Default Branch

2089713f94 · [rocm-libraries] ROCm/rocm-libraries#8227 (commit 75c30d5) · Updated 2026-06-26 12:00:58 +00:00

Branches

d02f49f7e1 · ck example for bshd_bs2hd input to debug bwd v3 · Updated 2025-03-23 15:12:50 +00:00    ROCm

1721
51

3993fbbde2 · support bs3hd in ck example to debug v3 · Updated 2025-03-23 02:14:21 +00:00    ROCm

1721
51

12679c489c · turn off threadwise trace · Updated 2025-03-22 07:39:59 +00:00    ROCm

1751
17

8afe31d0a1 · fix KPaing check · Updated 2025-03-18 07:20:25 +00:00    ROCm

1727
148

b5706a2b92 · fix longindex_t · Updated 2025-03-18 04:12:08 +00:00    ROCm

1695
20

fbc2a5d046 · Enable determinism for TE bwd path · Updated 2025-03-17 17:10:06 +00:00    ROCm

1721
52

abf5b9295b · revert soring force arg · Updated 2025-03-17 08:12:35 +00:00    ROCm

1695
17

3452ddc849 · updated the 2 f8xi4 files output to multiply of 16 · Updated 2025-03-14 05:37:15 +00:00    ROCm

1695
2

52b1cd7780 · hotfix fmoe build issue (#1976) · Updated 2025-03-13 07:11:59 +00:00    ROCm

1689
0
Included

03d6f32bd8 · I have discovered that acc_store does not run for index range condition · Updated 2025-03-13 01:11:32 +00:00    ROCm

1751
15

7abcad29d4 · [CK TILE] GEMM pk_int4_t dequant B · Updated 2025-03-10 11:10:29 +00:00    ROCm

1700
1

006e01bb9a · fix typo · Updated 2025-03-10 10:10:50 +00:00    ROCm

1746
203

442bac3f8c · support scatter index type config · Updated 2025-03-10 09:56:15 +00:00    ROCm

1701
1

121e2ef09f · Enhance two fp8xint4 gemm kernels · Updated 2025-03-10 06:09:44 +00:00    ROCm

1700
1

73d8f8ce2c · added debug prints to gridwise and threadwise files · Updated 2025-03-07 09:47:52 +00:00    ROCm

2496
1

5d7c41af35 · merge develop · Updated 2025-03-06 05:27:11 +00:00    ROCm

1711
256

8d2fae714b · rm unrelated files · Updated 2025-03-04 02:33:22 +00:00    ROCm

1721
240

7e0bed7a82 · hack bias for padding · Updated 2025-03-03 17:23:32 +00:00    ROCm

1738
47

e95531e749 · data transfer · Updated 2025-02-28 03:04:21 +00:00    ROCm

1801
5

c4dfe25bcc · enable fa v3 bwd codegen · Updated 2025-02-27 11:32:10 +00:00    ROCm

1738
46