinAccessCount: 4
+CHECK-NEXT        MaxAccessCount: 4
+CHECK-NEXT        TotalSize: 48
+CHECK-NEXT        MinSize: 48
+CHECK-NEXT        MaxSize: 48
+CHECK-NEXT        AllocTimestamp: {{[0-9]+}}
+CHECK-NEXT        DeallocTimestamp: {{[0-9]+}}
+CHECK-NEXT        TotalLifetime: 0
+CHECK-NEXT        MinLifetime: 0
+CHECK-NEXT        MaxLifetime: 0
+CHECK-NEXT        AllocCpuId: 11
+CHECK-NEXT        DeallocCpuId: 11
+CHECK-NEXT        NumMigratedCpu: 0
+CHECK-NEXT        NumLifetimeOverlaps: 0
+CHECK-NEXT        NumSameAllocCpu: 0
+CHECK-NEXT        NumSameDeallocCpu: 0
+CHECK-NEXT        DataTypeId: 0
+CHECK-NEXT        TotalAccessDensity: 8
+CHECK-NEXT        MinAccessDensity: 8
+CHECK-NEXT        MaxAccessDensity: 8
+CHECK-NEXT        TotalLifetimeAccessDensity: 8000
+CHECK-NEXT        MinLifetimeAccessDensity: 8000
+CHECK-NEXT        MaxLifetimeAccessDensity: 8000
+CHECK-NEXT        AccessHistogramSize: 6
+CHECK-NEXT        AccessHistogram: {{[0-9]+}}
+CHECK-NEXT        AccessHistogramValues: -2 -0 -0 -0 -1 -1
diff --git a/llvm/tools/llvm-exegesis/lib/X86/Target.cpp b/llvm/tools/llvm-exegesis/lib/X86/Target.cpp
index 1659cfb..5dae6c0 100644
--- a/llvm/tools/llvm-exegesis/lib/X86/Target.cpp
+++ b/llvm/tools/llvm-exegesis/lib/X86/Target.cpp
@@ -30,11 +30,12 @@
 #include <memory>
 #include <string>
 #include <vector>
-#if defined(_MSC_VER) && (defined(_M_IX86) || defined(_M_X64))
+#if defined(_MSC_VER) && (defined(_M_IX86) || defined(_M_X64)) &&              \
+    !defined(_M_ARM64EC)
 #include <immintrin.h>
 #include <intrin.h>
 #endif
-#if defined(_MSC_VER) && defined(_M_X64)
+#if defined(_MSC_VER) && defined(_M_X64) && !defined(_M_ARM64EC)
 #include <float.h> // For _clearfp in ~X86SavedState().
 #endif
 
@@ -654,7 +655,7 @@ namespace {
 class X86SavedState : public ExegesisTarget::SavedState {
 public:
   X86SavedState() {
-#if defined(_MSC_VER) && defined(_M_X64)
+#if defined(_MSC_VER) && defined(_M_X64) && !defined(_M_ARM64EC)
     _fxsave64(FPState);
     Eflags = __readeflags();
 #elif defined(__GNUC__) && defined(__x86_64__)
@@ -668,7 +669,7 @@ public:
   ~X86SavedState() {
     // Restoring the X87 state does not flush pending exceptions, make sure
     // these exceptions are flushed now.
-#if defined(_MSC_VER) && defined(_M_X64)
+#if defined(_MSC_VER) && defined(_M_X64) && !defined(_M_ARM64EC)
     _clearfp();
     _fxrstor64(FPState);
     __writeeflags(Eflags);
@@ -682,7 +683,7 @@ public:
   }
 
 private:
-#if defined(__x86_64__) || defined(_M_X64)
+#if defined(__x86_64__) || defined(_M_X64) && !defined(_M_ARM64EC)
   alignas(16) char FPState[512];
   uint64_t Eflags;
 #endif
@@ -824,8 +825,9 @@ private:
       // For now, only do the check if we see an Intel machine because
       // the counter uses some intel-specific magic and it could
       // be confuse and think an AMD machine actually has LBR support.
-#if defined(__i386__) || defined(_M_IX86) || defined(__x86_64__) ||            \
-    defined(_M_X64)
+#if (defined(__i386__) || defined(_M_IX86) || defined(__x86_64__) ||           \
+     defined(_M_X64)) &&                                                       \
+    !defined(_M_ARM64EC)
     using namespace sys::detail::x86;
 
     if (getVendorSignature() == VendorSignatures::GENUINE_INTEL)
diff --git a/llvm/unittests/Analysis/InlineAdvisorPlugin/CMakeLists.txt b/llvm/unittests/Analysis/InlineAdvisorPlugin/CMakeLists.txt
index deabf11..d9da627 100644
--- a/llvm/unittests/Analysis/InlineAdvisorPlugin/CMakeLists.txt
+++ b/llvm/unittests/Analysis/InlineAdvisorPlugin/CMakeLists.txt
@@ -2,7 +2,7 @@
 # libraries, but expects them to exist in the process loading the plugin. This 
 # doesn't work with DLLs on Windows (where a shared library can't have undefined
 # references), so just skip this testcase on Windows.
-if ((NOT WIN32 OR LLVM_BUILD_LLVM_DYLIB) AND NOT CYGWIN)
+if ((NOT WIN32 AND NOT CYGWIN) OR LLVM_BUILD_LLVM_DYLIB)
   unset(LLVM_LINK_COMPONENTS)
   add_llvm_library(InlineAdvisorPlugin MODULE BUILDTREE_ONLY
     InlineAdvisorPlugin.cpp
diff --git a/llvm/unittests/Analysis/InlineOrderPlugin/CMakeLists.txt b/llvm/unittests/Analysis/InlineOrderPlugin/CMakeLists.txt
index 0b37ceb..941e18efc 100644
--- a/llvm/unittests/Analysis/InlineOrderPlugin/CMakeLists.txt
+++ b/llvm/unittests/Analysis/InlineOrderPlugin/CMakeLists.txt
@@ -2,7 +2,7 @@
 # libraries, but expects them to exist in the process loading the plugin. This 
 # doesn't work with DLLs on Windows (where a shared library can't have undefined
 # references), so just skip this testcase on Windows.
-if ((NOT WIN32 OR LLVM_BUILD_LLVM_DYLIB) AND NOT CYGWIN)
+if ((NOT WIN32 AND NOT CYGWIN) OR LLVM_BUILD_LLVM_DYLIB)
   unset(LLVM_LINK_COMPONENTS)
   add_llvm_library(InlineOrderPlugin MODULE BUILDTREE_ONLY
     InlineOrderPlugin.cpp
diff --git a/llvm/unittests/Frontend/HLSLBindingTest.cpp b/llvm/unittests/Frontend/HLSLBindingTest.cpp
index bd6a0ff..ca2f7b5 100644
--- a/llvm/unittests/Frontend/HLSLBindingTest.cpp
+++ b/llvm/unittests/Frontend/HLSLBindingTest.cpp
@@ -155,8 +155,10 @@ TEST(HLSLBindingTest, TestExactOverlap) {
 
