Skip to content

Pull requests: ROCm/rocMLIR

New pull request New
20 Open 1,632 Closed
20 Open 1,632 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

Heuristic for Split-K (full tuning only)
#1740 opened Feb 11, 2025 by dhernandez0 Loading…
@dhernandez0
1
Add bf16 to tuning runner
#1739 opened Feb 10, 2025 by djramic Loading…
Add fp8 convolution to the tuning runner
#1738 opened Feb 7, 2025 by djramic Loading…
4
Enable bf16 atomic add for gfx950
#1734 opened Feb 4, 2025 by dhernandez0 Loading…
@dhernandez0
[CI] Added gfx942 architecture to the 'Tune MLIR kernels' stage
#1733 opened Feb 3, 2025 by stefankoncarevic Loading…
@stefankoncarevic
1
[DO NOT SQUASH] Enable atomic add bf16 reduction and split-k for Navi4x
#1732 opened Feb 3, 2025 by dhernandez0 Loading…
@dhernandez0
1
Implement performance checking script
#1727 opened Jan 28, 2025 by dorde-antic Loading…
1
Move the TosaPartition pass to MHAL
#1726 opened Jan 27, 2025 by mirza-halilcevic Loading…
15
Improve performance of quantizelinear for int4
#1706 opened Dec 17, 2024 by dhernandez0 Loading…
@dhernandez0
13
Use convert-to-llvm pass for conversions to LLVMIR Dialect
#1679 opened Oct 9, 2024 by umangyadav Loading…
6
Use rocprofv2 instead of rocprof.
#1672 opened Oct 8, 2024 by pcf000 Loading…
21
Use target attached on gpu module when lowering
#1666 opened Sep 30, 2024 by umangyadav Loading…
@umangyadav
7
Move output swizzling pass before fusions
#1651 opened Sep 13, 2024 by krzysz00 Loading…
@dhernandez0
13
Mfma prune
#1577 opened Jul 18, 2024 by ethansaurusrex Loading…
4
Added quickTunerGen.py, the main driver of the quick tuner scripts
#1576 opened Jul 18, 2024 by ethansaurusrex Loading…
9
quickTunerPreproc.py, preprocessor script for quick tuner scripts
#1575 opened Jul 18, 2024 by ethansaurusrex Loading…
16
Unify accelerator and non-accelator paths
#1544 opened Jun 6, 2024 by djramic Loading…
18
Removed implicit split-k while moving from conv to gemm
#1527 opened May 21, 2024 by ravil-mobile • Draft
2
Double buffering improvements
#1511 opened May 13, 2024 by giuseros Loading…
6
Add back a few large GEMM tests
#1219 opened Aug 25, 2023 by krzysz00 Loading…
4
ProTip! Exclude everything labeled bug with -label:bug.