Skip to content

Pull requests: ROCm/rocMLIR

New pull request New
54 Open 2,036 Closed
54 Open 2,036 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

Initialize threadResource only once per thread
#2199 opened Jan 6, 2026 by umangyadav Loading…
@umangyadav
Add support for non-packed strides from MIGraphX
#2198 opened Jan 6, 2026 by justinrosner Loading…
1 of 2 tasks
1
5
Fix LDS vectorization logic to allow for bank conflict resolution when kPack > kPerThread
#2196 opened Jan 6, 2026 by umangyadav Loading…
@umangyadav
4
Remove word boundaries to match error anywhere in the string
#2194 opened Jan 5, 2026 by dorde-antic Loading…
1 task done
1
Add rocMLIR support for gfx1250 scaled wmma instructions
#2193 opened Jan 2, 2026 by justinrosner Loading…
3 tasks
23
Add assumeDimIsConstant() for BottomUpTMBuilder
#2192 opened Jan 2, 2026 by justinrosner Loading…
2 tasks
3
Fix Flash Attention precision loss on RDNA3 (gfx11xx)
#2189 opened Dec 30, 2025 by ppetrovi-amd • Draft
[rocmlir-tuning-driver] Add rotating buffers and inline instruction cache flush
#2188 opened Dec 30, 2025 by pabloantoniom Loading…
1 task
Sched group
#2186 opened Dec 25, 2025 by umangyadav • Draft
1 task
[WIP] Refactor GemmFeatures
#2185 opened Dec 24, 2025 by pabloantoniom • Draft
1 task
7
LDS transpose load: attention, extended wave configs, regular + transpose load
#2184 opened Dec 23, 2025 by stefankoncarevic Loading…
1 task
@stefankoncarevic
19
(Async)DirectToLDS support in WMMA
#2182 opened Dec 22, 2025 by pabloantoniom Loading…
1 task
6
Upstream merge dec18
#2180 opened Dec 18, 2025 by umangyadav Loading…
2 of 11 tasks
@umangyadav
1
Parallelize tuning across GPUs
#2179 opened Dec 18, 2025 by mirza-halilcevic Loading…
1 task done
36
Fix missing dialect registration and other issues
#2177 opened Dec 18, 2025 by pabloantoniom Loading…
1 task
2
Add squeeze, gather, and scatter_none ops to MIGraphX dialect
#2176 opened Dec 17, 2025 by justinrosner Loading…
1 of 2 tasks
22
GEMM+GEMM tuning
#2175 opened Dec 17, 2025 by dorde-antic Loading…
1 task done
3
MIGraphX support for Flash Decoding + KVCache
#2174 opened Dec 16, 2025 by justinrosner Loading…
3 tasks done
4
Add quick-tune support for Attention
#2169 opened Dec 11, 2025 by mirza-halilcevic Loading…
1 task
21
Use num_chiplets
#2168 opened Dec 11, 2025 by dhernandez0 Loading…
1 task done
@dhernandez0
37
Use tosa.matmul_t_block_scaled for the scaled GEMMs
#2164 opened Dec 9, 2025 by umangyadav Loading…
@umangyadav
24
Change XCD remapping logic
#2161 opened Dec 9, 2025 by dhernandez0 Loading…
1 task done
@dhernandez0
19
[DO NOT MERGE] Test MITuna's branch i2162
#2142 opened Dec 3, 2025 by dorde-antic Loading…
Extend problem key for reduction fusions
#2133 opened Nov 26, 2025 by justinrosner Loading…
1 of 2 tasks
65
Python tests
#2128 opened Nov 25, 2025 by dorde-antic • Draft
1 task
3
ProTip! Mix and match filters to narrow down what you’re looking for.