linux-kernel - Re: [ 117/135] NFS: return error from decode

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]

Date:	Thu, 20 Sep 2012 02:21:34 +0100
From:	Ben Hutchings <ben@...adent.org.uk>
To:	"Myklebust, Trond" <Trond.Myklebust@...app.com>
Cc:	Herton Ronaldo Krzesinski <herton.krzesinski@...onical.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"stable@...r.kernel.org" <stable@...r.kernel.org>,
	"torvalds@...ux-foundation.org" <torvalds@...ux-foundation.org>,
	"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
	"alan@...rguk.ukuu.org.uk" <alan@...rguk.ukuu.org.uk>,
	"Adamson, Dros" <Weston.Adamson@...app.com>
Subject: Re: [ 117/135] NFS: return error from decode_getfh in decode open

On Wed, 2012-09-19 at 15:53 +0000, Myklebust, Trond wrote:
> On Wed, 2012-09-19 at 04:39 +0100, Ben Hutchings wrote:
> > On Tue, 2012-09-18 at 22:26 -0300, Herton Ronaldo Krzesinski wrote:
> > > On Mon, Sep 17, 2012 at 01:38:22AM +0100, Ben Hutchings wrote:
> > > > 3.2-stable review patch.  If anyone has any objections, please let me know.
> > > > 
> > > > I'm not sure whether my expansion of the fix is correct here.
[...]
> > > I'm also not sure about the expansion of the fix here, not knowing very
> > > much about nfs. It seems that the code in some cases want to discard the
> > > status from decode_getfattr, for example nfs4_xdr_dec_rename is one case
> > > which does the same. Is it ok to return the status of decode_getfattr on
> > > nfs4_xdr_dec_open here? Or should it remain like it was before?
> > 
> > It looks like we try to get the file and directory attributes, and those
> > are nice to have but the open operation is still successful even if we
> > can't get them.  So my extra 'fixes' here are wrong.  Trond, is this
> > right?
> 
> Hi Ben,
> 
> We do not want to fail the RPC call in the case where the attribute
> decode failed; we already have ways to recover from that situation in
> the higher layers of the open code (we send a separate getattr RPC
> call). If, on the other hand, the filehandle decode fails, then we are
> screwed since we can't know exactly which file was just opened and this
> would be the reason for Dros's patch.
> 
> So please do leave the decode_getfattr() line as it was, and apply the
> change to the decode_getfh() line only.

Thanks for the confirmation; this is what I've done.

Ben.

-- 
Ben Hutchings
The world is coming to an end.	Please log off.

Download attachment "signature.asc" of type "application/pgp-signature" (829 bytes)