[ThinLTO] Pass down opt level to LTO backend and handle -O0 LTO in new PM

Summary:
The opt level was not being passed down to the ThinLTO backend when
invoked via clang (for distributed ThinLTO).

This exposed an issue where the new PM was asserting if the Thin or
regular LTO backend pipelines were invoked with -O0 (not a new issue,
could be provoked by invoking in-process *LTO backends via linker using
new PM and -O0). Fix this similar to the old PM where -O0 only does the
necessary lowering of type metadata (WPD and LowerTypeTest passes) and
then quits, rather than asserting.

Reviewers: xur

Subscribers: mehdi_amini, inglorion, eraman, hiraditya, steven_wu, dexonsmith, cfe-commits, llvm-commits, pcc

Tags: #clang, #llvm

Differential Revision: https://reviews.llvm.org/D61022

llvm-svn: 359025

diff --git a/clang/lib/CodeGen/BackendUtil.cpp b/clang/lib/CodeGen/BackendUtil.cpp
index 995df8c..51ee592 100644 (file)
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -1325,6 +1325,7 @@ static void runThinLTOBackend(ModuleSummaryIndex *CombinedIndex, Module *M,
   Conf.MAttrs = TOpts.Features;
   Conf.RelocModel = CGOpts.RelocationModel;
   Conf.CGOptLevel = getCGOptLevel(CGOpts);
+  Conf.OptLevel = CGOpts.OptimizationLevel;
   initTargetOptions(Conf.Options, CGOpts, TOpts, LOpts, HeaderOpts);
   Conf.SampleProfile = std::move(SampleProfile);
 

diff --git a/clang/test/CodeGen/thinlto-debug-pm.c b/clang/test/CodeGen/thinlto-debug-pm.c
index 2accde1..dc3bd33 100644 (file)
--- a/clang/test/CodeGen/thinlto-debug-pm.c
+++ b/clang/test/CodeGen/thinlto-debug-pm.c
@@ -1,10 +1,17 @@
-// Test to ensure -fdebug-pass-manager works when invoking the
-// ThinLTO backend path with the new PM.
+// Test to ensure the opt level is passed down to the ThinLTO backend.
 // REQUIRES: x86-registered-target
 // RUN: %clang_cc1 -o %t.o -flto=thin -fexperimental-new-pass-manager -triple x86_64-unknown-linux-gnu -emit-llvm-bc %s
 // RUN: llvm-lto -thinlto -o %t %t.o
-// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-obj -O2 -o %t2.o -x ir %t.o -fthinlto-index=%t.thinlto.bc -fdebug-pass-manager -fexperimental-new-pass-manager 2>&1 | FileCheck %s
-// CHECK: Running pass:
+
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-obj -O2 -o %t2.o -x ir %t.o -fthinlto-index=%t.thinlto.bc -fdebug-pass-manager -fexperimental-new-pass-manager 2>&1 | FileCheck %s --check-prefix=O2-NEWPM
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-obj -O0 -o %t2.o -x ir %t.o -fthinlto-index=%t.thinlto.bc -fdebug-pass-manager -fexperimental-new-pass-manager 2>&1 | FileCheck %s --check-prefix=O0-NEWPM
+// O2-NEWPM: Running pass: LoopVectorizePass
+// O0-NEWPM-NOT: Running pass: LoopVectorizePass
+
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-obj -O2 -o %t2.o -x ir %t.o -fthinlto-index=%t.thinlto.bc -mllvm -debug-pass=Structure 2>&1 | FileCheck %s --check-prefix=O2-OLDPM
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-obj -O0 -o %t2.o -x ir %t.o -fthinlto-index=%t.thinlto.bc -mllvm -debug-pass=Structure 2>&1 | FileCheck %s --check-prefix=O0-OLDPM
+// O2-OLDPM: Loop Vectorization
+// O0-OLDPM-NOT: Loop Vectorization
 
 void foo() {
 }

diff --git a/clang/test/CodeGen/thinlto-distributed-cfi-devirt.ll b/clang/test/CodeGen/thinlto-distributed-cfi-devirt.ll
index 2ecf149..0bc23eb 100644 (file)
--- a/clang/test/CodeGen/thinlto-distributed-cfi-devirt.ll
+++ b/clang/test/CodeGen/thinlto-distributed-cfi-devirt.ll
@@ -39,12 +39,12 @@
 ; CHECK-DIS: ^2 = typeid: (name: "_ZTS1A", summary: (typeTestRes: (kind: allOnes, sizeM1BitWidth: 7), wpdResolutions: ((offset: 0, wpdRes: (kind: branchFunnel)), (offset: 8, wpdRes: (kind: singleImpl, singleImplName: "_ZN1A1nEi"))))) ; guid = 7004155349499253778
 
 ; RUN: %clang_cc1 -triple x86_64-grtev4-linux-gnu \
-; RUN:   -emit-obj -fthinlto-index=%t.o.thinlto.bc \
+; RUN:   -emit-obj -fthinlto-index=%t.o.thinlto.bc -O2 \
 ; RUN:   -emit-llvm -o - -x ir %t.o | FileCheck %s --check-prefixes=CHECK-IR
 
 ; Check that backend does not fail generating native code.
 ; RUN: %clang_cc1 -triple x86_64-grtev4-linux-gnu \
-; RUN:   -emit-obj -fthinlto-index=%t.o.thinlto.bc \
+; RUN:   -emit-obj -fthinlto-index=%t.o.thinlto.bc -O2 \
 ; RUN:   -o %t.native.o -x ir %t.o
 
