| [WS] Update aref ops and lower_aref pass (#7479) | 10 个月前 |
| Pawel/hoist unpipelineable operands (#7082) | 11 个月前 |
| [TritonNvidiaGPU] Fix memory leak in TritonNvidiaGPU/Transforms/InterleaveTMem.cpp (#7924) | 9 个月前 |
| [NVIDIA] Add is_async flag to MMAv5 ops (#7590) | 10 个月前 |
| [NFC] Use RankedTensorType's clone and cloneWithEncoding member functions (#7464) | 10 个月前 |
| [BACKEND] Add optimization for local_load -> tmem_store layout (#7015) | 9 个月前 |
| [NFC] Use RankedTensorType's clone and cloneWithEncoding member functions (#7464) | 10 个月前 |
| [LAYOUTS] Implement toLinearLayout for TensorMemoryEncodingAttr (#7748) | 10 个月前 |
| [LAYOUTS] Implement generalized swizzling for convert_layout (#7565) | 10 个月前 |
| [NFC] Remove uses of deprecated GEN_PASS_CLASSES for TritonNvidiaGPU/Transforms (#6898) | 1 年前 |
| [BACKEND] Use immutable alloc for tma store (#7873) | 9 个月前 |
| Add support for padding option to TMA loads (#7993) | 9 个月前 |
| [Blackwell] Handle control flow in TMEM allocation (#7698) | 10 个月前 |
| [WS] Update aref ops and lower_aref pass (#7479) | 10 个月前 |