leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 29 Packages Projects Releases Wiki Activity

Collection of changes to fix clang build. (#1200)

Browse Source 
* Remove unused variables

* Qualify calls to make_fragment_? from templated base class.

Fixes clang build error.

* Add missing `#include <cstdio>`

* Various changes to fix clang compile errors.

* More changes to fix clang build.

Remaining issues:

- `params` initializer of `CollectiveEpilogue`.
- `ops` initializer of `Sm90VisitorImplBase`.
- `__usAtomicCAS` needs to be added to clang upstream.

* Fix remaining clang build issues.

* Qualify `cute::rank()` calls.

* Qualify some more calls that are otherwise ambiguous between `cute` and `std` namespace.

* Double-escape special registers in inline asm.

* small change

---------

Co-authored-by: Haicheng Wu <haichengw@nvidia.com>
This commit is contained in:
Christian Sigg
2023-12-08 14:42:12 -05:00
committed by GitHub
co-authored by Haicheng Wu
parent f4a0216601
commit e1483d5fa0
46 changed files with 308 additions and 273 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/cute/config.hpp
+4 -3
View File
@@ -30,7 +30,7 @@
**************************************************************************************************/
#pragma once
#if defined(__CUDA_ARCH__) || defined(_NVHPC_CUDA)
#if defined(__CUDACC__) || defined(_NVHPC_CUDA)
# define CUTE_HOST_DEVICE __forceinline__ __host__ __device__
# define CUTE_DEVICE __forceinline__ __device__
# define CUTE_HOST __forceinline__ __host__
@@ -46,10 +46,11 @@
# define CUTE_HOST_RTC CUTE_HOST
#endif
#if !defined(__CUDACC_RTC__) && (defined(__CUDA_ARCH__) || defined(_NVHPC_CUDA))
#if !defined(__CUDACC_RTC__) && !defined(__clang__) && \
(defined(__CUDA_ARCH__) || defined(_NVHPC_CUDA))
# define CUTE_UNROLL #pragma unroll
# define CUTE_NO_UNROLL #pragma unroll 1
#elif defined(__CUDACC_RTC__)
#elif defined(__CUDACC_RTC__) || defined(__clang__)
# define CUTE_UNROLL _Pragma("unroll")
# define CUTE_NO_UNROLL _Pragma("unroll 1")
#else