NVIDIA/cutlass
1
0
Fork 0
mirror of https://github.com/NVIDIA/cutlass.git synced 2026-07-17 00:57:25 +00:00
Code Issues Packages Projects Releases Wiki Activity
911 Commits 25 Branches 61 Tags
e6233cbac5d7c7a865c19c91cd684ceece19513c
Commit Graph

5 Commits

Author SHA1 Message Date
Jiaxuan Bai
e37889f489 Fix grouped GEMM with CUTLASS DSL 4.7 (#3369) 
The Hopper grouped GEMM wrapper selected the current NVVM TMA bindings only when the CUTLASS version prefix was exactly 4.6. A 4.7 wheel therefore fell back to the removed CpAsyncBulkTensorLoadMode/loadMode interface.

Detect the TMALoadMode capability directly, preserving the legacy fallback for older bindings. For example, 4.7 now uses the TMALoadMode/mode path and descriptor address-space cast.

Verification: Python syntax compilation and diff checks.
 2026-07-08 10:16:39 +08:00
brandonsun
d80a4e53b5 fix validation codes (#3303) 2026-06-05 20:16:02 +08:00
Anakin(Yancheng) Zheng
0e9ac0734c Fix example with upcoming release (#3293) 
* Fix example new releases

* Remove return
 2026-06-03 13:40:27 +08:00
Longsheng Du
5f06f5fc1a fix elect_sync api (#3262) 2026-05-26 08:50:00 +08:00
Junkai-Wu
cb37157db5 v4.5 tag update (#3202) 
* Python DSL examples reorganization.

* v4.5 tag update.
 2026-05-05 20:55:27 -04:00