[BugFix] Fix potential cuda-graph IMA (#21196)

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
2025-07-19 05:18:47 -04:00
parent 18e519ec86
commit 59f935300c
2 changed files with 6 additions and 6 deletions
--- a/vllm/v1/attention/backends/utils.py
+++ b/vllm/v1/attention/backends/utils.py
@@ -59,11 +59,6 @@ class CommonAttentionMetadata:
    block_table_tensor: torch.Tensor
    slot_mapping: torch.Tensor

-    def __post_init__(self):
-        # Fill unused with -1. Needed for reshape_and_cache in full cuda graph
-        # mode.
-        self.slot_mapping[self.num_actual_tokens:].fill_(-1)
-

 M = TypeVar("M")