CUTLASS 3.2.1 (#1113)

* Updates for 3.2.1 release.

* Minor fix in gemm op profiler for raster order.

* Add scheduler mapping for raster order in the kernels.
This commit is contained in:
ANIKET SHIVAM
2023-09-26 17:24:26 -04:00
committed by GitHub
parent e0aaa3c3b3
commit 90d3b0fb18
428 changed files with 22252 additions and 21761 deletions
+57 -13
View File
@@ -33,7 +33,6 @@
#include <cute/config.hpp>
#include <cute/util/type_traits.hpp>
#include <cute/numeric/integral_constant.hpp>
//
// CUDA compatible print and printf
@@ -119,16 +118,68 @@ get_format(double) {
CUTE_HOST_DEVICE
void
print(char const& c) {
print(char c) {
printf("%c", c);
}
template <class T,
__CUTE_REQUIRES(is_std_integral<T>::value)>
CUTE_HOST_DEVICE
void
print(T const& a) {
printf("%d", int(a));
print(signed char a) {
printf("%hhd", a);
}
CUTE_HOST_DEVICE
void
print(unsigned char a) {
printf("%hhu", a);
}
CUTE_HOST_DEVICE
void
print(short a) {
printf("%hd", a);
}
CUTE_HOST_DEVICE
void
print(unsigned short a) {
printf("%hu", a);
}
CUTE_HOST_DEVICE
void
print(int a) {
printf("%d", a);
}
CUTE_HOST_DEVICE
void
print(unsigned int a) {
printf("%u", a);
}
CUTE_HOST_DEVICE
void
print(long a) {
printf("%ld", a);
}
CUTE_HOST_DEVICE
void
print(unsigned long a) {
printf("%lu", a);
}
CUTE_HOST_DEVICE
void
print(long long a) {
printf("%lld", a);
}
CUTE_HOST_DEVICE
void
print(unsigned long long a) {
printf("%llu", a);
}
template <class... T>
@@ -138,13 +189,6 @@ print(char const* format, T const&... t) {
printf(format, t...);
}
template <class... T>
CUTE_HOST_DEVICE
void
print(T const&... t) {
(print(t), ...);
}
CUTE_HOST_DEVICE
void
print(char const* format) {