Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
96 Open 1,751 Closed
96 Open 1,751 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
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

Allow reshaping of empty tensors
#2369 opened Jun 7, 2024 by jacobhinkle • Draft
1
5
Use standard {uniform,normal} if params are 0, 1
#2368 opened Jun 7, 2024 by jacobhinkle • Draft
1
Sharded pointwise scheduler
#2366 opened Jun 7, 2024 by cowanmeg Loading…
6
Move checkConcretization for reshapes
#2363 opened Jun 7, 2024 by jacobhinkle Loading…
1
8
[DO NOT MERGE] Add scheduling support for SDPAOp
#2361 opened Jun 6, 2024 by Priya2698 • Draft
Tensor parallel MLP
#2360 opened Jun 6, 2024 by cowanmeg • Draft
prioritize moving buffer used by outer bcast
#2357 opened Jun 6, 2024 by liqiangxl • Draft
4
Adjust scale factor for half-precision rng uniform
#2355 opened Jun 6, 2024 by jacobhinkle Loading…
1
8
Host irs: add wait IR
#2338 opened Jun 4, 2024 by samnordmann Loading…
iter grouped multiple reductions
#2332 opened Jun 3, 2024 by liqiangxl Loading…
9
Host irs: use Communicator directly instead of wrapped inside PostOnStream
#2328 opened Jun 3, 2024 by samnordmann Loading…
13
Transpose Scheduler: Ignore reduction axis in merge
#2326 opened Jun 1, 2024 by Priya2698 • Draft
1
16
[WIP - CI Testing Only] Fix Issue 2258
#2318 opened May 29, 2024 by rdspring1 • Draft
1
Host irs: add streams
#2312 opened May 29, 2024 by samnordmann Loading…
30
[WIP DO NOT REVIEW] "Fold" operations to generalize Reductions
#2307 opened May 28, 2024 by jacobhinkle • Draft
step-2 of revising outer reduction heuristics, use block reduction if SM usage is high
#2302 opened May 24, 2024 by liqiangxl • Draft
Add a new SdpaFwdOp IR node for Flash Attention
#2294 opened May 23, 2024 by Priya2698 Loading…
28
[Python Frontend] Documentation
#2292 opened May 23, 2024 by rdspring1 • Draft
[User Schedule] Create TMA schedule example Python API Issues related to the Python API
#2288 opened May 22, 2024 by rdspring1 Loading…
6
step-1 of revising outer reduction heuristics, revise unroll & vectorization
#2281 opened May 22, 2024 by liqiangxl Loading…
start post reduction broadcast backward check from output
#2274 opened May 20, 2024 by liqiangxl Loading…
26
Multi-dimensional DeviceMesh
#2260 opened May 16, 2024 by cowanmeg Loading…
42
Allow FusionExecutorCache to take preallocated outputs
#2247 opened May 15, 2024 by samnordmann Loading…
21
[WIP] IdModel-based indexing idmodel
#2238 opened May 13, 2024 by naoyam • Draft
[WIP] Check that matmul epilogue contains only pointwise ops matmul Segmentation Issues related to nvFuser Segmentation Top-Down Matmul Dev
#2225 opened May 9, 2024 by jacobhinkle • Draft
1
ProTip! Add no:assignee to see everything that’s not assigned.