Re: list_head debugging patch

William Lee Irwin III (wli@holomorphy.com)
Thu, 29 May 2003 12:38:49 -0700


On Thursday 29 May 2003 15:08, William Lee Irwin III wrote:
>> This appears to get the kernel to crap its pants in very, very
>> short order. Given the number of things going wrong, I almost
>> wonder if I did something wrong. Things get real ugly, really,
>> really fast.

On Thu, May 29, 2003 at 09:22:43PM +0200, Morten Helgesen wrote:
> [snip]
> I gave this a go - booted without problems. I did some
> untaring/copying/deleting and didn`t see anything unusual, but a
> 'dbench 8' died right away.
[...]
> EIP is at clear_queue_congested+0x78/0xb0

clear_queue_congested() is doing an opportunistic check for list_empty()
without taking a lock. The patch basically changes list_empty() to look
at elements of the list instead of just pieces of the head. As opposed
to auditing for this, could you remove the __list_head_check() from
list_empty() and try again?

Thanks.

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