]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: f5636f8)
CUDA: directly include cuda/iterator (#22936)
authorOliver Simons <redacted>
Mon, 11 May 2026 10:16:38 +0000 (12:16 +0200)
committerGitHub <redacted>
Mon, 11 May 2026 10:16:38 +0000 (12:16 +0200)
commit8cef8201a1e0213662abbfcbcd3ff2eb773174df
tree349bea638ece36df17fb8bce75d3cf515a7724f4tree
parentf5636f8fc77e403596a8524d789c4a16350837accommit | diff
CUDA: directly include cuda/iterator (#22936)

Before, we relied on a transient import from `cub/cub.cuh`, which is
bad practice to do as cub may not always expose cuda/iterator
ggml/src/ggml-cuda/argsort.cu diff | blob | history
Packaging of ggml-org/llama.cpp