Re: [PATCH] gianfar: Fix invalid TX frames returned on error queue when time stamping.

[Richard Cochran <richardcochran@gmail.com>]

On Mon, Jan 09, 2012 at 08:16:01AM +0100, Manfred Rudigier wrote:
> From: David Miller [mailto:davem@davemloft.net]
> Sent: Thursday, January 05, 2012 19:27
> 
> >From: Manfred Rudigier <manfred.rudigier@omicron.at>
> >Date: Thu, 5 Jan 2012 15:50:21 +0100
> >
> >> +
> >> +		/* Keep sock if we must return a time stamp on the err queue */
> >> +		skb_new->sk = skb->sk;
> >
> >When I see something like this without any kind of reference counting or
> >similar, I am gravely concerned.
> 
> The skb_tstamp_tx function called during gfar_clean_tx_ring requires
> the skb->sk pointer to be set. Otherwise no time stamp can be queued
> on the socket error queue.
> What would be the correct way for doing this?

Hi Manfred,

You have to make sure the new clone holds a reference on skb->sk.
A recent, similar problem was discussed:

    http://comments.gmane.org/gmane.linux.network/208143

And the solution came in:

    commit da92b194cc36b5dc1fbd85206aeeffd80bee0c39
    Author: Richard Cochran <richardcochran@gmail.com>
    Date:   Fri Oct 21 00:49:15 2011 +0000

    net: hold sock reference while processing tx timestamps

HTH,

Richard