Forwarded: Re: [syzbot] [block?] WARNING in blk_mq_free_tag_set

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@xxxxxxxxxxxxxxx, syzkaller-bugs@xxxxxxxxxxxxxxxx.

***

Subject: Re: [syzbot] [block?] WARNING in blk_mq_free_tag_set
Author: raghunathpalla.0209@xxxxxxxxx

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git f0100363d8c3

diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c
index ed204b3f4b84..9fa0987c203b 100644
--- a/kernel/rcu/srcutree.c
+++ b/kernel/rcu/srcutree.c
@@ -701,10 +701,13 @@ void cleanup_srcu_struct(struct srcu_struct *ssp)
 	for_each_possible_cpu(cpu) {
 		struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu);
 
-		// Call srcu_barrier() before this cleanup_srcu_struct()
-		// to avoid triggering this WARN_ON().
-		if (WARN_ON(timer_delete_sync(&sdp->delay_work) &&
-			    rcu_segcblist_n_cbs(&sdp->srcu_cblist)) &&
+		// Deleting a pending delay timer that still has callbacks
+		// behind it is not an error: queueing the work here and
+		// flushing it below invokes them.  A caller that really did
+		// forget srcu_barrier() is caught by the WARN_ON() after the
+		// flush.
+		if (timer_delete_sync(&sdp->delay_work) &&
+		    rcu_segcblist_n_cbs(&sdp->srcu_cblist) &&
 		    rcu_cpu_beenfullyonline(sdp->cpu))
 			queue_work_on(sdp->cpu, rcu_gp_wq, &sdp->work);
 		flush_work(&sdp->work);