Skip to content

Pull requests: triton-lang/triton

New pull request New
43 Open 3,219 Closed
43 Open 3,219 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

[DO NOT MERGE] Add back "Hint compiler to preload kernel args"
#4830 opened Sep 30, 2024 by zhanglx13 • Draft
[WIP] Sink the 2nd tt.load after local_load's
#4823 opened Sep 28, 2024 by zhanglx13 • Draft
[AMD] instruction counting during TritonAMDGPUToLLVM pass
#4819 opened Sep 27, 2024 by ravil-mobile • Draft
36
[AMD] Implement dot layout to linear layout conversion
#4817 opened Sep 27, 2024 by oplavsic • Draft
23
[SWP] separate logic in assignMemoryLayouts that should belong to scheduling
#4814 opened Sep 27, 2024 by manman-ren Loading…
3
Add platform info into cache key for compatibility
#4809 opened Sep 26, 2024 by xw285cornell Loading…
5 of 7 tasks
5
[AMD] Define a view_slice operation
#4804 opened Sep 25, 2024 by hmalgewatta • Draft
5 of 7 tasks
23
[AMD] Fix shared layout order for batch dimension in pipeline passes
#4796 opened Sep 24, 2024 by binarman • Draft
7
Implement scaled_dot(mxfp8, fp8) via mma
#4795 opened Sep 24, 2024 by lezcano Loading…
21
[BACKEND] Support convert_layout with num_ctas > 1 Using Linear Layout
#4782 opened Sep 23, 2024 by Jokeren • Draft
[FRONTEND] Delete dead code of visit_FunctionDef's handling of default argument
#4778 opened Sep 21, 2024 by lijinpei Loading…
5 of 7 tasks
fix the bug that for block_k=16 mma, the compilation crash on Ampere.
#4768 opened Sep 20, 2024 by bingyizh233 Loading…
5 of 6 tasks
34
Use clang-12 to fix issue with LLVM_ABI_BREAKING_CHECKS
#4760 opened Sep 19, 2024 by karupayun Loading…
11
Add almalinux arm64 llvm build
#4738 opened Sep 17, 2024 by atalman • Draft
1
Refactor compiler specializations to consider backend
#4734 opened Sep 16, 2024 by giuseros Loading…
60
Refactor the C code template in third_party/nvidia/backend/driver.py
#4722 opened Sep 13, 2024 by sfzhu93 Loading…
5 tasks done
4
[AMD] Add buffer support
#4716 opened Sep 12, 2024 by giuseros Loading…
15
[Frontend] [BC breaking] Always follow C semantics on %
#4698 opened Sep 10, 2024 by lezcano Loading…
2
[SWP] add unit test for two performance warnings
#4689 opened Sep 10, 2024 by sfzhu93 • Draft
5 tasks done
14
Add the option to configure FTZ
#4626 opened Sep 1, 2024 by cchan Loading…
5
[SWP] Attempt to move all scheduling logic to a scheduling pass
#4618 opened Aug 30, 2024 by manman-ren • Draft
9
[WIP] [AMD] Emit AMD specific intrinsics for dot
#4594 opened Aug 28, 2024 by binarman • Draft
6 tasks
1
[SWP] add LoopScheduling pass and move scheduling heuristics from SWP to the new pass
#4592 opened Aug 27, 2024 by manman-ren • Draft
20
[frontend] added overflow checks in debug mode
#4589 opened Aug 27, 2024 by ptillet Loading…
2
[WIP] Optimize fma dot
#4581 opened Aug 26, 2024 by binarman • Draft
6 tasks
1
ProTip! Mix and match filters to narrow down what you’re looking for.