diff --git a/tests/unittest/_torch/attention/sparse/dsa/test_dsa_sparse_mla.py b/tests/unittest/_torch/attention/sparse/dsa/test_dsa_sparse_mla.py
index f3a459d21b4e..ed3af070b2a4 100644
--- a/tests/unittest/_torch/attention/sparse/dsa/test_dsa_sparse_mla.py
+++ b/tests/unittest/_torch/attention/sparse/dsa/test_dsa_sparse_mla.py
@@ -777,6 +777,12 @@ def yarn_get_mscale(scale=1, mscale=1):
         * kv_cache_tokens_per_block
         * max_num_contexts
     )
+    # The static token-sparse context FMHA may load KV rows beyond the selected ones, and its
+    # TMA descriptor does not bound rows to the pool. A pool sized only for these short
+    # sequences (a single 64-token block for context_sequence_lengths=[10]) can therefore be
+    # read past its end, which faults whenever the following address range is unmapped. Keep
+    # at least TopK rows in the pool, as any real deployment does.
+    max_tokens = max(max_tokens, SPARSE_TOPK)
 
     pretrained_config = SimpleNamespace(
         rms_norm_eps=1e-6,