Skip to content

[Clang][Driver] Only enable internalization for OpenMP target offloading with ThinLTO on AMDGPU #138547

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

Merged
merged 1 commit into from
May 5, 2025
Merged

[Clang][Driver] Only enable internalization for OpenMP target offloading with ThinLTO on AMDGPU #138547

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
11 changes: 5 additions & 6 deletions clang/lib/Driver/ToolChains/Clang.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9284,12 +9284,6 @@ void LinkerWrapper::ConstructJob(Compilation &C, const JobAction &JA,
CmdArgs.push_back(Args.MakeArgString(
"--device-linker=" + TC->getTripleString() + "=" + Arg));

// Enable internalization for AMDGPU.
if (TC->getTriple().isAMDGPU())
CmdArgs.push_back(
Args.MakeArgString("--device-linker=" + TC->getTripleString() +
"=-plugin-opt=-amdgpu-internalize-symbols"));

// Forward the LTO mode relying on the Driver's parsing.
if (C.getDriver().getOffloadLTOMode() == LTOK_Full)
CmdArgs.push_back(Args.MakeArgString(
Expand All @@ -9304,6 +9298,11 @@ void LinkerWrapper::ConstructJob(Compilation &C, const JobAction &JA,
CmdArgs.push_back(
Args.MakeArgString("--device-linker=" + TC->getTripleString() +
"=-plugin-opt=-avail-extern-to-local"));
if (Kind == Action::OFK_OpenMP) {
CmdArgs.push_back(
Args.MakeArgString("--device-linker=" + TC->getTripleString() +
"=-plugin-opt=-amdgpu-internalize-symbols"));
}
}
}
}
Expand Down
6 changes: 0 additions & 6 deletions clang/test/Driver/hip-options.hip
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -259,9 +259,3 @@
// RUN: --offload-arch=gfx1100 --offload-new-driver --offload-jobs=0x4 %s 2>&1 | \
// RUN: FileCheck -check-prefix=INVJOBS %s
// INVJOBS: clang: error: invalid integral value '0x4' in '--offload-jobs=0x4'

// Check that internalization is enabled for offloading on AMDGPU.
// RUN: %clang -### -Werror --target=x86_64-unknown-linux-gnu -nogpuinc -nogpulib \
// RUN: --offload-arch=gfx1100 --offload-new-driver %s 2>&1 | \
// RUN: FileCheck -check-prefix=INTERNALIZATION %s
// INTERNALIZATION: --device-linker=amdgcn-amd-amdhsa=-plugin-opt=-amdgpu-internalize-symbols
8 changes: 1 addition & 7 deletions clang/test/Driver/openmp-offload-gpu.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -388,15 +388,9 @@
// THINLTO-GFX906: --device-compiler=amdgcn-amd-amdhsa=-flto=thin
// THINLTO-GFX906-SAME: --device-linker=amdgcn-amd-amdhsa=-plugin-opt=-force-import-all
// THINLTO-GFX906-SAME: --device-linker=amdgcn-amd-amdhsa=-plugin-opt=-avail-extern-to-local
// THINLTO-GFX906-SAME: --device-linker=amdgcn-amd-amdhsa=-plugin-opt=-amdgpu-internalize-symbols
//
// RUN: %clang -### --target=x86_64-unknown-linux-gnu -fopenmp=libomp \
// RUN: --offload-arch=sm_52 -foffload-lto=thin -nogpulib -nogpuinc %s 2>&1 \
// RUN: | FileCheck --check-prefix=THINLTO-SM52 %s
// THINLTO-SM52: --device-compiler=nvptx64-nvidia-cuda=-flto=thin

// Check that internalization is enabled for OpenMP offloading on AMDGPU.
//
// RUN: %clang -### --target=x86_64-unknown-linux-gnu -fopenmp=libomp \
// RUN: --offload-arch=gfx906 -nogpulib -nogpuinc %s 2>&1 \
// RUN: | FileCheck --check-prefix=INTERNALIZATION-GFX906 %s
// INTERNALIZATION-GFX906: --device-linker=amdgcn-amd-amdhsa=-plugin-opt=-amdgpu-internalize-symbols