Re: [TIP] BUG kmalloc-4096: Poison overwritten (ath5k_rx_skb_alloc)

[Bob Copeland]

On Sun, Feb 22, 2009 at 9:47 AM, Frederic Weisbecker <fweisbec@xxxxxxxxx> wrote:
> On Sun, Feb 22, 2009 at 12:20:36PM +0000, Sitsofe Wheeler wrote:
>> On Sun, Feb 22, 2009 at 01:01:21PM +0100, Jiri Slaby wrote:
>> > The unsupported jumbo message might be a clue. When we jump to the next:
>> > label, the buffer is at the end of the list in software, while in
>> > hardware it isn't. In theory, we might hit the bug with rx buffers
>> > exhaustion, because the test (bf_last == bf) doesn't work as expected then.
>>
>> This seems to be happening somewhat regularly now - I've got a small
>> collections of the warnings (I'll include them below in case they are
>> any help):

If this is a recent phenomenon, can you try reverting my patch,
fcf6b1bca8cdfefc986909b57277af4628955bd8?

This was the last patch to touch the rx path in a meaningful way.  I can't
think of anything there that would cause a use-after-free in the change,
maybe removing the "bf->skb = NULL" line, but that's the software struct,
not the hardware dma descriptor.

-- 
Bob Copeland %% www.bobcopeland.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/