]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
overview / pkg / ggml / sources / whisper.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: b092e95)
vulkan: mul_mat_id coopmat2 optimizations (llama/15546)
authorJeff Bolz <redacted>
Sun, 31 Aug 2025 07:06:43 +0000 (02:06 -0500)
committerGeorgi Gerganov <redacted>
Sat, 20 Sep 2025 10:42:45 +0000 (13:42 +0300)
commit191def71ced9cd38ff1a77d75e079820d35b3d23
tree3fd61ef1bc7b61bc86da5210d8cf34b5d1a74d1atree
parentb092e95aaa2743c5e7331a540b6850275fbaba6ccommit | diff
vulkan: mul_mat_id coopmat2 optimizations (llama/15546)

* vulkan: mul_mat_id coopmat2 optimizations

Add a path for when the tile fits in BN/2, similar to what we have for mul_mat.

Only call fetch_scales/store_scales once per QUANT_K block, and once at the
beginning in case start_k is not aligned.

* Also add a path for BN/4 - worth a couple more percent
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
ggml/src/ggml-vulkan/vulkan-shaders/mul_mm_cm2.comp diff | blob | history
Packaging of ggerganov/whisper.cpp