biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Attention] MLA decode optimizations (#12528)

Browse Source 
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
Signed-off-by: simon-mo <xmo@berkeley.edu>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: simon-mo <simon.mo@hey.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Zhuohan Li <zhuohan123@gmail.com>
Co-authored-by: Tyler Michael Smith <tysmith@redhat.com>
Co-authored-by: Alexander Matveev <59768536+alexm-neuralmagic@users.noreply.github.com>
Co-authored-by: simon-mo <xmo@berkeley.edu>
This commit is contained in:
Lucas Wilkinson
2025-01-31 02:49:37 -05:00
committed by GitHub
parent a1fc18c030
commit cabaf4eff3
31 changed files with 2266 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
csrc/torch_bindings.cpp
View File

@@ -463,6 +463,15 @@ TORCH_LIBRARY_EXPAND(CONCAT(TORCH_EXTENSION_NAME, _cache_ops), cache_ops) {
cache_ops.impl("reshape_and_cache_flash", torch::kCUDA,
&reshape_and_cache_flash);
// Concat kv_c and k_pe and cache them.
cache_ops.def(
"concat_and_cache_mla(Tensor kv_c, Tensor k_pe,"
" Tensor! kv_cache,"
" Tensor slot_mapping,"
" str kv_cache_dtype,"
" Tensor scale) -> ()");
cache_ops.impl("concat_and_cache_mla", torch::kCUDA, &concat_and_cache_mla);
// Convert the key and value cache to fp8 data type.
cache_ops.def(
"convert_fp8(Tensor! dst_cache, Tensor src_cache, float scale, "