From: Dmitry Monakhov <dmonakhov@openvz.org>
To: linux-ext4@vger.kernel.org
Cc: tytso@mit.edu, jack@suse.cz, lczerner@redhat.com,
Dmitry Monakhov <dmonakhov@openvz.org>
Subject: [PATCH 08/10] ext4: endless truncate due to nonlocked dio readers V2
Date: Mon, 24 Sep 2012 15:44:18 +0400 [thread overview]
Message-ID: <1348487060-19598-9-git-send-email-dmonakhov@openvz.org> (raw)
In-Reply-To: <1348487060-19598-1-git-send-email-dmonakhov@openvz.org>
If we have enough aggressive DIO readers, truncate and other dio
waiters will wait forever inside inode_dio_wait(). It is reasonable
to disable nonlock DIO read optimization during truncate.
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
---
fs/ext4/inode.c | 8 ++++++--
1 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 32e9701..d3f86e7 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -4330,9 +4330,13 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
if (attr->ia_valid & ATTR_SIZE) {
if (attr->ia_size != inode->i_size) {
truncate_setsize(inode, attr->ia_size);
- /* Inode size will be reduced, wait for dio in flight */
- if (orphan)
+ /* Inode size will be reduced, wait for dio in flight.
+ * Temproraly disable unlocked DIO to prevent livelock */
+ if (orphan) {
+ ext4_inode_block_unlocked_dio(inode);
inode_dio_wait(inode);
+ ext4_inode_resume_unlocked_dio(inode);
+ }
}
ext4_truncate(inode);
}
--
1.7.7.6
next prev parent reply other threads:[~2012-09-24 11:44 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-24 11:44 [PATCH 00/10] ext4: Bunch of DIO/AIO fixes V3 Dmitry Monakhov
2012-09-24 11:44 ` [PATCH 01/10] ext4: ext4_inode_info diet Dmitry Monakhov
2012-09-26 12:28 ` Jan Kara
2012-09-24 11:44 ` [PATCH 02/10] ext4: give i_aiodio_unwritten more appropriate name Dmitry Monakhov
2012-09-26 12:32 ` Jan Kara
2012-09-24 11:44 ` [PATCH 03/10] ext4: fix unwritten counter leakage Dmitry Monakhov
2012-09-26 13:07 ` Jan Kara
2012-09-27 12:19 ` Dmitry Monakhov
2012-09-27 12:34 ` Jan Kara
2012-09-27 12:54 ` Dmitry Monakhov
2012-09-27 13:07 ` Jan Kara
2012-09-24 11:44 ` [PATCH 04/10] ext4: completed_io locking cleanup V3 Dmitry Monakhov
2012-09-26 13:42 ` Jan Kara
2012-09-27 11:24 ` Dmitry Monakhov
2012-09-24 11:44 ` [PATCH 05/10] ext4: serialize dio nonlocked reads with defrag workers V3 Dmitry Monakhov
2012-09-26 13:49 ` Jan Kara
2012-09-24 11:44 ` [PATCH 06/10] ext4: punch_hole should wait for DIO writers V2 Dmitry Monakhov
2012-09-26 13:56 ` Jan Kara
2012-09-24 11:44 ` [PATCH 07/10] ext4: serialize unlocked dio reads with truncate Dmitry Monakhov
2012-09-24 11:44 ` Dmitry Monakhov [this message]
2012-09-26 14:05 ` [PATCH 08/10] ext4: endless truncate due to nonlocked dio readers V2 Jan Kara
2012-09-27 15:11 ` Dmitry Monakhov
2012-09-27 15:23 ` Jan Kara
2012-09-24 11:44 ` [PATCH 09/10] ext4: serialize truncate with owerwrite DIO workers V2 Dmitry Monakhov
2012-09-24 11:44 ` [PATCH 10/10] ext4: fix ext_remove_space for punch_hole case Dmitry Monakhov
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=1348487060-19598-9-git-send-email-dmonakhov@openvz.org \
--to=dmonakhov@openvz.org \
--cc=jack@suse.cz \
--cc=lczerner@redhat.com \
--cc=linux-ext4@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).