Skip to content

Pull requests: facebookexperimental/triton

New pull request New
89 Open 1,080 Closed
89 Open 1,080 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[AutoWS][PartitionSchedule] Route post-loop TMEM ops to default parti… CLA Signed This label is managed by the Meta Open Source bot.
#1179 opened Mar 28, 2026 by manman-ren Loading…
[autoWS] Add skills to debug accuracy or compare ttgir for performance CLA Signed This label is managed by the Meta Open Source bot.
#1178 opened Mar 28, 2026 by manman-ren • Draft
[AutoWS] [WIP] Support pipelining TMA wait store CLA Signed This label is managed by the Meta Open Source bot.
#1176 opened Mar 27, 2026 by njriasan Loading…
1
[AutoWS] Support rotating TMEM Waits using SWP with a provided "safety" attribute CLA Signed This label is managed by the Meta Open Source bot.
#1172 opened Mar 26, 2026 by njriasan Loading…
2
[TLX][AMD] Fix async_load, local_load, and async_token for AMD pipelines
#1170 opened Mar 26, 2026 by Hardcode84 • Draft
1
Split-K matmul in stock Triton (A100) (#1169) CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1169 opened Mar 26, 2026 by stashuk-olek Loading…
1
[WIP][TLX][lint] Add Z3-based barrier deadlock detection pass CLA Signed This label is managed by the Meta Open Source bot.
#1158 opened Mar 26, 2026 by sfzhu93 • Draft
[WIP][TLX][lint] Add barrier execution order analysis pass CLA Signed This label is managed by the Meta Open Source bot.
#1157 opened Mar 26, 2026 by sfzhu93 • Draft
[WIP] [AutoWS] Add an initial implementation for subtiledOperator CLA Signed This label is managed by the Meta Open Source bot.
#1155 opened Mar 25, 2026 by njriasan Loading…
1
[WIP][TLX] Enable 2-CTA for FA bwd CLA Signed This label is managed by the Meta Open Source bot.
#1153 opened Mar 25, 2026 by htyu • Draft
1
[TLX] Support column-major A and B inputs in blackwell_gemm_ws and hopper_gemm_ws CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1145 opened Mar 25, 2026 by htyu Loading…
1
[WIP] barrier analysis for ttgir CLA Signed This label is managed by the Meta Open Source bot.
#1144 opened Mar 25, 2026 by Sibylau • Draft
1
[tlx] Add TMA-pipelined skinny GEMM kernel with split-K for hopper_gemm_ws (#1140) CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1140 opened Mar 24, 2026 by stashuk-olek Loading…
1
[AutoWS] Enable TMEM load subslicing for BLOCK_M=256 CLA Signed This label is managed by the Meta Open Source bot.
#1135 opened Mar 24, 2026 by jananisriram • Draft
1
[TLX] Disable USE_WARP_BARRIER in FA bwd and add bwd correctness test CLA Signed This label is managed by the Meta Open Source bot.
#1133 opened Mar 24, 2026 by htyu Loading…
1
[WIP] [AutoWS] AutoWS Release Milestone CLA Signed This label is managed by the Meta Open Source bot.
#1130 opened Mar 23, 2026 by njriasan • Draft
[AutoWS] Add per-partition warp count overrides CLA Signed This label is managed by the Meta Open Source bot.
#1128 opened Mar 23, 2026 by ethche Loading…
[Triton] Support 2-CTA in Triton with cross-CTA sync CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1120 opened Mar 20, 2026 by tissue3 Loading…
8
[WIP] Introduce a persistent B GEMM kernel CLA Signed This label is managed by the Meta Open Source bot.
#1114 opened Mar 20, 2026 by njriasan Loading…
µTLX (aka utlx, aka micro-tlx) CLA Signed This label is managed by the Meta Open Source bot.
#1112 opened Mar 19, 2026 by plotfi Loading…
[autoWS] Add prologue peeling for load partition innermost loop CLA Signed This label is managed by the Meta Open Source bot.
#1109 opened Mar 19, 2026 by Sibylau Loading…
1
[TLX] [BE] Refactor test_tlx.py to allow better scaling CLA Signed This label is managed by the Meta Open Source bot.
#1107 opened Mar 18, 2026 by njriasan Loading…
7
[TLX] Insert cluster barrier before ret for all clustered kernels (#1105) CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1105 opened Mar 18, 2026 by htyu Loading…
1
[triton][beta] Fix smem_budget query for offline compilation CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1104 opened Mar 18, 2026 by Sibylau Loading…
3
[Triton] [Numerics] Add a prototype for specifying bitwise consistent reductions CLA Signed This label is managed by the Meta Open Source bot.
#1100 opened Mar 17, 2026 by njriasan • Draft
ProTip! Mix and match filters to narrow down what you’re looking for.