CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-17 11:37:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

CUDA/HIP: Share the same unified memory allocation logic. (#12934)

Browse Source 
Replace compile-time `GGML_HIP_UMA` with environment variable `GGML_CUDA_ENABLE_UNIFIED_MEMORY`. This unifies the usage on NVIDIA and AMD GPUs, and allows a single binary to be shared between integrated and dedicated GPUs.
This commit is contained in:
David Huang
2025-04-15 17:20:38 +08:00
committed by GitHub
parent 510676475f
commit 84778e9770
6 changed files with 22 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/src/ggml-hip/CMakeLists.txt
View File

@@ -89,10 +89,6 @@ endif()
add_compile_definitions(GGML_USE_HIP)
if (GGML_HIP_UMA)
add_compile_definitions(GGML_HIP_UMA)
endif()
if (GGML_CUDA_FORCE_MMQ)
add_compile_definitions(GGML_CUDA_FORCE_MMQ)
endif()