From mboxrd@z Thu Jan  1 00:00:00 1970
Date: Thu, 25 Jun 2020 21:55:35 -0400
From: sl@stanleylieber.com
To: 9front@9front.org
Subject: Re: [9front] vmx(1) vs openbsd
In-Reply-To: <61D0EA46F9773C79C149E55F5711DABB@eigenstate.org>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
List-ID: <9front.9front.org>
List-Help: <http://lists.9front.org>
X-Glyph: ➈
X-Bullshit: advanced transactional configuration service 
Message-ID: <20200626015535.V4hnN8QSYXqIUc1qoz0D-gm4lkGBdMTcjuTSHRPPI2w@z>

>> http://plan9.stanleylieber.com/hardware/thinkpad/x210/pre-1st-gen-1920x1200/sysinfo
>> 
>> vmx -M 2G -n ether0 -d /dev/sdE1/data -v vesa:1280x800 /tmp/bsd.67 device=sd0a
>> 
>> http://plan9.stanleylieber.com/vmx/img/20200624.vmx.openbsd67.png
>> 
>> vmx(1) itself did not crash.  is there something maybe wonky with my machine's
>> support for intel virtualization extensions?  i had no trouble booting openbsd
>>  on my x250[0], but this is the first time i've tried to boot a local disk
>> guest from a tls-booted host.
>> 
>> sl
>> 
>> [0] http://plan9.stanleylieber.com/hardware/thinkpad/x250/20cm-cto1ww/sysinfo
> 
> xgetbv is an instruction that we probably don't emulate yet.
> Seems to be somethign that OpenBSD calls as part of the
> spectre-class attack mitigations. (Specifically, MDS).
> 
> We can probably work around it by lying around cpuid,
> but it'd be better to fix it right. Not sure what that
> means yet.

i believe the last time i booted this openbsd installation in vmx(1) it
was using a slightly older version of the 6.6 kernel. the current 6.6
kernel on the ftp sites (presumably patched?) fails the same way.

sl