Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

dequantize + matrix multiplication CUDA kernels #2043

Closed
wants to merge 2 commits into from
Closed

dequantize + matrix multiplication CUDA kernels #2043

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2d55023
dequantize + matrix multiplication CUDA kernels
JohannesGaessler Jun 26, 2023
b90c80b
Add __restrict__ to dequantize_mul_mat kernels
JohannesGaessler Jul 8, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,6 +71,7 @@ option(LLAMA_CUBLAS "llama: use cuBLAS"
set(LLAMA_CUDA_DMMV_X "32" CACHE STRING "llama: x stride for dmmv CUDA kernels")
set(LLAMA_CUDA_DMMV_Y "1" CACHE STRING "llama: y block size for dmmv CUDA kernels")
option(LLAMA_CUDA_DMMV_F16 "llama: use 16 bit floats for dmmv CUDA kernels" OFF)
option(LLAMA_CUDA_DMM "llama: use dequantize mul mat CUDA kernels" OFF)
set(LLAMA_CUDA_KQUANTS_ITER "2" CACHE STRING "llama: iters./thread per block for Q2_K/Q6_K")
option(LLAMA_CLBLAST "llama: use CLBlast" OFF)
option(LLAMA_METAL "llama: use Metal" OFF)
Expand Down Expand Up @@ -251,6 +252,9 @@ if (LLAMA_CUBLAS)
if (LLAMA_CUDA_DMMV_F16)
add_compile_definitions(GGML_CUDA_DMMV_F16)
endif()
if (LLAMA_CUDA_DMM)
add_compile_definitions(GGML_CUDA_DMM)
endif()
add_compile_definitions(K_QUANTS_PER_ITERATION=${LLAMA_CUDA_KQUANTS_ITER})

if (LLAMA_STATIC)
Expand Down
3 changes: 3 additions & 0 deletions Makefile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -179,6 +179,9 @@ endif # LLAMA_CUDA_DMMV_Y
ifdef LLAMA_CUDA_DMMV_F16
NVCCFLAGS += -DGGML_CUDA_DMMV_F16
endif # LLAMA_CUDA_DMMV_F16
ifdef LLAMA_CUDA_DMM
NVCCFLAGS += -DGGML_CUDA_DMM
endif # LLAMA_CUDA_DMM
ifdef LLAMA_CUDA_KQUANTS_ITER
NVCCFLAGS += -DK_QUANTS_PER_ITERATION=$(LLAMA_CUDA_KQUANTS_ITER)
else
Expand Down
Loading