Skip to content

Pull requests: iree-org/iree

New pull request New
113 Open 16,989 Closed
113 Open 16,989 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

[LLVMGPU] Vectorize map_scatter in LLVMGPUTileAndFuse pipeline
#21890 opened Sep 8, 2025 by Max191 Loading…
[CI] Fix mi308 spec for sharktank_models tests
#21888 opened Sep 8, 2025 by Groverkss Loading…
[GPU] Remove MMAScheduleAttr
#21884 opened Sep 7, 2025 by Groverkss Loading…
3
[Codegen] Upgrade Transforms and Utils to free create functions. NFC.
#21882 opened Sep 7, 2025 by kuhar Loading…
Upgrade GlobalOpt, InputConversion, ExternalInterfacess to free create function. NFC.
#21878 opened Sep 7, 2025 by kuhar Loading…
Integrate LLVM at cf44f19e57210844334e05cf7ed79fb5695cc68f
#21875 opened Sep 6, 2025 by Muzammiluddin-Syed-ECE Loading…
[Codegen][ROCm] Add repro instructions for .rocmasm files
#21874 opened Sep 6, 2025 by kuhar Loading…
6
Display a warning when we spill SGPRs or VGPRs
#21863 opened Sep 5, 2025 by sebvince Loading…
17
Add UB dialect python bindings
#21862 opened Sep 5, 2025 by Hardcode84 • Draft
[iree][codegen] Add the #iree_gpu.optimize_occupancy attribute
#21861 opened Sep 5, 2025 by fabianmcg • Draft
Revert "Disable failing ARM-SME tests. (#21715)"
#21860 opened Sep 5, 2025 by banach-space Loading…
Add gfx950 ukernel patterns
#21856 opened Sep 5, 2025 by sebvince Loading…
17
[LLVMGPU][Codegen] Emit packed chain FMA from select multi_reductions and contracts
#21855 opened Sep 5, 2025 by efric Loading…
1
3
[Dispatch Creation] Rework dispatch formation logic
#21854 opened Sep 5, 2025 by IanWood1 • Draft
[wip][Codegen][LLVMGPU] Adjust thread load size in reduction dim
#21853 opened Sep 4, 2025 by newling • Draft
2
Bump version to 3.8.0 after 3.7.0 release.
#21852 opened Sep 4, 2025 by sa-faizal Loading…
1
[DispatchCreation] Sink bit-extend producers into split-reduction forall loop
#21850 opened Sep 4, 2025 by rkayaith • Draft
5
[Codegen] Added Vector rewrite for erf ; FastMath
#21849 opened Sep 4, 2025 by keshavvinayak01 Loading…
5
[codegen] more consumer fusion
#21848 opened Sep 4, 2025 by jtuyls Loading…
2
3
Implement iree_gpu.coalesced_gather_dma op
#21846 opened Sep 4, 2025 by lialan Loading…
1
20
[CPU] Do not fuse producers if it can cause invalid dominance property.
#21845 opened Sep 3, 2025 by hanhanW Loading…
7
[iree][codegen] Add #iree_codegen.denormal_fp_math to set denormals behavior
#21840 opened Sep 3, 2025 by fabianmcg Loading…
19
[Codegen][GPU] Enable TileAndFuse for matmul by default
#21834 opened Sep 3, 2025 by jerryyin • Draft
1
Guard more optimization flags under --iree-opt-level.
#21827 opened Sep 2, 2025 by hanhanW • Draft
4
[GPU] Do not do c promotion for unaligned (I)GEMMs
#21823 opened Sep 2, 2025 by nirvedhmeshram • Draft
3
ProTip! no:milestone will show everything without a milestone.