public inbox for linux-scsi@vger.kernel.org
 help / color / mirror / Atom feed
From: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
To: target-devel <target-devel@vger.kernel.org>
Cc: linux-scsi <linux-scsi@vger.kernel.org>,
	lkml <linux-kernel@vger.kernel.org>,
	Rob Millner <rlm@daterainc.com>, Chu Yuan Lin <cyl@datera.io>,
	James Shen <jcs@datera.io>,
	Nicholas Bellinger <nab@linux-iscsi.org>
Subject: [PATCH 0/2] target: Bug-fixes for v4.11-rc
Date: Thu, 30 Mar 2017 08:29:52 +0000	[thread overview]
Message-ID: <1490862594-2712-1-git-send-email-nab@linux-iscsi.org> (raw)

From: Nicholas Bellinger <nab@linux-iscsi.org>

Hi all,

Here are two additional target bug-fixes that have been found
by the DATERA Q/A + automation team during extended longevity
and stress testing atop v4.1.y stable code.

The first is a iscsi-target specific TMR reference leak during
session shutdown when se_cmd quiesce occured before hand-off
back to iscsi-target, that results in se_tmr_req descriptors
being left after se_cmd was freed.

This would manifest as kernel paging OOPsen during LUN_RESET
after the leak occured, because the associated se_tmr_req had
not been released even though the pre-allocated parent se_cmd
descriptor had already been freed during session shutdown.

The second is a race between se_lun configfs unlink shutdown
when se_lun->lun_ref percpu_ref RCU grace period is happening,
and user-space attempts to add a new NodeACL mappedlun configfs
symlink while se_lun shutdown is occuring.

This would manifest as NULL pointer dereference OOPsen within
target_stat_scsi_att_intr_port_show_attr_dev() after a new
NodeACL mappedlun was added to a se_lun that had already
been shutdown.

Both have been verified on v4.1.y stable code, and forward
ported to v4.11-rc.

Please review.

--nab

Nicholas Bellinger (2):
  iscsi-target: Fix TMR reference leak during session shutdown
  target: Avoid mappedlun symlink creation during lun shutdown

 drivers/target/iscsi/iscsi_target_util.c     | 12 +++++++-----
 drivers/target/target_core_fabric_configfs.c |  5 +++++
 drivers/target/target_core_tpg.c             |  4 ++++
 include/target/target_core_base.h            |  1 +
 4 files changed, 17 insertions(+), 5 deletions(-)

-- 
1.9.1

             reply	other threads:[~2017-03-30  8:29 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-30  8:29 Nicholas A. Bellinger [this message]
2017-03-30  8:29 ` [PATCH 1/2] iscsi-target: Fix TMR reference leak during session shutdown Nicholas A. Bellinger
2017-03-30 17:08   ` Bart Van Assche
2017-04-02 22:38     ` Nicholas A. Bellinger
2017-03-30  8:29 ` [PATCH 2/2] target: Avoid mappedlun symlink creation during lun shutdown Nicholas A. Bellinger

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=1490862594-2712-1-git-send-email-nab@linux-iscsi.org \
    --to=nab@linux-iscsi.org \
    --cc=cyl@datera.io \
    --cc=jcs@datera.io \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=rlm@daterainc.com \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox