From: Richard Cochran <[email protected]>
Date: Mon, 20 Jun 2011 09:48:07 +0200

> Because the socket buffer is freed in the completion interrupt, it is not
> safe to access it after submitting it to the hardware.
> 
> Signed-off-by: Richard Cochran <[email protected]>

Since this isn't actually a bonafide bug fix I've applied this to net-next-2.6

_______________________________________________
stable mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/stable

Reply via email to