Dongdong Li 0469d5fccd [OPTIMIZER] Remove extra wgmma_wait_group in flash attention (#2399) ...

Co-authored-by: dongdongl <dongdongl@nvidia.com>

2023-10-26 16:35:36 +00:00

..

Analysis

[BACKEND] Remove ttg.cmp and ttg.select and replace by arith op (#2526)

2023-10-23 19:35:46 -07:00

Conversion

[BACKEND] Dedup elementwise in LLVM IR based on constancy (#2512)

2023-10-25 11:25:29 -04:00

lib

upgrade llvm to b1115f8c (NFC) (#2403)

2023-10-16 16:38:49 -07:00

LLVMIR

[BACKEND] Add LLVM pre-processing pass to break struct types (#2285)

2023-09-13 10:03:29 -07:00

NVGPU

upgrade llvm to b1115f8c (NFC) (#2403)

2023-10-16 16:38:49 -07:00

Triton

[BACKEND] Remove ttg.cmp and ttg.select and replace by arith op (#2526)

2023-10-23 19:35:46 -07:00

TritonGPU

[OPTIMIZER] Remove extra wgmma_wait_group in flash attention (#2399)

2023-10-26 16:35:36 +00:00

TritonNvidiaGPU

[BACKEND] Remove ttg.cmp and ttg.select and replace by arith op (#2526)

2023-10-23 19:35:46 -07:00

CMakeLists.txt

[TESTS] Now using downloaded FileCheck directly without copying it (#1293)

2023-03-07 09:11:08 -08:00

lit.cfg.py

[BACKEND] Add LLVM pre-processing pass to break struct types (#2285)

2023-09-13 10:03:29 -07:00

lit.site.cfg.py.in

Merge triton-mlir branch - Complete rewrite of the backend from scratch (#1004)

2022-12-21 01:30:50 -08:00