All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mike Christie <michael.christie@oracle.com>
To: mlombard@redhat.com, martin.petersen@oracle.com,
	mgurtovoy@nvidia.com, sagi@grimberg.me, d.bogdanov@yadro.com,
	linux-scsi@vger.kernel.org, target-devel@vger.kernel.org
Cc: Mike Christie <michael.christie@oracle.com>
Subject: [PATCH v2 05/13] scsi: target: iscsit: stop/wait on cmds during conn close
Date: Wed, 11 Jan 2023 21:08:24 -0600	[thread overview]
Message-ID: <20230112030832.110143-6-michael.christie@oracle.com> (raw)
In-Reply-To: <20230112030832.110143-1-michael.christie@oracle.com>

This fixes a bug added in:

commit f36199355c64 ("scsi: target: iscsi: Fix cmd abort fabric stop
race")

If we have multiple sessions to the same se_device we can hit a race where
a LUN_RESET on one session cleans up the se_cmds from under another
session which is being closed. This results in the closing session freeing
its conn/session structs while they are still in use.

The bug is:

1. Session1 has IO se_cmd1.
2. Session2 can also have se_cmds for IO and optionally TMRs for ABORTS
but then gets a LUN_RESET.
3. The LUN_RESET on session2 sees the se_cmds on session1 and during
the drain stages marks them all with CMD_T_ABORTED.
4. session1 is now closed so iscsit_release_commands_from_conn only sees
se_cmds with the CMD_T_ABORTED bit set and returns immediately even
though we have outstanding commands.
5. session1's connection and session are freed.
6. The backend request for se_cmd1 completes and it accesses the freed
connection/session.

This hooks the iscsit layer into the cmd counter code, so we can wait for
all outstanding se_cmds before freeing the connection.

Fixes: f36199355c64 ("scsi: target: iscsi: Fix cmd abort fabric stop race")
Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/target/iscsi/iscsi_target.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 11115c207844..83b007141229 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -4245,6 +4245,16 @@ static void iscsit_release_commands_from_conn(struct iscsit_conn *conn)
 		iscsit_free_cmd(cmd, true);
 
 	}
+
+	/*
+	 * Wait on commands that were cleaned up via the aborted_task path.
+	 * LLDs that implement iscsit_wait_conn will already have waited for
+	 * commands.
+	 */
+	if (!conn->conn_transport->iscsit_wait_conn) {
+		target_stop_cmd_counter(conn->cmd_cnt);
+		target_wait_for_cmds(conn->cmd_cnt);
+	}
 }
 
 static void iscsit_stop_timers_for_cmds(
-- 
2.31.1


  parent reply	other threads:[~2023-01-12  3:09 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-12  3:08 [PATCH v2 00/13] target task management fixes Mike Christie
2023-01-12  3:08 ` [PATCH v2 01/13] scsi: target: Move sess cmd counter to new struct Mike Christie
2023-01-12  3:08 ` [PATCH v2 02/13] scsi: target: Move cmd counter allocation Mike Christie
2023-01-12  3:08 ` [PATCH v2 03/13] scsi: target: Pass in cmd counter to use during cmd setup Mike Christie
2023-01-12  3:08 ` [PATCH v2 04/13] scsi: target: iscsit/isert: Alloc per conn cmd counter Mike Christie
2023-01-12  3:08 ` Mike Christie [this message]
2023-01-12  3:08 ` [PATCH v2 06/13] scsi: target: iscsit: Fix TAS handling during conn cleanup Mike Christie
2023-01-12  3:08 ` [PATCH v2 07/13] scsi: target: Fix multiple LUN_RESET handling Mike Christie
2023-01-12  3:08 ` [PATCH v2 08/13] scsi: target: drop tas arg from __transport_wait_for_tasks Mike Christie
2023-01-12  3:08 ` [PATCH v2 09/13] scsi: target: iscsit: Fix isert disconnect handling during login Mike Christie
2023-01-13 14:08   ` Dmitry Bogdanov
2023-01-18  2:14     ` Mike Christie
2023-01-23  9:53       ` Sagi Grimberg
2023-01-24  2:40         ` michael.christie
2023-01-12  3:08 ` [PATCH v2 10/13] scsi: target: iscsit: Add helper to check when cmd has failed Mike Christie
2023-01-12  3:08 ` [PATCH v2 11/13] scsi: target: Treat CMD_T_FABRIC_STOP like CMD_T_STOP Mike Christie
2023-01-13 14:15   ` Dmitry Bogdanov
2023-01-17 11:52     ` Dmitry Bogdanov
2023-01-17 18:45       ` Mike Christie
2023-01-17 19:55         ` Dmitry Bogdanov
2023-01-18 19:12           ` Mike Christie
2023-01-12  3:08 ` [PATCH v2 12/13] scsi: target: iscsit: Cleanup isert commands at conn closure Mike Christie
2023-01-12  3:08 ` [PATCH v2 13/13] IB/isert: Fix hang in target_wait_for_cmds Mike Christie
2023-01-23  9:50   ` Sagi Grimberg
2023-01-12  3:13 ` [PATCH v2 00/13] target task management fixes Mike Christie
2023-01-23  9:54   ` Sagi Grimberg

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=20230112030832.110143-6-michael.christie@oracle.com \
    --to=michael.christie@oracle.com \
    --cc=d.bogdanov@yadro.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=mgurtovoy@nvidia.com \
    --cc=mlombard@redhat.com \
    --cc=sagi@grimberg.me \
    --cc=target-devel@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.