Skip to content

Pull requests: ROCm/FlyDSL

New pull request New
50 Open 939 Closed
50 Open 939 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

[Ext][Coop] Add more warp-level primitives
#1167 opened Sep 18, 2026 by sjfeng1999 Collaborator Loading…
1 task
4
[dont merge][llvm] cherry commits for unclausevmem patch
#1166 opened Sep 18, 2026 by jli-melchior Collaborator Loading…
1 task
[Misc] Fix 5 Dependabot vulnerabilities
#1164 opened Sep 18, 2026 by Phil-amd Member Loading…
[DSL][Runtime][Kernel] Add in-kernel wave tracing (fx.ktrace) for gfx942/gfx950
#1162 opened Sep 18, 2026 by Phil-amd Member Loading…
@Phil-amd
feat(review): add local coderfeli review watcher
#1158 opened Sep 17, 2026 by jhinpan Collaborator Loading…
feat(review): harden review engine for unattended runs
#1157 opened Sep 17, 2026 by jhinpan Collaborator Loading…
gfx950 flex attention with user score/mask mods
#1155 opened Sep 17, 2026 by RichardChamberlain1 Loading…
3 of 4 tasks
@RichardChamberlain1
[wip][llvm][skill] Add an llvm skill for picking and verifying LLVM knobs
#1154 opened Sep 17, 2026 by Phil-amd Member Loading…
@Phil-amd
[llvm] Drop four LLVM tuning knobs that never reached codegen
#1153 opened Sep 17, 2026 by Phil-amd Member Loading…
@Phil-amd
7
[LLVM] bump llvm which improve the sequence for initial unclaused vmem
#1145 opened Sep 16, 2026 by jli-melchior Collaborator Loading…
1 task
[Build] Find ROCm through rocm-sdk before falling back to /opt/rocm*
#1142 opened Sep 15, 2026 by xinyazhang Contributor Loading…
1 task done
2
Opus fa4 align bf16 perf
#1139 opened Sep 15, 2026 by coderfeli Collaborator Loading…
1 task
4
[Kernel][Perf] Fix and tune gfx950 dense FP8 attention
#1137 opened Sep 15, 2026 by coderfeli Collaborator Loading…
1 task done
1
[Kernel] Refactor preshuffle GEMM indexing with layout algebra
#1131 opened Sep 14, 2026 by coderfeli Collaborator Loading…
Support variadic operands in fully expanded tiled GEMM
#1128 opened Sep 12, 2026 by coderfeli Collaborator Loading…
[Kernel][Perf] Optimize gfx950 head64 prefill attention
#1126 opened Sep 11, 2026 by michael604work Loading…
1
[Perf] gemm_bf16 gfx1250: cross-tile carry to hide K-tile prologue ds…
#1125 opened Sep 11, 2026 by amd-hhashemi Loading…
1 task
1
Reject global->LDS direct loads on gfx11
#1122 opened Sep 11, 2026 by mgehre-amd Contributor Loading…
refactor communication ops multi-gpu
#1116 opened Sep 10, 2026 by yanboshao Contributor Loading…
1 task
[Bugfix][Kernel] Fix fused RoPE for head dimension 96
#1113 opened Sep 9, 2026 by tangzzycc Loading…
1 task done
1
[DO NOT MERGE] Add MegaMoE engineering skill
#1111 opened Sep 9, 2026 by GwilliamHu Member • Draft
[LLVM] Fix true16 lowering for packed FP8 conversions
#1110 opened Sep 9, 2026 by big-yellow-duck Contributor Loading…
[DSL] Expose F32_FP8 conversion from packed_i32
#1109 opened Sep 9, 2026 by big-yellow-duck Contributor Loading…
3
[Bugfix][Benchmark] Fix compiler API usage in Softmax and RMSNorm benchmarks
#1108 opened Sep 9, 2026 by tangzzycc Loading…
1 task
1
[Skills] Add flydsl-code-review skill and workflow
#1106 opened Sep 8, 2026 by zhiding512 Collaborator Loading…
33
ProTip! Type g i on any issue or pull request to go back to the issue listing page.