seize: use separate checkpoint_devices function

Move `run_plugins(CHECKPOINT_DEVICES)` out of `collect_pstree()` to
ensure that the function's sole responsibility is to use the cgroup
freezer for the process tree. This allows us to avoid a time-out
error when checkpointing applications with large GPU state.

Suggested-by: Andrei Vagin <avagin@google.com>
Suggested-by: Jesus Ramos <jeramos@nvidia.com>
Signed-off-by: Radostin Stoyanov <rstoyanov@fedoraproject.org>

criu/cr-dump.c

@@ -1963,6 +1963,9 @@ int cr_pre_dump_tasks(pid_t pid)
 	if (collect_pstree())
 		goto err;
 
+	if (checkpoint_devices())
+		goto err;
+
 	if (collect_pstree_ids_predump())
 		goto err;
 
@@ -2192,6 +2195,9 @@ int cr_dump_tasks(pid_t pid)
 	if (collect_pstree())
 		goto err;
 
+	if (checkpoint_devices())
+		goto err;
+
 	if (collect_pstree_ids())
 		goto err;
 

criu/include/seize.h

@@ -2,6 +2,7 @@
 #define __CR_SEIZE_H__
 
 extern int collect_pstree(void);
+extern int checkpoint_devices(void);
 struct pstree_item;
 extern void pstree_switch_state(struct pstree_item *root_item, int st);
 extern const char *get_real_freezer_state(void);

criu/seize.c

@@ -1017,7 +1017,6 @@ int collect_pstree(void)
 	pid_t pid = root_item->pid->real;
 	int ret, exit_code = -1;
 	struct proc_status_creds creds;
-	struct pstree_item *iter;
 
 	timing_start(TIME_FREEZING);
 
@@ -1078,6 +1077,21 @@ int collect_pstree(void)
 		goto err;
 	}
 
+	exit_code = 0;
+	timing_stop(TIME_FREEZING);
+	timing_start(TIME_FROZEN);
+
+err:
+	/* Freezing stage finished in time - disable timer. */
+	alarm(0);
+	return exit_code;
+}
+
+int checkpoint_devices(void)
+{
+	struct pstree_item *iter;
+	int ret, exit_code = -1;
+
 	for_each_pstree_item(iter) {
 		if (!task_alive(iter))
 			continue;
@@ -1087,11 +1101,6 @@ int collect_pstree(void)
 	}
 
 	exit_code = 0;
-	timing_stop(TIME_FREEZING);
-	timing_start(TIME_FROZEN);
-
 err:
-	/* Freezing stage finished in time - disable timer. */
-	alarm(0);
 	return exit_code;
-}
+}