[clang] 1435c8e - Reapply "[LinkerWrapper] Clean up options after proper forwarding" (#126495)
Joseph Huber via cfe-commits
cfe-commits at lists.llvm.org
Fri Feb 14 07:56:51 PST 2025
Author: Joseph Huber
Date: 2025-02-14T09:56:46-06:00
New Revision: 1435c8ed95fa10a55c2f924984141e427b89c330
URL: https://github.com/llvm/llvm-project/commit/1435c8ed95fa10a55c2f924984141e427b89c330
DIFF: https://github.com/llvm/llvm-project/commit/1435c8ed95fa10a55c2f924984141e427b89c330.diff
LOG: Reapply "[LinkerWrapper] Clean up options after proper forwarding" (#126495)
Summary:
The test failed because it no longer passed Rpass by default without
LTO. I think that's desirable as it matches the standard behavior.
This reverts commit 6fd99de31864a5ef84ae8613b3a9034e05293461.
Added:
Modified:
clang/lib/Driver/ToolChains/Clang.cpp
clang/test/Driver/linker-wrapper.c
clang/test/Driver/openmp-offload.c
clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp
clang/tools/clang-linker-wrapper/LinkerWrapperOpts.td
offload/test/offloading/bug51781.c
Removed:
################################################################################
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index 55ec3db0ee994..96af466e067a8 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -9282,6 +9282,14 @@ void LinkerWrapper::ConstructJob(Compilation &C, const JobAction &JA,
for (StringRef Arg : LinkerArgs)
CmdArgs.push_back(Args.MakeArgString(
"--device-linker=" + TC->getTripleString() + "=" + Arg));
+
+ // Forward the LTO mode relying on the Driver's parsing.
+ if (C.getDriver().getOffloadLTOMode() == LTOK_Full)
+ CmdArgs.push_back(Args.MakeArgString(
+ "--device-compiler=" + TC->getTripleString() + "=-flto=full"));
+ else if (C.getDriver().getOffloadLTOMode() == LTOK_Thin)
+ CmdArgs.push_back(Args.MakeArgString(
+ "--device-compiler=" + TC->getTripleString() + "=-flto=thin"));
}
}
@@ -9289,6 +9297,9 @@ void LinkerWrapper::ConstructJob(Compilation &C, const JobAction &JA,
Args.MakeArgString("--host-triple=" + getToolChain().getTripleString()));
if (Args.hasArg(options::OPT_v))
CmdArgs.push_back("--wrapper-verbose");
+ if (Arg *A = Args.getLastArg(options::OPT_cuda_path_EQ))
+ CmdArgs.push_back(
+ Args.MakeArgString(Twine("--cuda-path=") + A->getValue()));
// Construct the link job so we can wrap around it.
Linker->ConstructJob(C, JA, Output, Inputs, Args, LinkingOutput);
diff --git a/clang/test/Driver/linker-wrapper.c b/clang/test/Driver/linker-wrapper.c
index df0a1d1e9a84d..7586b87743bf5 100644
--- a/clang/test/Driver/linker-wrapper.c
+++ b/clang/test/Driver/linker-wrapper.c
@@ -21,16 +21,16 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run \
// RUN: --linker-path=/usr/bin/ld %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=NVPTX-LINK
-// NVPTX-LINK: clang{{.*}} -o {{.*}}.img --target=nvptx64-nvidia-cuda -march=sm_70 -O2 -flto {{.*}}.o {{.*}}.o
+// NVPTX-LINK: clang{{.*}} -o {{.*}}.img --target=nvptx64-nvidia-cuda -march=sm_70 {{.*}}.o {{.*}}.o
// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=nvptx64-nvidia-cuda,arch=sm_70 \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=nvptx64-nvidia-cuda,arch=sm_70
// RUN: %clang -cc1 %s -triple x86_64-unknown-linux-gnu -emit-obj -o %t.o -fembed-offload-object=%t.out
-// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --device-debug -O0 \
+// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --device-compiler=-g \
// RUN: --linker-path=/usr/bin/ld %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=NVPTX-LINK-DEBUG
-// NVPTX-LINK-DEBUG: clang{{.*}} -o {{.*}}.img --target=nvptx64-nvidia-cuda -march=sm_70 -O2 -flto {{.*}}.o {{.*}}.o -g
+// NVPTX-LINK-DEBUG: clang{{.*}} --target=nvptx64-nvidia-cuda -march=sm_70 {{.*}}-g
// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=amdgcn-amd-amdhsa,arch=gfx908 \
@@ -39,16 +39,16 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run \
// RUN: --linker-path=/usr/bin/ld %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=AMDGPU-LINK
-// AMDGPU-LINK: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -O2 -flto -Wl,--no-undefined {{.*}}.o {{.*}}.o
