All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Bob Peterson <rpeterso@redhat.com>,
	Andreas Gruenbacher <agruenba@redhat.com>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.10 16/38] gfs2: fix a deadlock on withdraw-during-mount
Date: Wed, 16 Jun 2021 17:33:25 +0200	[thread overview]
Message-ID: <20210616152835.912416286@linuxfoundation.org> (raw)
In-Reply-To: <20210616152835.407925718@linuxfoundation.org>

From: Bob Peterson <rpeterso@redhat.com>

[ Upstream commit 865cc3e9cc0b1d4b81c10d53174bced76decf888 ]

Before this patch, gfs2 would deadlock because of the following
sequence during mount:

mount
   gfs2_fill_super
      gfs2_make_fs_rw <--- Detects IO error with glock
         kthread_stop(sdp->sd_quotad_process);
            <--- Blocked waiting for quotad to finish

logd
   Detects IO error and the need to withdraw
   calls gfs2_withdraw
      gfs2_make_fs_ro
         kthread_stop(sdp->sd_quotad_process);
            <--- Blocked waiting for quotad to finish

gfs2_quotad
   gfs2_statfs_sync
      gfs2_glock_wait <---- Blocked waiting for statfs glock to be granted

glock_work_func
   do_xmote <---Detects IO error, can't release glock: blocked on withdraw
      glops->go_inval
      glock_blocked_by_withdraw
         requeue glock work & exit <--- work requeued, blocked by withdraw

This patch makes a special exception for the statfs system inode glock,
which allows the statfs glock UNLOCK to proceed normally. That allows the
quotad daemon to exit during the withdraw, which allows the logd daemon
to exit during the withdraw, which allows the mount to exit.

Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/gfs2/glock.c | 24 +++++++++++++++++++++---
 1 file changed, 21 insertions(+), 3 deletions(-)

diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index ea2f2de44806..59130cbbd995 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -569,6 +569,16 @@ out_locked:
 	spin_unlock(&gl->gl_lockref.lock);
 }
 
+static bool is_system_glock(struct gfs2_glock *gl)
+{
+	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
+	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
+
+	if (gl == m_ip->i_gl)
+		return true;
+	return false;
+}
+
 /**
  * do_xmote - Calls the DLM to change the state of a lock
  * @gl: The lock state
@@ -658,17 +668,25 @@ skip_inval:
 	 * to see sd_log_error and withdraw, and in the meantime, requeue the
 	 * work for later.
 	 *
+	 * We make a special exception for some system glocks, such as the
+	 * system statfs inode glock, which needs to be granted before the
+	 * gfs2_quotad daemon can exit, and that exit needs to finish before
+	 * we can unmount the withdrawn file system.
+	 *
 	 * However, if we're just unlocking the lock (say, for unmount, when
 	 * gfs2_gl_hash_clear calls clear_glock) and recovery is complete
 	 * then it's okay to tell dlm to unlock it.
 	 */
 	if (unlikely(sdp->sd_log_error && !gfs2_withdrawn(sdp)))
 		gfs2_withdraw_delayed(sdp);
-	if (glock_blocked_by_withdraw(gl)) {
-		if (target != LM_ST_UNLOCKED ||
-		    test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags)) {
+	if (glock_blocked_by_withdraw(gl) &&
+	    (target != LM_ST_UNLOCKED ||
+	     test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags))) {
+		if (!is_system_glock(gl)) {
 			gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD);
 			goto out;
+		} else {
+			clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
 		}
 	}
 
