CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-15 11:17:31 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml : remove k_quants_per_iteration macro

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2024-07-04 21:19:09 +03:00
parent 436787f170
commit e48fd74b45
12 changed files with 110 additions and 283 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/CMakeLists.txt
View File

@@ -120,8 +120,6 @@ option(GGML_CUDA_FORCE_CUBLAS "ggml: always use cuBLAS instead of
set (GGML_CUDA_DMMV_X "32" CACHE STRING "ggml: x stride for dmmv CUDA kernels")
set (GGML_CUDA_MMV_Y "1" CACHE STRING "ggml: y block size for mmv CUDA kernels")
option(GGML_CUDA_F16 "ggml: use 16 bit floats for some calculations" OFF)
set (GGML_CUDA_KQUANTS_ITER "2" CACHE STRING
"ggml: iters./thread per block for Q2_K/Q6_K")
set (GGML_CUDA_PEER_MAX_BATCH_SIZE "128" CACHE STRING
"ggml: max. batch size for using peer access")
option(GGML_CUDA_NO_PEER_COPY "ggml: do not use peer to peer copies" OFF)