Skip to content

vulkan: fix flash attention garbage output on MoltenVK with AMD GPUs #23218

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

Closed
csfercoci wants to merge 1 commit into from
Closed

vulkan: fix flash attention garbage output on MoltenVK with AMD GPUs #23218

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
24 changes: 14 additions & 10 deletions ggml/src/ggml-vulkan/ggml-vulkan.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2950,8 +2950,13 @@ static vk_fa_tuning_params get_fa_tuning_params_scalar(const vk_device& device,
vk_fa_tuning_params result{};
result.path = FA_SCALAR;

if (device->vendor_id == VK_VENDOR_ID_INTEL) {
// Disable subgroup use due to performance issues when enforcing subgroup sizes
if (device->vendor_id == VK_VENDOR_ID_INTEL
#ifdef __APPLE__
|| (device->vendor_id == VK_VENDOR_ID_AMD && device->driver_id == vk::DriverId::eMoltenvk)
#endif
) {
// Disable subgroup use — Intel has performance issues with subgroup sizes,
// MoltenVK/AMD has broken subgroupShuffleXor (Metal translation bug)
result.subgroup_size = 32;
result.disable_subgroups = true;
} else if (device->vendor_id == VK_VENDOR_ID_AMD && device->architecture != AMD_GCN) {
Expand Down Expand Up @@ -5240,23 +5245,22 @@ static vk_device ggml_vk_get_device(size_t idx) {
(vk11_props.subgroupSupportedOperations & vk::SubgroupFeatureFlagBits::eBasic);
device->subgroup_arithmetic = (vk11_props.subgroupSupportedStages & vk::ShaderStageFlagBits::eCompute) &&
(vk11_props.subgroupSupportedOperations & vk::SubgroupFeatureFlagBits::eArithmetic);
#ifdef __APPLE__
// Workaround for subgroup arithmetic failing on MoltenVK with AMD GPUs (issue 15846)
if (device->vendor_id == VK_VENDOR_ID_AMD) {
device->subgroup_arithmetic = false;
}
#endif
device->subgroup_shuffle = (vk11_props.subgroupSupportedStages & vk::ShaderStageFlagBits::eCompute) &&
(vk11_props.subgroupSupportedOperations & vk::SubgroupFeatureFlagBits::eShuffle);
device->subgroup_clustered = (vk11_props.subgroupSupportedStages & vk::ShaderStageFlagBits::eCompute) &&
(vk11_props.subgroupSupportedOperations & vk::SubgroupFeatureFlagBits::eClustered);

device->subgroup_ballot = (vk11_props.subgroupSupportedStages & vk::ShaderStageFlagBits::eCompute) &&
(vk11_props.subgroupSupportedOperations & vk::SubgroupFeatureFlagBits::eBallot);

device->subgroup_vote = (vk11_props.subgroupSupportedStages & vk::ShaderStageFlagBits::eCompute) &&
(vk11_props.subgroupSupportedOperations & vk::SubgroupFeatureFlagBits::eVote);

#ifdef __APPLE__
// Workaround for subgroup arithmetic failing on MoltenVK with AMD GPUs (issue 15846)
if (device->vendor_id == VK_VENDOR_ID_AMD) {
device->subgroup_arithmetic = false;
}
#endif

const bool force_disable_f16 = getenv("GGML_VK_DISABLE_F16") != nullptr;

device->fp16 = !force_disable_f16 && fp16_storage && fp16_compute;
Expand Down