linux-ext4 - Re: [bug report] ext4, project: expand inode extra size if possible

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20191213113237.GF15474@quack2.suse.cz>
Date:   Fri, 13 Dec 2019 12:32:37 +0100
From:   Jan Kara <jack@...e.cz>
To:     Dan Carpenter <dan.carpenter@...cle.com>
Cc:     miaoxie@...wei.com, linux-ext4@...r.kernel.org
Subject: Re: [bug report] ext4, project: expand inode extra size if possible

On Mon 02-12-19 12:41:04, Dan Carpenter wrote:
> Hello Miao Xie,
> 
> The patch c03b45b853f5: "ext4, project: expand inode extra size if
> possible" from Aug 6, 2017, leads to the following static checker
> warning:
> 
>     fs/ext4/inode.c:5708 ext4_expand_extra_isize()
>     warn: inconsistent returns 'EXT4_I(inode)->xattr_sem'.
>       Locked on  : 5708
>       Unlocked on: 5708
> 
> fs/ext4/inode.c
>   5681          handle = ext4_journal_start(inode, EXT4_HT_INODE,
>   5682                                      EXT4_DATA_TRANS_BLOCKS(inode->i_sb));
>   5683          if (IS_ERR(handle)) {
>   5684                  error = PTR_ERR(handle);
>   5685                  brelse(iloc->bh);
>   5686                  return error;
>   5687          }
>   5688  
>   5689          ext4_write_lock_xattr(inode, &no_expand);
>   5690  
>   5691          BUFFER_TRACE(iloc->bh, "get_write_access");
>   5692          error = ext4_journal_get_write_access(handle, iloc->bh);
>   5693          if (error) {
>   5694                  brelse(iloc->bh);
>   5695                  goto out_stop;
> 
> Shouldn't this goto the ext4_write_unlock_xattr()?

Yes, it should AFAICT. Thanks for spotting this. Care to send a patch?

								Honza

> 
>   5696          }
>   5697  
>   5698          error = __ext4_expand_extra_isize(inode, new_extra_isize, iloc,
>   5699                                            handle, &no_expand);
>   5700  
>   5701          rc = ext4_mark_iloc_dirty(handle, inode, iloc);
>   5702          if (!error)
>   5703                  error = rc;
>   5704  
>   5705          ext4_write_unlock_xattr(inode, &no_expand);
>   5706  out_stop:
>   5707          ext4_journal_stop(handle);
>   5708          return error;
>   5709  }
> 
> 
> regards,
> dan carpenter
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR