svn commit: r255323 - in head/sys: amd64/conf i386/conf

[Bryan Venteicher]

On Sat, Sep 7, 2013 at 2:06 AM, Gleb Smirnoff <glebius at freebsd.org> wrote:

> On Fri, Sep 06, 2013 at 03:46:11PM -0500, Bryan Venteicher wrote:
> B> On Fri, Sep 6, 2013 at 3:24 PM, Bryan Venteicher <bryanv at freebsd.org>
> wrote:
> B>
> B> > Author: bryanv
> B> > Date: Fri Sep  6 20:24:21 2013
> B> > New Revision: 255323
> B> > URL: http://svnweb.freebsd.org/changeset/base/255323
> B> >
> B> > Log:
> B> >   Add vmx device to the i386 and amd64 NOTES files
> B> >
> B> >
> B> FWIW - I'm on the fence about adding vmx to GENERIC for 10.0. IIRC,
> B> VMware's vmxnet3 driver returns BUS_PROBE_VENDOR so the two drivers
> should
> B> coexist. This is assuming VMware updates the driver for 10 ... which I'm
> B> guessing isn't likely and was a large reason I added this driver in the
> B> first place.
> B>
> B> LMK if anybody has strong thoughts either way.
>
> May be it is worth providing a special kernel config for VMware
> environment?
>
>
Add QEMU, HYPERV, BHYVE, and VIRTUALBOX to that list. I don't think it is
unreasonable for us to provide such configs, but I think we are lacking
most of the infrastructure to make it useful.


> When GENERIC is booted in vmware it has a load of stuff that would be
> never used. And if a GENERIC with vmx is booted on hardware, vmx won't
> be ever used.
>
>
I agree it is far from ideal, but that's just the nature of what we've made
GENERIC.


>  --
> Totus tuus, Glebius.
>