mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
	akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
	"Theodore Tso" <tytso@mit.edu>, Jan Kara <jack@suse.cz>
Subject: [13/91] jbd: fix fsync() tid wraparound bug
Date: Wed, 15 Jun 2011 17:15:10 -0700	[thread overview]
Message-ID: <20110616001610.589473656@clark.kroah.org> (raw)
In-Reply-To: <20110616001900.GA25375@kroah.com>

2.6.32-longterm review patch.  If anyone has any objections, please let us know.

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

From: Ted Ts'o <tytso@mit.edu>

commit d9b01934d56a96d9f4ae2d6204d4ea78a36f5f36 upstream.

If an application program does not make any changes to the indirect
blocks or extent tree, i_datasync_tid will not get updated.  If there
are enough commits (i.e., 2**31) such that tid_geq()'s calculations
wrap, and there isn't a currently active transaction at the time of
the fdatasync() call, this can end up triggering a BUG_ON in
fs/jbd/commit.c:

	J_ASSERT(journal->j_running_transaction != NULL);

It's pretty rare that this can happen, since it requires the use of
fdatasync() plus *very* frequent and excessive use of fsync().  But
with the right workload, it can.

We fix this by replacing the use of tid_geq() with an equality test,
since there's only one valid transaction id that is valid for us to
start: namely, the currently running transaction (if it exists).

Reported-by: Martin_Zielinski@McAfee.com
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/jbd/journal.c |   16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -435,9 +435,12 @@ int __log_space_left(journal_t *journal)
 int __log_start_commit(journal_t *journal, tid_t target)
 {
 	/*
-	 * Are we already doing a recent enough commit?
+	 * The only transaction we can possibly wait upon is the
+	 * currently running transaction (if it exists).  Otherwise,
+	 * the target tid must be an old one.
 	 */
-	if (!tid_geq(journal->j_commit_request, target)) {
+	if (journal->j_running_transaction &&
+	    journal->j_running_transaction->t_tid == target) {
 		/*
 		 * We want a new commit: OK, mark the request and wakup the
 		 * commit thread.  We do _not_ do the commit ourselves.
@@ -449,7 +452,14 @@ int __log_start_commit(journal_t *journa
 			  journal->j_commit_sequence);
 		wake_up(&journal->j_wait_commit);
 		return 1;
-	}
+	} else if (!tid_geq(journal->j_commit_request, target))
+		/* This should never happen, but if it does, preserve
+		   the evidence before kjournald goes into a loop and
+		   increments j_commit_sequence beyond all recognition. */
+		WARN_ONCE(1, "jbd: bad log_start_commit: %u %u %u %u\n",
+		    journal->j_commit_request, journal->j_commit_sequence,
+		    target, journal->j_running_transaction ?
+		    journal->j_running_transaction->t_tid : 0);
 	return 0;
 }
 



