Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
161 Open 3,194 Closed
161 Open 3,194 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

Expand ComputeAtMap exact and almost_exact disjointsets by IdModel
#4178 opened Apr 2, 2025 by naoyam • Draft
Expand rope benchmark to sweep across sequence length
#4177 opened Apr 2, 2025 by jjsjann123 Loading…
1
5
Remove the legacy analysis code as IdModel is built by default
#4174 opened Apr 2, 2025 by naoyam Loading…
2
Embedding shmoo
#4173 opened Apr 2, 2025 by jjsjann123 Loading…
3
allgather loop split, contig + noncontig
#4170 opened Apr 2, 2025 by Priya2698 • Draft
3
Support computeWith in TensorIndexer idmodel
#4163 opened Apr 1, 2025 by naoyam Loading…
1
6
[WIP] Support scalar outputs from fusions #3947
#4162 opened Apr 1, 2025 by csarofeen • Draft
1
Create Statement, Expr, and Val bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API
#4157 opened Mar 30, 2025 by rdspring1 Loading…
20
Create direct_bindings_api extension Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API
#4156 opened Mar 30, 2025 by rdspring1 Loading…
10
Insert block sync if it doesn't exist in WarAsyncWaitInserter
#4151 opened Mar 26, 2025 by jacobhinkle Loading…
1
9
[Host Ir] stream lowering, first milestone
#4148 opened Mar 26, 2025 by samnordmann • Draft
1
[Host irs] Stream lowering of single device fusions
#4147 opened Mar 26, 2025 by samnordmann Loading…
4
[Host ir] support for set reduce and binary op
#4146 opened Mar 26, 2025 by samnordmann Loading…
4
[Host Ir] refactor and cleanup lowering and segmentation
#4145 opened Mar 26, 2025 by samnordmann Loading…
5
[Host irs] alias and preallocated output support
#4144 opened Mar 26, 2025 by samnordmann Loading…
2
Add unit tests for single tile Blackwell MMA
#4137 opened Mar 25, 2025 by zasdfgbnm • Draft
1
Fix bugs with HostIr integration into FusionExecutorCache
#4136 opened Mar 25, 2025 by nsarka Loading…
@nsarka
23
Don't use IterType::gatherscatter with gather
#4131 opened Mar 21, 2025 by naoyam • Draft
2
Update README.md
#4118 opened Mar 20, 2025 by nWEIdia Loading…
2
Add missing device meshes in tests
#4117 opened Mar 20, 2025 by Priya2698 Loading…
6
Move opcheck to C++20 concepts for DynamicType lib
#4109 opened Mar 19, 2025 by zasdfgbnm • Draft
1
Move opcheck in DynamicType lib to C++20
#4107 opened Mar 19, 2025 by zasdfgbnm • Draft
3
Change c10::irange to iota, part 1
#4105 opened Mar 19, 2025 by zasdfgbnm • Draft
7
Cache data members in FusionKernelRuntime to avoid large step overheads for fusions with lots of args and segments
#4099 opened Mar 18, 2025 by kevinstephano • Draft
3
cache unmappable buffer in smem
#4090 opened Mar 17, 2025 by liqiangxl • Draft
ProTip! Add no:assignee to see everything that’s not assigned.