Skip to content

Pull requests: ROCm/composable_kernel

New pull request New
64 Open 3,132 Closed
64 Open 3,132 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

Shuffle fix for gfx950
#3491 opened Dec 25, 2025 by ThomasNing Loading…
1 of 7 tasks
Enable gptoss sink
#3490 opened Dec 25, 2025 by LJ-underdog Loading…
7 tasks done
9
[CK_TILE] MX FLATMM Fix M Padding
#3489 opened Dec 25, 2025 by DDEle Loading…
7 tasks done
enable f8 tests
#3488 opened Dec 25, 2025 by joyeamd Loading…
7 tasks
Update/add to qr_ks_vs_whole_k_prefetch pipeline
#3485 opened Dec 24, 2025 by qianfengz Loading…
Ensure Consistent 64-Bit Unsigned Handling Across OS Platforms
#3483 opened Dec 23, 2025 by afagaj Loading…
7 tasks done
[CK profiler] Perform verification on GPU when using GPU reference
#3482 opened Dec 23, 2025 by johannes-graner Loading…
3 of 7 tasks
add double buffering support for gemm abquant
#3481 opened Dec 23, 2025 by kensclin Loading…
1 of 7 tasks
DO NOT MERGE Testing
#3472 opened Dec 19, 2025 by amd-anclark Loading…
[CK_BUILDER] validation organization: streamhpc
#3471 opened Dec 19, 2025 by Snektron Loading…
Replace grouped conv bwd wei wmmaV3 bilin/scale bf16f32bf16 support with bf16bf16bf16 organization: streamhpc
#3470 opened Dec 19, 2025 by krithalith Loading…
7 tasks
[CK_TILE] Refactor UniversalGemm::MakeGemmTensorViews to separate descriptor and view creation
#3467 opened Dec 19, 2025 by amd-meskelin • Draft
7 tasks
Add support for direct store in epilogue and padding support for wave transfer without transpose organization: streamhpc
#3465 opened Dec 19, 2025 by EnricoDeg Loading…
6 of 7 tasks
@EnricoDeg
6
[CI, CK examples] Disable time_kernel for CI tests and examples
#3464 opened Dec 19, 2025 by johannes-graner Loading…
7 tasks
[CK TILE ENGINE] Enable gfx950 GPU architecture for supported combination
#3463 opened Dec 19, 2025 by ThruptiRajLakshmanaGowda • Draft
1 of 7 tasks
2
[CK_Tile] Support for group size 128 for Preshuffle quant for 2d block scale gemm
#3462 opened Dec 19, 2025 by amd-khushbu Loading…
7 tasks
Grouped convolution backward data WMMA v3 implementation organization: streamhpc
#3460 opened Dec 19, 2025 by ApoorvaKalyani Loading…
4 of 7 tasks
27
WIP: Build CK on Windows
#3458 opened Dec 18, 2025 by vpietila-amd • Draft
7 tasks
[CK_TILE] Allow UniversalGemmKernel::RunGemm to be called using tensor descriptors
#3457 opened Dec 18, 2025 by amd-meskelin • Draft
7 tasks
[CKTILE] Support A/B Quantization in Blockscale Grouped Gemm
#3452 opened Dec 18, 2025 by kyle-256 Loading…
1 of 7 tasks
[CK_Tile] Support for various group sizes Preshuffle quant for 2d block scale gemm
#3445 opened Dec 17, 2025 by amd-khushbu Loading…
1 of 7 tasks
4
[FMHA] Batch Prefill Support Improvements: Change KV Cache Layout & Large Page Size Support
#3442 opened Dec 16, 2025 by Jeff-Huang Loading…
7 tasks
9
Merge group related improvements for convolution operations organization: streamhpc
#3439 opened Dec 16, 2025 by zsotakal Loading…
7 tasks
@zsotakal
5
Adding remaining flavors for grouped conv fwd
#3436 opened Dec 16, 2025 by wj-laskowski • Draft
6 of 7 tasks
[CK_TILE]MoE a8w4
#3435 opened Dec 16, 2025 by Zzz9990 Loading…
7 tasks
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.