]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: fix FA VKQ accumulator overflow (llama/17746)
authorJohannes Gäßler <redacted>
Fri, 5 Dec 2025 08:18:10 +0000 (09:18 +0100)
committerGeorgi Gerganov <redacted>
Thu, 11 Dec 2025 13:32:54 +0000 (15:32 +0200)
commitbeccb8b75365f024cdfb90f2ffe2c81736831910
tree45d484a0d3d51dd487d214fdde250f757d46b658
parentb96dffa007fd33b8036a2961eb432c8f881fa00e
CUDA: fix FA VKQ accumulator overflow (llama/17746)
src/ggml-cuda/fattn-common.cuh
src/ggml-cuda/fattn-mma-f16.cuh
src/ggml-cuda/fattn-tile.cuh
src/ggml-cuda/fattn-vec.cuh
src/ggml-cuda/fattn-wmma-f16.cu