Merge remote-tracking branch 'nccl/master' into develop
[ROCm/rccl commit: 3f94267f21]
Этот коммит содержится в:
@@ -1073,7 +1073,7 @@ struct Apply_LoadMultimem {
|
||||
#endif
|
||||
|
||||
#if defined(RCCL_BFLOAT16)
|
||||
#if NCCL_CUDA_ARCH_FAMILY_SPECIFIC == 1000 || NCCL_CUDA_ARCH_FAMILY_SPECIFIC == 1010 || NCCL_CUDA_ARCH_SPECIFIC == 1200 || NCCL_CUDA_ARCH_SPECIFIC == 1210
|
||||
#if NCCL_CUDA_ARCH_SPECIFIC == 1000 || NCCL_CUDA_ARCH_SPECIFIC == 1010 || NCCL_CUDA_ARCH_FAMILY_SPECIFIC == 1000 || NCCL_CUDA_ARCH_FAMILY_SPECIFIC == 1010 || NCCL_CUDA_ARCH_SPECIFIC == 1200 || NCCL_CUDA_ARCH_SPECIFIC == 1210
|
||||
DEFINE_Apply_LoadMultimem_sum_v4_and_xparts(__nv_fp8_e4m3, e4m3x4, 4)
|
||||
DEFINE_Apply_LoadMultimem_minmax_v4_and_xparts(__nv_fp8_e4m3, e4m3x4, 4)
|
||||
DEFINE_Apply_LoadMultimem_sum_v4_and_xparts(__nv_fp8_e5m2, e5m2x4, 4)
|
||||
|
||||
Ссылка в новой задаче
Block a user