From: Hannes Reinecke <hare@suse.de>
To: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>,
Johannes Thumshirn <jth@kernel.org>,
James Bottomley <james.bottomley@hansenpartnership.com>,
Chad Dupuis <chad.dupuis@cavium.com>,
linux-scsi@vger.kernel.org
Subject: [PATCH 3/7] libfc: Do not take rdata->rp_mutex when processing a -FC_EX_CLOSED ELS response.
Date: Fri, 30 Sep 2016 09:04:10 +0200 [thread overview]
Message-ID: <1475219054-31241-4-git-send-email-hare@suse.de> (raw)
In-Reply-To: <1475219054-31241-1-git-send-email-hare@suse.de>
From: Chad Dupuis <chad.dupuis@cavium.com>
When an ELS response handler receives a -FC_EX_CLOSED, the rdata->rp_mutex is
already held which can lead to a deadlock condition like the following stack trace:
[<ffffffffa04d8f18>] fc_rport_plogi_resp+0x28/0x200 [libfc]
[<ffffffffa04cfa1a>] fc_invoke_resp+0x6a/0xe0 [libfc]
[<ffffffffa04d0c08>] fc_exch_mgr_reset+0x1b8/0x280 [libfc]
[<ffffffffa04d87b3>] fc_rport_logoff+0x43/0xd0 [libfc]
[<ffffffffa04ce73d>] fc_disc_stop+0x6d/0xf0 [libfc]
[<ffffffffa04ce7ce>] fc_disc_stop_final+0xe/0x20 [libfc]
[<ffffffffa04d55f7>] fc_fabric_logoff+0x17/0x70 [libfc]
The other ELS handlers need to follow the FLOGI response handler and simply do
a kref_put against the fc_rport_priv struct and exit when receving a
-FC_EX_CLOSED response.
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
---
drivers/scsi/libfc/fc_rport.c | 32 ++++++++++++++++++++++++--------
1 file changed, 24 insertions(+), 8 deletions(-)
diff --git a/drivers/scsi/libfc/fc_rport.c b/drivers/scsi/libfc/fc_rport.c
index a0ceba1..ff33ae6 100644
--- a/drivers/scsi/libfc/fc_rport.c
+++ b/drivers/scsi/libfc/fc_rport.c
@@ -952,10 +952,13 @@ static void fc_rport_plogi_resp(struct fc_seq *sp, struct fc_frame *fp,
u16 cssp_seq;
u8 op;
- mutex_lock(&rdata->rp_mutex);
-
FC_RPORT_DBG(rdata, "Received a PLOGI %s\n", fc_els_resp_type(fp));
+ if (fp == ERR_PTR(-FC_EX_CLOSED))
+ goto put;
+
+ mutex_lock(&rdata->rp_mutex);
+
if (rdata->rp_state != RPORT_ST_PLOGI) {
FC_RPORT_DBG(rdata, "Received a PLOGI response, but in state "
"%s\n", fc_rport_state(rdata));
@@ -994,6 +997,7 @@ out:
fc_frame_free(fp);
err:
mutex_unlock(&rdata->rp_mutex);
+put:
kref_put(&rdata->kref, lport->tt.rport_destroy);
}
@@ -1079,10 +1083,13 @@ static void fc_rport_prli_resp(struct fc_seq *sp, struct fc_frame *fp,
u8 op;
u8 resp_code = 0;
- mutex_lock(&rdata->rp_mutex);
-
FC_RPORT_DBG(rdata, "Received a PRLI %s\n", fc_els_resp_type(fp));
+ if (fp == ERR_PTR(-FC_EX_CLOSED))
+ goto put;
+
+ mutex_lock(&rdata->rp_mutex);
+
if (rdata->rp_state != RPORT_ST_PRLI) {
FC_RPORT_DBG(rdata, "Received a PRLI response, but in state "
"%s\n", fc_rport_state(rdata));
@@ -1150,6 +1157,7 @@ out:
fc_frame_free(fp);
err:
mutex_unlock(&rdata->rp_mutex);
+put:
kref_put(&rdata->kref, rdata->local_port->tt.rport_destroy);
}
@@ -1230,10 +1238,13 @@ static void fc_rport_rtv_resp(struct fc_seq *sp, struct fc_frame *fp,
struct fc_rport_priv *rdata = rdata_arg;
u8 op;
- mutex_lock(&rdata->rp_mutex);
-
FC_RPORT_DBG(rdata, "Received a RTV %s\n", fc_els_resp_type(fp));
+ if (fp == ERR_PTR(-FC_EX_CLOSED))
+ goto put;
+
+ mutex_lock(&rdata->rp_mutex);
+
if (rdata->rp_state != RPORT_ST_RTV) {
FC_RPORT_DBG(rdata, "Received a RTV response, but in state "
"%s\n", fc_rport_state(rdata));
@@ -1275,6 +1286,7 @@ out:
fc_frame_free(fp);
err:
mutex_unlock(&rdata->rp_mutex);
+put:
kref_put(&rdata->kref, rdata->local_port->tt.rport_destroy);
}
@@ -1374,10 +1386,13 @@ static void fc_rport_adisc_resp(struct fc_seq *sp, struct fc_frame *fp,
struct fc_els_adisc *adisc;
u8 op;
- mutex_lock(&rdata->rp_mutex);
-
FC_RPORT_DBG(rdata, "Received a ADISC response\n");
+ if (fp == ERR_PTR(-FC_EX_CLOSED))
+ goto put;
+
+ mutex_lock(&rdata->rp_mutex);
+
if (rdata->rp_state != RPORT_ST_ADISC) {
FC_RPORT_DBG(rdata, "Received a ADISC resp but in state %s\n",
fc_rport_state(rdata));
@@ -1412,6 +1427,7 @@ out:
fc_frame_free(fp);
err:
mutex_unlock(&rdata->rp_mutex);
+put:
kref_put(&rdata->kref, rdata->local_port->tt.rport_destroy);
}
--
1.8.5.6
next prev parent reply other threads:[~2016-09-30 7:04 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-30 7:04 [PATCH 0/7] libfc: sanitize kref and state machine fixes Hannes Reinecke
2016-09-30 7:04 ` [PATCH 1/7] libfc: Revisit kref handling Hannes Reinecke
2016-09-30 7:23 ` Johannes Thumshirn
2016-09-30 7:04 ` [PATCH 2/7] libfc: Fixup disc_mutex handling Hannes Reinecke
2016-09-30 7:25 ` Johannes Thumshirn
2016-09-30 7:04 ` Hannes Reinecke [this message]
2016-09-30 7:26 ` [PATCH 3/7] libfc: Do not take rdata->rp_mutex when processing a -FC_EX_CLOSED ELS response Johannes Thumshirn
2016-09-30 7:04 ` [PATCH 4/7] libfc: Do not drop down to FLOGI for fc_rport_login() Hannes Reinecke
2016-09-30 7:26 ` Johannes Thumshirn
2016-09-30 7:04 ` [PATCH 5/7] libfc: Do not login if the port is already started Hannes Reinecke
2016-09-30 7:04 ` [PATCH 6/7] libfc: don't advance state machine for incoming FLOGI Hannes Reinecke
2016-09-30 7:04 ` [PATCH 7/7] fcoe: Harden CVL handling when we have not logged into the fabric Hannes Reinecke
2016-09-30 7:27 ` Johannes Thumshirn
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1475219054-31241-4-git-send-email-hare@suse.de \
--to=hare@suse.de \
--cc=chad.dupuis@cavium.com \
--cc=hch@lst.de \
--cc=james.bottomley@hansenpartnership.com \
--cc=jth@kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).