diff options
| -rw-r--r-- | README.md | 2 |
1 files changed, 1 insertions, 1 deletions
@@ -532,7 +532,7 @@ Building the program with BLAS support may lead to some performance improvements cmake -B build -DLLAMA_HIPBLAS=ON -DAMDGPU_TARGETS=gfx1030 -DCMAKE_BUILD_TYPE=Release \ && cmake --build build --config Release -- -j 16 ``` - On Linux it is also possible to use unified memory architecture (UMA) to share main memory between the CPU and integrated GPU by setting `-DLLAMA_HIP_UMA=ON"`. + On Linux it is also possible to use unified memory architecture (UMA) to share main memory between the CPU and integrated GPU by setting `-DLLAMA_HIP_UMA=ON`. However, this hurts performance for non-integrated GPUs (but enables working with integrated GPUs). - Using `make` (example for target gfx1030, build with 16 CPU threads): |