Skip to content

Pull requests: ROCm/composable_kernel

New pull request New
67 Open 1,681 Closed
67 Open 1,681 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

Enable ClangBuildAnalizer when doing ninja build traces.
#2009 opened Mar 22, 2025 by illsilin Loading…
4
[CK_TILE] Multiple-D GEMM example
#2008 opened Mar 21, 2025 by mozga-amd • Draft
7 tasks
add a fast compilation path for static for (0..N) compilation time
#2005 opened Mar 21, 2025 by tenpercent Loading…
1 of 7 tasks
use clang builtin for compile-time sequence indexing compilation time
#2003 opened Mar 20, 2025 by tenpercent Loading…
2 of 7 tasks
[CK_TILE] Remove scratch usage from universal gemm
#2001 opened Mar 20, 2025 by jakpiase Loading…
3 of 7 tasks
WIP: Introduce MX GEMM for FP8 data type
#2000 opened Mar 20, 2025 by andriy-ca Loading…
4 tasks done
26
Updated int4 moe xlops 16x16 optimization
#1998 opened Mar 20, 2025 by mtgu0705 Loading…
7 tasks
Testx
#1997 opened Mar 20, 2025 by coderfeli • Draft
7 tasks
Split up data_type header.
#1996 opened Mar 20, 2025 by illsilin Loading…
1 task done
9
Add padding to weight-preshuffled
#1995 opened Mar 19, 2025 by ltqin Loading…
6
[New] Build up the feature of CK Tile GEMM CodeGen
#1994 opened Mar 18, 2025 by amd-khushbu Loading…
1 of 7 tasks
@ThomasNing @amd-khushbu
60
use fast path for sequence generation in old CK compilation time
#1993 opened Mar 18, 2025 by tenpercent Loading…
2 of 7 tasks
Post-merge changes for fully async args copy in ck grouped gemm
#1991 opened Mar 18, 2025 by aledudek Loading…
5 of 7 tasks
13
Enable determinism for TE bwd path
#1985 opened Mar 17, 2025 by wangye805 Loading…
7 tasks
Docs: Add precision support reference page ci:docs-only Skip most non-doc CI for this PR documentation Improvements or additions to documentation
#1973 opened Mar 12, 2025 by adeljo-amd Loading…
7 tasks
@adeljo-amd
20
[CK TILE] GEMM pk_int4_t dequant B
#1962 opened Mar 10, 2025 by bartekxk • Draft
7 tasks
@bartekxk
[CK_TILE] Batched transpose 2d
#1960 opened Mar 10, 2025 by xm35p4fu6 Loading…
1 of 7 tasks
@xm35p4fu6
42
[CK_TILE] Add 2:4 structured sparsity support for fp16 gemm
#1957 opened Mar 7, 2025 by jakpiase Loading…
3 of 7 tasks
13
Fix binary_inf in NumericLimits<float>
#1943 opened Mar 4, 2025 by mirza-halilcevic Loading…
support both FP8 interpretations at the same time
#1942 opened Mar 4, 2025 by jeffdaily Loading…
7 tasks
6
[CK-TILE DOC] Ck-tile grouped GEMM documentation
#1939 opened Mar 4, 2025 by mozga-amd • Draft
7 tasks
30
Fix set_slice_tile API and sequence hisogram function
#1938 opened Mar 4, 2025 by aska-0096 Loading…
4 of 6 tasks
1
simiplify fmha v3 bwd codegen
#1921 opened Feb 27, 2025 by slippedJim Loading…
7 tasks
Add readme of fmha
#1917 opened Feb 26, 2025 by asleepzzz • Draft
7 tasks
Fix batched transpose
#1914 opened Feb 25, 2025 by xm35p4fu6 • Draft
1 of 7 tasks
ProTip! Add no:assignee to see everything that’s not assigned.