Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: NetBSD-current under Hyper-V (Déjà vu...)

You can download the full Server 2012 for evaluation. I just used an mail address for the registration. The licence is for 180
days, but can be rearmed for 30 days more - twice. You can also enable
the Hyper-V role on a Windows 8 (professional or enterprise); the
functionality is the same as the server version sans the resilience
features. Alternatively you can use the free stand-alone Hyper-V
server, but you will need W8 or W2012svr for the management console.

I'll build tomorrow a kernel with these settings.


On 6 May 2013 03:19, Chuck Silvers <> wrote:
> On Wed, Apr 24, 2013 at 03:51:19PM +0100, Chavdar Ivanov wrote:
>> Hi,
>> Five years to the day, I post the same. I'd agree that it is not
>> particularly important, google returns mostly rehashes of my post (
>> )when I search for 'NetBSD Hyper-V', so I guess I am one of the very
>> few trying to do that.
>> Basically, one has to boot netbsd -12 (both i386 and amd64) in order
>> to get usable system, otherwise one gets the the timeout messages
>> described in the above reference and no communication. In addition
>> there is something wrong with some keyboard timeouts, I guess - long
>> repeats are initiated very easily in this case, one has to be very
>> careful not to overstay the keypress more than a fraction of a second.
>> With ACPI and SMP disabled the tulip driver (legacy network adapter in
>> Microsoft's parlance) weems to work OK (only 'tlp: receive ruing
>> overrun' messages are seen from time to time, but the communication is
>> OK). Solaris/Illumos with the dnet driver works fine, FreeBSD is a
>> mixed bag (works on the full Hyper-V 3.0, refuses on the Windows 8
>> variety), OpenBSD is not working as well (but I haven't tried similar
>> to NetBSDs workarounds yet).
>> There is an ongoing project to port the synthetic driver and the rest
>> to FreeBSD - see ; this is
>> presently for FreeBSD 8.2, I don't know if it will be very hard to
>> port it to NetBSD (and that certainly is beyond my abilities, I could
>> help with testing of course).
>> I tried to go through the tulip code, but couldn't find anything
>> obvious (and as it is connected with the watchdog and the ACPI code,
>> suspected that the problem is deeper than the driver code itself ).
> I've been working on the ACPI code a bit recently, so maybe I can help
> with that aspect.   could you send the boot output from a kernel with
> is there some way I could get a hyper-V setup without having to spend money?
> -Chuck


Home | Main Index | Thread Index | Old Index