Skip to content

Pull requests: triton-lang/triton

New pull request New
47 Open 3,429 Closed
47 Open 3,429 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Frontend] Print out config when autotune fails
#5117 opened Nov 11, 2024 by htyu Loading…
[GitHub] Add issue templates and slightly refine PR template
#5115 opened Nov 11, 2024 by peterbell10 • Draft
2
Fix barrier insertion after assert op
#5114 opened Nov 11, 2024 by anmyachev Loading…
1
[AMD] Enable B scale for scaled_dot
#5112 opened Nov 11, 2024 by antiagainst • Draft
Fix llvm-build for almalinux
#5111 opened Nov 11, 2024 by pbchekin Loading…
5 of 7 tasks
1
[DRAFT][PROTON] Add proton.state utility
#5110 opened Nov 10, 2024 by Jokeren • Draft
[BACKEND][NVIDIA] Change getTotalElemsPerThreadForOperand to use getRepForOperand
#5105 opened Nov 8, 2024 by ggengnv Loading…
1
5
[BACKEND]Refactor convertLayoutOp by expanding dot operands of 2d tensors to 3d
#5104 opened Nov 8, 2024 by yiqian1 • Draft
4
Remove unknown distribution option: test_suite
#5098 opened Nov 7, 2024 by anmyachev Loading…
[CI] remove unused inductor workflows
#5073 opened Nov 5, 2024 by leseb Loading…
4 of 7 tasks
4
[AMD] Add atomicRMW dpp logic
#5072 opened Nov 5, 2024 by joviliast • Draft
13
[Triton][Allocation] Enable getScratchValueSize specialization
#5070 opened Nov 5, 2024 by victor-eds Loading…
5 of 7 tasks
24
[DRAFT][AMD] Introduce OptimizeAtomicLayouts pass
#5028 opened Oct 31, 2024 by joviliast • Draft
5
[AMD] Support Cross-Lane Reduction With DPP
#5019 opened Oct 30, 2024 by knwng • Draft
4 of 7 tasks
35
[AMD] Add tritonamdgpu-block-pingpong pass
#5018 opened Oct 30, 2024 by jungpark-mlir • Draft
9
[AMD][BACKEND] Switch to code object v5
#5005 opened Oct 29, 2024 by AlexAUT • Draft
5 of 7 tasks
4
[BACKEND][NVIDIA] Add DotOp Hoisting Pass for WGMMA and Add Lowering for SMEM-to-MMAv3 DotOp Copy
#5003 opened Oct 28, 2024 by ggengnv • Draft
53
[AMD] getBackwardSlice variant with handling for op regions
#4993 opened Oct 25, 2024 by karthik-man • Draft
5 of 7 tasks
7
[AMD][prototype] Transpose between global load and local store for non-TN layouts
#4984 opened Oct 23, 2024 by jtang10 • Draft
1 of 7 tasks
Added an option to dump IR to files
#4932 opened Oct 16, 2024 by arakhmati-openai • Draft
Fix order
#4914 opened Oct 15, 2024 by rawnhenry Loading…
3 of 7 tasks
9
[SWP] Fix a bug in SWP that did not correctly compute the number of loop iterations
#4887 opened Oct 10, 2024 by sfzhu93 Loading…
5 tasks done
13
[AMD][Pipeliner] Improve clustering and add prefetch
#4881 opened Oct 9, 2024 by sjw36 Loading…
27
[tool][proton] Intra kernel profiling support
#4861 opened Oct 7, 2024 by fywkevin • Draft
5 of 7 tasks
2
Fixing tensor broadcast
#4860 opened Oct 7, 2024 by maxim-m66 Loading…
5 of 7 tasks
7
ProTip! Add no:assignee to see everything that’s not assigned.