[3/N][KV-Cache Layout Refactor] Standardize Mamba cache; drop get_transfer_cache_regions (#44456)

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
This commit is contained in:
Lucas Wilkinson
2026-07-21 09:16:15 +00:00
committed by GitHub
co-authored by Claude OpenAI Codex
parent eb44b3aaa4
commit 6700813f86
10 changed files with 115 additions and 225 deletions
@@ -125,8 +125,7 @@ def test_register_kv_caches(backend):
own dedicated tensors.
Uses the real GPUModelRunner.initialize_kv_cache_tensors to produce
kv_caches, which automatically applies
_update_hybrid_attention_mamba_layout for hybrid models.
the raw per-layer kv_caches registered by the connector.
Verifies that the canonicalized CanonicalKVCaches has the correct
block tensors, tensor_idx references, and page sizes across all groups.