diff options
| -rw-r--r-- | llvm/include/llvm/Transforms/IPO/WholeProgramDevirt.h | 2 | ||||
| -rw-r--r-- | llvm/lib/LTO/LTO.cpp | 11 | ||||
| -rw-r--r-- | llvm/lib/Transforms/IPO/WholeProgramDevirt.cpp | 6 | ||||
| -rw-r--r-- | llvm/test/ThinLTO/X86/Inputs/devirt_promote.ll | 39 | ||||
| -rw-r--r-- | llvm/test/ThinLTO/X86/devirt_promote.ll | 72 |
5 files changed, 120 insertions, 10 deletions
diff --git a/llvm/include/llvm/Transforms/IPO/WholeProgramDevirt.h b/llvm/include/llvm/Transforms/IPO/WholeProgramDevirt.h index b30d098acab..22435e4ed1e 100644 --- a/llvm/include/llvm/Transforms/IPO/WholeProgramDevirt.h +++ b/llvm/include/llvm/Transforms/IPO/WholeProgramDevirt.h @@ -251,7 +251,7 @@ void runWholeProgramDevirtOnIndex( /// devirt target names for any locals that were exported. void updateIndexWPDForExports( ModuleSummaryIndex &Summary, - StringMap<FunctionImporter::ExportSetTy> &ExportLists, + function_ref<bool(StringRef, GlobalValue::GUID)> isExported, std::map<ValueInfo, std::vector<VTableSlotSummary>> &LocalWPDTargetsMap); } // end namespace llvm diff --git a/llvm/lib/LTO/LTO.cpp b/llvm/lib/LTO/LTO.cpp index 9e16cba44fd..6c5858b942a 100644 --- a/llvm/lib/LTO/LTO.cpp +++ b/llvm/lib/LTO/LTO.cpp @@ -1304,11 +1304,6 @@ Error LTO::runThinLTO(AddStreamFn AddStream, NativeObjectCache Cache, ComputeCrossModuleImport(ThinLTO.CombinedIndex, ModuleToDefinedGVSummaries, ImportLists, ExportLists); - // Update local devirtualized targets that were exported by cross-module - // importing - updateIndexWPDForExports(ThinLTO.CombinedIndex, ExportLists, - LocalWPDTargetsMap); - // Figure out which symbols need to be internalized. This also needs to happen // at -O0 because summary-based DCE is implemented using internalization, and // we must apply DCE consistently with the full LTO module in order to avoid @@ -1338,6 +1333,12 @@ Error LTO::runThinLTO(AddStreamFn AddStream, NativeObjectCache Cache, ExportList->second.count(GUID)) || ExportedGUIDs.count(GUID); }; + + // Update local devirtualized targets that were exported by cross-module + // importing or by other devirtualizations marked in the ExportedGUIDs set. + updateIndexWPDForExports(ThinLTO.CombinedIndex, isExported, + LocalWPDTargetsMap); + auto isPrevailing = [&](GlobalValue::GUID GUID, const GlobalValueSummary *S) { return ThinLTO.PrevailingModuleForGUID[GUID] == S->modulePath(); diff --git a/llvm/lib/Transforms/IPO/WholeProgramDevirt.cpp b/llvm/lib/Transforms/IPO/WholeProgramDevirt.cpp index 4490a7ef62a..4055fe04999 100644 --- a/llvm/lib/Transforms/IPO/WholeProgramDevirt.cpp +++ b/llvm/lib/Transforms/IPO/WholeProgramDevirt.cpp @@ -713,7 +713,7 @@ void runWholeProgramDevirtOnIndex( void updateIndexWPDForExports( ModuleSummaryIndex &Summary, - StringMap<FunctionImporter::ExportSetTy> &ExportLists, + function_ref<bool(StringRef, GlobalValue::GUID)> isExported, std::map<ValueInfo, std::vector<VTableSlotSummary>> &LocalWPDTargetsMap) { for (auto &T : LocalWPDTargetsMap) { auto &VI = T.first; @@ -721,9 +721,7 @@ void updateIndexWPDForExports( assert(VI.getSummaryList().size() == 1 && "Devirt of local target has more than one copy"); auto &S = VI.getSummaryList()[0]; - const auto &ExportList = ExportLists.find(S->modulePath()); - if (ExportList == ExportLists.end() || - !ExportList->second.count(VI.getGUID())) + if (!isExported(S->modulePath(), VI.getGUID())) continue; // It's been exported by a cross module import. diff --git a/llvm/test/ThinLTO/X86/Inputs/devirt_promote.ll b/llvm/test/ThinLTO/X86/Inputs/devirt_promote.ll new file mode 100644 index 00000000000..bcd49aefe15 --- /dev/null +++ b/llvm/test/ThinLTO/X86/Inputs/devirt_promote.ll @@ -0,0 +1,39 @@ +target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128" +target triple = "x86_64-grtev4-linux-gnu" + +%struct.A = type { i32 (...)