Default Branch

604c56bc0e · [rocm-libraries] ROCm/rocm-libraries#7850 (commit e8f2756) · Updated 2026-06-29 18:51:17 +00:00

Branches

f728c90546 · Merge branch 'develop' into aviralgoel/memory_pipeline_refactor_2 · Updated 2026-01-23 20:38:05 +00:00    ROCm

493
8

797aff8d49 · update test codes · Updated 2026-01-23 08:59:50 +00:00    ROCm

532
8

37a3f0fb28 · initial draft of split out the permute_n_epilogue · Updated 2026-01-22 09:08:44 +00:00    ROCm

508
1

ef44c88506 · initial draft · Updated 2026-01-22 08:50:40 +00:00    ROCm

508
1

ece69df994 · Improve execution time of batch prefill kernel with vectorized KV cache layout · Updated 2026-01-22 06:00:32 +00:00    ROCm

558
2

19a156aa0a · Apply clang-format with -style=file · Updated 2026-01-22 03:17:44 +00:00    ROCm

511
6

eed4270cb7 · Apply clang-format with -style=file · Updated 2026-01-22 03:14:00 +00:00    ROCm

511
3

8d32b38fbd · fix format · Updated 2026-01-22 03:09:42 +00:00    ROCm

507
3

25f759fc22 · mxfp4 non reduce version · Updated 2026-01-22 02:29:37 +00:00    ROCm

556
2

e548a3f280 · reproduce tolerance gfx11 · Updated 2026-01-21 11:34:41 +00:00    ROCm

577
2

557a8d3f21 · Get the test to compile · Updated 2026-01-21 09:57:01 +00:00    ROCm

532
12

6c28037024 · [CK_TILE] Fix Int32 Overflow in Deterministic FMHA BWD · Updated 2026-01-20 17:25:24 +00:00    ROCm

708
1

aa2a5778e6 · Apply clang-format to sequence helper tests · Updated 2026-01-20 17:20:03 +00:00    ROCm

526
23

4010341092 · Add max error metric · Updated 2026-01-20 13:47:47 +00:00    ROCm

518
5

6e7f79d3dc · add irregular tail vectorloads · Updated 2026-01-19 15:25:40 +00:00    ROCm

562
1

95ca0c373a · Merge branch 'develop' into meskelin/refactor-makegemmtensorviews · Updated 2026-01-19 12:33:08 +00:00    ROCm

524
5

00849ac2e2 · Replace lambdas with named functors in transform_tensor_descriptor · Updated 2026-01-17 03:45:36 +00:00    ROCm

526
4

d7e7fbdcff · Add generate_identity_sequences helper for common pattern · Updated 2026-01-17 03:45:31 +00:00    ROCm

526
3

0568a7a03c · Add static_for_indexed for reduced template instantiation · Updated 2026-01-16 21:16:32 +00:00    ROCm

534
11

ee426bea45 · still debugging: speculating soemthing with cshuffle epilogue · Updated 2026-01-16 20:53:14 +00:00    ROCm

652
50