mirror of
https://github.com/ml-explore/mlx.git
synced 2025-12-16 01:49:05 +08:00
Faster general unary op (#2472)
* faster general unary op * faster general ops + reorg * fix + comment * binary two * copy general
This commit is contained in:
@@ -8,7 +8,6 @@ target_sources(
|
||||
PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/allocator.cpp
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/arange.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/arg_reduce.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/binary.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/binary_two.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/compiled.cpp
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/copy.cu
|
||||
@@ -45,12 +44,14 @@ target_sources(
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/softmax.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/sort.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/ternary.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/unary.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/utils.cpp
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/quantized/affine_quantize.cu
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/quantized/quantized.cpp
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/worker.cpp)
|
||||
|
||||
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/binary)
|
||||
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/unary)
|
||||
|
||||
if(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 12.9.0)
|
||||
target_sources(
|
||||
mlx PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/gemms/cublas_gemm_batched_12_9.cu)
|
||||
|
||||
Reference in New Issue
Block a user