   // Since the bindings overlap exactly we need sigil values to differentiate
   // them.
-  char ID1;
-  char ID2;
+  // Note: We initialize these to 0 to suppress a -Wuninitialized-const-pointer,
+  // but we really are just using the stack addresses here.
+  char ID1 = 0;
+  char ID2 = 0;
 
   // StructuredBuffer<float> A  : register(t5);
   // StructuredBuffer<float> B  : register(t5);
diff --git a/llvm/unittests/Transforms/Vectorize/VPlanVerifierTest.cpp b/llvm/unittests/Transforms/Vectorize/VPlanVerifierTest.cpp
index 2091220..9e06c42 100644
--- a/llvm/unittests/Transforms/Vectorize/VPlanVerifierTest.cpp
+++ b/llvm/unittests/Transforms/Vectorize/VPlanVerifierTest.cpp
@@ -312,9 +312,14 @@ TEST_F(VPVerifierTest, NonHeaderPHIInHeader) {
 #endif
   EXPECT_FALSE(verifyVPlanIsValid(Plan));
 #if GTEST_HAS_STREAM_REDIRECTION
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
   EXPECT_STREQ(
       "Found non-header PHI recipe in header VPBB: IR   <badref> = phi i32 \n",
       ::testing::internal::GetCapturedStderr().c_str());
+#else
+  EXPECT_STREQ("Found non-header PHI recipe in header VPBB",
+               ::testing::internal::GetCapturedStderr().c_str());
+#endif
 #endif
 }
 
diff --git a/llvm/utils/gn/secondary/llvm/lib/Analysis/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/Analysis/BUILD.gn
index 3d08c3f..d394923 100644
--- a/llvm/utils/gn/secondary/llvm/lib/Analysis/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/Analysis/BUILD.gn
@@ -8,6 +8,7 @@ static_library("Analysis") {
     "//llvm/include/llvm/Config:config",
     "//llvm/lib/BinaryFormat",
     "//llvm/lib/IR",
+    "//llvm/lib/Frontend/HLSL",
     "//llvm/lib/ProfileData",
     "//llvm/lib/Support",
     "//llvm/lib/TargetParser",
diff --git a/llvm/utils/gn/secondary/llvm/lib/Frontend/HLSL/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/Frontend/HLSL/BUILD.gn
index 4c1c613..fce564e 100644
--- a/llvm/utils/gn/secondary/llvm/lib/Frontend/HLSL/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/Frontend/HLSL/BUILD.gn
@@ -6,6 +6,7 @@ static_library("HLSL") {
   ]
   sources = [
     "CBuffer.cpp",
+    "HLSLBinding.cpp",
     "HLSLResource.cpp",
     "HLSLRootSignature.cpp",
     "RootSignatureMetadata.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn b/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn
index c29277c..6890c48 100644
--- a/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn
@@ -13,6 +13,7 @@ unittest("LLVMFrontendTests") {
     "//llvm/lib/Testing/Support",
   ]
   sources = [
+    "HLSLBindingTest.cpp",
     "HLSLRootSignatureDumpTest.cpp",
     "HLSLRootSignatureRangesTest.cpp",
     "OpenACCTest.cpp",
diff --git a/llvm/utils/lit/lit/Test.py b/llvm/utils/lit/lit/Test.py
index 1bd5ba8..7290977 100644
--- a/llvm/utils/lit/lit/Test.py
+++ b/llvm/utils/lit/lit/Test.py
@@ -247,6 +247,9 @@ class Test:
         # and will be honored when the test result is supplied.
         self.xfails = []
 