+// AMDGPU-LINK: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -Wl,--no-undefined {{.*}}.o {{.*}}.o
// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.amdgpu.bc,kind=openmp,triple=amdgcn-amd-amdhsa,arch=gfx1030 \
// RUN: --image=file=%t.amdgpu.bc,kind=openmp,triple=amdgcn-amd-amdhsa,arch=gfx1030
// RUN: %clang -cc1 %s -triple x86_64-unknown-linux-gnu -emit-obj -o %t.o -fembed-offload-object=%t.out
-// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --save-temps -O2 \
+// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --device-compiler=--save-temps \
// RUN: --linker-path=/usr/bin/ld %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=AMDGPU-LTO-TEMPS
-// AMDGPU-LTO-TEMPS: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx1030 -O2 -flto -Wl,--no-undefined {{.*}}.o -save-temps
+// AMDGPU-LTO-TEMPS: clang{{.*}} --target=amdgcn-amd-amdhsa -mcpu=gfx1030 {{.*}}-save-temps
// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=x86_64-unknown-linux-gnu \
@@ -59,7 +59,7 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: --linker-path=/usr/bin/ld.lld --whole-archive %t.a --no-whole-archive \
// RUN: %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=CPU-LINK
-// CPU-LINK: clang{{.*}} -o {{.*}}.img --target=x86_64-unknown-linux-gnu -O2 -flto -Wl,--no-undefined {{.*}}.o {{.*}}.o -Wl,-Bsymbolic -shared -Wl,--whole-archive {{.*}}.a -Wl,--no-whole-archive
+// CPU-LINK: clang{{.*}} -o {{.*}}.img --target=x86_64-unknown-linux-gnu -Wl,--no-undefined {{.*}}.o {{.*}}.o -Wl,-Bsymbolic -shared -Wl,--whole-archive {{.*}}.a -Wl,--no-whole-archive
// RUN: %clang -cc1 %s -triple x86_64-unknown-linux-gnu -emit-obj -o %t.o
// RUN: clang-linker-wrapper --dry-run --host-triple=x86_64-unknown-linux-gnu -mllvm -openmp-opt-disable \
@@ -148,7 +148,7 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --clang-backend \
// RUN: --linker-path=/usr/bin/ld %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=CLANG-BACKEND
-// CLANG-BACKEND: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -O2 -flto -Wl,--no-undefined {{.*}}.o
+// CLANG-BACKEND: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -Wl,--no-undefined {{.*}}.o
// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=nvptx64-nvidia-cuda,arch=sm_70
@@ -171,8 +171,8 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run \
// RUN: --linker-path=/usr/bin/ld %t-on.o %t-off.o %t.a -o a.out 2>&1 | FileCheck %s --check-prefix=AMD-TARGET-ID
-// AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack+ -O2 -flto -Wl,--no-undefined {{.*}}.o {{.*}}.o
-// AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack- -O2 -flto -Wl,--no-undefined {{.*}}.o {{.*}}.o
+// AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack+ -Wl,--no-undefined {{.*}}.o {{.*}}.o
+// AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack- -Wl,--no-undefined {{.*}}.o {{.*}}.o
// RUN: clang-offload-packager -o %t-lib.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=amdgcn-amd-amdhsa,arch=generic
@@ -187,8 +187,8 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run \
// RUN: --linker-path=/usr/bin/ld %t1.o %t2.o %t.a -o a.out 2>&1 | FileCheck %s --check-prefix=ARCH-ALL
-// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a -O2 -flto -Wl,--no-undefined {{.*}}.o {{.*}}.o
-// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -O2 -flto -Wl,--no-undefined {{.*}}.o {{.*}}.o
+// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a -Wl,--no-undefined {{.*}}.o {{.*}}.o
+// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -Wl,--no-undefined {{.*}}.o {{.*}}.o
// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=x86_64-unknown-linux-gnu \
diff --git a/clang/test/Driver/openmp-offload.c b/clang/test/Driver/openmp-offload.c
index 6f56ae00ba065..2cf2643af6c15 100644
--- a/clang/test/Driver/openmp-offload.c
+++ b/clang/test/Driver/openmp-offload.c
@@ -208,3 +208,13 @@
// RUN: -fsyntax-only %s 2>&1 | FileCheck -check-prefix=CHK-SYNTAX-ONLY-ARGS %s
// CHK-SYNTAX-ONLY-ARGS: "-cc1" "-triple" "powerpc64le-ibm-linux-gnu"{{.*}}"-fsyntax-only"
// CHK-SYNTAX-ONLY-ARGS: "-cc1" "-triple" "powerpc64le-unknown-linux"{{.*}}"-fsyntax-only"
+
+//
+// Ensure `-foffload-lto` is forwarded properly.
