Should 9.3 carry a warning about NEW_XORG

[John Marshall]

On Fri, 04 Jul 2014, 15:31 -0400, Ed Maste wrote:
> In HEAD syscons(4) and vt(4) are now both compiled in by default in
> the GENERIC kernel.  Syscons remains the default at the moment; you
> can set the loader tunable kern.vty=vt to choose vt(4) instead.  Vt is
> selected automatically if booting via UEFI on amd64.
> 
> Both stable/10 and stable/9 require a recompile to use vt(4).  The
> plan is to merge these changes from HEAD in time for 10.1, but 9.3
> will not have them.

So, perhaps 9.3 should remain "opt in" like 9.2 and not build NEW_XORG
by default?

> Note that vt(4) enables vty switching from Xorg, but WITH_NEW_XORG
> generally should not depend on vt(4).  X should work fine, just
> without the ability to switch back to a vty.  If WITH_NEW_XORG fails
> on certain hardware I think it'll be independent of the use of sc(v)
> vs vt(4).

Thank you, Ed, for taking the time to explain all that.  I had been
under the impression that NEW_XORG depended on vt(4), which was why I
modified my 9.3 kernel.  sc(4) worked fine after my initial upgrade,
vt(4) works fine; it's just X that doesn't.

Perhaps my "X no longer works" scenario is due to "certain hardware"?
Is there a list somewhere of hardware on which NEW_XORG will not work,
so that folks running 9.2 with that hardware can set WITHOUT_NEW_XORG
BEFORE they upgrade to 9.3 and save themselves grief?

-- 
John Marshall
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 181 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20140705/cfc40d55/attachment.sig>