+        # Exclude this test if it's xfail.
+        self.exclude_xfail = False
+
         # If true, ignore all items in self.xfails.
         self.xfail_not = False
 
diff --git a/llvm/utils/lit/lit/TestRunner.py b/llvm/utils/lit/lit/TestRunner.py
index 73db67a..e7cd707 100644
--- a/llvm/utils/lit/lit/TestRunner.py
+++ b/llvm/utils/lit/lit/TestRunner.py
@@ -2175,6 +2175,8 @@ def parseIntegratedTestScript(test, additional_parsers=[], require_script=True):
     assert parsed["DEFINE:"] == script
     assert parsed["REDEFINE:"] == script
     test.xfails += parsed["XFAIL:"] or []
+    if test.exclude_xfail and test.isExpectedToFail():
+        return lit.Test.Result(Test.EXCLUDED, "excluding XFAIL tests")
     test.requires += parsed["REQUIRES:"] or []
     test.unsupported += parsed["UNSUPPORTED:"] or []
     if parsed["ALLOW_RETRIES:"]:
diff --git a/llvm/utils/lit/lit/cl_arguments.py b/llvm/utils/lit/lit/cl_arguments.py
index 3292554..e889515 100644
--- a/llvm/utils/lit/lit/cl_arguments.py
+++ b/llvm/utils/lit/lit/cl_arguments.py
@@ -304,6 +304,16 @@ def parse_args():
         default=os.environ.get("LIT_XFAIL_NOT", ""),
     )
     selection_group.add_argument(
+        "--exclude-xfail",
+        help="exclude XFAIL tests (unless they are in the --xfail-not list). "
+        "Note: This option is implemented in "
+        "lit.TestRunner.parseIntegratedTestScript and so will have no effect on "
+        "test formats that do not call that and do not implement the option "
+        "separately.",
+        default=False,
+        action="store_true",
+    )
+    selection_group.add_argument(
         "--num-shards",
         dest="numShards",
         metavar="M",
diff --git a/llvm/utils/lit/lit/main.py b/llvm/utils/lit/lit/main.py
index 0939838..9650a0e 100755
--- a/llvm/utils/lit/lit/main.py
+++ b/llvm/utils/lit/lit/main.py
@@ -240,6 +240,8 @@ def mark_xfail(selected_tests, opts):
             t.xfails += "*"
         if test_file in opts.xfail_not or test_full_name in opts.xfail_not:
             t.xfail_not = True
+        if opts.exclude_xfail:
+            t.exclude_xfail = True
 
 
 def mark_excluded(discovered_tests, selected_tests):
diff --git a/llvm/utils/lit/tests/Inputs/xfail-cl/true-xfail-conditionally.txt b/llvm/utils/lit/tests/Inputs/xfail-cl/true-xfail-conditionally.txt
new file mode 100644
index 0000000..6fdecd6
--- /dev/null
+++ b/llvm/utils/lit/tests/Inputs/xfail-cl/true-xfail-conditionally.txt
@@ -0,0 +1,2 @@
+# XFAIL: this-does-not-exist
+# RUN: true
\ No newline at end of file
diff --git a/llvm/utils/lit/tests/xfail-cl.py b/llvm/utils/lit/tests/xfail-cl.py
index ef1bb04..f1e0e33 100644
--- a/llvm/utils/lit/tests/xfail-cl.py
+++ b/llvm/utils/lit/tests/xfail-cl.py
@@ -5,6 +5,18 @@
 # RUN:   %{inputs}/xfail-cl \
 # RUN: | FileCheck --check-prefix=CHECK-FILTER %s
 
+# RUN: %{lit} --xfail 'false.txt;false2.txt;top-level-suite :: b :: test.txt' \
+# RUN:   --exclude-xfail \
+# RUN:   %{inputs}/xfail-cl \
+# RUN: | FileCheck --check-prefixes=CHECK-EXCLUDED,CHECK-EXCLUDED-NOOVERRIDE %s
+
+# RUN: %{lit} --xfail 'false.txt;false2.txt;top-level-suite :: b :: test.txt' \
+# RUN:   --xfail-not 'true-xfail.txt' \
+# RUN:   --exclude-xfail \
+# RUN:   %{inputs}/xfail-cl \
+# RUN: | FileCheck --check-prefixes=CHECK-EXCLUDED,CHECK-EXCLUDED-OVERRIDE %s
+
+
 # RUN: env LIT_XFAIL='false.txt;false2.txt;top-level-suite :: b :: test.txt' \
 # RUN:   LIT_XFAIL_NOT='true-xfail.txt;top-level-suite :: a :: test-xfail.txt' \
 # RUN: %{lit} %{inputs}/xfail-cl \
@@ -23,7 +35,7 @@
 
 # END.
 
-# CHECK-FILTER: Testing: 10 tests, {{[0-9]*}} workers
+# CHECK-FILTER: Testing: 11 tests, {{[0-9]*}} workers
 # CHECK-FILTER-DAG: {{^}}PASS: top-level-suite :: a :: test.txt
 # CHECK-FILTER-DAG: {{^}}XFAIL: top-level-suite :: b :: test.txt
 # CHECK-FILTER-DAG: {{^}}XFAIL: top-level-suite :: a :: false.txt
@@ -37,3 +49,17 @@
 
 # CHECK-OVERRIDE: Testing: 1 tests, {{[0-9]*}} workers
 # CHECK-OVERRIDE: {{^}}PASS: top-level-suite :: true-xfail.txt
+
+# CHECK-EXCLUDED: Testing: 11 tests, {{[0-9]*}} workers
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: a :: false.txt
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: a :: test-xfail.txt
+# CHECK-EXCLUDED-DAG: {{^}}PASS: top-level-suite :: a :: test.txt
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: b :: false.txt
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: b :: test-xfail.txt
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: b :: test.txt
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: false.txt
+# CHECK-EXCLUDED-DAG: {{^}}EXCLUDED: top-level-suite :: false2.txt
+# CHECK-EXCLUDED-DAG: {{^}}PASS: top-level-suite :: true-xfail-conditionally.txt
+# CHECK-EXCLUDED-NOOVERRIDE-DAG: {{^}}EXCLUDED: top-level-suite :: true-xfail.txt
+# CHECK-EXCLUDED-OVERRIDE-DAG: {{^}}PASS: top-level-suite :: true-xfail.txt
+# CHECK-EXCLUDED-DAG: {{^}}PASS: top-level-suite :: true.txt
diff --git a/mlir/include/mlir/Dialect/SCF/IR/SCFOps.td b/mlir/include/mlir/Dialect/SCF/IR/SCFOps.td
index 2d15544..0c1c15b 100644
--- a/mlir/include/mlir/Dialect/SCF/IR/SCFOps.td
+++ b/mlir/include/mlir/Dialect/SCF/IR/SCFOps.td
@@ -87,6 +87,9 @@ def ExecuteRegionOp : SCF_Op<"execute_region", [
     be accessed inside the op. The op's region can have multiple blocks and the
     blocks can have multiple distinct terminators. Values returned from this op's
     region define the op's results.
+    The optional 'no_inline' flag can be set to request the ExecuteRegionOp to be
+    preserved as much as possible and not being inlined in the parent block until
+    an explicit lowering step.
 
