ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-04-20 17:11:17 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

CUDA: compress mode option and default to size (llama/12029)

Browse Source 
cuda 12.8 added the option to specify stronger compression for binaries, so we now default to "size".
This commit is contained in:
Erik Scholz 2025-03-01 12:57:22 +01:00 committed by Georgi Gerganov
parent c98681e6d5
commit b442dcd598
2 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
ggml/CMakeLists.txt
View File

@ -155,6 +155,9 @@ option(GGML_CUDA_NO_VMM "ggml: do not try to use CUDA VMM"
option(GGML_CUDA_FA "ggml: compile ggml FlashAttention CUDA kernels" ON)
option(GGML_CUDA_FA_ALL_QUANTS "ggml: compile all quants for FlashAttention" OFF)
option(GGML_CUDA_GRAPHS "ggml: use CUDA graphs (llama.cpp only)" ${GGML_CUDA_GRAPHS_DEFAULT})
set (GGML_CUDA_COMPRESSION_MODE "size" CACHE STRING
"ggml: cuda link binary compression mode; requires cuda 12.8+")
set_property(CACHE GGML_CUDA_COMPRESSION_MODE PROPERTY STRINGS "none;speed;balance;size")
option(GGML_HIP "ggml: use HIP" OFF)
option(GGML_HIP_GRAPHS "ggml: use HIP graph, experimental, slow" OFF)

9
ggml/src/ggml-cuda/CMakeLists.txt
View File

@ -102,6 +102,15 @@ if (CUDAToolkit_FOUND)
set(CUDA_FLAGS -use_fast_math)
if (CUDAToolkit_VERSION VERSION_GREATER_EQUAL "12.8")
# Options are:
# - none (not recommended)
# - speed (nvcc's default)
# - balance
# - size
list(APPEND CUDA_FLAGS -compress-mode=${GGML_CUDA_COMPRESSION_MODE})
endif()
if (GGML_FATAL_WARNINGS)
list(APPEND CUDA_FLAGS -Werror all-warnings)
endif()