v4.2 release. (#2587)

* Fix default cluster callback values to 1 to avoid profiler failure when these values are not set in command line.

* v4.2 release.
This commit is contained in:
Junkai-Wu
2025-08-22 18:11:24 -04:00
committed by GitHub
parent 11cad1f67b
commit a49a78ffef
351 changed files with 28182 additions and 2032 deletions
+2 -2
View File
@@ -391,9 +391,9 @@ cute::array<T,N> reverse(cute::array<T,N> const& t)
//
// Specialize tuple-related functionality for cute::array
//
#include "cutlass/cutlass.h"
#if defined(__CUDACC_RTC__)
#include <cuda/std/tuple>
#include CUDA_STD_HEADER(tuple)
#else
#include <tuple>
#endif
+2 -2
View File
@@ -555,9 +555,9 @@ void fill(array_subbyte<T,N>& a, T const& value)
//
// Specialize tuple-related functionality for cute::array_subbyte
//
#include "cutlass/cutlass.h"
#if defined(__CUDACC_RTC__)
#include <cuda/std/tuple>
#include CUDA_STD_HEADER(tuple)
#else
#include <tuple>
#endif
+2 -2
View File
@@ -77,9 +77,9 @@ find(type_list<T...> const&) noexcept {
//
// Specialize tuple-related functionality for cute::type_list
//
#include "cutlass/cutlass.h"
#if defined(__CUDACC_RTC__)
#include <cuda/std/tuple>
#include CUDA_STD_HEADER(tuple)
#else
#include <tuple>
#endif