     Example:
 
@@ -98,6 +101,14 @@ def ExecuteRegionOp : SCF_Op<"execute_region", [
       }
     }
 
+    // the same as above but with no_inline attribute
+    scf.for %i = 0 to 128 step %c1 {
+      %y = scf.execute_region -> i32 no_inline {
+        %x = load %A[%i] : memref<128xi32>
+        scf.yield %x : i32
+      }
+    }
+
     affine.for %i = 0 to 100 {
       "foo"() : () -> ()
       %v = scf.execute_region -> i64 {
@@ -119,6 +130,10 @@ def ExecuteRegionOp : SCF_Op<"execute_region", [
     ```
   }];
 
+  let arguments = (ins
+    UnitAttr:$no_inline
+  );
+
   let results = (outs Variadic<AnyType>);
 
   let regions = (region AnyRegion:$region);
diff --git a/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVBase.td b/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVBase.td
index 9038326..9c74cff0 100644
--- a/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVBase.td
+++ b/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVBase.td
@@ -4448,6 +4448,7 @@ def SPIRV_OC_OpUMulExtended                   : I32EnumAttrCase<"OpUMulExtended"
 def SPIRV_OC_OpSMulExtended                   : I32EnumAttrCase<"OpSMulExtended", 152>;
 def SPIRV_OC_OpIsNan                          : I32EnumAttrCase<"OpIsNan", 156>;
 def SPIRV_OC_OpIsInf                          : I32EnumAttrCase<"OpIsInf", 157>;
+def SPIRV_OC_OpIsFinite                       : I32EnumAttrCase<"OpIsFinite", 158>;
 def SPIRV_OC_OpOrdered                        : I32EnumAttrCase<"OpOrdered", 162>;
 def SPIRV_OC_OpUnordered                      : I32EnumAttrCase<"OpUnordered", 163>;
 def SPIRV_OC_OpLogicalEqual                   : I32EnumAttrCase<"OpLogicalEqual", 164>;
@@ -4630,7 +4631,8 @@ def SPIRV_OpcodeAttr :
       SPIRV_OC_OpVectorTimesMatrix, SPIRV_OC_OpMatrixTimesVector,
       SPIRV_OC_OpMatrixTimesMatrix, SPIRV_OC_OpDot, SPIRV_OC_OpIAddCarry,
       SPIRV_OC_OpISubBorrow, SPIRV_OC_OpUMulExtended, SPIRV_OC_OpSMulExtended,
-      SPIRV_OC_OpIsNan, SPIRV_OC_OpIsInf, SPIRV_OC_OpOrdered, SPIRV_OC_OpUnordered,
+      SPIRV_OC_OpIsNan, SPIRV_OC_OpIsInf, SPIRV_OC_OpIsFinite,
+      SPIRV_OC_OpOrdered, SPIRV_OC_OpUnordered,
       SPIRV_OC_OpLogicalEqual, SPIRV_OC_OpLogicalNotEqual, SPIRV_OC_OpLogicalOr,
       SPIRV_OC_OpLogicalAnd, SPIRV_OC_OpLogicalNot, SPIRV_OC_OpSelect,
       SPIRV_OC_OpIEqual, SPIRV_OC_OpINotEqual, SPIRV_OC_OpUGreaterThan,
diff --git a/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVLogicalOps.td b/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVLogicalOps.td
index ab535d7..9331fc5 100644
--- a/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVLogicalOps.td
+++ b/mlir/include/mlir/Dialect/SPIRV/IR/SPIRVLogicalOps.td
@@ -403,6 +403,28 @@ def SPIRV_INotEqualOp : SPIRV_LogicalBinaryOp<"INotEqual",
 
 // -----
 
+def SPIRV_IsFiniteOp : SPIRV_LogicalUnaryOp<"IsFinite", SPIRV_Float, []> {
+  let summary = "Result is true if x is an IEEE Finite, otherwise result is false";
+
+  let description = [{
+    Result Type must be a scalar or vector of Boolean type.
+
+    x must be a scalar or vector of floating-point type.  It must have the
+    same number of components as Result Type.
+
+    Results are computed per component.
+
+    #### Example:
+
+    ```mlir
+    %2 = spirv.IsFinite %0: f32
+    %3 = spirv.IsFinite %1: vector<4xf32>
+    ```
+  }];
+}
+
+// -----
+
 def SPIRV_IsInfOp : SPIRV_LogicalUnaryOp<"IsInf", SPIRV_Float, []> {
   let summary = "Result is true if x is an IEEE Inf, otherwise result is false";
 
@@ -418,7 +440,7 @@ def SPIRV_IsInfOp : SPIRV_LogicalUnaryOp<"IsInf", SPIRV_Float, []> {
 
     ```mlir
     %2 = spirv.IsInf %0: f32
-    %3 = spirv.IsInf %1: vector<4xi32>
+    %3 = spirv.IsInf %1: vector<4xf32>
     ```
   }];
 }
@@ -442,7 +464,7 @@ def SPIRV_IsNanOp : SPIRV_LogicalUnaryOp<"IsNan", SPIRV_Float, []> {
 
