Skip to content

[SYCL][AMDGPU] Use uint64_t type for SYCLReqdWorkGroupSize product type #22622

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

Open
wenju-he wants to merge 2 commits into
base: sycl
Choose a base branch
from
+30 −3
Open

[SYCL][AMDGPU] Use uint64_t type for SYCLReqdWorkGroupSize product type #22622

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8155ee0
[SYCL][AMDGPU] Use uint64_t type for SYCLReqdWorkGroupSize product type
wenju-he Jul 14, 2026
3087647
add test
wenju-he Jul 17, 2026
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
6 changes: 3 additions & 3 deletions clang/lib/CodeGen/Targets/AMDGPU.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -352,10 +352,10 @@ void AMDGPUTargetCodeGenInfo::setFunctionDeclAttributes(
M.handleAMDGPUFlatWorkGroupSizeAttr(F, FlatWGS, ReqdWGS);
} else if (M.getLangOpts().SYCLIsDevice) {
if (const auto *SYCLReqdWGS = FD->getAttr<SYCLReqdWorkGroupSizeAttr>()) {
auto GetDim = [](std::optional<llvm::APSInt> V) -> unsigned {
return V ? (unsigned)V->getZExtValue() : 1;
auto GetDim = [](std::optional<llvm::APSInt> V) -> uint64_t {
return V ? V->getZExtValue() : 1;
};
unsigned Size = GetDim(SYCLReqdWGS->getXDimVal()) *
uint64_t Size = GetDim(SYCLReqdWGS->getXDimVal()) *
GetDim(SYCLReqdWGS->getYDimVal()) *
GetDim(SYCLReqdWGS->getZDimVal());
std::string AttrVal = llvm::utostr(Size) + "," + llvm::utostr(Size);
Expand Down
27 changes: 27 additions & 0 deletions clang/test/CodeGenSYCL/reqd-work-group-size-overflow.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,27 @@
// RUN: %clang_cc1 -fsycl-is-device -internal-isystem %S/Inputs -triple amdgcn-amd-amdhsa -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s

// The product of the reqd_work_group_size dimensions below is 2^32, which
// overflows a 32-bit unsigned integer. Check that the resulting
// amdgpu-flat-work-group-size function attribute reflects the full 64-bit
// product instead of a wrapped-around 32-bit value.

#include "Inputs/sycl.hpp"

using namespace sycl;
queue q;

class FunctorOverflow {
public:
[[sycl::reqd_work_group_size(65536, 65536)]] void operator()() const {}
};

int main() {
q.submit([&](handler &h) {
FunctorOverflow f;
h.single_task<class kernel_overflow>(f);
});
return 0;
}

// CHECK: define {{.*}} void @{{.*}}kernel_overflow() #[[ATTR:[0-9]+]]
// CHECK: attributes #[[ATTR]] = { {{.*}}"amdgpu-flat-work-group-size"="4294967296,4294967296"{{.*}} }
Loading