ltqin bf07a0150e [rocm-libraries] ROCm/rocm-libraries#7723 (commit 4ed6c51) ...

[CK Tile] Enable LSE output for fp8bf16 V3 FMHA kernels
 (#7723)

###  Motivation
The V3 pipeline (qr_async_trload_v3) for fp8bf16 FMHA kernels did not
support LSE (Log-Sum-Exp) output. This PR enables LSE output support for
fp8bf16 V3 FMHA kernels, allowing users to retrieve attention statistics
alongside attention outputs.
### Technical Details
    - StandardAttention: lse = softmax_scale * m + log(l)
- LogitsSoftCap: lse = (m / log2(e)) + log(l)

### Test Plan
Run FMHA forward example with fp8bf16 precision and LSE output enabled:
- Test 1: Basic LSE functionality
./build/bin/tile_example_fmha_fwd -v=1 -b=1 -h=8 -s=1024 -d=128
-prec=fp8bf16 -init=3 -qscale=1 -lse=1
- Test 2: LSE with LogitsSoftCap (CMakeList should remove Logits filter)
./build/bin/tile_example_fmha_fwd -v=1 -b=1 -h=8 -s=1024 -d=128
-prec=fp8bf16 -init=3 -qscale=1 -lse=1 -logits_soft_cap=30.0

2026-05-28 15:58:54 +00:00

..

ck

[rocm-libraries] ROCm/rocm-libraries#6761 (commit d19f6f1)

2026-05-27 18:55:15 +00:00

ck_tile

[rocm-libraries] ROCm/rocm-libraries#7723 (commit 4ed6c51)

2026-05-28 15:58:54 +00:00

rapidjson

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00