ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-30 19:31:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

Better CPU prompt processing performance for SWA models (#696)

Browse Source 
* This does the trick for PP

* Compute mask bounds when creating the mask

* Set mask bounds for all supported SWA models

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow
2025-08-17 10:30:27 +03:00
committed by GitHub
parent 4bf5c8184b
commit 93a4f6089f
5 changed files with 140 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/include/ggml.h
View File

@@ -2043,6 +2043,10 @@ extern "C" {
struct ggml_tensor * a,
struct ggml_tensor * sinks);
GGML_API void ggml_flash_attn_ext_add_bounds(
struct ggml_tensor * a,
struct ggml_tensor * bounds);
// TODO: needs to be adapted to ggml_flash_attn_ext
GGML_API struct ggml_tensor * ggml_flash_attn_back(
struct ggml_context * ctx,