Skip to content

Pull requests: ROCm/FlyDSL

New pull request New
25 Open 360 Closed
25 Open 360 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

[FlyOp] Add stages args for MmaMakeFragment to simplify multi-stages allocation
#437 opened Apr 25, 2026 by sjfeng1999 Collaborator Loading…
1 task
4
optimized fast path by reducing register use and vectorize on normal …
#436 opened Apr 24, 2026 by kudomcho Loading…
Refine hgemm split-k barrier
#435 opened Apr 24, 2026 by xytpai Contributor Loading…
Adds Grouped and Batched GEMM kernels with blockscaling matching DeepGEMM API
#433 opened Apr 23, 2026 by aryaman-gupta Loading…
4 tasks done
Add A16W4 MoE GEMM stage2 kernel (BF16 activations x MXFP4 weights)
#431 opened Apr 23, 2026 by apicciau • Draft
1 task
@apicciau
Add RDNA4 MoE WMMA kernel path
#430 opened Apr 23, 2026 by vivienfanghuagood Collaborator • Draft
Kefan.cao/fix universal copy
#429 opened Apr 23, 2026 by kefan203 Contributor Loading…
1 task
3
CI: add standalone ATOM integration workflow
#428 opened Apr 22, 2026 by gyohuangxin Member Loading…
3 of 4 tasks
9
Add TopK Gating Softmax Kernel
#426 opened Apr 22, 2026 by amd-wsung102 Loading…
1 task done
1
Add BF16xFP4 MoE GEMM stage1 kernel
#424 opened Apr 21, 2026 by apicciau • Draft
1 task
feat: align quant and fused kernels with Triton in FlyDSL
#421 opened Apr 21, 2026 by cschenjunlin Loading…
1 of 7 tasks
2
Pr/a16wi4 group splitk
#420 opened Apr 21, 2026 by yadaish Collaborator • Draft
1 task
feat: support mori IR JIT compilation with shmem
#418 opened Apr 21, 2026 by yanboshao Contributor Loading…
1 task
10
[AOT] Add dump_to_object, export_to_c, and load_module for AOT export
#414 opened Apr 20, 2026 by coderfeli Collaborator Loading…
Add gfx950 (MI355X) preload tuning table for preshuffle GEMM
#411 opened Apr 16, 2026 by andyluo7 Contributor Loading…
Unify MoE test perf timing to CUDA Event bench and fix minor issues
#408 opened Apr 16, 2026 by XingerZhu Collaborator Loading…
1 task
6
[Kernel][MI350] GDR prefill K5 vk implementation
#405 opened Apr 16, 2026 by huizzhan • Draft
1 task
gemm a16w16 flydsl implementation (WIP)
#401 opened Apr 14, 2026 by omuhamma • Draft
1 task
@omuhamma
Add initial Windows support
#395 opened Apr 13, 2026 by 0xDELUXA • Draft
1 task done
2
[MI450][Kernel] add Deepseek MHA bf16 kernel verified on MI450
#393 opened Apr 13, 2026 by jli-melchior Contributor Loading…
1 task
support glibc2.28 and python 3.10~3.14
#386 opened Apr 13, 2026 by coderfeli Collaborator Loading…
Add hgemm_splitk+allreduce prologue/epilogue fusion kernels
#354 opened Apr 7, 2026 by xytpai Contributor • Draft
[FLYDSL]: if dispatch dynamic tests refactor
#346 opened Apr 3, 2026 by xudoyuan Contributor Loading…
1 task
12
Fujun.han/multi vendor support
#345 opened Apr 3, 2026 by Peter9606 Contributor Loading…
[Feature] Add JAX integration for FlyDSL kernels
#257 opened Mar 21, 2026 by wenchenvincent Loading…
1 task done
11
ProTip! Follow long discussions with comments:>50.