     ```mlir
     %2 = spirv.IsNan %0: f32
-    %3 = spirv.IsNan %1: vector<4xi32>
+    %3 = spirv.IsNan %1: vector<4xf32>
     ```
   }];
 }
diff --git a/mlir/include/mlir/Dialect/Vector/IR/VectorOps.td b/mlir/include/mlir/Dialect/Vector/IR/VectorOps.td
index 3885439..5d45508 100644
--- a/mlir/include/mlir/Dialect/Vector/IR/VectorOps.td
+++ b/mlir/include/mlir/Dialect/Vector/IR/VectorOps.td
@@ -2595,6 +2595,7 @@ def Vector_MaskOp : Vector_Op<"mask", [
 
 def Vector_TransposeOp :
   Vector_Op<"transpose", [Pure,
+    DeclareOpInterfaceMethods<InferIntRangeInterface, ["inferResultRanges"]>,
     DeclareOpInterfaceMethods<VectorUnrollOpInterface, ["getShapeForUnroll"]>,
     PredOpTrait<"operand and result have same element type",
                  TCresVTEtIsSameAsOpBase<0, 0>>]> {
@@ -2876,7 +2877,10 @@ def Vector_ScanOp :
 // VectorStepOp
 //===----------------------------------------------------------------------===//
 
-def Vector_StepOp : Vector_Op<"step", [Pure]> {
+def Vector_StepOp : Vector_Op<"step", [
+    Pure,
+    DeclareOpInterfaceMethods<InferIntRangeInterface, ["inferResultRanges"]>
+  ]> {
   let summary = "A linear sequence of values from 0 to N";
   let description = [{
     A `step` operation produces an index vector, i.e. a 1-D vector of values of
diff --git a/mlir/lib/Conversion/MathToSPIRV/MathToSPIRV.cpp b/mlir/lib/Conversion/MathToSPIRV/MathToSPIRV.cpp
index a877ad2..1787e0a 100644
--- a/mlir/lib/Conversion/MathToSPIRV/MathToSPIRV.cpp
+++ b/mlir/lib/Conversion/MathToSPIRV/MathToSPIRV.cpp
@@ -488,7 +488,12 @@ namespace mlir {
 void populateMathToSPIRVPatterns(const SPIRVTypeConverter &typeConverter,
                                  RewritePatternSet &patterns) {
   // Core patterns
-  patterns.add<CopySignPattern>(typeConverter, patterns.getContext());
+  patterns
+      .add<CopySignPattern,
+           CheckedElementwiseOpPattern<math::IsInfOp, spirv::IsInfOp>,
+           CheckedElementwiseOpPattern<math::IsNaNOp, spirv::IsNanOp>,
+           CheckedElementwiseOpPattern<math::IsFiniteOp, spirv::IsFiniteOp>>(
+          typeConverter, patterns.getContext());
 
   // GLSL patterns
   patterns
diff --git a/mlir/lib/Dialect/SCF/IR/SCF.cpp b/mlir/lib/Dialect/SCF/IR/SCF.cpp
index 759e58b..0262a1b 100644
--- a/mlir/lib/Dialect/SCF/IR/SCF.cpp
+++ b/mlir/lib/Dialect/SCF/IR/SCF.cpp
@@ -137,6 +137,9 @@ ParseResult ExecuteRegionOp::parse(OpAsmParser &parser,
   if (parser.parseOptionalArrowTypeList(result.types))
     return failure();
 
+  if (succeeded(parser.parseOptionalKeyword("no_inline")))
+    result.addAttribute("no_inline", parser.getBuilder().getUnitAttr());
+
   // Introduce the body region and parse it.
   Region *body = result.addRegion();
   if (parser.parseRegion(*body, /*arguments=*/{}, /*argTypes=*/{}) ||
@@ -148,8 +151,9 @@ ParseResult ExecuteRegionOp::parse(OpAsmParser &parser,
 
 void ExecuteRegionOp::print(OpAsmPrinter &p) {
   p.printOptionalArrowTypeList(getResultTypes());
-
   p << ' ';
+  if (getNoInline())
+    p << "no_inline ";
   p.printRegion(getRegion(),
                 /*printEntryBlockArgs=*/false,
                 /*printBlockTerminators=*/true);
@@ -184,7 +188,7 @@ struct SingleBlockExecuteInliner : public OpRewritePattern<ExecuteRegionOp> {
 
   LogicalResult matchAndRewrite(ExecuteRegionOp op,
                                 PatternRewriter &rewriter) const override {
-    if (!op.getRegion().hasOneBlock())
+    if (!op.getRegion().hasOneBlock() || op.getNoInline())
       return failure();
     replaceOpWithRegion(rewriter, op, op.getRegion());
     return success();
diff --git a/mlir/lib/Dialect/Vector/IR/VectorOps.cpp b/mlir/lib/Dialect/Vector/IR/VectorOps.cpp
index 8789f55..a21b5ba 100644
--- a/mlir/lib/Dialect/Vector/IR/VectorOps.cpp
+++ b/mlir/lib/Dialect/Vector/IR/VectorOps.cpp
@@ -5916,14 +5916,13 @@ OpFoldResult ShapeCastOp::fold(FoldAdaptor adaptor) {
   }
 
   // shape_cast(constant) -> constant
-  if (auto splatAttr =
-          llvm::dyn_cast_if_present<SplatElementsAttr>(adaptor.getSource()))
-    return splatAttr.reshape(getType());
+  if (auto denseAttr =
+          dyn_cast_if_present<DenseElementsAttr>(adaptor.getSource()))
+    return denseAttr.reshape(getType());
 
