Linux-ext4 Archive on lore.kernel.org
 help / color / Atom feed
From: Dmitry Monakhov <dmonakhov@openvz.org>
To: linux-fsdevel@vger.kernel.org
Cc: linux-ext4@vger.kernel.org, jack@suse.cz,
	Dmitry Monakhov <dmtrmonakhov@yandex-team.ru>
Subject: [PATCH 2/2] fs/quota: Check that quota is not dirty before release
Date: Thu, 31 Oct 2019 10:39:20 +0000
Message-ID: <20191031103920.3919-2-dmonakhov@openvz.org> (raw)
In-Reply-To: <20191031103920.3919-1-dmonakhov@openvz.org>

From: Dmitry Monakhov <dmtrmonakhov@yandex-team.ru>

There is a race window where quota was redirted once we drop dq_list_lock inside dqput(),
but before we grab dquot->dq_lock inside dquot_release()

TASK1                                                       TASK2 (chowner)
->dqput()
  we_slept:
    spin_lock(&dq_list_lock)
    if (dquot_dirty(dquot)) {
          spin_unlock(&dq_list_lock);
          dquot->dq_sb->dq_op->write_dquot(dquot);
          goto we_slept
    if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) {
          spin_unlock(&dq_list_lock);
          dquot->dq_sb->dq_op->release_dquot(dquot);
                                                            dqget()
							    mark_dquot_dirty()
							    dqput()
          goto we_slept;
        }
So dquot dirty quota will be released by TASK1, but on next we_sleept loop
we detect this and call ->write_dquot() for it.
XFSTEST: https://github.com/dmonakhov/xfstests/commit/440a80d4cbb39e9234df4d7240aee1d551c36107

Signed-off-by: Dmitry Monakhov <dmtrmonakhov@yandex-team.ru>
---
 fs/ocfs2/quota_global.c  |  2 +-
 fs/quota/dquot.c         |  2 +-
 include/linux/quotaops.h | 10 ++++++++++
 3 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index 7a92219..eda8348 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -728,7 +728,7 @@ static int ocfs2_release_dquot(struct dquot *dquot)
 
 	mutex_lock(&dquot->dq_lock);
 	/* Check whether we are not racing with some other dqget() */
-	if (atomic_read(&dquot->dq_count) > 1)
+	if (dquot_is_busy(dquot))
 		goto out;
 	/* Running from downconvert thread? Postpone quota processing to wq */
 	if (current == osb->dc_task) {
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index b492b9e..72d24a5 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -497,7 +497,7 @@ int dquot_release(struct dquot *dquot)
 
 	mutex_lock(&dquot->dq_lock);
 	/* Check whether we are not racing with some other dqget() */
-	if (atomic_read(&dquot->dq_count) > 1)
+	if (dquot_is_busy(dquot))
 		goto out_dqlock;
 	if (dqopt->ops[dquot->dq_id.type]->release_dqblk) {
 		ret = dqopt->ops[dquot->dq_id.type]->release_dqblk(dquot);
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index 185d948..91e0b76 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -54,6 +54,16 @@ static inline struct dquot *dqgrab(struct dquot *dquot)
 	atomic_inc(&dquot->dq_count);
 	return dquot;
 }
+
+static inline bool dquot_is_busy(struct dquot *dquot)
+{
+	if (test_bit(DQ_MOD_B, &dquot->dq_flags))
+		return true;
+	if (atomic_read(&dquot->dq_count) > 1)
+		return true;
+	return false;
+}
+
 void dqput(struct dquot *dquot);
 int dquot_scan_active(struct super_block *sb,
 		      int (*fn)(struct dquot *dquot, unsigned long priv),
-- 
2.7.4


  reply index

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-31 10:39 [PATCH 1/2] fs/quota: fix livelock in dquot_writeback_dquots Dmitry Monakhov
2019-10-31 10:39 ` Dmitry Monakhov [this message]
2019-10-31 18:00   ` [PATCH 2/2] fs/quota: Check that quota is not dirty before release Jan Kara
2019-10-31 18:11     ` Jan Kara

Reply instructions:

You may reply publically 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=20191031103920.3919-2-dmonakhov@openvz.org \
    --to=dmonakhov@openvz.org \
    --cc=dmtrmonakhov@yandex-team.ru \
    --cc=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@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

Linux-ext4 Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-ext4/0 linux-ext4/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-ext4 linux-ext4/ https://lore.kernel.org/linux-ext4 \
		linux-ext4@vger.kernel.org
	public-inbox-index linux-ext4

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-ext4


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git