Skip to content

fix(hip): force VEC FA path for quantized KV (fixes RDNA2 max_blocks_per_sm assert) #13

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
TheTom wants to merge 1 commit into
base: main
Choose a base branch
from
+13 −0
Open

fix(hip): force VEC FA path for quantized KV (fixes RDNA2 max_blocks_per_sm assert) #13

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
13 changes: 13 additions & 0 deletions ggml/src/ggml-cuda/fattn.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -464,6 +464,19 @@ static best_fattn_kernel ggml_cuda_get_best_fattn_kernel(const int device, const
// For small batch sizes the vector kernel may be preferable over the kernels optimized for large batch sizes:
const bool can_use_vector_kernel = Q->ne[0] <= 256 && Q->ne[0] % 64 == 0 && K->ne[1] % FATTN_KQ_STRIDE == 0;

#ifdef GGML_USE_HIP
// HIP/ROCm: the TILE/MMA/WMMA FA paths allocate unbounded f16 temp buffers
// for quantized KV types (K_f16, V_f16 in launch_fattn). The pool retains
// peak allocation size, so the temp buffer VRAM exceeds KV compression savings.
// This causes quantized KV to OOM before f16 on the same context length.
// Force VEC path which does inline dequant with zero temp buffer overhead.
// Trade-off: prefill is slower (sequential query processing).
// Limitation: head_dim > 256 cannot use VEC (falls through to TILE).
if ((ggml_is_quantized(K->type) || ggml_is_quantized(V->type)) && can_use_vector_kernel) {
return BEST_FATTN_KERNEL_VEC;
}
#endif // GGML_USE_HIP

// If Turing tensor cores are available, use them:
if (turing_mma_available(cc) && Q->ne[0] != 40 && Q->ne[0] != 72) {
if (can_use_vector_kernel) {
Expand Down