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, Azat Khuzhin <a3at.mail@gmail.com>,
	Christoph Hellwig <hch@lst.de>, NeilBrown <neilb@suse.de>,
	Jens Axboe <axboe@fb.com>
Subject: [PATCH 4.0 23/60] block: destroy bdi before blockdev is unregistered.
Date: Fri, 15 May 2015 16:14:24 -0700	[thread overview]
Message-ID: <20150515231019.522351414@linuxfoundation.org> (raw)
In-Reply-To: <20150515231018.815421141@linuxfoundation.org>

4.0-stable review patch.  If anyone has any objections, please let me know.

------------------

From: NeilBrown <neilb@suse.de>

commit 6cd18e711dd8075da9d78cfc1239f912ff28968a upstream.

Because of the peculiar way that md devices are created (automatically
when the device node is opened), a new device can be created and
registered immediately after the
	blk_unregister_region(disk_devt(disk), disk->minors);
call in del_gendisk().

Therefore it is important that all visible artifacts of the previous
device are removed before this call.  In particular, the 'bdi'.

Since:
commit c4db59d31e39ea067c32163ac961e9c80198fd37
Author: Christoph Hellwig <hch@lst.de>
    fs: don't reassign dirty inodes to default_backing_dev_info

moved the
   device_unregister(bdi->dev);
call from bdi_unregister() to bdi_destroy() it has been quite easy to
lose a race and have a new (e.g.) "md127" be created after the
blk_unregister_region() call and before bdi_destroy() is ultimately
called by the final 'put_disk', which must come after del_gendisk().

The new device finds that the bdi name is already registered in sysfs
and complains

> [ 9627.630029] WARNING: CPU: 18 PID: 3330 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x5a/0x70()
> [ 9627.630032] sysfs: cannot create duplicate filename '/devices/virtual/bdi/9:127'

We can fix this by moving the bdi_destroy() call out of
blk_release_queue() (which can happen very late when a refcount
reaches zero) and into blk_cleanup_queue() - which happens exactly when the md
device driver calls it.

Then it is only necessary for md to call blk_cleanup_queue() before
del_gendisk().  As loop.c devices are also created on demand by
opening the device node, we make the same change there.

Fixes: c4db59d31e39ea067c32163ac961e9c80198fd37
Reported-by: Azat Khuzhin <a3at.mail@gmail.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NeilBrown <neilb@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 block/blk-core.c     |    2 ++
 block/blk-sysfs.c    |    2 --
 drivers/block/loop.c |    2 +-
 drivers/md/md.c      |    4 ++--
 4 files changed, 5 insertions(+), 5 deletions(-)

--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -552,6 +552,8 @@ void blk_cleanup_queue(struct request_qu
 		q->queue_lock = &q->__queue_lock;
 	spin_unlock_irq(lock);
 
+	bdi_destroy(&q->backing_dev_info);
+
 	/* @q is and will stay empty, shutdown and put */
 	blk_put_queue(q);
 }
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -522,8 +522,6 @@ static void blk_release_queue(struct kob
 
 	blk_trace_shutdown(q);
 
-	bdi_destroy(&q->backing_dev_info);
-
 	ida_simple_remove(&blk_queue_ida, q->id);
 	call_rcu(&q->rcu_head, blk_free_queue_rcu);
 }
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1672,8 +1672,8 @@ out:
 
 static void loop_remove(struct loop_device *lo)
 {
-	del_gendisk(lo->lo_disk);
 	blk_cleanup_queue(lo->lo_queue);
+	del_gendisk(lo->lo_disk);
 	blk_mq_free_tag_set(&lo->tag_set);
 	put_disk(lo->lo_disk);
 	kfree(lo);
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -4754,12 +4754,12 @@ static void md_free(struct kobject *ko)
 	if (mddev->sysfs_state)
 		sysfs_put(mddev->sysfs_state);
 
+	if (mddev->queue)
+		blk_cleanup_queue(mddev->queue);
 	if (mddev->gendisk) {
 		del_gendisk(mddev->gendisk);
 		put_disk(mddev->gendisk);
 	}
-	if (mddev->queue)
-		blk_cleanup_queue(mddev->queue);
 
