Skip to content

Pull requests: iree-org/iree

New pull request New
214 Open 18,964 Closed
214 Open 18,964 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Codegen][CPU] Widen *_CAST* lhs/rhs before broadcasting.
#24518 opened May 20, 2026 by bjacob Collaborator • Draft
[Codegen][CPU] Flatten contiguous trailing dims of transfers before unrolling.
#24517 opened May 20, 2026 by bjacob Collaborator • Draft
[Codegen][CPU] Route inner_tiled broadcast into m_bcst-foldable slot.
#24516 opened May 20, 2026 by bjacob Collaborator • Draft
[HAL][AMDGPU] Support raw HSACO custom kernargs
#24512 opened May 20, 2026 by AWoloszyn Contributor Loading…
[HAL][AMDGPU] Export host allocations and device properties
#24511 opened May 20, 2026 by AWoloszyn Contributor Loading…
[VectorDistribute] Enable intrinsic usage for unaligned attention
#24510 opened May 20, 2026 by sommerlukas Contributor • Draft
[VectorDistribute] LDS promotion with async DMA
#24509 opened May 20, 2026 by sommerlukas Contributor Loading…
[Codegen] Use default read semantics for LinalgExt scatter
#24504 opened May 16, 2026 by moomoohorse321 Loading…
1
[Codegen][NFC] Move DropSchedulePass into Codegen/Common
#24501 opened May 16, 2026 by Alex-Wengg Contributor Loading…
4
[Codegen] Replace custom extract slice swap pattern with upstream ver…
#24487 opened May 14, 2026 by FlintWangacc Contributor Loading…
[do not review] ci test for gfx950 vdmfma 2
#24486 opened May 14, 2026 by efric Member • Draft
Guard against NULL frontier_tracker in standalone HIP/CUDA devices
#24474 opened May 13, 2026 by Abhishek-Varma Contributor Loading…
feat(hal): implement functions to normalize buffer references and buffer bindings to subspan buffers
#24472 opened May 13, 2026 by schuermans-roofline Contributor Loading…
[Codegen] Add phase group information into a new enum for shared mem info
#24471 opened May 13, 2026 by Muzammiluddin-Syed-ECE Contributor Loading…
21
[Codegen][AMDGPU] Add --iree-llvmgpu-use-global-transpose-load flag and ConfigUtils wiring
#24469 opened May 13, 2026 by nirvedhmeshram Contributor • Draft
[Codegen][AMDGPU] Add UseGlobalTransposeLoad promotion attr and operand promotion for RDNA4
#24468 opened May 13, 2026 by nirvedhmeshram Contributor • Draft
[WIP] [LinalgExt] Fix attention NaN for fully-masked rows
#24447 opened May 12, 2026 by keshavvinayak01 Contributor • Draft
1
[Codegen] Split StableHLO/JAX select-style argmax reductions
#24445 opened May 12, 2026 by YuWei-CH Contributor Loading…
29
[LLVMGPU] Handle single-workgroup scatter dispatches
#24442 opened May 12, 2026 by YuWei-CH Contributor Loading…
6
[ci test] [do not review] gfx950 vdmfma
#24441 opened May 11, 2026 by efric Member • Draft
[LinalgExt] Scalarize rank-0 score payloads in online attention
#24429 opened May 11, 2026 by keshavvinayak01 Contributor • Draft
[Codegen][GPU] Phase-aware LDS bank conflict analysis and XOR swizzle search
#24428 opened May 11, 2026 by Muzammiluddin-Syed-ECE Contributor Loading…
46
[Torch][LinalgExt] Lower flex_attention masks before online_attention
#24426 opened May 9, 2026 by keshavvinayak01 Contributor Loading…
19
[StableHLO] Guard convolution widen operand fusion
#24424 opened May 9, 2026 by YuWei-CH Contributor Loading…
Add chipDetails for RDNA3
#24419 opened May 8, 2026 by RattataKing Contributor Loading…
1
ProTip! Add no:assignee to see everything that’s not assigned.