Add seqlens_k bounds validation in GroupQueryAttention to prevent GEMM OOB #28031
Add seqlens_k bounds validation in GroupQueryAttention to prevent GEM…
a49b8389
Add non-prompt regression test for seqlens_k underflow guard
6da73057
Address Copilot review: INVALID_ARGUMENT errors, total_seq_len check,…
b2f6dfd0
Add seqlens_k bounds validation in GroupQueryAttention to prevent GEM…
5edfbfc1
Address review: int64 cast, total_seq_len validation, boundary tests
5d6f96d6
Add SafeInt defense-in-depth and INT32_MAX test for GQA seqlens_k
11f5d838
Use SafeInt for memory allocation size calculation in GQAAttentionBase
d3fddd2f
vraspar
enabled auto-merge (squash) 25 days ago
vraspar
merged
7c56fa83
into main 25 days ago
vraspar
deleted the vraspar/fix-gqa-seqlens-k-oob branch 25 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub