]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: f709c7a)
mmap: add Haiku support by skipping RLIMIT_MEMLOCK check (#18819)
authorDaniel Benjaminsson <redacted>
Wed, 14 Jan 2026 07:11:05 +0000 (08:11 +0100)
committerGitHub <redacted>
Wed, 14 Jan 2026 07:11:05 +0000 (09:11 +0200)
commitd34aa07193d27aa04da9a77c63ee125ec614714a
tree3617208770f2dd683263c89ffea5c27ac388c37etree
parentf709c7a33fd89dd4de3f0eeccbfd2da8297fe9d3commit | diff
mmap: add Haiku support by skipping RLIMIT_MEMLOCK check (#18819)

Haiku OS does not support RLIMIT_MEMLOCK, similar to visionOS/tvOS.
Skip the resource limit check on Haiku to allow mlock functionality
to work without compile errors.

Tested on Haiku with NVIDIA RTX 3080 Ti using Vulkan backend.
src/llama-mmap.cpp diff | blob | history
Packaging of ggml-org/llama.cpp