]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 0dbcabd)
llama : add thread safety test (#14035)
authorDiego Devesa <redacted>
Mon, 16 Jun 2025 15:11:43 +0000 (08:11 -0700)
committerGitHub <redacted>
Mon, 16 Jun 2025 15:11:43 +0000 (08:11 -0700)
commit6adc3c3ebc029af058ac950a8e2a825fdf18ecc6
tree9640673c4751564564992758c499d3e69167c286tree
parent0dbcabde8c006d5cf781ca0fe071c41559572a72commit | diff
llama : add thread safety test (#14035)

* llama : add thread safety test

* llamafile : remove global state

* llama : better LLAMA_SPLIT_MODE_NONE logic

when main_gpu < 0 GPU devices are not used

---------

Co-authored-by: Georgi Gerganov <redacted>
.github/workflows/build.yml diff | blob | history
ci/run.sh diff | blob | history
common/common.cpp diff | blob | history
ggml/src/ggml-cpu/ggml-cpu-impl.h diff | blob | history
ggml/src/ggml-cpu/ggml-cpu.c diff | blob | history
ggml/src/ggml-cpu/llamafile/sgemm.cpp diff | blob | history
src/llama.cpp diff | blob | history
tests/CMakeLists.txt diff | blob | history
tests/test-thread-safety.cpp [new file with mode: 0644] blob
Packaging of ggml-org/llama.cpp