Skip to content

HIP: enable vec fattn on RDNA4 #14323

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 2 commits into from
Jun 22, 2025
Merged

HIP: enable vec fattn on RDNA4 #14323

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
f4e081c
CUDA: fix codestyle nit
IMbackK Jun 22, 2025
3ae60e7
HIP: enable vec fattn on RDNA4
IMbackK Jun 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
14 changes: 12 additions & 2 deletions ggml/src/ggml-cuda/common.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,8 +241,18 @@ static bool fp16_mma_available(const int cc) {
#if defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__) && !defined(GGML_HIP_ROCWMMA_FATTN)
return false;
#else
return (GGML_CUDA_CC_IS_NVIDIA(cc) && ggml_cuda_highest_compiled_arch(cc) >= GGML_CUDA_CC_VOLTA) ||
GGML_CUDA_CC_IS_CDNA(cc) || GGML_CUDA_CC_IS_RDNA3(cc) || GGML_CUDA_CC_IS_RDNA4(cc);
if ((GGML_CUDA_CC_IS_NVIDIA(cc) && ggml_cuda_highest_compiled_arch(cc) >= GGML_CUDA_CC_VOLTA) ||
GGML_CUDA_CC_IS_CDNA(cc) || GGML_CUDA_CC_IS_RDNA3(cc)) {
return true;
} else if (GGML_CUDA_CC_IS_RDNA4(cc)) {
#if defined(GGML_HIP_ROCWMMA_FATTN) && defined(GGML_HIP_ROCWMMA_FATTN_GFX12)
return true;
#else
return false;
#endif // defined(GGML_HIP_ROCWMMA_FATTN) && defined(GGML_HIP_ROCWMMA_FATTN_GFX12)
} else {
return false;
}
#endif // defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__) && !defined(GGML_HIP_ROCWMMA_FATTN)
}

Expand Down
7 changes: 2 additions & 5 deletions ggml/src/ggml-cuda/ggml-cuda.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,8 +100,7 @@ int ggml_cuda_get_device() {
static cudaError_t ggml_cuda_device_malloc(void ** ptr, size_t size, int device) {
ggml_cuda_set_device(device);
cudaError_t err;
if (getenv("GGML_CUDA_ENABLE_UNIFIED_MEMORY") != nullptr)
{
if (getenv("GGML_CUDA_ENABLE_UNIFIED_MEMORY") != nullptr) {
err = cudaMallocManaged(ptr, size);
#if defined(GGML_USE_HIP)
if (err == hipSuccess) {
Expand All @@ -119,9 +118,7 @@ static cudaError_t ggml_cuda_device_malloc(void ** ptr, size_t size, int device)
err = cudaMalloc(ptr, size);
}
#endif // defined(GGML_USE_HIP)
}
else
{
} else {
err = cudaMalloc(ptr, size);
}
return err;
Expand Down
Loading