v4.3 update. (#2709)

* v4.3 update.

* Update the cute_dsl_api changelog's doc link

* Update version to 4.3.0

* Update the example link

* Update doc to encourage user to install DSL from requirements.txt

---------

Co-authored-by: Larry Wu <larwu@nvidia.com>
This commit is contained in:
Junkai-Wu
2025-10-22 02:26:30 +08:00
committed by GitHub
parent e6e2cc29f5
commit b1d6e2c9b3
244 changed files with 59272 additions and 10455 deletions

View File

@@ -156,7 +156,8 @@
# define CUTE_ARCH_TMA_SM120_ENABLED
#endif
#if (defined(CUTLASS_ARCH_MMA_SM120_ENABLED) || defined(CUTLASS_ARCH_MMA_SM120A_ENABLED))
#if (defined(CUTLASS_ARCH_MMA_SM120_ENABLED) || defined(CUTLASS_ARCH_MMA_SM120A_ENABLED) ||\
defined(CUTLASS_ARCH_MMA_SM121_ENABLED) || defined(CUTLASS_ARCH_MMA_SM121A_ENABLED))
# if (__CUDACC_VER_MAJOR__ > 12 || (__CUDACC_VER_MAJOR__ == 12 && __CUDACC_VER_MINOR__ >= 8))
# define CUTE_ARCH_F8F6F4_MMA_ENABLED
# define CUTE_ARCH_MXF8F6F4_MMA_ENABLED
@@ -165,15 +166,6 @@
# endif
#endif
#if (defined(CUTLASS_ARCH_MMA_SM121_ENABLED) || defined(CUTLASS_ARCH_MMA_SM121A_ENABLED))
# if (__CUDACC_VER_MAJOR__ > 12 || (__CUDACC_VER_MAJOR__ == 12 && __CUDACC_VER_MINOR__ >= 9))
# define CUTE_ARCH_F8F6F4_MMA_ENABLED
# define CUTE_ARCH_MXF8F6F4_MMA_ENABLED
# define CUTE_ARCH_MXF4NVF4_2X_UE8M0_MMA_ENABLED
# define CUTE_ARCH_MXF4NVF4_4X_UE4M3_MMA_ENABLED
# endif
#endif
#if defined(CUTLASS_ARCH_MMA_SM100F_ENABLED) || defined(CUTLASS_ARCH_MMA_SM103F_ENABLED)
# define CUTE_ARCH_LDSM_SM100A_ENABLED
# define CUTE_ARCH_STSM_SM100A_ENABLED

View File

@@ -516,7 +516,7 @@ constexpr uint64_t parse_int_digits(uint64_t result, int digit, Ts... digits)
// var has type cute::constant<int,32>.
//
template <char... digits>
constexpr cute::constant<int,detail::parse_int_digits(0, (digits - '0')...)> operator "" _c()
constexpr cute::constant<int,detail::parse_int_digits(0, (digits - '0')...)> operator""_c()
{
static_assert((('0' <= digits && digits <= '9') && ...),
"Expected 0 <= digit <= 9 for each digit of the integer.");