Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Pass -foffload-lto instead of -flto for cuda/hip kernels in clangLinkerWrapper #16605

Merged
merged 2 commits into from
Jan 29, 2025
Merged

Pass -foffload-lto instead of -flto for cuda/hip kernels in clangLinkerWrapper #16605

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ccf221a
Pass -offload-lto instead of -lto for cuda/hip kernels and enable fai…
omarahmed1111 Jan 13, 2025
a0f881c
Merge branch 'sycl' into fix-NewOffloadDrivers-failed-tests
omarahmed1111 Jan 15, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 9 additions & 9 deletions clang/test/Driver/linker-wrapper.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ __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 -O2 -foffload-lto {{.*}}.o {{.*}}.o

// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=nvptx64-nvidia-cuda,arch=sm_70 \
Expand All @@ -30,7 +30,7 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --device-debug -O0 \
// 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{{.*}} -o {{.*}}.img --target=nvptx64-nvidia-cuda -march=sm_70 -O2 -foffload-lto {{.*}}.o {{.*}}.o -g

// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=amdgcn-amd-amdhsa,arch=gfx908 \
Expand All @@ -39,7 +39,7 @@ __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 -O2 -foffload-lto -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 \
Expand All @@ -48,7 +48,7 @@ __attribute__((visibility("protected"), used)) int x;
// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run --save-temps -O2 \
// 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 {{.*}} -save-temps
// AMDGPU-LTO-TEMPS: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx1030 -O2 -foffload-lto -Wl,--no-undefined {{.*}} -save-temps

// RUN: clang-offload-packager -o %t.out \
// RUN: --image=file=%t.elf.o,kind=openmp,triple=x86_64-unknown-linux-gnu \
Expand Down Expand Up @@ -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 -O2 -foffload-lto -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
Expand All @@ -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+ -O2 -foffload-lto -Wl,--no-undefined {{.*}}.o {{.*}}.o
// AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack- -O2 -foffload-lto -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
Expand All @@ -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 -O2 -foffload-lto -Wl,--no-undefined {{.*}}.o {{.*}}.o
// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -O2 -foffload-lto -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 \
Expand Down
7 changes: 6 additions & 1 deletion clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1533,12 +1533,17 @@ Expected<StringRef> clang(ArrayRef<StringRef> InputFiles, const ArgList &Args,
};

// 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",
Args.MakeArgString("--plugin-opt=" + StringRef(Arg->getValue()))});

if (Triple.isNVPTX() || Triple.isAMDGPU()) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is there an upstream pr for this based on previous feedback?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I thought we are going to do this after merging this. I will open one for that.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok as long as you'll make an upstream PR lgtm to merge here first

CmdArgs.push_back("-foffload-lto");
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should this go upstream?

Copy link
Contributor

@asudarsa asudarsa Jan 14, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1 on adding this change to upstream code. I am also wondering why it is not an issue in the upstream compiler and why we see it only with SYCL offloading.

Also @omarahmed1111 can you please comment on the test failure?

Thanks

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The test failure for loop_extended is a flaky failure not related to this PR, the failed test disabled on this PR: #16612 . I updated the PR, so should be solved now.

} else {
CmdArgs.push_back("-flto");
}

if (!Triple.isNVPTX() && !Triple.isSPIRV())
CmdArgs.push_back("-Wl,--no-undefined");

Expand Down
2 changes: 0 additions & 2 deletions sycl/test-e2e/NewOffloadDriver/multisource.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,6 @@
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
// XFAIL: cuda
// XFAIL-TRACKER: https://github.com/intel/llvm/issues/16413
// Separate kernel sources and host code sources
// Test with `--offload-new-driver`
// RUN: %{build} --offload-new-driver -c -o %t.kernel.o -DINIT_KERNEL -DCALC_KERNEL
Expand Down
3 changes: 0 additions & 3 deletions sycl/test-e2e/NewOffloadDriver/split-per-source-main.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,9 +1,6 @@
// RUN: %{build} -Wno-error=unused-command-line-argument -fsycl-device-code-split=per_source -I %S/Inputs -o %t.out %S/Inputs/split-per-source-second-file.cpp \
// RUN: --offload-new-driver -fsycl-dead-args-optimization
// RUN: %{run} %t.out
//
// XFAIL: cuda
// XFAIL-TRACKER: https://github.com/intel/llvm/issues/16413

#include "Inputs/split-per-source.h"

Expand Down
2 changes: 0 additions & 2 deletions sycl/test-e2e/NewOffloadDriver/sycl-external-with-optional-features.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,6 @@
// RUN: %{build} -DSOURCE2 --offload-new-driver -c -o %t2.o
// RUN: %clangxx -Wno-error=unused-command-line-argument -fsycl -fsycl-targets=%{sycl_triple} --offload-new-driver %t1.o %t2.o -o %t.exe
// RUN: %{run} %t.exe
// XFAIL: cuda
// XFAIL-TRACKER: https://github.com/intel/llvm/issues/16413

#ifdef SOURCE1
#include <iostream>
Expand Down
Loading