]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 9a3ea68)
vulkan: Call ggml_vk_buffer_write_2d from ggml_vk_buffer_copy (#16793)
authorJeff Bolz <redacted>
Wed, 29 Oct 2025 08:53:04 +0000 (03:53 -0500)
committerGitHub <redacted>
Wed, 29 Oct 2025 08:53:04 +0000 (09:53 +0100)
commitf549b0007dbdd683215820f7229ce180a12b191d
tree87765287f4ad28b7a183005d0c8e76b693bf4aa3tree
parent9a3ea685b937c0f0cbfda2e50004ea54bf187512commit | diff
vulkan: Call ggml_vk_buffer_write_2d from ggml_vk_buffer_copy (#16793)

This lets the copy to the destination device use the host-visible
vidmem optimization.
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
Packaging of ggml-org/llama.cpp