From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org, Vyacheslav Dubeyko <slava@dubeyko.com>,
Sergey Alexandrov <splavgm@gmail.com>,
Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: [ 23/66] nilfs2: fix fix very long mount time issue
Date: Sun, 17 Feb 2013 22:50:24 +0000 [thread overview]
Message-ID: <20130217225004.343966423@decadent.org.uk> (raw)
In-Reply-To: <20130217225001.621306883@decadent.org.uk>
3.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vyacheslav Dubeyko <slava@dubeyko.com>
commit a9bae189542e71f91e61a4428adf6e5a7dfe8063 upstream.
There exists a situation when GC can work in background alone without
any other filesystem activity during significant time.
The nilfs_clean_segments() method calls nilfs_segctor_construct() that
updates superblocks in the case of NILFS_SC_SUPER_ROOT and
THE_NILFS_DISCONTINUED flags are set. But when GC is working alone the
nilfs_clean_segments() is called with unset THE_NILFS_DISCONTINUED flag.
As a result, the update of superblocks doesn't occurred all this time
and in the case of SPOR superblocks keep very old values of last super
root placement.
SYMPTOMS:
Trying to mount a NILFS2 volume after SPOR in such environment ends with
very long mounting time (it can achieve about several hours in some
cases).
REPRODUCING PATH:
1. It needs to use external USB HDD, disable automount and doesn't
make any additional filesystem activity on the NILFS2 volume.
2. Generate temporary file with size about 100 - 500 GB (for example,
dd if=/dev/zero of=<file_name> bs=1073741824 count=200). The size of
file defines duration of GC working.
3. Then it needs to delete file.
4. Start GC manually by means of command "nilfs-clean -p 0". When you
start GC by means of such way then, at the end, superblocks is updated
by once. So, for simulation of SPOR, it needs to wait sometime (15 -
40 minutes) and simply switch off USB HDD manually.
5. Switch on USB HDD again and try to mount NILFS2 volume. As a
result, NILFS2 volume will mount during very long time.
REPRODUCIBILITY: 100%
FIX:
This patch adds checking that superblocks need to update and set
THE_NILFS_DISCONTINUED flag before nilfs_clean_segments() call.
Reported-by: Sergey Alexandrov <splavgm@gmail.com>
Signed-off-by: Vyacheslav Dubeyko <slava@dubeyko.com>
Tested-by: Vyacheslav Dubeyko <slava@dubeyko.com>
Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Tested-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
fs/nilfs2/ioctl.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index fdb1807..f385935 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
@@ -664,8 +664,11 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp,
if (ret < 0)
printk(KERN_ERR "NILFS: GC failed during preparation: "
"cannot read source blocks: err=%d\n", ret);
- else
+ else {
+ if (nilfs_sb_need_update(nilfs))
+ set_nilfs_discontinued(nilfs);
ret = nilfs_clean_segments(inode->i_sb, argv, kbufs);
+ }
nilfs_remove_all_gcinodes(nilfs);
clear_nilfs_gc_running(nilfs);
next prev parent reply other threads:[~2013-02-17 22:56 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-17 22:50 [ 00/66] 3.2.39-stable review Ben Hutchings
2013-02-17 22:50 ` [ 01/66] drm/i915: Fix misplaced intel_mark_page_flip_active() Ben Hutchings
2013-02-17 22:50 ` [ 02/66] xhci: Fix isoc TD encoding Ben Hutchings
