linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: Dmitry Monakhov <dmonakhov@openvz.org>
Cc: linux-ext4@vger.kernel.org, tytso@mit.edu, jack@suse.cz,
	wenqing.lz@taobao.com
Subject: Re: [PATCH 5/7] ext4: serialize unlocked dio reads with truncate
Date: Mon, 10 Sep 2012 11:54:50 +0200	[thread overview]
Message-ID: <20120910095450.GG22903@quack.suse.cz> (raw)
In-Reply-To: <1347211634-11509-6-git-send-email-dmonakhov@openvz.org>

On Sun 09-09-12 21:27:12, Dmitry Monakhov wrote:
> Current serialization will works only for DIO which holds
> i_mutex, but nonlocked DIO following race is possible:
> 
> dio_nolock_read_task            truncate_task
> 				->ext4_setattr()
> 				 ->inode_dio_wait()
> ->ext4_ext_direct_IO
>   ->ext4_ind_direct_IO
>     ->__blockdev_direct_IO
>       ->ext4_get_block
> 				 ->truncate_setsize()
> 				 ->ext4_truncate()
> 				 #alloc truncated blocks
> 				 #to other inode
>       ->submit_io()
>      #INFORMATION LEAK
> 
> In order to serialize with unlocked DIO reads we have to
> rearrange wait sequence
> 1) update i_size first
> 2) if i_size about to be reduced wait for outstanding DIO requests
> 3) and only after that truncate inode blocks
  Looks good. You can add:
Reviewed-by: Jan Kara <jack@suse.cz>

								Honza
> 
> Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
> ---
>  fs/ext4/inode.c |    7 +++++--
>  1 files changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
> index ffb4a27..93e6b09 100644
> --- a/fs/ext4/inode.c
> +++ b/fs/ext4/inode.c
> @@ -4283,7 +4283,6 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
>  	}
>  
>  	if (attr->ia_valid & ATTR_SIZE) {
> -		inode_dio_wait(inode);
>  
>  		if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) {
>  			struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
> @@ -4332,8 +4331,12 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
>  	}
>  
>  	if (attr->ia_valid & ATTR_SIZE) {
> -		if (attr->ia_size != i_size_read(inode))
> +		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_dio_wait(inode);
> +		}
>  		ext4_truncate(inode);
>  	}
>  
> -- 
> 1.7.7.6
> 
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR

  reply	other threads:[~2012-09-10  9:54 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-09 17:27 [PATCH 0/7] ext4: Bunch of DIO/AIO fixes Dmitry Monakhov
2012-09-09 17:27 ` [PATCH 1/7] ext4: ext4_inode_info diet Dmitry Monakhov
2012-09-13 10:50   ` Zheng Liu
2012-09-13 11:15     ` Dmitry Monakhov
2012-09-15 15:53       ` Theodore Ts'o
2012-09-09 17:27 ` [PATCH 2/7] ext4: completed_io locking cleanup Dmitry Monakhov
2012-09-10  9:23   ` Jan Kara
2012-09-10 10:19     ` Dmitry Monakhov
2012-09-13 10:48   ` Zheng Liu
2012-09-09 17:27 ` [PATCH 3/7] ext4: serialize dio nolocked reads with defrag workers V2 Dmitry Monakhov
2012-09-10  9:31   ` Jan Kara
2012-09-10 10:00     ` Jan Kara
2012-09-09 17:27 ` [PATCH 4/7] ext4: fsync should wait for DIO writers Dmitry Monakhov
2012-09-10  9:51   ` Jan Kara
2012-09-10 10:56     ` Dmitry Monakhov
2012-09-12 14:02       ` Jan Kara
2012-09-12  5:40     ` Zheng Liu
2012-09-13 10:46   ` Zheng Liu
2012-09-13 11:01     ` Dmitry Monakhov
2012-09-13 12:36       ` Zheng Liu
2012-09-09 17:27 ` [PATCH 5/7] ext4: serialize unlocked dio reads with truncate Dmitry Monakhov
2012-09-10  9:54   ` Jan Kara [this message]
2012-09-09 17:27 ` [PATCH 6/7] ext4: endless truncate due to nonlocked dio readers V2 Dmitry Monakhov
2012-09-13 10:41   ` Zheng Liu
2012-09-13 12:07     ` Jan Kara
2012-09-13 12:57       ` Zheng Liu
2012-09-13 14:34         ` Jan Kara
2012-09-13 23:31           ` Zheng Liu
2012-09-09 17:27 ` [PATCH 7/7] ext4: serialize truncate with owerwrite DIO workers V2 Dmitry Monakhov
2012-09-13 10:37   ` Zheng Liu

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=20120910095450.GG22903@quack.suse.cz \
    --to=jack@suse.cz \
    --cc=dmonakhov@openvz.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=tytso@mit.edu \
    --cc=wenqing.lz@taobao.com \
    /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).