Skip to content

Pull requests: triton-lang/triton

New pull request New
283 Open 8,145 Closed
283 Open 8,145 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

fix: include IR overrides in autotune cache keys
#10880 opened Jul 14, 2026 by ShiroKSH Loading…
[SANITIZER] Track compiler-owned shared scratch in ConSan
#10879 opened Jul 14, 2026 by pawelszczerbuk Contributor Loading…
[AMD] Add RDNA4m target
#10878 opened Jul 14, 2026 by ptrojahn Contributor • Draft
Catch LLVM diagnostic and raise Python exception
#10877 opened Jul 14, 2026 by woct0rdho Contributor Loading…
5 of 7 tasks
only adjust the remat cost for expensive op
#10876 opened Jul 14, 2026 by yanxuer-999 Contributor • Draft
7 tasks
[KERNELS] topk: don't specialize on bitmatrix strides (fixes per-batch-size recompiles)
#10875 opened Jul 14, 2026 by sbeurnier Loading…
1
1
[AMD] Add tests for compiler fence in AtomicRMWOp lowering on gfx1250
#10874 opened Jul 14, 2026 by tyb0807 Contributor Loading…
Avoid output TMA for Blackwell matmul accumulation
#10873 opened Jul 14, 2026 by yongjik Collaborator Loading…
[AMD] emulate descriptor reduce
#10871 opened Jul 14, 2026 by zwu-2025 Contributor • Draft
[CI] Skip non-zero padding test on GFX1250
#10870 opened Jul 14, 2026 by zwu-2025 Contributor Loading…
2
[NVIDIA] Test cluster barrier lowering with assertions
#10867 opened Jul 14, 2026 by jeffniu-openai Collaborator Loading…
[Gluon] Add TMEM accumulator overlap example
#10865 opened Jul 13, 2026 by adstraw Collaborator • Draft
4
Scope ConSan cluster visibility to warp partitions
#10864 opened Jul 13, 2026 by pawelszczerbuk Contributor • Draft
Fix CLC scheduler slot reuse in multi-CTA Gluon matmuls
#10863 opened Jul 13, 2026 by pawelszczerbuk Contributor • Draft
6
[Blackwell] [Gluon] Support Blackwell 256-bit global load/store LLVM lowering
#10861 opened Jul 13, 2026 by njriasan Collaborator Loading…
1
[Backend] LLVM bump to llvm/llvm-project@41c4167c
#10859 opened Jul 13, 2026 by ravil-mobile Contributor • Draft
[Blackwell] Fix incorrectly pipelined back-to-back MMAs sharing the same accumulator
#10858 opened Jul 13, 2026 by masahi Collaborator Loading…
1
6
[BACKEND] Fix multibuffer descriptor subslices
#10856 opened Jul 12, 2026 by jeffniu-openai Collaborator Loading…
1 task
1
[FPSAN] Model Float2 inline assembly
#10854 opened Jul 12, 2026 by jeffniu-openai Collaborator • Draft
2
[TRITONGPU] Cost descriptor layouts for partial reductions
#10853 opened Jul 12, 2026 by guoriyue Contributor • Draft
1
[BACKEND] Fix device_print(..., hex=True) for floating-point values
#10850 opened Jul 11, 2026 by guoriyue Contributor Loading…
5 of 7 tasks
2
[LLVM] Always set the target machine when arch is specified
#10849 opened Jul 11, 2026 by zhanglx13 Collaborator • Draft
[triton-ext] Add Triton headers/definitions to wheel
#10847 opened Jul 10, 2026 by abrown Contributor • Draft
1
[TRITONGPU] Propagate assigned CGA layouts to loads
#10837 opened Jul 9, 2026 by Jokeren Contributor • Draft
[BC breaking][FRONTEND] Remove block pointer support
#10833 opened Jul 9, 2026 by ThomasRaoux Collaborator Loading…
3
ProTip! Add no:assignee to see everything that’s not assigned.