Skip to content

AMDGPU: Allow readonly features to be written to IR when there is no target #148141

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 9 commits into from
Jul 29, 2025
Merged

AMDGPU: Allow readonly features to be written to IR when there is no target #148141

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
c9cf987
AMDGPU: Remove "gws" from the “read-only” target feature list
changpeng Jul 11, 2025
5beb2e8
AMDGPU: Remove "gws" from the “read-only” target feature list
changpeng Jul 11, 2025
d78dfd9
AMDGPU: Remove "gws" from the “read-only” target feature list
changpeng Jul 11, 2025
de11611
Merge branch 'llvm:main' into readonly
changpeng Jul 15, 2025
afeec4b
Merge branch 'llvm:main' into readonly
changpeng Jul 16, 2025
532226f
Merge branch 'llvm:main' into readonly
changpeng Jul 18, 2025
57ea48f
Merge branch 'llvm:main' into readonly
changpeng Jul 21, 2025
56aa23d
AMDGPU: Mark features as read-only only when target supports them
changpeng Jul 21, 2025
870f9a9
AMDGPU: Allow readonly features to be written to IR when there is no …
changpeng Jul 22, 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
7 changes: 5 additions & 2 deletions clang/lib/Basic/Targets/AMDGPU.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -266,8 +266,11 @@ AMDGPUTargetInfo::AMDGPUTargetInfo(const llvm::Triple &Triple,

MaxAtomicPromoteWidth = MaxAtomicInlineWidth = 64;
CUMode = !(GPUFeatures & llvm::AMDGPU::FEATURE_WGP);
for (auto F : {"image-insts", "gws", "vmem-to-lds-load-insts"})
ReadOnlyFeatures.insert(F);

for (auto F : {"image-insts", "gws", "vmem-to-lds-load-insts"}) {
if (GPUKind != llvm::AMDGPU::GK_NONE)
ReadOnlyFeatures.insert(F);
}
HalfArgsAndReturns = true;
}

Expand Down
2 changes: 1 addition & 1 deletion clang/test/CodeGenOpenCL/amdgpu-features-readonly.cl
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
// RUN: %clang_cc1 -triple amdgcn -target-feature +gws -o /dev/null %s 2>&1 \
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx942 -target-feature +gws -o /dev/null %s 2>&1 \
// RUN: | FileCheck --check-prefix=GWS %s

// GWS: warning: feature flag '+gws' is ignored since the feature is read only [-Winvalid-command-line-argument]
Expand Down
16 changes: 16 additions & 0 deletions clang/test/CodeGenOpenCL/amdgpu-readonly-features-written-with-no-target.cl
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
// REQUIRES: amdgpu-registered-target

// Check the readonly feature will can be written to the IR
// if there is no target specified.

// RUN: %clang_cc1 -triple amdgcn -emit-llvm -o - %s | FileCheck --check-prefix=NOCPU %s
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx942 -emit-llvm -o - %s | FileCheck --check-prefix=GFX942 %s
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1100 -emit-llvm -o - %s | FileCheck --check-prefix=GFX1100 %s
// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1200 -emit-llvm -o - %s | FileCheck --check-prefix=GFX1200 %s

__attribute__((target("gws,image-insts,vmem-to-lds-load-insts"))) void test() {}

// NOCPU: "target-features"="+gws,+image-insts,+vmem-to-lds-load-insts"
// GFX942: "target-features"="+16-bit-insts,+atomic-buffer-global-pk-add-f16-insts,+atomic-ds-pk-add-16-insts,+atomic-fadd-rtn-insts,+atomic-flat-pk-add-16-insts,+atomic-global-pk-add-bf16-inst,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+fp8-conversion-insts,+fp8-insts,+gfx8-insts,+gfx9-insts,+gfx90a-insts,+gfx940-insts,+mai-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize64,+xf32-insts"
// GFX1100: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot12-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32"
// GFX1200: "target-features"="+16-bit-insts,+atomic-buffer-global-pk-add-f16-insts,+atomic-buffer-pk-add-bf16-inst,+atomic-ds-pk-add-16-insts,+atomic-fadd-rtn-insts,+atomic-flat-pk-add-16-insts,+atomic-global-pk-add-bf16-inst,+ci-insts,+dl-insts,+dot10-insts,+dot11-insts,+dot12-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+fp8-conversion-insts,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx12-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32"