2013-02-17 22:50 ` [ 03/66] xhci: Fix TD size for isochronous URBs Ben Hutchings
2013-02-17 22:50 ` [ 04/66] USB: XHCI: fix memory leak of URB-private data Ben Hutchings
2013-02-17 22:50 ` [ 05/66] usb: Prevent dead ports when xhci is not enabled Ben Hutchings
2013-02-17 22:50 ` [ 06/66] sched/rt: Use root_domain of rt_rq not current processor Ben Hutchings
2013-02-17 22:50 ` [ 07/66] rtlwifi: Fix the usage of the wrong variable in usb.c Ben Hutchings
2013-02-17 22:50 ` [ 08/66] drm/radeon: add quirk for RV100 board Ben Hutchings
2013-02-17 22:50 ` [ 09/66] USB: qcserial: add Telit Gobi QDL device Ben Hutchings
2013-02-17 22:50 ` [ 10/66] USB: option: add support for Telit LE920 Ben Hutchings
2013-02-17 22:50 ` [ 11/66] mwifiex: fix incomplete scan in case of IE parsing error Ben Hutchings
2013-02-17 22:50 ` [ 12/66] USB: EHCI: fix bug in scheduling periodic split transfers Ben Hutchings
2013-02-17 22:50 ` [ 13/66] x86-64: Replace left over sti/cli in ia32 audit exit code Ben Hutchings
2013-02-17 22:50 ` [ 14/66] Bluetooth: Fix handling of unexpected SMP PDUs Ben Hutchings
2013-02-17 22:50 ` [ 15/66] drm/radeon: Calling object_unrefer() when creating fb failure Ben Hutchings
2013-02-17 22:50 ` [ 16/66] drm/radeon: prevent crash in the ring space allocation Ben Hutchings
2013-02-17 22:50 ` [ 17/66] can: c_can: Set reserved bit in IFx_MASK2 to 1 on write Ben Hutchings
2013-02-17 22:50 ` [ 18/66] USB: ftdi_sio: add PID/VID entries for ELV WS 300 PC II Ben Hutchings
2013-02-17 22:50 ` [ 19/66] USB: option: add Changhong CH690 Ben Hutchings
2013-02-17 22:50 ` [ 20/66] USB: ftdi_sio: add Zolix FTDI PID Ben Hutchings
2013-02-17 22:50 ` [ 21/66] USB: storage: Define a new macro for USB storage match rules Ben Hutchings
2013-02-17 22:50 ` [ 22/66] USB: storage: optimize to match the Huawei USB storage devices and support new switch command Ben Hutchings
2013-02-17 22:50 ` Ben Hutchings [this message]
2013-02-17 22:50 ` [ 24/66] drivers/rtc/rtc-isl1208.c: call rtc_update_irq() from the alarm irq handler Ben Hutchings
2013-02-17 22:50 ` [ 25/66] drivers/rtc/rtc-pl031.c: fix the missing operation on enable Ben Hutchings
2013-02-17 22:50 ` [ 26/66] virtio_console: Dont access uninitialized data Ben Hutchings
2013-02-17 22:50 ` [ 27/66] wireless: rt2x00: rt{2500,73}usb.c put back duplicate id Ben Hutchings
2013-02-17 22:50 ` [ 28/66] Wireless: rt2x00: Add device id for Sweex LW323 to rt2800usb.c Ben Hutchings
2013-02-17 22:50 ` [ 29/66] rt2800usb: Add support for 2001:3c1e (D-Link DWA-125 rev B1) USB Wi-Fi adapter Ben Hutchings
2013-02-17 22:50 ` [ 30/66] [media] gspca_kinect: add Kinect for Windows USB id Ben Hutchings
2013-02-17 22:50 ` [ 31/66] ahci: support the STA2X11 I/O Hub Ben Hutchings
2013-02-17 22:50 ` [ 32/66] [libata] ahci: Add support for Enmotus Bobcat device Ben Hutchings
2013-02-17 22:50 ` [ 33/66] ptrace/x86: Introduce set_task_blockstep() helper Ben Hutchings
2013-02-17 22:50 ` [ 34/66] ptrace/x86: Partly fix set_task_blockstep()->update_debugctlmsr() logic Ben Hutchings
2013-02-17 22:50 ` [ 35/66] ptrace: introduce signal_wake_up_state() and ptrace_signal_wake_up() Ben Hutchings
2013-02-17 22:50 ` [ 36/66] ptrace: ensure arch_ptrace/ptrace_request can never race with SIGKILL Ben Hutchings
2013-02-17 22:50 ` [ 37/66] wake_up_process() should be never used to wakeup a TASK_STOPPED/TRACED task Ben Hutchings
