[ROCm][Cleanup] Remove stale AITER FA hybrid KV-cache TODO (#44178)

Signed-off-by: Tuukka Sarvi <tuukka.sarvi@amd.com> Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>

Tuukka Sarvi committed Jun 16, 2026 at 17:28 UTC 3d34f8cbdcc9545571f5c2ec78e0327b1e5779e4

1 file changed -2

vllm/v1/attention/backends/rocm_aiter_fa.py

-2

@@ -1408,8 +1408,6 @@ class AiterFlashAttentionImpl(AttentionImpl):

1408 assert k_scale is not None and v_scale is not None, (

1409 "k_scale and v_scale are required for shuffled update"

1410 )

1411 - # TODO: Add correct KV cache handling for hybrid model. KV cache

1412 - # may not be contiguous if mamba state exists.

1411 reshape_and_cache_shuffle_triton(

1412 key,

1413 value,