leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 27 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
examples/51_hopper_gett/51_hopper_gett.cu
+9 -9
View File
@@ -186,15 +186,15 @@ main(int argc, char const* argv[]) {
using ElementEpilogue = float;
// The following constexpr values set the max number of modes in each MNKL mode
constexpr int MaxRank_M = rank(RowModeStridesA{}); // Max row modes
constexpr int MaxRank_N = rank(ColModeStridesB{}); // Max column modes
constexpr int MaxRank_K = rank(RedModeStridesA{}); // Max contraction modes
constexpr int MaxRank_L = rank(BatModeStridesA{}); // Max batch modes
static_assert(rank(RowModeStridesA{}) == rank(RowModeStridesC{}));
static_assert(rank(ColModeStridesB{}) == rank(RowModeStridesC{}));
static_assert(rank(RedModeStridesA{}) == rank(RedModeStridesB{}));
static_assert(rank(BatModeStridesA{}) == rank(BatModeStridesC{}));
static_assert(rank(BatModeStridesB{}) == rank(BatModeStridesC{}));
constexpr int MaxRank_M = cute::rank(RowModeStridesA{}); // Max row modes
constexpr int MaxRank_N = cute::rank(ColModeStridesB{}); // Max column modes
constexpr int MaxRank_K = cute::rank(RedModeStridesA{}); // Max contraction modes
constexpr int MaxRank_L = cute::rank(BatModeStridesA{}); // Max batch modes
static_assert(cute::rank(RowModeStridesA{}) == cute::rank(RowModeStridesC{}));
static_assert(cute::rank(ColModeStridesB{}) == cute::rank(RowModeStridesC{}));
static_assert(cute::rank(RedModeStridesA{}) == cute::rank(RedModeStridesB{}));
static_assert(cute::rank(BatModeStridesA{}) == cute::rank(BatModeStridesC{}));
static_assert(cute::rank(BatModeStridesB{}) == cute::rank(BatModeStridesC{}));
// Parse command line to get modes, extents, and strides
cutlass::GettCommandLine cmd;