From 1fcdcc28b119a6608774d52de905931bd5f8a43d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Johannes=20G=C3=A4=C3=9Fler?= Date: Thu, 25 May 2023 23:07:29 +0200 Subject: cuda : performance optimizations (#1530) * xor hack * block y dim * loop unrolling * Fixed cmake LLAMA_CUDA_BY option * Removed hipblas compatibility code * Define GGML_CUDA_DMMV_BLOCK_Y if not defined * Fewer iters, more ops per iter * Renamed DMMV X/Y compilation options --- CMakeLists.txt | 54 +++++++++++++++++++++++++++++------------------------- 1 file changed, 29 insertions(+), 25 deletions(-) (limited to 'CMakeLists.txt') diff --git a/CMakeLists.txt b/CMakeLists.txt index 39db2e3..31c5bd9 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -37,42 +37,44 @@ endif() # # general -option(LLAMA_STATIC "llama: static link libraries" OFF) -option(LLAMA_NATIVE "llama: enable -march=native flag" OFF) -option(LLAMA_LTO "llama: enable link time optimization" OFF) +option(LLAMA_STATIC "llama: static link libraries" OFF) +option(LLAMA_NATIVE "llama: enable -march=native flag" OFF) +option(LLAMA_LTO "llama: enable link time optimization" OFF) # debug -option(LLAMA_ALL_WARNINGS "llama: enable all compiler warnings" ON) -option(LLAMA_ALL_WARNINGS_3RD_PARTY "llama: enable all compiler warnings in 3rd party libs" OFF) -option(LLAMA_GPROF "llama: enable gprof" OFF) +option(LLAMA_ALL_WARNINGS "llama: enable all compiler warnings" ON) +option(LLAMA_ALL_WARNINGS_3RD_PARTY "llama: enable all compiler warnings in 3rd party libs" OFF) +option(LLAMA_GPROF "llama: enable gprof" OFF) # sanitizers -option(LLAMA_SANITIZE_THREAD "llama: enable thread sanitizer" OFF) -option(LLAMA_SANITIZE_ADDRESS "llama: enable address sanitizer" OFF) -option(LLAMA_SANITIZE_UNDEFINED "llama: enable undefined sanitizer" OFF) +option(LLAMA_SANITIZE_THREAD "llama: enable thread sanitizer" OFF) +option(LLAMA_SANITIZE_ADDRESS "llama: enable address sanitizer" OFF) +option(LLAMA_SANITIZE_UNDEFINED "llama: enable undefined sanitizer" OFF) # instruction set specific -option(LLAMA_AVX "llama: enable AVX" ON) -option(LLAMA_AVX2 "llama: enable AVX2" ON) -option(LLAMA_AVX512 "llama: enable AVX512" OFF) -option(LLAMA_AVX512_VBMI "llama: enable AVX512-VBMI" OFF) -option(LLAMA_AVX512_VNNI "llama: enable AVX512-VNNI" OFF) -option(LLAMA_FMA "llama: enable FMA" ON) +option(LLAMA_AVX "llama: enable AVX" ON) +option(LLAMA_AVX2 "llama: enable AVX2" ON) +option(LLAMA_AVX512 "llama: enable AVX512" OFF) +option(LLAMA_AVX512_VBMI "llama: enable AVX512-VBMI" OFF) +option(LLAMA_AVX512_VNNI "llama: enable AVX512-VNNI" OFF) +option(LLAMA_FMA "llama: enable FMA" ON) # in MSVC F16C is implied with AVX2/AVX512 if (NOT MSVC) - option(LLAMA_F16C "llama: enable F16C" ON) + option(LLAMA_F16C "llama: enable F16C" ON) endif() # 3rd party libs -option(LLAMA_ACCELERATE "llama: enable Accelerate framework" ON) -option(LLAMA_BLAS "llama: use BLAS" OFF) -option(LLAMA_BLAS_VENDOR "llama: BLA_VENDOR from https://cmake.org/cmake/help/latest/module/FindBLAS.html#blas-lapack-vendors" Generic) -option(LLAMA_CUBLAS "llama: use cuBLAS" OFF) -option(LLAMA_CLBLAST "llama: use CLBlast" OFF) - -option(LLAMA_BUILD_TESTS "llama: build tests" ${LLAMA_STANDALONE}) -option(LLAMA_BUILD_EXAMPLES "llama: build examples" ${LLAMA_STANDALONE}) -option(LLAMA_BUILD_SERVER "llama: build server example" OFF) +option(LLAMA_ACCELERATE "llama: enable Accelerate framework" ON) +option(LLAMA_BLAS "llama: use BLAS" OFF) +option(LLAMA_BLAS_VENDOR "llama: BLA_VENDOR from https://cmake.org/cmake/help/latest/module/FindBLAS.html#blas-lapack-vendors" Generic) +option(LLAMA_CUBLAS "llama: use cuBLAS" OFF) +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_CLBLAST "llama: use CLBlast" OFF) + +option(LLAMA_BUILD_TESTS "llama: build tests" ${LLAMA_STANDALONE}) +option(LLAMA_BUILD_EXAMPLES "llama: build examples" ${LLAMA_STANDALONE}) +option(LLAMA_BUILD_SERVER "llama: build server example" OFF) # # Build info header @@ -184,6 +186,8 @@ if (LLAMA_CUBLAS) set(GGML_CUDA_SOURCES ggml-cuda.cu ggml-cuda.h) add_compile_definitions(GGML_USE_CUBLAS) + add_compile_definitions(GGML_CUDA_DMMV_X=${LLAMA_CUDA_DMMV_X}) + add_compile_definitions(GGML_CUDA_DMMV_Y=${LLAMA_CUDA_DMMV_Y}) if (LLAMA_STATIC) set(LLAMA_EXTRA_LIBS ${LLAMA_EXTRA_LIBS} CUDA::cudart_static CUDA::cublas_static CUDA::cublasLt_static) -- cgit v1.2.3