From: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
To: Andreas Hasenack <ahasenack-y7mWNqJcIDpfJ/NunPodnw@public.gmane.org>
Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Subject: Re: exception 12
Date: Thu, 22 Feb 2007 19:16:46 +0200 [thread overview]
Message-ID: <45DDCFFE.9060706@qumranet.com> (raw)
In-Reply-To: <200702221506.50268.ahasenack-y7mWNqJcIDpfJ/NunPodnw@public.gmane.org>
Andreas Hasenack wrote:
> On Tuesday 20 February 2007 13:54:53 Avi Kivity wrote:
>
>> Andreas Hasenack wrote:
>>
>>> kernel 2.6.20 vanilla
>>> kvm 14 kernel modules and userspace + "qemu kvm target" patch
>>> everything built with gcc 4
>>>
>>> Is this supposed to work? The vdisk.img was created with qemu-img
>>> following the howto (http://kvm.qumranet.com/kvmwiki/HOWTO). I don't have
>>> the iso for a boot image, so inserted an ubuntu 6.06 alternate install cd
>>> in the drive (/dev/cdrom).
>>>
>> This looks like another case of missing real mode emulation on Intel
>> hosts. If you have access to an AMD machine with svm support, you might
>>
>
> Is this a shortcoming in the Intel implementation? Something serious, i.e., is
> Intel just "playing" with virtualization while AMD has it done correctly?
>
>
>
The AMD virtualization support makes it fairly easy to support real
mode. The Intel implementation requires significant virtual machine
monitor support for proper real mode.
Real mode has no impact on performance, only on the boot stages. It's
certainly incorrect to say Intel is only "playing" with virtualization;
it's just a design choice of an obscure feature. But in this particular
area, Intel has made life hard on VMM developers.
There are other things that are much simpler on the AMD implementation;
it's generally easier to work with. Since they've arrived later on the
scene, they can learn from Intel's mistakes.
Having said all that, the missing real mode support is a kvm bug, not an
Intel one. We have to work with whatever the hardware provides and
supply what it doesn't.
--
error compiling committee.c: too many arguments to function
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
prev parent reply other threads:[~2007-02-22 17:16 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-02-20 15:16 exception 12 Andreas Hasenack
[not found] ` <200702201316.01276.ahasenack-y7mWNqJcIDpfJ/NunPodnw@public.gmane.org>
2007-02-20 15:17 ` Andreas Hasenack
2007-02-20 15:54 ` Avi Kivity
[not found] ` <45DB19CD.4010702-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-02-20 16:58 ` Andreas Hasenack
[not found] ` <200702201458.15783.ahasenack-y7mWNqJcIDpfJ/NunPodnw@public.gmane.org>
2007-02-20 16:59 ` Avi Kivity
2007-02-22 17:06 ` Andreas Hasenack
[not found] ` <200702221506.50268.ahasenack-y7mWNqJcIDpfJ/NunPodnw@public.gmane.org>
2007-02-22 17:16 ` Avi Kivity [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=45DDCFFE.9060706@qumranet.com \
--to=avi-atkuwr5tajbwk0htik3j/w@public.gmane.org \
--cc=ahasenack-y7mWNqJcIDpfJ/NunPodnw@public.gmane.org \
--cc=kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox