]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
CUDA: fix race conditions FlashAttention kernels (llama/13438)
authorJohannes Gäßler <redacted>
Sat, 10 May 2025 20:22:48 +0000 (22:22 +0200)
committerGeorgi Gerganov <redacted>
Tue, 13 May 2025 10:59:21 +0000 (13:59 +0300)
commit6db0e01db69f9cc1bc7d5b17f18fad3eb672eed0
tree9fce8b7500ca23fc325f3bedd777f3931bd35235
parent16f3546f38cf4d58cd915c1066b86edac5fc4fef
CUDA: fix race conditions FlashAttention kernels (llama/13438)
ggml/src/ggml-cuda/fattn-mma-f16.cuh
ggml/src/ggml-cuda/fattn-vec-f16.cuh