+//
+// RUN: %clang -### --target=powerpc64le-linux -fopenmp=libomp -fopenmp-targets=powerpc64le-ibm-linux-gnu \
+// RUN: -foffload-lto %s 2>&1 | FileCheck -check-prefix=CHK-DEVICE-LTO-FULL %s
+// CHK-DEVICE-LTO-FULL: clang-linker-wrapper{{.*}} "--device-compiler=powerpc64le-ibm-linux-gnu=-flto=full"
+// RUN: %clang -### --target=powerpc64le-linux -fopenmp=libomp -fopenmp-targets=powerpc64le-ibm-linux-gnu \
+// RUN: -foffload-lto=thin %s 2>&1 | FileCheck -check-prefix=CHK-DEVICE-LTO-THIN %s
+// CHK-DEVICE-LTO-THIN: clang-linker-wrapper{{.*}} "--device-compiler=powerpc64le-ibm-linux-gnu=-flto=thin"
diff --git a/clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp b/clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp
index aa43b2f5f2a1b..e78763faad73e 100644
--- a/clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp
+++ b/clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp
@@ -483,7 +483,6 @@ Expected<StringRef> clang(ArrayRef<StringRef> InputFiles, const ArgList &Args) {
if (!TempFileOrErr)
return TempFileOrErr.takeError();
- StringRef OptLevel = Args.getLastArgValue(OPT_opt_level, "O2");
SmallVector<StringRef, 16> CmdArgs{
*ClangPath,
"--no-default-config",
@@ -496,10 +495,7 @@ Expected<StringRef> clang(ArrayRef<StringRef> InputFiles, const ArgList &Args) {
Triple.isAMDGPU() ? CmdArgs.push_back(Args.MakeArgString("-mcpu=" + Arch))
: CmdArgs.push_back(Args.MakeArgString("-march=" + Arch));
- CmdArgs.push_back(Args.MakeArgString("-" + OptLevel));
-
// Forward all of the `--offload-opt` and similar options to the device.
- CmdArgs.push_back("-flto");
for (auto &Arg : Args.filtered(OPT_offload_opt_eq_minus, OPT_mllvm))
CmdArgs.append(
{"-Xlinker",
@@ -548,29 +544,12 @@ Expected<StringRef> clang(ArrayRef<StringRef> InputFiles, const ArgList &Args) {
CmdArgs.append({"-Xlinker", Args.MakeArgString(
"-mllvm=" + StringRef(Arg->getValue()))});
- if (Args.hasArg(OPT_debug))
- CmdArgs.push_back("-g");
-
- if (SaveTemps)
- CmdArgs.push_back("-save-temps");
-
if (SaveTemps && linkerSupportsLTO(Args))
CmdArgs.push_back("-Wl,--save-temps");
if (Args.hasArg(OPT_embed_bitcode))
CmdArgs.push_back("-Wl,--lto-emit-llvm");
- if (Verbose)
- CmdArgs.push_back("-v");
-
- if (!CudaBinaryPath.empty())
- CmdArgs.push_back(Args.MakeArgString("--cuda-path=" + CudaBinaryPath));
-
- for (StringRef Arg : Args.getAllArgValues(OPT_ptxas_arg))
- llvm::copy(
- SmallVector<StringRef>({"-Xcuda-ptxas", Args.MakeArgString(Arg)}),
- std::back_inserter(CmdArgs));
-
for (StringRef Arg : Args.getAllArgValues(OPT_linker_arg_EQ))
CmdArgs.append({"-Xlinker", Args.MakeArgString(Arg)});
for (StringRef Arg : Args.getAllArgValues(OPT_compiler_arg_EQ))
diff --git a/clang/tools/clang-linker-wrapper/LinkerWrapperOpts.td b/clang/tools/clang-linker-wrapper/LinkerWrapperOpts.td
index 57d918db0a73c..17fb9db35fe39 100644
--- a/clang/tools/clang-linker-wrapper/LinkerWrapperOpts.td
+++ b/clang/tools/clang-linker-wrapper/LinkerWrapperOpts.td
@@ -17,11 +17,9 @@ def cuda_path_EQ : Joined<["--"], "cuda-path=">,
Flags<[WrapperOnlyOption]>, MetaVarName<"<dir>">,
HelpText<"Set the system CUDA path">;
def host_triple_EQ : Joined<["--"], "host-triple=">,
- Flags<[WrapperOnlyOption]>, MetaVarName<"<triple>">,
- HelpText<"Triple to use for the host compilation">;
-def opt_level : Joined<["--"], "opt-level=">,
