> > Removing another signature of detecting virtualization and increasing
> > compatibility would be negligible gain? Just asking...
> I don't think we are going to try and defeat the NVidia virtualization
> checks, and I can probably assure you that they would patch them as
> fast as we bypassed them.
No-no, that's not what I meant. It is not about nVidia, just general
quality of the hypervisor. Ideally, running in the VM should to be
indistinguishable from running on bare metal, except things like
CPUID, peripheral ID strings, etc (that can be easily changed by
recompiling). For instance, suppose someone is doing security
research -- studying clever malware for example; you'd want to run
in VM, but want to have as few ways as possible for the subject to
find out that it is running in a VM. Or Internet security... Just
some examples. FreeBSD was always (as far as I remember, anyway) in
the lead in research (as a platform) and quality, so I figured it'd
be nice to keep it this way.
> This is officially an item on my plate. Functional hardware that works
> doing VGA passthrough under ESXi was brought up about 2 weeks ago so I
> am past that stage. I have Quadro FX3800 on a Supermicro X9DAi.
Great! I'm glad to see there is interest in GPU passthrough on FreeBSD.
> You have helped some in uncovering the next set of issues, but my
> plate is very large, and seems to have grown appendages that are
> holding all sorts of things :-)
Yeah, it's a dream to have a nice, flat plates for projects :)
> I am still coming up to speed on the bhyve code, so it wont be
> a half hour fix by me, but it well get fixed.
Count me in for help. I haven't got lots of time to play with this
(or at least shouldn't be spending so much time), but can aid with
testing, and also can provide access to a machine with an AMD card
(or both nVidia & AMD).
> Rod Grimes
[SorAlx] ridin' VN2000 Classic LT
email@example.com mailing list
To unsubscribe, send any mail to