From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Eryu Guan <eguan@redhat.com>,
Brian Foster <bfoster@redhat.com>,
"Darrick J. Wong" <darrick.wong@oracle.com>
Subject: [PATCH 4.4 48/53] xfs: wait on new inodes during quotaoff dquot release
Date: Mon, 5 Jun 2017 18:17:46 +0200 [thread overview]
Message-ID: <20170605153040.237371154@linuxfoundation.org> (raw)
In-Reply-To: <20170605153037.105331684@linuxfoundation.org>
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brian Foster <bfoster@redhat.com>
commit e20c8a517f259cb4d258e10b0cd5d4b30d4167a0 upstream.
The quotaoff operation has a race with inode allocation that results
in a livelock. An inode allocation that occurs before the quota
status flags are updated acquires the appropriate dquots for the
inode via xfs_qm_vop_dqalloc(). It then inserts the XFS_INEW inode
into the perag radix tree, sometime later attaches the dquots to the
inode and finally clears the XFS_INEW flag. Quotaoff expects to
release the dquots from all inodes in the filesystem via
xfs_qm_dqrele_all_inodes(). This invokes the AG inode iterator,
which skips inodes in the XFS_INEW state because they are not fully
constructed. If the scan occurs after dquots have been attached to
an inode, but before XFS_INEW is cleared, the newly allocated inode
will continue to hold a reference to the applicable dquots. When
quotaoff invokes xfs_qm_dqpurge_all(), the reference count of those
dquot(s) remain elevated and the dqpurge scan spins indefinitely.
To address this problem, update the xfs_qm_dqrele_all_inodes() scan
to wait on inodes marked on the XFS_INEW state. We wait on the
inodes explicitly rather than skip and retry to avoid continuous
retry loops due to a parallel inode allocation workload. Since
quotaoff updates the quota state flags and uses a synchronous
transaction before the dqrele scan, and dquots are attached to
inodes after radix tree insertion iff quota is enabled, one INEW
waiting pass through the AG guarantees that the scan has processed
all inodes that could possibly hold dquot references.
Reported-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
fs/xfs/xfs_qm_syscalls.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -764,5 +764,6 @@ xfs_qm_dqrele_all_inodes(
uint flags)
{
ASSERT(mp->m_quotainfo);
- xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, NULL);
+ xfs_inode_ag_iterator_flags(mp, xfs_dqrele_inode, flags, NULL,
+ XFS_AGITER_INEW_WAIT);
}
next prev parent reply other threads:[~2017-06-05 16:21 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-05 16:16 [PATCH 4.4 00/53] 4.4.71-stable review Greg Kroah-Hartman
2017-06-05 16:16 ` [PATCH 4.4 01/53] sparc: Fix -Wstringop-overflow warning Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 02/53] dccp/tcp: do not inherit mc_list from parent Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 03/53] ipv6/dccp: do not inherit ipv6_mc_list " Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 04/53] s390/qeth: handle sysfs error during initialization Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 05/53] s390/qeth: unbreak OSM and OSN support Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 06/53] s390/qeth: avoid null pointer dereference on OSN Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 07/53] s390/qeth: add missing hash table initializations Greg Kroah-Hartman
2017-06-06 8:11 ` Julian Wiedmann
2017-06-06 8:52 ` Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 08/53] tcp: avoid fragmenting peculiar skbs in SACK Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 09/53] sctp: fix src address selection if using secondary addresses for ipv6 Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 10/53] sctp: do not inherit ipv6_{mc|ac|fl}_list from parent Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 11/53] tcp: eliminate negative reordering in tcp_clean_rtx_queue Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 12/53] net: Improve handling of failures on link and route dumps Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 13/53] ipv6: Prevent overrun when parsing v6 header options Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 14/53] ipv6: Check ip6_find_1stfragopt() return value properly Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 15/53] bridge: netlink: check vlan_default_pvid range Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 17/53] bridge: start hello_timer when enabling KERNEL_STP in br_stp_start Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 18/53] ipv6: fix out of bound writes in __ip6_append_data() Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 19/53] be2net: Fix offload features for Q-in-Q packets Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 20/53] virtio-net: enable TSO/checksum offloads for Q-in-Q vlans Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 21/53] tcp: avoid fastopen API to be used on AF_UNSPEC Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 22/53] sctp: fix ICMP processing if skb is non-linear Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 23/53] ipv4: add reference counting to metrics Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 24/53] netem: fix skb_orphan_partial() Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 25/53] net: phy: marvell: Limit errata to 88m1101 Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 26/53] vlan: Fix tcp checksum offloads in Q-in-Q vlans Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 27/53] i2c: i2c-tiny-usb: fix buffer not being DMA capable Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 28/53] mmc: sdhci-iproc: suppress spurious interrupt with Multiblock read Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 29/53] HID: wacom: Have wacom_tpc_irq guard against possible NULL dereference Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 30/53] scsi: mpt3sas: Force request partial completion alignment Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 33/53] pcmcia: remove left-over %Z format Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 34/53] ALSA: hda - apply STAC_9200_DELL_M22 quirk for Dell Latitude D430 Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 35/53] slub/memcg: cure the brainless abuse of sysfs attributes Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 36/53] drm/gma500/psb: Actually use VBT mode when it is found Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 37/53] mm/migrate: fix refcount handling when !hugepage_migration_supported() Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 38/53] mlock: fix mlock count can not decrease in race condition Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 39/53] PCI/PM: Add needs_resume flag to avoid suspend complete optimization Greg Kroah-Hartman
2017-06-06 16:19 ` Ben Hutchings
2017-06-07 9:49 ` Greg Kroah-Hartman
2017-06-08 9:58 ` Imre Deak
2017-06-05 16:17 ` [PATCH 4.4 40/53] xfs: Fix missed holes in SEEK_HOLE implementation Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 41/53] xfs: fix off-by-one on max nr_pages in xfs_find_get_desired_pgoff() Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 42/53] xfs: fix over-copying of getbmap parameters from userspace Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 43/53] xfs: handle array index overrun in xfs_dir2_leaf_readbuf() Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 44/53] xfs: prevent multi-fsb dir readahead from reading random blocks Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 45/53] xfs: fix up quotacheck buffer list error handling Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 46/53] xfs: support ability to wait on new inodes Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 47/53] xfs: update ag iterator to support " Greg Kroah-Hartman
2017-06-05 16:17 ` Greg Kroah-Hartman [this message]
2017-06-05 16:17 ` [PATCH 4.4 49/53] xfs: fix indlen accounting error on partial delalloc conversion Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 50/53] xfs: bad assertion for delalloc an extent that start at i_size Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 51/53] xfs: fix unaligned access in xfs_btree_visit_blocks Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 52/53] xfs: in _attrlist_by_handle, copy the cursor back to userspace Greg Kroah-Hartman
2017-06-05 16:17 ` [PATCH 4.4 53/53] xfs: only return -errno or success from attr ->put_listent Greg Kroah-Hartman
2017-06-05 20:35 ` [PATCH 4.4 00/53] 4.4.71-stable review Shuah Khan
2017-06-05 22:07 ` Guenter Roeck
2017-06-06 7:19 ` 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=20170605153040.237371154@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=bfoster@redhat.com \
--cc=darrick.wong@oracle.com \
--cc=eguan@redhat.com \
--cc=linux-kernel@vger.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 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).