- Flags<[WrapperOnlyOption]>, MetaVarName<"<O0, O1, O2, or O3>">,
- HelpText<"Optimization level for LTO">;
+ Flags<[WrapperOnlyOption]>,
+ MetaVarName<"<triple>">,
+ HelpText<"Triple to use for the host compilation">;
def device_linker_args_EQ : Joined<["--"], "device-linker=">,
Flags<[WrapperOnlyOption]>, MetaVarName<"<value> or <triple>=<value>">,
HelpText<"Arguments to pass to the device linker invocation">;
@@ -34,18 +32,8 @@ def dry_run : Flag<["--"], "dry-run">,
def verbose : Flag<["--"], "wrapper-verbose">,
Flags<[WrapperOnlyOption]>, HelpText<"Verbose output from tools">;
def embed_bitcode : Flag<["--"], "embed-bitcode">,
- Flags<[WrapperOnlyOption]>, HelpText<"Embed linked bitcode in the module">;
-def debug : Flag<["--"], "device-debug">, Flags<[WrapperOnlyOption]>,
- HelpText<"Use debugging">;
-def ptxas_arg : Joined<["--"], "ptxas-arg=">,
- Flags<[WrapperOnlyOption]>,
- HelpText<"Argument to pass to the 'ptxas' invocation">;
-def pass_remarks_EQ : Joined<["--"], "pass-remarks=">,
- Flags<[WrapperOnlyOption]>, HelpText<"Pass remarks for LTO">;
-def pass_remarks_missed_EQ : Joined<["--"], "pass-remarks-missed=">,
- Flags<[WrapperOnlyOption]>, HelpText<"Pass remarks for LTO">;
-def pass_remarks_analysis_EQ : Joined<["--"], "pass-remarks-analysis=">,
- Flags<[WrapperOnlyOption]>, HelpText<"Pass remarks for LTO">;
+ Flags<[WrapperOnlyOption]>,
+ HelpText<"Embed linked bitcode in the module">;
def print_wrapped_module : Flag<["--"], "print-wrapped-module">,
Flags<[WrapperOnlyOption]>,
HelpText<"Print the wrapped module's IR for testing">;
diff --git a/offload/test/offloading/bug51781.c b/offload/test/offloading/bug51781.c
index 17b7499a7606e..2f30b035afbbe 100644
--- a/offload/test/offloading/bug51781.c
+++ b/offload/test/offloading/bug51781.c
@@ -5,7 +5,7 @@
// SPMDize. There is no main thread, so there's no issue.
//
-// RUN: %libomptarget-compile-generic -O2 -Rpass=openmp-opt > %t.spmd 2>&1
+// RUN: %libomptarget-compile-generic -O2 -foffload-lto -Rpass=openmp-opt > %t.spmd 2>&1
// RUN: %fcheck-nvptx64-nvidia-cuda -check-prefix=SPMD -input-file=%t.spmd
// RUN: %fcheck-amdgcn-amd-amdhsa -check-prefix=SPMD -input-file=%t.spmd
// RUN: %libomptarget-run-generic 2>&1 | %fcheck-generic
@@ -15,7 +15,7 @@
// Use the custom state machine, which must avoid the same barrier problem as
// the generic state machine.
//
-// RUN: %libomptarget-compile-generic -O2 -Rpass=openmp-opt \
+// RUN: %libomptarget-compile-generic -O2 -foffload-lto -Rpass=openmp-opt \
// RUN: -mllvm -openmp-opt-disable-spmdization > %t.custom 2>&1
// RUN: %fcheck-nvptx64-nvidia-cuda -check-prefix=CUSTOM -input-file=%t.custom
// RUN: %fcheck-amdgcn-amd-amdhsa -check-prefix=CUSTOM -input-file=%t.custom
@@ -24,7 +24,7 @@
// Repeat with reduction clause, which has managed to break the custom state
// machine in the past.
//
-// RUN: %libomptarget-compile-generic -O2 -Rpass=openmp-opt -DADD_REDUCTION \
+// RUN: %libomptarget-compile-generic -O2 -foffload-lto -Rpass=openmp-opt -DADD_REDUCTION \
// RUN: -mllvm -openmp-opt-disable-spmdization > %t.custom 2>&1
// RUN: %fcheck-nvptx64-nvidia-cuda -check-prefix=CUSTOM -input-file=%t.custom
// RUN: %fcheck-amdgcn-amd-amdhsa -check-prefix=CUSTOM -input-file=%t.custom
More information about the cfe-commits
mailing list