]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 52bb63c)
Vulkan Intel Fixes, Optimizations and Debugging Flags (#5301)
author0cc4m <redacted>
Sat, 3 Feb 2024 17:15:00 +0000 (18:15 +0100)
committerGitHub <redacted>
Sat, 3 Feb 2024 17:15:00 +0000 (18:15 +0100)
commite920ed393d989ed35625ddaf182ebb52cda07fcd
treec693f5c20d9adee08ebb95e28619cb3824eadc69tree
parent52bb63c7082c859c3f1dfc527227e6a95b299c7ccommit | diff
Vulkan Intel Fixes, Optimizations and Debugging Flags (#5301)

* Fix Vulkan on Intel ARC

Optimize matmul for Intel ARC

Add Vulkan dequant test

* Add Vulkan debug and validate flags to Make and CMakeLists.txt

* Enable asynchronous transfers in Vulkan backend

* Fix flake8

* Disable Vulkan async backend functions for now

* Also add Vulkan run tests command to Makefile and CMakeLists.txt
CMakeLists.txt diff | blob | history
Makefile diff | blob | history
ggml-vulkan-shaders.hpp diff | blob | history
ggml-vulkan.cpp diff | blob | history
ggml_vk_generate_shaders.py diff | blob | history
Packaging of ggml-org/llama.cpp