]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
overview / pkg / ggml / sources / whisper.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: ab5d4d9)
vulkan: Handle GPUs with less shared memory (llama/10468)
authorJeff Bolz <redacted>
Wed, 27 Nov 2024 07:30:27 +0000 (01:30 -0600)
committerGeorgi Gerganov <redacted>
Sun, 8 Dec 2024 18:14:35 +0000 (20:14 +0200)
commitb3301f7d82e0d0f8adf7f4067e74f3e2b7c163d5
treeb3603b15ad7860e2b415280a8b2da92f7cedcb2ctree
parentab5d4d93ecdc3dceaa40d66e9d8b250c12f32faacommit | diff
vulkan: Handle GPUs with less shared memory (llama/10468)

There have been reports of failure to compile on systems with <= 32KB
of shared memory (e.g. #10037). This change makes the large tile size
fall back to a smaller size if necessary, and makes mul_mat_id fall
back to CPU if there's only 16KB of shared memory.
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
Packaging of ggerganov/whisper.cpp