Re: [PATCH 3/3] p54: zero-out rx_status

[Christian Lamparter <chunkeey@googlemail.com>]

On Tuesday, December 11, 2012 04:30:13 PM Johannes Berg wrote:
> On Tue, 2012-12-11 at 16:25 +0100, Johannes Berg wrote:
> 
> > > If this issue just popped up now, I suspect that 
> > > something else is silently corrupting our SKBs 
> > > [or can anybody see how rt2x00 hit this issue?]
> 
> Oh, ok I assumed that rt2x00 RX SKBs are cleanly allocated, but now that
> I trace the code further I see that while the SKBs are on the queue
> inside rt2x00 they have something called "struct skb_frame_desc" on the
> skb->cb, so this is what's causing the rt2x00 issue. This should be
> mentioned in the rt2x00 commit, and all other drivers should be fine.
Ah, you are right. I assumed that rt2x00_queue stuff was only important
for TX. But it seems rt2x00 pci devices use them as well for RX.

Regards,
	Chr