   // shape_cast(poison) -> poison
-  if (llvm::dyn_cast_if_present<ub::PoisonAttr>(adaptor.getSource())) {
+  if (llvm::dyn_cast_if_present<ub::PoisonAttr>(adaptor.getSource()))
     return ub::PoisonAttr::get(getContext());
-  }
 
   return {};
 }
@@ -6316,6 +6315,11 @@ std::optional<SmallVector<int64_t, 4>> TransposeOp::getShapeForUnroll() {
   return llvm::to_vector<4>(getResultVectorType().getShape());
 }
 
+void TransposeOp::inferResultRanges(ArrayRef<ConstantIntRanges> argRanges,
+                                    SetIntRangeFn setResultRanges) {
+  setResultRanges(getResult(), argRanges.front());
+}
+
 namespace {
 
 // Rewrites two back-to-back TransposeOp operations into a single TransposeOp.
@@ -7198,6 +7202,23 @@ Value mlir::vector::makeArithReduction(OpBuilder &b, Location loc,
 }
 
 //===----------------------------------------------------------------------===//
+// StepOp
+//===----------------------------------------------------------------------===//
+
+void StepOp::inferResultRanges(ArrayRef<ConstantIntRanges> argRanges,
+                               SetIntRangeFn setResultRanges) {
+  auto resultType = cast<VectorType>(getType());
+  if (resultType.isScalable()) {
+    return;
+  }
+  unsigned bitwidth = ConstantIntRanges::getStorageBitwidth(resultType);
+  APInt zero(bitwidth, 0);
+  APInt high(bitwidth, resultType.getDimSize(0) - 1);
+  ConstantIntRanges result = {zero, high, zero, high};
+  setResultRanges(getResult(), result);
+}
+
+//===----------------------------------------------------------------------===//
 // Vector Masking Utilities
 //===----------------------------------------------------------------------===//
 
diff --git a/mlir/test/Conversion/MathToSPIRV/math-to-fpclassify-spirv.mlir b/mlir/test/Conversion/MathToSPIRV/math-to-fpclassify-spirv.mlir
new file mode 100644
index 0000000..3e5f592
--- /dev/null
+++ b/mlir/test/Conversion/MathToSPIRV/math-to-fpclassify-spirv.mlir
@@ -0,0 +1,27 @@
+// RUN: mlir-opt --convert-math-to-spirv %s | FileCheck %s
+
+module attributes {
+  spirv.target_env = #spirv.target_env<#spirv.vce<v1.0, [Shader], []>, #spirv.resource_limits<>>
+} {
+
+  // CHECK-LABEL: @fpclassify
+  func.func @fpclassify(%x: f32, %v: vector<4xf32>) {
+    // CHECK: spirv.IsFinite %{{.*}} : f32
+    %0 = math.isfinite %x : f32
+    // CHECK: spirv.IsFinite %{{.*}} : vector<4xf32>
+    %1 = math.isfinite %v : vector<4xf32>
+
+    // CHECK: spirv.IsNan %{{.*}} : f32
+    %2 = math.isnan %x : f32
+    // CHECK: spirv.IsNan %{{.*}} : vector<4xf32>
+    %3 = math.isnan %v : vector<4xf32>
+
+    // CHECK: spirv.IsInf %{{.*}} : f32
+    %4 = math.isinf %x : f32
+    // CHECK: spirv.IsInf %{{.*}} : vector<4xf32>
+    %5 = math.isinf %v : vector<4xf32>
+
+    return
+  }
+
+}
diff --git a/mlir/test/Dialect/SCF/canonicalize.mlir b/mlir/test/Dialect/SCF/canonicalize.mlir
index 12d30e17..308cf150 100644
--- a/mlir/test/Dialect/SCF/canonicalize.mlir
+++ b/mlir/test/Dialect/SCF/canonicalize.mlir
@@ -1440,8 +1440,8 @@ func.func @propagate_into_execute_region() {
 
 // -----
 
-// CHECK-LABEL: func @execute_region_elim
-func.func @execute_region_elim() {
+// CHECK-LABEL: func @execute_region_inline
+func.func @execute_region_inline() {
   affine.for %i = 0 to 100 {
     "test.foo"() : () -> ()
     %v = scf.execute_region -> i64 {
@@ -1461,8 +1461,30 @@ func.func @execute_region_elim() {
 
 // -----
 
-// CHECK-LABEL: func @func_execute_region_elim
-func.func @func_execute_region_elim() {
+// CHECK-LABEL: func @execute_region_no_inline
+func.func @execute_region_no_inline() {
+  affine.for %i = 0 to 100 {
+    "test.foo"() : () -> ()
+    %v = scf.execute_region -> i64 no_inline {
+      %x = "test.val"() : () -> i64
+      scf.yield %x : i64
+    }
+    "test.bar"(%v) : (i64) -> ()
+  }
+  return
+}
+
+// CHECK-NEXT:     affine.for %arg0 = 0 to 100 {
+// CHECK-NEXT:       "test.foo"() : () -> ()
+// CHECK-NEXT:       scf.execute_region
+// CHECK-NEXT:       %[[VAL:.*]] = "test.val"() : () -> i64
+// CHECK-NEXT:       scf.yield %[[VAL]] : i64
+// CHECK-NEXT:     }
+
+// -----
+
+// CHECK-LABEL: func @func_execute_region_inline
+func.func @func_execute_region_inline() {
     "test.foo"() : () -> ()
     %v = scf.execute_region -> i64 {
       %c = "test.cmp"() : () -> i1
@@ -1496,8 +1518,8 @@ func.func @func_execute_region_elim() {
 
