From 7f63b888c79b2c8cd4f86b3628f1d6f57f8439b9 Mon Sep 17 00:00:00 2001
From: John Baldwin <jhb@FreeBSD.org>
Date: Thu, 27 Jun 2019 19:36:30 +0000
Subject: [PATCH] Hold an explicit reference on the socket for the aiotx task.

Previously, the aiotx task relied on the aio jobs in the queue to hold
a reference on the socket.  However, when the last job is completed,
there is nothing left to hold a reference to the socket buffer lock
used to check if the queue is empty.  In addition, if the last job on
the queue is cancelled, the task can run with no queued jobs holding a
reference to the socket buffer lock the task uses to notice the queue
is empty.

Fix these races by holding an explicit reference on the socket when
the task is queued and dropping that reference when the task
completes.

Reviewed by:	np
MFC after:	1 week
Sponsored by:	Chelsio Communications
Differential Revision:	https://reviews.freebsd.org/D20539
---
 sys/dev/cxgbe/tom/t4_cpl_io.c | 25 ++++++++++++++-----------
 sys/dev/cxgbe/tom/t4_tom.h    |  2 +-
 2 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/sys/dev/cxgbe/tom/t4_cpl_io.c b/sys/dev/cxgbe/tom/t4_cpl_io.c
index 27ef745c764c..fd4f376d8676 100644
--- a/sys/dev/cxgbe/tom/t4_cpl_io.c
+++ b/sys/dev/cxgbe/tom/t4_cpl_io.c
@@ -74,7 +74,7 @@ __FBSDID("$FreeBSD$");
 #include "tom/t4_tom.h"
 
 static void	t4_aiotx_cancel(struct kaiocb *job);
-static void	t4_aiotx_queue_toep(struct toepcb *toep);
+static void	t4_aiotx_queue_toep(struct socket *so, struct toepcb *toep);
 
 static size_t
 aiotx_mbuf_pgoff(struct mbuf *m)
@@ -785,7 +785,7 @@ t4_push_frames(struct adapter *sc, struct toepcb *toep, int drop)
 					if (sowwakeup) {
 						if (!TAILQ_EMPTY(
 						    &toep->aiotx_jobq))
-							t4_aiotx_queue_toep(
+							t4_aiotx_queue_toep(so,
 							    toep);
 						sowwakeup_locked(so);
 					} else
@@ -829,7 +829,7 @@ t4_push_frames(struct adapter *sc, struct toepcb *toep, int drop)
 		}
 		if (sowwakeup) {
 			if (!TAILQ_EMPTY(&toep->aiotx_jobq))
-				t4_aiotx_queue_toep(toep);
+				t4_aiotx_queue_toep(so, toep);
 			sowwakeup_locked(so);
 		} else
 			SOCKBUF_UNLOCK(sb);
@@ -1821,7 +1821,7 @@ do_fw4_ack(struct sge_iq *iq, const struct rss_header *rss, struct mbuf *m)
 				tls_ofld->sb_off -= plen;
 			}
 			if (!TAILQ_EMPTY(&toep->aiotx_jobq))
-				t4_aiotx_queue_toep(toep);
+				t4_aiotx_queue_toep(so, toep);
 			sowwakeup_locked(so);	/* unlocks so_snd */
 		}
 		SOCKBUF_UNLOCK_ASSERT(sb);
@@ -2195,10 +2195,10 @@ static void
 t4_aiotx_task(void *context, int pending)
 {
 	struct toepcb *toep = context;
-	struct inpcb *inp = toep->inp;
-	struct socket *so = inp->inp_socket;
+	struct socket *so;
 	struct kaiocb *job;
 
+	so = toep->aiotx_so;
 	CURVNET_SET(toep->vnet);
 	SOCKBUF_LOCK(&so->so_snd);
 	while (!TAILQ_EMPTY(&toep->aiotx_jobq) && sowriteable(so)) {
@@ -2209,15 +2209,17 @@ t4_aiotx_task(void *context, int pending)
 
 		t4_aiotx_process_job(toep, so, job);
 	}
-	toep->aiotx_task_active = false;
+	toep->aiotx_so = NULL;
 	SOCKBUF_UNLOCK(&so->so_snd);
 	CURVNET_RESTORE();
 
 	free_toepcb(toep);
+	SOCK_LOCK(so);
+	sorele(so);
 }
 
 static void
-t4_aiotx_queue_toep(struct toepcb *toep)
+t4_aiotx_queue_toep(struct socket *so, struct toepcb *toep)
 {
 
 	SOCKBUF_LOCK_ASSERT(&toep->inp->inp_socket->so_snd);
@@ -2225,9 +2227,10 @@ t4_aiotx_queue_toep(struct toepcb *toep)
 	CTR3(KTR_CXGBE, "%s: queueing aiotx task for tid %d, active = %s",
 	    __func__, toep->tid, toep->aiotx_task_active ? "true" : "false");
 #endif
-	if (toep->aiotx_task_active)
+	if (toep->aiotx_so != NULL)
 		return;
-	toep->aiotx_task_active = true;
+	soref(so);
+	toep->aiotx_so = so;
 	hold_toepcb(toep);
 	soaio_enqueue(&toep->aiotx_task);
 }
@@ -2284,7 +2287,7 @@ t4_aio_queue_aiotx(struct socket *so, struct kaiocb *job)
 		panic("new job was cancelled");
 	TAILQ_INSERT_TAIL(&toep->aiotx_jobq, job, list);
 	if (sowriteable(so))
-		t4_aiotx_queue_toep(toep);
+		t4_aiotx_queue_toep(so, toep);
 	SOCKBUF_UNLOCK(&so->so_snd);
 	return (0);
 }
diff --git a/sys/dev/cxgbe/tom/t4_tom.h b/sys/dev/cxgbe/tom/t4_tom.h
index db9e50cd39e1..d4ff4490abad 100644
--- a/sys/dev/cxgbe/tom/t4_tom.h
+++ b/sys/dev/cxgbe/tom/t4_tom.h
@@ -194,7 +194,7 @@ struct toepcb {
 
 	TAILQ_HEAD(, kaiocb) aiotx_jobq;
 	struct task aiotx_task;
-	bool aiotx_task_active;
+	struct socket *aiotx_so;
 
 	/* Tx software descriptor */
 	uint8_t txsd_total;