tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-19 22:36:13 -04:00
Code Issues Packages Projects Releases Wiki Activity

MUSA: support ARM64 and enable dp4a .etc (#11843)

Browse Source 
* MUSA:  support ARM64 and enable __dp4a .etc

* fix cross entropy loss op for musa

* update

* add cc info log for musa

* add comment for the MUSA .cc calculation block

---------

Co-authored-by: Bodhi Hu <huaishun.hu@mthreads.com>
This commit is contained in:
Bodhi
2025-02-21 15:46:23 +08:00
committed by GitHub
parent ee02ad02c5
commit 0b3863ff95
7 changed files with 25 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/build.md
View File

@@ -206,6 +206,14 @@ This provides GPU acceleration using the MUSA cores of your Moore Threads MTT GP
cmake --build build --config Release
```
For static build:
```bash
cmake -B build -DGGML_MUSA=ON \
-DBUILD_SHARED_LIBS=OFF -DCMAKE_POSITION_INDEPENDENT_CODE=ON
cmake --build build --config Release
```
The environment variable [`MUSA_VISIBLE_DEVICES`](https://docs.mthreads.com/musa-sdk/musa-sdk-doc-online/programming_guide/Z%E9%99%84%E5%BD%95/) can be used to specify which GPU(s) will be used.
The environment variable `GGML_CUDA_ENABLE_UNIFIED_MEMORY=1` can be used to enable unified memory in Linux. This allows swapping to system RAM instead of crashing when the GPU VRAM is exhausted.