** } +%struct.B = type { %struct.A } + +@_ZTV1B = constant { [4 x i8*] } { [4 x i8*] [i8* null, i8* undef, i8* bitcast (i32 (%struct.B*, i32)* @_ZN1B1fEi to i8*), i8* bitcast (i32 (%struct.A*, i32)* @_ZN1A1nEi to i8*)] }, !type !0, !type !1 + +define i32 @_ZN1B1fEi(%struct.B* %this, i32 %a) #0 { + ret i32 0; +} + +define internal i32 @_ZN1A1nEi(%struct.A* %this, i32 %a) #0 { + ret i32 0; +} + +define i32 @test2(%struct.B* %obj, i32 %a) { +entry: + %0 = bitcast %struct.B* %obj to i8*** + %vtable2 = load i8**, i8*** %0 + %1 = bitcast i8** %vtable2 to i8* + %p2 = call i1 @llvm.type.test(i8* %1, metadata !"_ZTS1B") + call void @llvm.assume(i1 %p2) + + %fptrptr = getelementptr i8*, i8** %vtable2, i32 1 + %2 = bitcast i8** %fptrptr to i32 (%struct.B*, i32)** + %fptr33 = load i32 (%struct.B*, i32)*, i32 (%struct.B*, i32)** %2, align 8 + + %call4 = tail call i32 %fptr33(%struct.B* nonnull %obj, i32 %a) + ret i32 %call4 +} + +attributes #0 = { noinline optnone } + +declare i1 @llvm.type.test(i8*, metadata) +declare void @llvm.assume(i1) + +!0 = !{i64 16, !"_ZTS1A"} +!1 = !{i64 16, !"_ZTS1B"} diff --git a/llvm/test/ThinLTO/X86/devirt_promote.ll b/llvm/test/ThinLTO/X86/devirt_promote.ll new file mode 100644 index 00000000000..563ed994157 --- /dev/null +++ b/llvm/test/ThinLTO/X86/devirt_promote.ll @@ -0,0 +1,72 @@ +; REQUIRES: x86-registered-target + +; Test devirtualization requiring promotion of local targets, where the +; promotion is required by one devirtualization and needs to be updated +; for a second devirtualization in the defining module as a post-pass +; update. + +; Generate unsplit module with summary for ThinLTO index-based WPD. +; RUN: opt -thinlto-bc -o %t3.o %s +; RUN: opt -thinlto-bc -o %t4.o %p/Inputs/devirt_promote.ll + +; RUN: llvm-lto2 run %t3.o %t4.o -save-temps -use-new-pm -pass-remarks=. \ +; RUN: -wholeprogramdevirt-print-index-based \ +; RUN: -o %t5 \ +; RUN: -r=%t3.o,test,px \ +; RUN: -r=%t4.o,_ZN1B1fEi,p \ +; RUN: -r=%t4.o,test2,px \ +; RUN: -r=%t4.o,_ZTV1B,px \ +; RUN: 2>&1 | FileCheck %s --check-prefix=REMARK --check-prefix=PRINT +; RUN: llvm-dis %t5.1.4.opt.bc -o - | FileCheck %s --check-prefix=CHECK-IR1 +; RUN: llvm-dis %t5.2.4.opt.bc -o - | FileCheck %s --check-prefix=CHECK-IR2 +; RUN: llvm-nm %t5.1 | FileCheck %s --check-prefix=NM-INDEX1 +; RUN: llvm-nm %t5.2 | FileCheck %s --check-prefix=NM-INDEX2 + +; NM-INDEX1: U _ZN1A1nEi.llvm. + +; Make sure that not only did _ZN1A1nEi get promoted (due to the +; devirtualization in the other module) but the reference due to the +; devirtualization in its defining module should be to the promoted +; symbol. +; NM-INDEX2-NOT: U _ZN1A1nEi +; NM-INDEX2: T _ZN1A1nEi.llvm. +; NM-INDEX2-NOT: U _ZN1A1nEi + +; We should devirt call to _ZN1A1nEi once in importing module and once +; in original (exporting) module. +; REMARK-COUNT-2: single-impl: devirtualized a call to _ZN1A1nEi.llvm. + +target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128" +target triple = "x86_64-grtev4-linux-gnu" + +%struct.A = type { i32 (...)** } + +; CHECK-IR1-LABEL: define i32 @test +define i32 @test(%struct.A* %obj, i32 %a) { +entry: + %0 = bitcast %struct.A* %obj to i8*** + %vtable = load i8**, i8*** %0 + %1 = bitcast i8** %vtable to i8* + %p = call i1 @llvm.type.test(i8* %1, metadata !"_ZTS1A") + call void @llvm.assume(i1 %p) + %fptrptr = getelementptr i8*, i8** %vtable, i32 1 + %2 = bitcast i8** %fptrptr to i32 (%struct.A*, i32)** + %fptr1 = load i32 (%struct.A*, i32)*, i32 (%struct.A*, i32)** %2, align 8 + + ; Check that the call was devirtualized. + ; CHECK-IR1: %call = tail call i32 bitcast (void ()* @_ZN1A1nEi + %call = tail call i32 %fptr1(%struct.A* nonnull %obj, i32 %a) + + ret i32 %call +} +; CHECK-IR1-LABEL: ret i32 +; CHECK-IR1-LABEL: } + +; CHECK-IR2: define i32 @test2 +; Check that the call was devirtualized. +; CHECK-IR2: %call4 = tail call i32 @_ZN1A1nEi + +declare i1 @llvm.type.test(i8*, metadata) +declare void @llvm.assume(i1) + +attributes #0 = { noinline optnone } |
