v3.8.0 update (#2082)

* 3.8 update

* fix Markus' name

---------

Co-authored-by: yuzhai <yuzhai@nvidia.com>
This commit is contained in:
Yujia Zhai
2025-02-06 21:33:40 -05:00
committed by GitHub
co-authored by yuzhai
parent affd1b693d
commit 833f6990e0
168 changed files with 24945 additions and 3436 deletions
@@ -37,6 +37,7 @@
#include "cutlass/cutlass.h"
#include "cutlass/detail/collective.hpp"
#include "cutlass/library/library.h"
#include "cutlass/library/util.h"
#include "cutlass/transform/kernel/sparse_gemm_compressor.hpp" // StructuredSparseCompressor
#include "cutlass/transform/device/transform_universal_adapter.hpp" // TransformUniversalAdapter
#include "cutlass/util/packed_stride.hpp" // make_cute_packed_stride
@@ -45,14 +46,6 @@
///////////////////////////////////////////////////////////////////////////////////////////////////
#define CUDA_CHECK(cuda_error) \
{ \
if (cuda_error != cudaSuccess) { \
printf("cudaError %s in %s:%d\n", cudaGetErrorString(cuda_error), __func__, __LINE__ ); \
return Status::kInvalid; \
} \
}
namespace cutlass::library {
///////////////////////////////////////////////////////////////////////////////////////////////////