]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 5174d72)
CUDA: loop over ne2*ne3 in case it overflows (#19538)
authorAman Gupta <redacted>
Fri, 13 Feb 2026 11:31:40 +0000 (17:01 +0530)
committerGitHub <redacted>
Fri, 13 Feb 2026 11:31:40 +0000 (17:01 +0530)
commit5065da554e8e78e7f2dff35dede259c8dd56c9f3
tree5a995da83130257200bead1a5b7c7afe13930549tree
parent5174d7206fc743efa4da0d02e8668903a3eb8429commit | diff
CUDA: loop over ne2*ne3 in case it overflows (#19538)

* CUDA: loop over ne2*ne3 in case it overflows

* use fastdiv
ggml/src/ggml-cuda/convert.cu diff | blob | history
Packaging of ggml-org/llama.cpp