linux-kernel - Re: [RFC PATCH v2 0/3] fix overlayfs locks and leases

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] [day] [month] [year] [list]

Message-ID: <1469132373.2189.5.camel@poochiereds.net>
Date:	Thu, 21 Jul 2016 16:19:33 -0400
From:	Jeff Layton <jlayton@...chiereds.net>
To:	Miklos Szeredi <mszeredi@...hat.com>,
	"J. Bruce Fields" <bfields@...ldses.org>,
	Al Viro <viro@...iv.linux.org.uk>
Cc:	linux-unionfs@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-fsdevel@...r.kernel.org
Subject: Re: [RFC PATCH v2 0/3] fix overlayfs locks and leases

On Thu, 2016-07-21 at 15:53 +0200, Miklos Szeredi wrote:
> I've split out the writecount handling and changed it around so that
> underlying layers are consistent and yet leases work correctly on
> overlayfs.
> 
> Also pushed to the tip of
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs.git overlayfs-next
> 
> Thanks,
> Miklos
> ---
> 
> Miklos Szeredi (3):
>   locks: fix file locking on overlayfs
>   vfs: make argument of d_real_inode() const
>   vfs: do get_write_access() on upper layer of overlayfs
> 
>  fs/locks.c              | 53 ++++++++++++++++++++++++++++---------------------
>  fs/namespace.c          |  2 +-
>  fs/open.c               | 17 +++++++++++++---
>  fs/overlayfs/super.c    |  2 +-
>  include/linux/dcache.h  |  5 +++--
>  include/linux/fs.h      | 16 +++++++++++++--
>  include/uapi/linux/fs.h |  1 +
>  7 files changed, 64 insertions(+), 32 deletions(-)
> 

Looks pretty sane overall.

Also, when I mentioned accessing the writable layer in openwrt in the
last set, I forgot that you typically only do reads on it, so the
writecount wouldn't be affected in the case of accessing to do backups.
So, I'm not sure I had a legit objection to the earlier patch, but I
think this looks a little cleaner anyway:

Acked-by: Jeff Layton <jlayton@...chiereds.net>