CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

Vulkan Intel Fixes, Optimizations and Debugging Flags (#5301)

Browse Source 
* 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
This commit is contained in:
0cc4m
2024-02-03 18:15:00 +01:00
committed by GitHub
parent 52bb63c708
commit e920ed393d
5 changed files with 1257 additions and 10330 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
Makefile
View File

@@ -457,6 +457,18 @@ ifdef LLAMA_VULKAN_CHECK_RESULTS
MK_CPPFLAGS += -DGGML_VULKAN_CHECK_RESULTS
endif
ifdef LLAMA_VULKAN_DEBUG
MK_CPPFLAGS += -DGGML_VULKAN_DEBUG
endif
ifdef LLAMA_VULKAN_VALIDATE
MK_CPPFLAGS += -DGGML_VULKAN_VALIDATE
endif
ifdef LLAMA_VULKAN_RUN_TESTS
MK_CPPFLAGS += -DGGML_VULKAN_RUN_TESTS
endif
ggml-vulkan.o: ggml-vulkan.cpp ggml-vulkan.h
$(CXX) $(CXXFLAGS) -c $< -o $@
endif # LLAMA_VULKAN