2013-02-17 22:50 ` [ 38/66] Bluetooth: Fix sending HCI commands after reset Ben Hutchings
2013-02-17 22:50 ` [ 39/66] HID: usbhid: quirk for Formosa IR receiver Ben Hutchings
2013-02-17 22:50 ` [ 40/66] kernel/resource.c: fix stack overflow in __reserve_region_with_split() Ben Hutchings
2013-02-17 22:50 ` [ 41/66] net: prevent setting ttl=0 via IP_TTL Ben Hutchings
2013-02-17 22:50 ` [ 42/66] ipv6: fix the noflags test in addrconf_get_prefix_route Ben Hutchings
2013-02-17 22:50 ` [ 43/66] MAINTAINERS: Stephen Hemminger email change Ben Hutchings
2013-02-17 22:50 ` [ 44/66] ipv6: fix header length calculation in ip6_append_data() Ben Hutchings
2013-02-17 22:50 ` [ 45/66] isdn/gigaset: fix zero size border case in debug dump Ben Hutchings
2013-02-17 22:50 ` [ 46/66] netxen: fix off by one bug in netxen_release_tx_buffer() Ben Hutchings
2013-02-17 22:50 ` [ 47/66] r8169: remove the obsolete and incorrect AMD workaround Ben Hutchings
2013-02-17 22:50 ` [ 48/66] net: loopback: fix a dst refcounting issue Ben Hutchings
2013-02-17 22:50 ` [ 49/66] pktgen: correctly handle failures when adding a device Ben Hutchings
2013-02-17 22:50 ` [ 50/66] ipv6: do not create neighbor entries for local delivery Ben Hutchings
2013-02-17 22:50 ` [ 51/66] packet: fix leakage of tx_ring memory Ben Hutchings
2013-02-17 22:50 ` [ 52/66] atm/iphase: rename fregt_t -> ffreg_t Ben Hutchings
2013-02-17 22:50 ` [ 53/66] sctp: refactor sctp_outq_teardown to insure proper re-initalization Ben Hutchings
2013-02-17 22:50 ` [ 54/66] net: sctp: sctp_setsockopt_auth_key: use kzfree instead of kfree Ben Hutchings
2013-02-17 22:50 ` [ 55/66] net: sctp: sctp_endpoint_free: zero out secret key data Ben Hutchings
2013-02-17 22:50 ` [ 56/66] xen/netback: shutdown the ring if it contains garbage Ben Hutchings
2013-02-17 22:50 ` [ 57/66] xen/netback: dont leak pages on failure in xen_netbk_tx_check_gop Ben Hutchings
2013-02-17 22:50 ` [ 58/66] xen/netback: free already allocated memory on failure in xen_netbk_get_requests Ben Hutchings
2013-02-17 22:51 ` [ 59/66] netback: correct netbk_tx_err to handle wrap around Ben Hutchings
2013-02-17 22:51 ` [ 60/66] tcp: frto should not set snd_cwnd to 0 Ben Hutchings
2013-02-17 22:51 ` [ 61/66] tcp: fix for zero packets_in_flight was too broad Ben Hutchings
2013-02-17 22:51 ` [ 62/66] tcp: fix MSG_SENDPAGE_NOTLAST logic Ben Hutchings
2013-02-17 22:51 ` [ 63/66] bridge: Pull ip header into skb->data before looking into ip header Ben Hutchings
2013-02-17 22:51 ` [ 64/66] tg3: Avoid null pointer dereference in tg3_interrupt in netconsole mode Ben Hutchings
2013-02-17 22:51 ` [ 65/66] tg3: Fix crc errors on jumbo frame receive Ben Hutchings
2013-02-17 22:51 ` [ 66/66] x86/xen: dont assume %ds is usable in xen_iret for 32-bit PVOPS Ben Hutchings
2013-02-18 0:02 ` [ 00/66] 3.2.39-stable review Ben Hutchings
2013-02-19 22:25 ` Satoru Takeuchi
2013-02-20 2:59 ` Ben Hutchings
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=20130217225004.343966423@decadent.org.uk \
--to=ben@decadent.org.uk \
--cc=akpm@linux-foundation.org \
--cc=konishi.ryusuke@lab.ntt.co.jp \
--cc=linux-kernel@vger.kernel.org \
--cc=slava@dubeyko.com \
--cc=splavgm@gmail.com \
--cc=stable@vger.kernel.org \
--cc=torvalds@linux-foundation.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