 	kfree(mddev);
 }



  parent reply	other threads:[~2015-05-15 23:20 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-15 23:14 [PATCH 4.0 00/60] 4.0.4-stable review Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 01/60] x86/spinlocks: Fix regression in spinlock contention detection Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 02/60] ACPI / SBS: Add 5 us delay to fix SBS hangs on MacBook Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 03/60] x86/PCI/ACPI: Make all resources except [io 0xcf8-0xcff] available on PCI bus Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 04/60] ACPI / PNP: add two IDs to list for PNPACPI device enumeration Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 05/60] ocfs2: dlm: fix race between purge and get lock resource Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 06/60] rtc: armada38x: fix concurrency access in armada38x_rtc_set_time Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 07/60] nilfs2: fix sanity check of btree level in nilfs_btree_root_broken() Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 08/60] RDMA/CMA: Canonize IPv4 on IPV6 sockets properly Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 09/60] gpio: sysfs: fix memory leaks and device hotplug Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 10/60] mnt: Fix fs_fully_visible to verify the root directory is visible Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 11/60] mm/memory-failure: call shake_page() when error hits thp tail page Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 12/60] path_openat(): fix double fput() Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 13/60] writeback: use |1 instead of +1 to protect against div by zero Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 14/60] mm: soft-offline: fix num_poisoned_pages counting on concurrent events Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 15/60] [media] marvell-ccic: fix YCbCr ordering Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 16/60] vfio: Fix runaway interruptible timeout Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 17/60] xen/events: Clear cpu_evtchn_mask before resuming Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 18/60] xen/xenbus: Update xenbus event channel on resume Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 19/60] xen/console: Update console " Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 20/60] xen/events: Set irq_info->evtchn before binding the channel to CPU in __startup_pirq() Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 21/60] xen-pciback: Add name prefix to global permissive variable Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 22/60] Revert "dm crypt: fix deadlock when async crypto algorithm returns -EBUSY" Greg Kroah-Hartman
2015-05-15 23:14 ` Greg Kroah-Hartman [this message]
2015-05-15 23:14 ` [PATCH 4.0 24/60] blk-mq: fix race between timeout and CPU hotplug Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 25/60] blk-mq: fix CPU hotplug handling Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 26/60] ARM: dts: OMAP3-N900: Add microphone bias voltages Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 27/60] ARM: dts: imx25: Add #pwm-cells to pwm4 Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 28/60] ARM: dts: imx6: phyFLEX: USB VBUS control is active-high Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 29/60] ARM: dts: imx28: Fix AUART4 TX-DMA interrupt name Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 30/60] ARM: dts: imx23-olinuxino: Fix dr_mode of usb0 Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 31/60] ARM: dts: imx23-olinuxino: Fix polarity of LED GPIO Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 32/60] ARM: mvebu: armada-xp-openblocks-ax3-4: Disable internal RTC Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 33/60] ARM: OMAP2+: Fix omap off idle power consumption creeping up Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 34/60] ARM: net fix emit_udiv() for BPF_ALU | BPF_DIV | BPF_K intruction Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 35/60] ARM: ux500: Move GPIO regulator for SD-card into board DTSs Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 36/60] ARM: ux500: Enable GPIO regulator for SD-card for HREF boards Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 37/60] ARM: ux500: Enable GPIO regulator for SD-card for snowball Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 40/60] drm/i915: Add missing MacBook Pro models with dual channel LVDS Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 41/60] drm/i915/dp: there is no audio on port A Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 42/60] drm/amdkfd: allow unregister process with queues Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 43/60] drm/amdkfd: Initialize sdma vm when creating sdma queue Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 45/60] drm/radeon: dont setup audio on asics that dont support it Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 50/60] pinctrl: Dont just pretend to protect pinctrl_maps, do it for real Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 51/60] mmc: card: Dont access RPMB partitions for normal read/write Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 52/60] mmc: core: add missing pm event in mmc_pm_notify to fix hib restore Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 53/60] mmc: sh_mmcif: Fix timeout value for command request Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 54/60] ARM: 8307/1: psci: move psci firmware calls out of line Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 55/60] sound/oss: fix deadlock in sequencer_ioctl(SNDCTL_SEQ_OUTOFBAND) Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 56/60] coredump: accept any write method Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 57/60] ACPICA: Tables: Change acpi_find_root_pointer() to use acpi_physical_address Greg Kroah-Hartman
2015-05-15 23:14 ` [PATCH 4.0 58/60] ACPICA: Utilities: Cleanup to enforce ACPI_PHYSADDR_TO_PTR()/ACPI_PTR_TO_PHYSADDR() Greg Kroah-Hartman
2015-05-15 23:15 ` [PATCH 4.0 59/60] ACPICA: Utilities: Cleanup to convert physical address printing formats Greg Kroah-Hartman
2015-05-15 23:15 ` [PATCH 4.0 60/60] ACPICA: Utilities: Cleanup to remove useless ACPI_PRINTF/FORMAT_xxx helpers Greg Kroah-Hartman
2015-05-16  3:16 ` [PATCH 4.0 00/60] 4.0.4-stable review Guenter Roeck
2015-05-16 15:26   ` Greg Kroah-Hartman
2015-05-16  3:16 ` Shuah Khan
2015-05-16 15:27   ` Greg Kroah-Hartman

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=20150515231019.522351414@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=a3at.mail@gmail.com \
    --cc=axboe@fb.com \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=neilb@suse.de \
    --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.