Patchwork [07/11] ext4: serialize unlocked dio reads with truncate

login
register
mail settings
Submitter Dmitri Monakho
Date Sept. 28, 2012, 3:44 p.m.
Message ID <1348847051-6746-8-git-send-email-dmonakhov@openvz.org>
Download mbox | patch
Permalink /patch/187815/
State Accepted
Headers show

Comments

Dmitri Monakho - Sept. 28, 2012, 3:44 p.m.
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

Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
---
 fs/ext4/inode.c |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)
Theodore Ts'o - Sept. 29, 2012, 4:49 a.m.
On Fri, Sep 28, 2012 at 07:44:07PM +0400, Dmitry Monakhov wrote:
>  	if (attr->ia_valid & ATTR_SIZE) {
> -		if (attr->ia_size != i_size_read(inode))
> +		if (attr->ia_size != inode->i_size) {

Why this change?   Why are you not using i_size_read() here?

    	 	       	       	   - Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Dmitri Monakho - Sept. 29, 2012, 11:43 a.m.
On Sat, 29 Sep 2012 00:49:04 -0400, "Theodore Ts'o" <tytso@mit.edu> wrote:
> On Fri, Sep 28, 2012 at 07:44:07PM +0400, Dmitry Monakhov wrote:
> >  	if (attr->ia_valid & ATTR_SIZE) {
> > -		if (attr->ia_size != i_size_read(inode))
> > +		if (attr->ia_size != inode->i_size) {
> 
> Why this change?   Why are you not using i_size_read() here?
Since we hold i_mutex no one is able to change i_size, and it
is safe to access it directly.
> 
>     	 	       	       	   - Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index da52855..583cb3f 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -4255,7 +4255,6 @@  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);
@@ -4304,8 +4303,12 @@  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);
 	}