Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 13 Nov 2009 10:13:24 -0800
From:      Mark Atkinson <atkin901@gmail.com>
To:        freebsd-current@freebsd.org
Subject:   Re: 8.0RC2 amd64 - kernel panic running make buildworld
Message-ID:  <hdk7k4$ip3$1@ger.gmane.org>
In-Reply-To: <4AFD1150.1080205@icyb.net.ua>
References:  <1031257439203@webmail57.yandex.ru>	<20091105184925.16b55c43@ernst.jennejohn.org>	<31221257446063@webmail71.yandex.ru>	<20091106101943.5a763f43@ernst.jennejohn.org>	<41361257585651@webmail39.yandex.ru>	<20091107115256.3df62bc3@ernst.jennejohn.org>	<1257618758.1511.14.camel@RabbitsDen>	<6511257846119@webmail85.yandex.ru>	<20091110105856.1270038e@ernst.jennejohn.org>	<1257864452.46072.25.camel@RabbitsDen>	<20091110162205.48abcffe@ernst.jennejohn.org>	<4AF99D53.9030005@icyb.net.ua>	<hdc73v$4rt$1@ger.gmane.org>	<941257966918@webmail42.yandex.ru>	<hdf5u4$qfr$1@ger.gmane.org>	<4AFC14BE.7020106@icyb.net.ua> <hdhb9l$vqf$1@ger.gmane.org> <4AFD1150.1080205@icyb.net.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
Andriy Gapon wrote:
> on 12/11/2009 17:57 Mark Atkinson said the following:
>>> superpages and no machine check - works
>>> machine check and no superpages - works
>>> machine check and superpages - problem
>> That's not quite the same for sure, definitely try replacing the memory
>> first if you haven't already.
> 
> I am not sure why would you say this.
> I still see more similarities than differences:
> 1. in all cases it's family 10h CPUs
> 2. in all cases turning off super-pages helps
> 3. in all case failure seems to happen through MCA mechanism
> 
> The hardware is quite tested too.
> 

Sorry, I only mentioned the difference in that in my case and
the case you listed as:

'superpages and no machine check - works'

Always ends up in either

* hardware reset on my machine (watchdog or other), or
* bus error during compilation

when doing buildworld with superpages on my affected machine. You seem
to list here that turning off superpages always helps, so that matches
my behavior.  I tried looking back through your posts to find the
discrepancy, but didn't garner enough detail.

All the best,


Mark




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?hdk7k4$ip3$1>