Am Mittwoch, 29. August 2012 schrieb Camaleón:
> On Tue, 28 Aug 2012 20:01:41 +0200, Martin Steigerwald wrote:
> > Am Sonntag, 26. August 2012 schrieb Camaleón:
> >> On Sun, 26 Aug 2012 18:47:34 +0200, Sven Hartge wrote:
> >> > Camaleón <noelamac@gmail.com> wrote:
> >> >> On Sun, 26 Aug 2012 18:26:19 +0200, Sven Hartge wrote:
> >> >>> Hmm. Could you post the output of
> >> >>>
> >> >>> grep flags /proc/cpuinfo
> >> >>>
> >> >>> so that we can see which capabilities the kernel thinks are
> >> >>> available on you CPU,
> >> >>
> >> >> That will be of little help because CPU will expose the VT-x
> >> >> capabilities despite the BIOS have them turned off.
> >> >
> >> > Oh?
> >> >
> >> > I must admit, I never owned a board/CPU combination, where the CPU
> >> > had virtualization features but the board lacked support for them.
> >> > So I assumed the capabilities list would reflect this.
> >>
> >> I experienced such situation: a BIOS revision that did not expose
> >> any VT- x options while the CPU supported them. A BIOS update
> >> solved it and a new menu for VT-x was added to the updated BIOS.
> >>
> >> But regardless this I think that "cpuinfo" is about CPU and not BIOS
> >> capabilities; from here you can't really know what to expect.
> >
> > May dmidecode be able to step in there?
> >
> > It shows the DMI / SMBIOS tables.
>
> I never have had a positive experience with dmidecode because it
> provides untrustworthy output :-/

In my experience it depends heavily on the BIOS. But for this ThinkPad
T520 the output appears to be pretty accurate for me. On the FTS Esprimo
workstation at work the output appears to be quite lacking of detail.

Whether everything is correct? Well…

But since it somehow seems to relate to the BIOS I thought it might help
here.

--
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 201208291617.26143.Martin@lichtvoll.de">http://lists.debian.org/201208291617.26143.Martin@lichtvoll.de

On Wed, 29 Aug 2012 16:17:25 +0200, Martin Steigerwald wrote:

> Am Mittwoch, 29. August 2012 schrieb Camaleón:

(...)

>> >> But regardless this I think that "cpuinfo" is about CPU and not BIOS
>> >> capabilities; from here you can't really know what to expect.
>> >
>> > May dmidecode be able to step in there?
>> >
>> > It shows the DMI / SMBIOS tables.
>>
>> I never have had a positive experience with dmidecode because it
>> provides untrustworthy output :-/
>
> In my experience it depends heavily on the BIOS.

Having an updated version of the utility it can also help but yet again,
the results are not always the expected.

(side note: in the Spanish mailing list we've been recently discussing
about this issue... a user was a bit perplexed for the results of "lshw" -
which uses dmidecode- that showed his micro only had enabled one of the
two cores while the rest of the tools said otherwise)

> But for this ThinkPad T520 the output appears to be pretty accurate for
> me. On the FTS Esprimo workstation at work the output appears to be
> quite lacking of detail.
>
> Whether everything is correct? Well…
>
> But since it somehow seems to relate to the BIOS I thought it might help
> here.

It can be also useless -even if dmidecode is right- because the current
BIOS revision could be not exposing the VT-x capabilities if they were
disabled on purpose by the manufacturer (but can be enabled by updating
the BIOS firmware).

I think this is one of those specific/complex situations where I prefer
to contact the motherboard manufacturer because only they can tell you
about the board options and capabilities.

Greetings,

--
Camaleón


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: http://lists.debian.org/k1lbqr$koe$8@ger.gmane.org

On Sun, Aug 26, 2012 at 6:36 PM, Camaleón <noelamac@gmail.com> wrote:
> On Sun, 26 Aug 2012 18:26:19 +0200, Sven Hartge wrote:
>
>> Artifex Maximus <artifexor@gmail.com> wrote:
>>> On Sat, Aug 25, 2012 at 2:37 AM, Sven Hartge <sven@svenhartge.de>
>>> wrote:
>>>> Artifex Maximus <artifexor@gmail.com> wrote:
>>>>> On Fri, Aug 24, 2012 at 3:51 PM, Camaleón <noelamac@gmail.com> wrote:
>>>>>> On Wed, 22 Aug 2012 20:21:50 +0200, Artifex Maximus wrote:
>>>>
>>>>>>> The question is I need to upgrade my CPU to any VT capable CPU or
>>>>>>> is there any solution staying at the current CPU?
>>>>>>
>>>>>> Well, you can stil install a 32-bits guest OS but for 64-bits you
>>>>>> need a VT-x enabled micro, yes.
>>>>
>>>>> OK, now a lovely little E8400 power my machine which is VT-x and VT-d
>>>>> capable.
>>>>
>>>> Did you enable the support for those features inside the BIOS setup?
>>>> If not, they may be masked and unusable for Linux and VirtualBox.
>>
>>> Even it is a quite old motherboard "Intel Virtualization Technology"
>>> BIOS options is enabled. I cannot change that option though. My
>>> motherboard just for info is a Fujitsu-Siemens G31T-M2 which is some
>>> ECS OEM motherboard with G31 chipset.
>>
>> Hmm. Could you post the output of
>>
>> grep flags /proc/cpuinfo
>>
>> so that we can see which capabilities the kernel thinks are available on
>> you CPU,
>
> That will be of little help because CPU will expose the VT-x capabilities
> despite the BIOS have them turned off.

Someone recommend the XenClient CD. I've tried that and says that VTx
is disabled even it is enabled in BIOS. :-(

> @Artifex, contact your motherboard manufacturer and ask specifically for
> this. I see nothing at the manual or specs regarding VT-x for your board¹.
>
> ¹http://www.ecs.com.tw/ECSWebSite/Product/Product_Detail.aspx?DetailID=812&CategoryID=1&Menu ID=16&LanID=0

Thanks. Mine is Fujitsu-Siemens (manufactured by ECS) G31T-M2 which is
slightly different from any other which I found on ECS site. So no
BIOS there. I already have the latest (dated 2008) FuSi BIOS for my
board. I'll contact with FuSi hope they will help me.

Bye,
a


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: http://lists.debian.org/CAPkuXvHpUkPiJSkK=zu6yx-er4SkvZawb626cNM5x2PN5Qdzuw@mail.gmail.com