Early drop to debugger with DEBUG_MEMGUARD

[Jeff Roberson]

On Mon, 12 Aug 2013, David Wolfskill wrote:

> On Tue, Aug 13, 2013 at 08:29:44AM +0300, Konstantin Belousov wrote:
>> ...
>> The r254025 indeed introduced the problem, and Davide pointed out you a
>> workaround for the assertion triggering.
>
> Right; I tried one of those -- I hope I got it right...
>
>> Proper fix for the memguard requires a policy of M_NEXTFIT or like, to
>> avoid a reuse of the previous allocated range as long as possible.

That's why I passed a start address as a lower bound to vmem_xalloc.  I 
would like to eventually implement nextfit.

>
> Ah.
>
>> But, you have some further issue even after the assertion was silenced,
>> isn't it ?

I will fix this today and do some stress tests with memguard on.  Sorry 
for the difficulty.

Thanks,
Jeff

>
> Yes; please see
> <http://docs.FreeBSD.org/cgi/mid.cgi?20130812160154.GF1570> for a copy
> of the message that shows the resulting panic.  (Or see previous
> messages i this thread, if that's easier.)  It looks (from my naive
> perspective) as if mti_zone hadn't been initialized (properly?  at
> all?).
>
> In any case, I remain willing to test, subject to Internet connectivity
> flakiness where I am now and other demands on my time.
>
> Peace,
> david
> -- 
> David H. Wolfskill				david at catwhisker.org
> Taliban: Evil men with guns afraid of truth from a 14-year old girl.
>
> See http://www.catwhisker.org/~david/publickey.gpg for my public key.
>