  parent reply	other threads:[~2011-06-16  7:28 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-16  0:19 [00/91] 2.6.32.42-longterm review Greg KH
2011-06-16  0:14 ` [01/91] ftrace: Only update the function code on write to filter files Greg KH
2011-06-16  0:14 ` [02/91] kmemleak: Do not return a pointer to an object that kmemleak did not get Greg KH
2011-06-16  0:15 ` [03/91] [CPUFREQ] CPU hotplug, re-create sysfs directory and symlinks Greg KH
2011-06-16  0:15 ` [04/91] [CPUFREQ] Fix memory leak in cpufreq_stat Greg KH
2011-06-16  0:15 ` [05/91] powerpc/kexec: Fix memory corruption from unallocated slaves Greg KH
2011-06-16  0:15 ` [06/91] powerpc/oprofile: Handle events that raise an exception without overflowing Greg KH
2011-06-16  0:15 ` [07/91] block: rescan partitions on invalidated devices on -ENOMEDIA Greg KH
2011-06-16  0:15 ` [08/91] block: add proper state guards to __elv_next_request Greg KH
2011-06-16  0:15 ` [09/91] mtd: mtdconcat: fix NAND OOB write Greg KH
2011-06-16  0:15 ` [10/91] x86, 64-bit: Fix copy_[to/from]_user() checks for the Greg KH
2011-06-16  0:15 ` [11/91] ext3: Fix fs corruption when make_indexed_dir() fails Greg KH
2011-06-16  0:15 ` [12/91] jbd: Fix forever sleeping process in do_get_write_access() Greg KH
2011-06-16  0:15 ` Greg KH [this message]
2011-06-16  0:15 ` [14/91] ext4: release page cache in ext4_mb_load_buddy error path Greg KH
2011-06-16  0:15 ` [15/91] [SCSI] Fix Ultrastor asm snippet Greg KH
2011-06-16  0:15 ` [16/91] x86, amd: Do not enable ARAT feature on AMD processors below Greg KH
2011-06-16  0:15 ` [17/91] x86, amd: Use _safe() msr access for GartTlbWlk disable code Greg KH
2011-06-16  0:15 ` [18/91] rcu: Fix unpaired rcu_irq_enter() from locking selftests Greg KH
2011-06-16  0:15 ` [19/91] staging: usbip: fix wrong endian conversion Greg KH
2011-06-16  0:15 ` [20/91] Fix for buffer overflow in ldm_frag_add not sufficient Greg KH
2011-06-16  0:15 ` [21/91] seqlock: Dont smp_rmb in seqlock reader spin loop Greg KH
2011-06-16  0:15 ` [22/91] ALSA: HDA: Use one dmic only for Dell Studio 1558 Greg KH
2011-06-16  0:15 ` [23/91] ASoC: Ensure output PGA is enabled for line outputs in Greg KH
2011-06-16  0:15 ` [24/91] ASoC: Add some missing volume update bit sets for wm_hubs Greg KH
2011-06-16  0:15 ` [25/91] mm/page_alloc.c: prevent unending loop in Greg KH
2011-06-16  0:15 ` [26/91] PCI: allow matching of prefetchable resources to Greg KH
2011-06-16  0:15 ` [27/91] loop: limit max_part module param to DISK_MAX_PARTS Greg KH
2011-06-16  0:15 ` [28/91] loop: handle on-demand devices correctly Greg KH
2011-06-16  0:15 ` [29/91] USB: CP210x Add 4 Device IDs for AC-Services Devices Greg KH
2011-06-16  0:15 ` [30/91] USB: moto_modem: Add USB identifier for the Motorola VE240 Greg KH
2011-06-16  0:15 ` [31/91] USB: serial: ftdi_sio: adding support for TavIR STK500 Greg KH
2011-06-16  0:15 ` [32/91] USB: gamin_gps: Fix for data transfer problems in native Greg KH
2011-06-16  0:15 ` [33/91] usb/gadget: at91sam9g20 fix end point max packet size Greg KH
2011-06-16  0:15 ` [34/91] usb: gadget: rndis: dont test against req->length Greg KH
2011-06-16  0:15 ` [35/91] xhci: Fix full speed bInterval encoding Greg KH
2011-06-16  0:15 ` [36/91] OHCI: fix regression caused by nVidia shutdown workaround Greg KH
2011-06-16  0:15 ` [37/91] p54usb: add zoom 4410 usbid Greg KH
2011-06-16  0:15 ` [38/91] eCryptfs: Allow 2 scatterlist entries for encrypted Greg KH
2011-06-16  0:15 ` [39/91] UBIFS: fix a rare memory leak in ro to rw remounting path Greg KH
2011-06-16  0:15 ` [40/91] i8k: Avoid lahf in 64-bit code Greg KH
2011-06-16  0:15 ` [53/91] UBIFS: fix shrinker object count reports Greg KH
2011-06-16  0:15 ` [54/91] UBIFS: fix memory leak on error path Greg KH
2011-06-16  0:15 ` [55/91] nbd: limit module parameters to a sane value Greg KH
2011-06-17 20:36 ` [Stable-review] [00/91] 2.6.32.42-longterm review Nikola Ciprich
2011-06-20 22:28   ` Greg KH

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=20110616001610.589473656@clark.kroah.org \
    --to=gregkh@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=jack@suse.cz \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable-review@kernel.org \
    --cc=stable@kernel.org \
    --cc=torvalds@linux-foundation.org \
    --cc=tytso@mit.edu \
    /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

all inboxes | Powered by JetHome®