Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 26 Aug 2003 09:51:28 +0200 (CEST)
From:      Lukas Ertl <l.ertl@univie.ac.at>
To:        Mark Tinguely <tinguely@casselton.net>
Cc:        current@freebsd.org
Subject:   Re: Another pmap related panic
Message-ID:  <20030826094910.M19846@leelou.in.tern>
In-Reply-To: <200308221408.h7ME8FtZ014370@casselton.net>
References:  <200308221408.h7ME8FtZ014370@casselton.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 22 Aug 2003, Mark Tinguely wrote:

> >
> >  I got another pmap related panic on my HTT SMP machine.  If I don't get
> >  that completely wrong, it dies again after accessing the return value of
> >  pmap_pte_quick().
>
> I haven't buried myself in the 5.x pmap/vm code, but I did a visual inspection
> of the Bosko Milekic "PSE changes" and they do look correct to me -- good job.
>
> Do these crashes happen with the HTT turned off?
>
> It appears interesting that both sources of your panics are in a pv_list loop.
> I guess I don't know the "vm_page_queue_mtx" locking well enough, but I see
> asserts if "MA_OWNED", but I see how these are set to limit one processor
> in the queue.

I have now applied Boskos Patch to a recent -CURRENT and also turned on
INVARIANTS - same panic, same place in pmap_is_modified().

A UP kernel seems to be ok, but I would need to let it run longer.

I'm starting to think of bad RAM.

regards,
le

-- 
Lukas Ertl                             eMail: l.ertl@univie.ac.at
UNIX Systemadministrator               Tel.:  (+43 1) 4277-14073
Vienna University Computer Center      Fax.:  (+43 1) 4277-9140
University of Vienna                   http://mailbox.univie.ac.at/~le/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030826094910.M19846>