tests: kill zombie IPC child processes after join timeout

If a spawned process gets stuck during CUDA context teardown (the
Python 3.12 + CUDA 12.9.1 hang pattern from issue #2004), the existing
join(timeout=CHILD_TIMEOUT_SEC) returns but leaves the child alive. That
zombie holds open IPC handles, causing the ipc_memory_resource fixture's
mr.close() to block indefinitely and tie up the runner for hours.

Kill any process that is still alive after its join timeout so the IPC
handle is released before fixture teardown runs. The test still fails
(exit code != 0 or completed == False), just quickly instead of hanging.

Fixes #2004

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: aryanputta

cuda_core/tests/memory_ipc/test_send_buffers.py

@@ -39,6 +39,13 @@ def test_main(self, ipc_device, nmrs):
 
             # Wait for the child process.
             process.join(timeout=CHILD_TIMEOUT_SEC)
+            if process.is_alive():
+                # Child is stuck (CUDA context teardown can hang under certain
+                # driver/Python combos — see issue #2004). Kill it so the
+                # IPC handle is released and the fixture teardown doesn't
+                # block the runner for hours.
+                process.kill()
+                process.join()
             assert process.exitcode == 0
 
             # Verify that the buffers were modified.
@@ -96,10 +103,20 @@ def test_main(self, ipc_device, ipc_memory_resource):
         proc_c.start()
 
         # Wait for C to signal completion then clean up.
-        event_c.wait(timeout=CHILD_TIMEOUT_SEC)
+        completed = event_c.wait(timeout=CHILD_TIMEOUT_SEC)
         event_b.set()  # b can finish now
         proc_b.join(timeout=CHILD_TIMEOUT_SEC)
         proc_c.join(timeout=CHILD_TIMEOUT_SEC)
+
+        # Kill any processes that are still alive. Without this, a child stuck
+        # in CUDA context teardown (issue #2004: Python 3.12 + CUDA 12.9.1)
+        # holds IPC handles and blocks fixture teardown indefinitely.
+        for p in (proc_b, proc_c):
+            if p.is_alive():
+                p.kill()
+                p.join()
+
+        assert completed, "process C did not complete within timeout"
         assert proc_b.exitcode == 0
         assert proc_c.exitcode == 0