happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix ci failures on turing and hip

This commit is contained in:
Patrick Buckley 2026-03-15 04:15:58 -07:00
parent adc7d74a07
commit d6b945ef5d
3 changed files with 14 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/src/ggml-cuda/convert.cuh
View File

@ -42,7 +42,11 @@ template<typename dst_t, typename src_t>
} else if constexpr(std::is_same_v<src_t, float2> && std::is_same_v<dst_t, half2>) {
return __float22half2_rn(x);
} else if constexpr(std::is_same_v<src_t, nv_bfloat162> && std::is_same_v<dst_t, float2>) {
#if !defined(GGML_USE_HIP) && defined(__CUDA_ARCH__) && __CUDA_ARCH__ >= 800
return __bfloat1622float2(x);
#else
return make_float2(__bfloat162float(__low2bfloat16(x)), __bfloat162float(__high2bfloat16(x)));
#endif
} else if constexpr(std::is_same_v<src_t, float2> && std::is_same_v<dst_t, nv_bfloat162>) {
// bypass compile error on cuda 12.0.1
#ifdef GGML_USE_HIP

11
ggml/src/ggml-hip/CMakeLists.txt
View File

@ -74,12 +74,11 @@ if (GGML_CUDA_FA_ALL_QUANTS)
list(APPEND GGML_SOURCES_ROCM ${SRCS})
add_compile_definitions(GGML_CUDA_FA_ALL_QUANTS)
else()
file(GLOB SRCS "../ggml-cuda/template-instances/fattn-vec*q4_0-q4_0.cu")
list(APPEND GGML_SOURCES_ROCM ${SRCS})
file(GLOB SRCS "../ggml-cuda/template-instances/fattn-vec*q8_0-q8_0.cu")
list(APPEND GGML_SOURCES_ROCM ${SRCS})
file(GLOB SRCS "../ggml-cuda/template-instances/fattn-vec*f16-f16.cu")
list(APPEND GGML_SOURCES_ROCM ${SRCS})
list(APPEND GGML_SOURCES_ROCM
../ggml-cuda/template-instances/fattn-vec-instance-f16-f16.cu
../ggml-cuda/template-instances/fattn-vec-instance-q4_0-q4_0.cu
../ggml-cuda/template-instances/fattn-vec-instance-q8_0-q8_0.cu
../ggml-cuda/template-instances/fattn-vec-instance-bf16-bf16.cu)
endif()
ggml_add_backend_library(ggml-hip

11
ggml/src/ggml-musa/CMakeLists.txt
View File

@ -48,12 +48,11 @@ if (MUSAToolkit_FOUND)
list(APPEND GGML_SOURCES_MUSA ${SRCS})
add_compile_definitions(GGML_CUDA_FA_ALL_QUANTS)
else()
file(GLOB SRCS "../ggml-cuda/template-instances/fattn-vec*q4_0-q4_0.cu")
list(APPEND GGML_SOURCES_MUSA ${SRCS})
file(GLOB SRCS "../ggml-cuda/template-instances/fattn-vec*q8_0-q8_0.cu")
list(APPEND GGML_SOURCES_MUSA ${SRCS})
file(GLOB SRCS "../ggml-cuda/template-instances/fattn-vec*f16-f16.cu")
list(APPEND GGML_SOURCES_MUSA ${SRCS})
list(APPEND GGML_SOURCES_MUSA
../ggml-cuda/template-instances/fattn-vec-instance-f16-f16.cu
../ggml-cuda/template-instances/fattn-vec-instance-q4_0-q4_0.cu
../ggml-cuda/template-instances/fattn-vec-instance-q8_0-q8_0.cu
../ggml-cuda/template-instances/fattn-vec-instance-bf16-bf16.cu)
endif()
set_source_files_properties(${GGML_SOURCES_MUSA} PROPERTIES LANGUAGE CXX)