-- 
2.30.2




  parent reply	other threads:[~2021-06-16 15:36 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-16 15:33 [PATCH 5.10 00/38] 5.10.45-rc1 review Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 01/38] net: ieee802154: fix null deref in parse dev addr Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 02/38] HID: quirks: Set INCREMENT_USAGE_ON_DUPLICATE for Saitek X65 Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 03/38] HID: a4tech: use A4_2WHEEL_MOUSE_HACK_B8 for A4TECH NB-95 Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 04/38] HID: hid-input: add mapping for emoji picker key Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 05/38] HID: hid-sensor-hub: Return error for hid_set_field() failure Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 06/38] HID: quirks: Add quirk for Lenovo optical mouse Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 07/38] HID: multitouch: set Stylus suffix for Stylus-application devices, too Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 08/38] HID: Add BUS_VIRTUAL to hid_connect logging Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 09/38] HID: usbhid: fix info leak in hid_submit_ctrl Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 10/38] drm/tegra: sor: Do not leak runtime PM reference Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 11/38] gpu: host1x: Split up client initalization and registration Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 12/38] drm/tegra: sor: Fully initialize SOR before registration Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 13/38] ARM: OMAP1: Fix use of possibly uninitialized irq variable Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 14/38] ARM: OMAP2+: Fix build warning when mmc_omap is not built Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 15/38] gfs2: Prevent direct-I/O write fallback errors from getting lost Greg Kroah-Hartman
2021-06-16 15:33 ` Greg Kroah-Hartman [this message]
2021-06-16 15:33 ` [PATCH 5.10 17/38] HID: gt683r: add missing MODULE_DEVICE_TABLE Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 18/38] riscv: Use -mno-relax when using lld linker Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 19/38] gfs2: Fix use-after-free in gfs2_glock_shrink_scan Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 20/38] scsi: target: core: Fix warning on realtime kernels Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 21/38] ethernet: myri10ge: Fix missing error code in myri10ge_probe() Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 22/38] scsi: qedf: Do not put host in qedf_vport_create() unconditionally Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 23/38] Bluetooth: Add a new USB ID for RTL8822CE Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 24/38] scsi: scsi_devinfo: Add blacklist entry for HPE OPEN-V Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 25/38] nvme-loop: reset queue count to 1 in nvme_loop_destroy_io_queues() Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 26/38] nvme-loop: clear NVME_LOOP_Q_LIVE when nvme_loop_configure_admin_queue() fails Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 27/38] nvme-loop: check for NVME_LOOP_Q_LIVE in nvme_loop_destroy_admin_queue() Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 28/38] nvme-loop: do not warn for deleted controllers during reset Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 29/38] net: ipconfig: Dont override command-line hostnames or domains Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 30/38] drm/amd/display: Allow bandwidth validation for 0 streams Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 31/38] drm/amdgpu: refine amdgpu_fru_get_product_info Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 32/38] drm/amd/display: Fix overlay validation by considering cursors Greg Kroah-Hartman
2021-06-16 16:02   ` Harry Wentland
2021-06-16 15:33 ` [PATCH 5.10 33/38] drm/amd/display: Fix potential memory leak in DMUB hw_init Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 34/38] drm/amd/amdgpu:save psp ring wptr to avoid attack Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 35/38] rtnetlink: Fix missing error code in rtnl_bridge_notify() Greg Kroah-Hartman
2021-06-18 23:58   ` Alexander Tsoy
2021-06-19  6:45     ` Greg Kroah-Hartman
2021-06-19 13:06       ` Alexander Tsoy
2021-06-16 15:33 ` [PATCH 5.10 36/38] net/x25: Return the correct errno code Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 37/38] net: " Greg Kroah-Hartman
2021-06-16 15:33 ` [PATCH 5.10 38/38] fib: " Greg Kroah-Hartman
2021-06-16 18:49 ` [PATCH 5.10 00/38] 5.10.45-rc1 review Jon Hunter
2021-06-16 21:07 ` Fox Chen
2021-06-16 22:40 ` Florian Fainelli
2021-06-17  6:51 ` Pavel Machek
2021-06-17  9:04 ` Naresh Kamboju
2021-06-17 10:27 ` Samuel Zou
2021-06-17 11:00 ` Sudip Mukherjee
2021-06-17 21:39 ` Guenter Roeck

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=20210616152835.912416286@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=agruenba@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rpeterso@redhat.com \
    --cc=sashal@kernel.org \
    --cc=stable@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.