Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
126 Open 2,266 Closed
126 Open 2,266 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

Create TmaCircularBufferInfo class
#3033 opened Sep 26, 2024 by rdspring1 Loading…
Add new test HopperSS.MultipleTile
#3032 opened Sep 26, 2024 by zasdfgbnm Loading…
Add new test HopperSS.SingleTileTransposed
#3031 opened Sep 26, 2024 by zasdfgbnm Loading…
1
Add new test HopperRS.MultipleTile
#3030 opened Sep 26, 2024 by zasdfgbnm Loading…
Add new test HopperRS.SingleTileTransposed
#3029 opened Sep 26, 2024 by zasdfgbnm Loading…
1
Indexing test cleanup
#3028 opened Sep 26, 2024 by naoyam Loading…
3
[Tracking] Add segmentation to python frontend user-scheduling
#3025 opened Sep 26, 2024 by rdspring1 • Draft
Enforce 128B alignment for TMA shared memory allocation.
#3023 opened Sep 26, 2024 by rdspring1 Loading…
2
[WIP] Accept multi-matmul fusions for automatic scheduling
#3018 opened Sep 25, 2024 by jacobhinkle • Draft
1
Set default standard back to 20
#3014 opened Sep 25, 2024 by zasdfgbnm Loading…
4
Only build core library on GitHub CI
#3013 opened Sep 24, 2024 by zasdfgbnm Loading…
Clean up schedules and tests (Part 2/2)
#3008 opened Sep 24, 2024 by csarofeen Loading…
10 tasks done
7
More test
#3007 opened Sep 24, 2024 by zasdfgbnm • Draft
Support non-conventional loop domains in TensorView::updateMaxProducerPosition. rope
#3003 opened Sep 24, 2024 by naoyam Loading…
6
Compute strides for Hopper Mma
#3002 opened Sep 24, 2024 by zasdfgbnm Loading…
33
Presegmentation pass to force segment breaks when inplace update can cause RW race
#2999 opened Sep 23, 2024 by Priya2698 Loading…
6
Only the first local rank prints segmented fusions.
#2996 opened Sep 23, 2024 by wujingyue Loading…
9
Multiple matmul scheduler: schedule epilogue, inlining, circular buffering
#2994 opened Sep 23, 2024 by jacobhinkle • Draft
2
Host irs: Coalescing
#2991 opened Sep 23, 2024 by samnordmann Loading…
removing obsolete code
#2982 opened Sep 20, 2024 by jjsjann123 • Draft
1
1
Host IRs: IfThenElse support
#2976 opened Sep 20, 2024 by samnordmann Loading…
1
Host irs: add p2p comms
#2970 opened Sep 19, 2024 by samnordmann Loading…
only set maxrregcount when want to use less than 255 registers
#2968 opened Sep 19, 2024 by liqiangxl Loading…
2
warp reduction in x and y dims
#2966 opened Sep 19, 2024 by liqiangxl • Draft
1
only do grid split when needed
#2965 opened Sep 19, 2024 by liqiangxl Loading…
9
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.