From 4d9ebd1b7fe0647f12b3fefa8fea7668381f581c Mon Sep 17 00:00:00 2001 From: Alberto Cabrera Date: Thu, 26 Sep 2024 16:37:55 +0100 Subject: [PATCH] sycl: initial cmake support of SYCL for AMD GPUs --- ggml/src/CMakeLists.txt | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/ggml/src/CMakeLists.txt b/ggml/src/CMakeLists.txt index cbc349500..97807efba 100644 --- a/ggml/src/CMakeLists.txt +++ b/ggml/src/CMakeLists.txt @@ -511,8 +511,8 @@ if (GGML_HIPBLAS) endif() if (GGML_SYCL) - if (NOT GGML_SYCL_TARGET MATCHES "^(INTEL|NVIDIA)$") - message(FATAL_ERROR "Invalid backend chosen, supported options are INTEL or NVIDIA") + if (NOT GGML_SYCL_TARGET MATCHES "^(INTEL|NVIDIA|AMD)$") + message(FATAL_ERROR "Invalid backend chosen, supported options are INTEL, NVIDIA, or AMD") endif() check_cxx_compiler_flag("-fsycl" SUPPORTS_SYCL) @@ -532,6 +532,9 @@ if (GGML_SYCL) list(APPEND GGML_CDEF_PUBLIC GGML_USE_SYCL) if (GGML_SYCL_F16) + if (GGML_SYCL_TARGET STREQUAL "AMD") + message(WARNING "AMD target does not entirely support FP16 in the SYCL backend.") + endif() add_compile_definitions(GGML_SYCL_F16) endif() @@ -543,6 +546,11 @@ if (GGML_SYCL) if (GGML_SYCL_TARGET STREQUAL "NVIDIA") add_compile_definitions(GGML_SYCL_WARP_SIZE=32) + elseif (GGML_SYCL_TARGET STREQUAL "AMD") + # INFO: Allowed Sub_group_siezs are not consistent through all + # hip targets. For example, 64 is used for certain models, but the backend + # does not support it. + add_compile_definitions(GGML_SYCL_WARP_SIZE=32) else() add_compile_definitions(GGML_SYCL_WARP_SIZE=16) endif() @@ -576,6 +584,12 @@ if (GGML_SYCL) elseif (GGML_SYCL_TARGET STREQUAL "NVIDIA") set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsycl-targets=nvptx64-nvidia-cuda") list(APPEND GGML_EXTRA_LIBS_PRIVATE sycl pthread m dl onemkl) + elseif (GGML_SYCL_TARGET STREQUAL "AMD") + if (GGML_SYCL_HIP_TARGET STREQUAL "") + message(ERROR "Can't enable SYCL hip backend, GGML_SYCL_HIP_TARGET has not been set.") + endif() + set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsycl-targets=amdgcn-amd-amdhsa -Xsycl-target-backend --offload-arch=${GGML_SYCL_HIP_TARGET}") + list(APPEND GGML_EXTRA_LIBS_PRIVATE sycl pthread m dl onemkl) endif() endif() endif()