Use compiler to determine best optimizations for platform · jerryrelmore/llama-cpp-python@dadfd96 · GitHub

jerryrelmore / llama-cpp-python Public

forked from abetlen/llama-cpp-python

• Notifications You must be signed in to change notification settings
• Fork 0
• Star 0

• Code
• Pull requests
• Actions
• Projects
• Security and quality
• Insights

Additional navigation options

Commit dadfd96

Browse filesBrowse the repository at this point in the historyBrowse files

abetlen

committed

Use compiler to determine best optimizations for platform

1 parent 4c0787b commit dadfd96
Copy full SHA for dadfd96

2 files changed

+9-1Lines changed: 9 additions & 1 deletion

File tree

Expand file treeCollapse file tree

Open diff view settings

Filter options

• .github/workflows
  • test.yaml
• CMakeLists.txt

Expand file treeCollapse file tree

Open diff view settings

Collapse file

‎.github/workflows/test.yaml‎

Copy file name to clipboardExpand all lines: .github/workflows/test.yaml

+1-1Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
@@ -73,7 +73,7 @@ jobs:
73	73	- name: Install dependencies
74	74	run: |
75	75	python3 -m pip install --upgrade pip
76		- CMAKE_ARGS="-DLLAMA_AVX=OFF -DLLAMA_AVX2=OFF -DLLAMA_AVX512=OFF -DLLAMA_AVX512_VBMI=OFF -DLLAMA_AVX512_VNNI=OFF -DLLAMA_FMA=OFF -DLLAMA_F16C=OFF -DLLAMA_ACCELERATE=OFF -DLLAMA_METAL=OFF" python3 -m pip install .[all] --verbose --config-settings=cmake.verbose=true --config-settings=logging.level=INFO
	76	+ python3 -m pip install .[all] --verbose
77	77	- name: Test with pytest
78	78	run: |
79	79	python3 -m pytest

Collapse file

‎CMakeLists.txt‎

Copy file name to clipboardExpand all lines: CMakeLists.txt

+8Lines changed: 8 additions & 0 deletions

Original file line number	Diff line number	Diff line change
@@ -4,6 +4,14 @@ project(llama_cpp)
4	4
5	5	option(BUILD_LLAMA_CPP "Build llama.cpp shared library and install alongside python package" ON)
6	6
	7	+ if (APPLE)
	8	+ set(LLAMA_AVX ON CACHE BOOL)
	9	+ set(LLAMA_AVX2 OFF CACHE BOOL)
	10	+ set(LLAMA_FMA OFF CACHE BOOL)
	11	+ set(LLAMA_F16C OFF CACHE BOOL)
	12	+ set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=native -mtune=native")
	13	+ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=native -mtune=native")
	14	+ endif()
7	15
8	16	if (BUILD_LLAMA_CPP)
9	17	set(BUILD_SHARED_LIBS "On")

0 commit comments

Comments

0 (0)