]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: f0152ef)
server : do not clear slots without unified KV cache (#24190)
authorfiesh <redacted>
Tue, 9 Jun 2026 07:45:16 +0000 (09:45 +0200)
committerGitHub <redacted>
Tue, 9 Jun 2026 07:45:16 +0000 (10:45 +0300)
commit961e9a3e46ca4cf7e6e86cfceb5b5e32084bf5f0
treedabd5dbe915b50fe4bca59ec250f2fdafb394073tree
parentf0152efe401acd5b329b5f62d87dc070a6d069f0commit | diff
server : do not clear slots without unified KV cache (#24190)

* Always export idle slots to RAM

Without this, a slot's VRAM cache may not be written to RAM.  If this
slot happens to be busy then later on, this triggers needless
preprocessing in another slot.

* cont : clean-up

---------

Co-authored-by: Christoph Weiss <redacted>
Co-authored-by: Georgi Gerganov <redacted>
common/arg.cpp diff | blob | history
tools/server/README.md diff | blob | history
tools/server/server-context.cpp diff | blob | history
Packaging of ggml-org/llama.cpp