fix: piecewise_cuda_graph get correct qo_indptr (#21452)

Co-authored-by: Avery Huang <averyh@nvidia.com>

test/registered/scheduler/test_abort.py

@@ -317,6 +317,8 @@ class TestAbortWithWaitingTimeout(WaitingTimeoutMixin, CustomTestCase):
                 timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
                 other_args=[
                     "--max-running-requests=1",
+                    # Disable PCG to avoid padding in flashinfer backend. Ref: https://github.com/sgl-project/sglang/pull/21452
+                    "--disable-piecewise-cuda-graph",
                 ],
             )
 

test/registered/scheduler/test_priority_scheduling.py

@@ -41,6 +41,8 @@ class TestPriorityScheduling(CustomTestCase):
                 "--max-queued-requests",  # Enforce max queued request number is 3
                 "3",
                 "--enable-priority-scheduling",  # Enable priority scheduling
+                # Disable PCG to avoid padding in flashinfer backend. Ref: https://github.com/sgl-project/sglang/pull/21452
+                "--disable-piecewise-cuda-graph",
             ),
             return_stdout_stderr=(cls.stdout, cls.stderr),
         )
@@ -247,6 +249,7 @@ class TestPrioritySchedulingMultipleRunningRequests(CustomTestCase):
                 "--max-queued-requests",  # Enforce max queued request number is 3
                 "3",
                 "--enable-priority-scheduling",  # Enable priority scheduling
+                "--disable-piecewise-cuda-graph",
             ),
             return_stdout_stderr=(cls.stdout, cls.stderr),
         )