[Kernel][ROCM] Upstream prefix prefill speed up for vLLM V1 (#13305)

Signed-off-by: Sage Moore <sage@neuralmagic.com>
Signed-off-by: root <root@banff-cyxtera-s73-5.ctr.dcgpu>
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Signed-off-by: root <root@banff-cyxtera-s65-4.amd.com>
Signed-off-by: maleksan85 <maleksan@amd.com>
Signed-off-by: <>
Co-authored-by: Sage Moore <sage@neuralmagic.com>
Co-authored-by: root <root@banff-cyxtera-s73-5.ctr.dcgpu>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: qli88 <qiang.li2@amd.com>
Co-authored-by: root <root@banff-cyxtera-s65-4.amd.com>

tests/core/block/e2e/test_correctness.py

@@ -195,15 +195,15 @@ def test_lookahead_greedy_equality_with_preemption(baseline_llm_generator,
     ])
 @pytest.mark.parametrize("per_test_common_llm_kwargs",
                          [{
-                             "block_size": 8,
+                             "block_size": 16,
                              "max_num_batched_tokens": 2,
                              "max_num_seqs": 2,
                          }, {
-                             "block_size": 8,
+                             "block_size": 16,
                              "max_num_batched_tokens": 3,
                              "max_num_seqs": 2,
                          }, {
-                             "block_size": 8,
+                             "block_size": 16,
                              "max_num_batched_tokens": 256,
                              "max_num_seqs": 10,
                          }])