 // -----
 
-// CHECK-LABEL: func @func_execute_region_elim_multi_yield
-func.func @func_execute_region_elim_multi_yield() {
+// CHECK-LABEL: func @func_execute_region_inline_multi_yield
+func.func @func_execute_region_inline_multi_yield() {
     "test.foo"() : () -> ()
     %v = scf.execute_region -> i64 {
       %c = "test.cmp"() : () -> i1
diff --git a/mlir/test/Dialect/SPIRV/IR/logical-ops.mlir b/mlir/test/Dialect/SPIRV/IR/logical-ops.mlir
index d6c3464..58b8288 100644
--- a/mlir/test/Dialect/SPIRV/IR/logical-ops.mlir
+++ b/mlir/test/Dialect/SPIRV/IR/logical-ops.mlir
@@ -33,6 +33,24 @@ func.func @inotequal_vector(%arg0: vector<4xi32>, %arg1: vector<4xi32>) -> vecto
 // -----
 
 //===----------------------------------------------------------------------===//
+// spirv.IsFinite
+//===----------------------------------------------------------------------===//
+
+func.func @isfinite_scalar(%arg0: f32) -> i1 {
+  // CHECK: spirv.IsFinite {{.*}} : f32
+  %0 = spirv.IsFinite %arg0 : f32
+  return %0 : i1
+}
+
+func.func @isfinite_vector(%arg0: vector<2xf32>) -> vector<2xi1> {
+  // CHECK: spirv.IsFinite {{.*}} : vector<2xf32>
+  %0 = spirv.IsFinite %arg0 : vector<2xf32>
+  return %0 : vector<2xi1>
+}
+
+// -----
+
+//===----------------------------------------------------------------------===//
 // spirv.IsInf
 //===----------------------------------------------------------------------===//
 
diff --git a/mlir/test/Dialect/Vector/canonicalize.mlir b/mlir/test/Dialect/Vector/canonicalize.mlir
index 9cfebd5..56996b5 100644
--- a/mlir/test/Dialect/Vector/canonicalize.mlir
+++ b/mlir/test/Dialect/Vector/canonicalize.mlir
@@ -1330,11 +1330,11 @@ func.func @fold_consecutive_broadcasts(%a : i32) -> vector<4x16xi32> {
 
 // -----
 
-// CHECK-LABEL: shape_cast_constant
+// CHECK-LABEL: shape_cast_splat_constant
 //       CHECK-DAG: %[[CST1:.*]] = arith.constant dense<1> : vector<3x4x2xi32>
 //       CHECK-DAG: %[[CST0:.*]] = arith.constant dense<2.000000e+00> : vector<20x2xf32>
 //       CHECK: return %[[CST0]], %[[CST1]] : vector<20x2xf32>, vector<3x4x2xi32>
-func.func @shape_cast_constant() -> (vector<20x2xf32>, vector<3x4x2xi32>) {
+func.func @shape_cast_splat_constant() -> (vector<20x2xf32>, vector<3x4x2xi32>) {
   %cst = arith.constant dense<2.000000e+00> : vector<5x4x2xf32>
   %cst_1 = arith.constant dense<1> : vector<12x2xi32>
   %0 = vector.shape_cast %cst : vector<5x4x2xf32> to vector<20x2xf32>
@@ -1344,6 +1344,36 @@ func.func @shape_cast_constant() -> (vector<20x2xf32>, vector<3x4x2xi32>) {
 
 // -----
 
+// Test of shape_cast's fold method:
+// shape_cast(constant) -> constant.
+//
+// CHECK-LABEL: @shape_cast_dense_int_constant
+//               CHECK: %[[CST:.*]] = arith.constant
+// CHECK-SAME{LITERAL}: dense<[[2, 3, 5], [7, 11, 13]]>
+//               CHECK: return %[[CST]] : vector<2x3xi8>
+func.func @shape_cast_dense_int_constant() -> vector<2x3xi8> {
+  %cst = arith.constant dense<[2, 3, 5, 7, 11, 13]> : vector<6xi8>
+  %0 = vector.shape_cast %cst : vector<6xi8> to vector<2x3xi8>
+  return %0 : vector<2x3xi8>
+}
+
+// -----
+
+// Test of shape_cast fold's method:
+// (shape_cast(const_x), const_x) -> (const_x_folded, const_x)
+//
+// CHECK-LABEL: @shape_cast_dense_float_constant
+//  CHECK-DAG: %[[CST0:.*]] = {{.*}}1.000000e+00, 2.000000e+00{{.*}} vector<1x2xf32>
+//  CHECK-DAG: %[[CST1:.*]] = {{.*}}1.000000e+00, 2.000000e+00{{.*}} vector<2xf32>
+//      CHECK: return %[[CST1]], %[[CST0]] : vector<2xf32>, vector<1x2xf32>
+func.func @shape_cast_dense_float_constant() -> (vector<2xf32>, vector<1x2xf32>){
+  %cst = arith.constant dense<[[1.0, 2.0]]> : vector<1x2xf32>
+  %0 = vector.shape_cast %cst : vector<1x2xf32> to vector<2xf32>
+  return %0, %cst : vector<2xf32>, vector<1x2xf32>
+}
+
+// -----
+
 // CHECK-LABEL: shape_cast_poison
 //       CHECK-DAG: %[[CST1:.*]] = ub.poison : vector<3x4x2xi32>
 //       CHECK-DAG: %[[CST0:.*]] = ub.poison : vector<20x2xf32>
diff --git a/mlir/test/Dialect/Vector/int-range-interface.mlir b/mlir/test/Dialect/Vector/int-range-interface.mlir
index 2563b48..b2f16bb 100644
--- a/mlir/test/Dialect/Vector/int-range-interface.mlir
+++ b/mlir/test/Dialect/Vector/int-range-interface.mlir
@@ -51,6 +51,15 @@ func.func @vector_shape_cast() -> vector<4x4xindex> {
   func.return %2 : vector<4x4xindex>
 }
 