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"

diff --git a/llvm/lib/Passes/PassBuilder.cpp b/llvm/lib/Passes/PassBuilder.cpp
index 32e31cb..1b4bffc 100644 (file)
--- a/llvm/lib/Passes/PassBuilder.cpp
+++ b/llvm/lib/Passes/PassBuilder.cpp
@@ -1046,10 +1046,16 @@ ModulePassManager PassBuilder::buildThinLTODefaultPipeline(
     //
     // Also, WPD has access to more precise information than ICP and can
     // devirtualize more effectively, so it should operate on the IR first.
+    //
+    // The WPD and LowerTypeTest passes need to run at -O0 to lower type
+    // metadata and intrinsics.
     MPM.addPass(WholeProgramDevirtPass(nullptr, ImportSummary));
     MPM.addPass(LowerTypeTestsPass(nullptr, ImportSummary));
   }
 
+  if (Level == O0)
+    return MPM;
+
   // Force any function attributes we want the rest of the pipeline to observe.
   MPM.addPass(ForceFunctionAttrsPass());
 
@@ -1075,9 +1081,16 @@ PassBuilder::buildLTOPreLinkDefaultPipeline(OptimizationLevel Level,
 ModulePassManager
 PassBuilder::buildLTODefaultPipeline(OptimizationLevel Level, bool DebugLogging,
                                      ModuleSummaryIndex *ExportSummary) {
-  assert(Level != O0 && "Must request optimizations for the default pipeline!");
   ModulePassManager MPM(DebugLogging);
 
+  if (Level == O0) {
+    // The WPD and LowerTypeTest passes need to run at -O0 to lower type
+    // metadata and intrinsics.
+    MPM.addPass(WholeProgramDevirtPass(ExportSummary, nullptr));
+    MPM.addPass(LowerTypeTestsPass(ExportSummary, nullptr));
+    return MPM;
+  }
+
   if (PGOOpt && PGOOpt->Action == PGOOptions::SampleUse) {
     // Load sample profile before running the LTO optimization pipeline.
     MPM.addPass(SampleProfileLoaderPass(PGOOpt->ProfileFile,

diff --git a/llvm/test/tools/gold/X86/opt-level.ll b/llvm/test/tools/gold/X86/opt-level.ll
index 33317ea..bd49e09 100644 (file)
--- a/llvm/test/tools/gold/X86/opt-level.ll
+++ b/llvm/test/tools/gold/X86/opt-level.ll
@@ -6,12 +6,25 @@
 ; RUN: %gold -plugin %llvmshlibdir/LLVMgold%shlibext -plugin-opt=save-temps \
 ; RUN:    -m elf_x86_64 \
 ; RUN:    -plugin-opt=O1 -r -o %t.o %t.bc
-; RUN: llvm-dis < %t.o.0.4.opt.bc -o - | FileCheck --check-prefix=CHECK-O1 %s
+; RUN: llvm-dis < %t.o.0.4.opt.bc -o - | FileCheck --check-prefix=CHECK-O1 --check-prefix=CHECK-O1-OLDPM %s
 ; RUN: %gold -plugin %llvmshlibdir/LLVMgold%shlibext -plugin-opt=save-temps \
 ; RUN:    -m elf_x86_64 \
 ; RUN:    -plugin-opt=O2 -r -o %t.o %t.bc
 ; RUN: llvm-dis < %t.o.0.4.opt.bc -o - | FileCheck --check-prefix=CHECK-O2 %s
 
+; RUN: %gold -plugin %llvmshlibdir/LLVMgold%shlibext -plugin-opt=save-temps \
+; RUN:    -m elf_x86_64 --plugin-opt=new-pass-manager \
+; RUN:    -plugin-opt=O0 -r -o %t.o %t.bc
+; RUN: llvm-dis < %t.o.0.4.opt.bc -o - | FileCheck --check-prefix=CHECK-O0 %s
+; RUN: %gold -plugin %llvmshlibdir/LLVMgold%shlibext -plugin-opt=save-temps \
+; RUN:    -m elf_x86_64 --plugin-opt=new-pass-manager \
+; RUN:    -plugin-opt=O1 -r -o %t.o %t.bc
+; RUN: llvm-dis < %t.o.0.4.opt.bc -o - | FileCheck --check-prefix=CHECK-O1 --check-prefix=CHECK-O1-NEWPM %s
+; RUN: %gold -plugin %llvmshlibdir/LLVMgold%shlibext -plugin-opt=save-temps \
+; RUN:    -m elf_x86_64 --plugin-opt=new-pass-manager \
+; RUN:    -plugin-opt=O2 -r -o %t.o %t.bc
+; RUN: llvm-dis < %t.o.0.4.opt.bc -o - | FileCheck --check-prefix=CHECK-O2 %s
+
 ; CHECK-O0: define internal void @foo(
 ; CHECK-O1: define internal void @foo(
 ; CHECK-O2-NOT: define internal void @foo(
@@ -36,7 +49,9 @@ f:
 
 end:
   ; CHECK-O0: phi
-  ; CHECK-O1: select
+  ; CHECK-O1-OLDPM: select
+  ; The new PM does not do as many optimizations at O1
+  ; CHECK-O1-NEWPM: phi
   %r = phi i32 [ 1, %t ], [ 2, %f ]
   ret i32 %r
 }