[Model][Hardware][AMD][Kernel] Part 2/2 -> Enable e2e QK Norm + RoPE + KV Cache runtime fusion for Qwen3-30B-A3B on ROCM_ATTN (#42755)
Signed-off-by: Jack Hu <Jack.Hu@amd.com> Signed-off-by: Jack Hu <jingxujaster@hotmail.com> Signed-off-by: Jack Hu <jack.hu@amd.com> Co-authored-by: Douglas Lehr <91553416+dllehr-amd@users.noreply.github.com>
This commit is contained in:
@@ -92,7 +92,8 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, rocm_ops) {
|
||||
" str kv_cache_dtype,"
|
||||
" Tensor k_scale, Tensor v_scale,"
|
||||
" Tensor? fp8_out_scale,"
|
||||
" str mfma_type) -> ()");
|
||||
" str mfma_type,"
|
||||
" bool use_interleaved_v_cache=False) -> ()");
|
||||
rocm_ops.impl("paged_attention", torch::kCUDA, &paged_attention);
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user