+// CHECK-LABEL: func @vector_transpose
+// CHECK: test.reflect_bounds {smax = 8 : index, smin = 7 : index, umax = 8 : index, umin = 7 : index}
+func.func @vector_transpose() -> vector<2x4xindex> {
+  %0 = test.with_bounds { smax = 8 : index, smin = 7 : index, umax = 8 : index, umin = 7 : index } : vector<4x2xindex>
+  %1 = vector.transpose %0, [1, 0] : vector<4x2xindex> to vector<2x4xindex>
+  %2 = test.reflect_bounds %1 : vector<2x4xindex>
+  func.return %2 : vector<2x4xindex>
+}
+
 // CHECK-LABEL: func @vector_extract
 // CHECK: test.reflect_bounds {smax = 6 : index, smin = 5 : index, umax = 6 : index, umin = 5 : index}
 func.func @vector_extract() -> index {
@@ -99,3 +108,11 @@ func.func @test_vector_extsi() -> vector<2xi32> {
   %2 = test.reflect_bounds %1 : vector<2xi32>
   func.return %2 : vector<2xi32>
 }
+
+// CHECK-LABEL: func @vector_step
+// CHECK: test.reflect_bounds {smax = 7 : index, smin = 0 : index, umax = 7 : index, umin = 0 : index}
+func.func @vector_step() -> vector<8xindex> {
+  %0 = vector.step : vector<8xindex>
+  %1 = test.reflect_bounds %0 : vector<8xindex>
+  func.return %1 : vector<8xindex>
+}
diff --git a/mlir/test/Target/SPIRV/logical-ops.mlir b/mlir/test/Target/SPIRV/logical-ops.mlir
index b200871..05cbddc 100644
--- a/mlir/test/Target/SPIRV/logical-ops.mlir
+++ b/mlir/test/Target/SPIRV/logical-ops.mlir
@@ -84,6 +84,8 @@ spirv.module Logical GLSL450 requires #spirv.vce<v1.0, [Shader], []> {
     %15 = spirv.IsNan %arg0 : f32
     // CHECK: spirv.IsInf
     %16 = spirv.IsInf %arg1 : f32
+    // CHECK: spirv.IsFinite
+    %17 = spirv.IsFinite %arg0 : f32
     spirv.Return
   }
 }
diff --git a/runtimes/CMakeLists.txt b/runtimes/CMakeLists.txt
index d0d2a53..bfb4341 100644
--- a/runtimes/CMakeLists.txt
+++ b/runtimes/CMakeLists.txt
@@ -36,7 +36,7 @@ list(INSERT CMAKE_MODULE_PATH 0
 # We order libraries to mirror roughly how they are layered, except that compiler-rt can depend
 # on libc++, so we put it after.
 set(LLVM_DEFAULT_RUNTIMES "libc;libunwind;libcxxabi;pstl;libcxx;compiler-rt;libclc;openmp;offload")
-set(LLVM_SUPPORTED_RUNTIMES "${LLVM_DEFAULT_RUNTIMES};llvm-libgcc;flang-rt")
+set(LLVM_SUPPORTED_RUNTIMES "${LLVM_DEFAULT_RUNTIMES};llvm-libgcc;flang-rt;libsycl")
 set(LLVM_ENABLE_RUNTIMES "" CACHE STRING
   "Semicolon-separated list of runtimes to build, or \"all\" (${LLVM_DEFAULT_RUNTIMES}). Supported runtimes are ${LLVM_SUPPORTED_RUNTIMES}.")
 if(LLVM_ENABLE_RUNTIMES STREQUAL "all" )
diff --git a/third-party/benchmark/src/cycleclock.h b/third-party/benchmark/src/cycleclock.h
index d4f1330..c0dffcf 100644
--- a/third-party/benchmark/src/cycleclock.h
+++ b/third-party/benchmark/src/cycleclock.h
@@ -79,7 +79,7 @@ inline BENCHMARK_ALWAYS_INLINE int64_t Now() {
   int64_t ret;
   __asm__ volatile("rdtsc" : "=A"(ret));
   return ret;
-#elif defined(__x86_64__) || defined(__amd64__)
+#elif (defined(__x86_64__) || defined(__amd64__)) && !defined(__arm64ec__)
   uint64_t low, high;
   __asm__ volatile("rdtsc" : "=a"(low), "=d"(high));
   return (high << 32) | low;
@@ -139,7 +139,7 @@ inline BENCHMARK_ALWAYS_INLINE int64_t Now() {
   struct timespec ts = {0, 0};
   clock_gettime(CLOCK_MONOTONIC, &ts);
   return static_cast<int64_t>(ts.tv_sec) * 1000000000 + ts.tv_nsec;
-#elif defined(__aarch64__)
+#elif defined(__aarch64__) || defined(__arm64ec__)
   // System timer of ARMv8 runs at a different frequency than the CPU's.
   // The frequency is fixed, typically in the range 1-50MHz.  It can be
   // read at CNTFRQ special register.  We assume the OS has set up
diff --git a/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel b/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel
index 118bf64..5930b85 100644
--- a/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel
@@ -1281,6 +1281,7 @@ cc_library(
         ":AnalysisFpExc",
         ":BinaryFormat",
         ":Core",
+        ":FrontendHLSL",
         ":Object",
         ":ProfileData",
         ":Support",
diff --git a/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel b/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel
index 813fd0c..c5fe783 100644
--- a/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel
@@ -133,6 +133,7 @@ cc_test(
         "//llvm:AsmParser",
         "//llvm:CodeGen",
         "//llvm:Core",
+        "//llvm:FrontendHLSL",
         "//llvm:Instrumentation",
         "//llvm:Passes",
         "//llvm:Support",