mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2025-01-12 19:50:17 +00:00
5cb6209de5
- cmake has a hardcoded `onemkl` that will cause linker failure for AMD path and needs to be changed - docfix: oneMKL compile needs `-DHIP_TARGETS` not `-DHIPTARGETS` - docfix: modified cmake build parameters to pass in more linker flags, otherwise linking will fail (might be slight overkill, but without adding `-DCMAKE_SHARED_LINKER_FLAGS` it fails |
||
|---|---|---|
| .. | ||
| include | ||
| src | ||
| .gitignore | ||
| CMakeLists.txt | ||