Subject: Re: and yet more weirdness with XFree86-4.5.0 on NetBSD/sparc
To: NetBSD/sparc Discussion List <port-sparc@NetBSD.org>
From: Greg A. Woods <woods@weird.com>
List: port-sparc
Date: 12/06/2006 12:05:43
--pgp-sign-Multipart_Wed_Dec__6_12:05:33_2006-1
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

At Wed, 6 Dec 2006 11:42:24 +0000,
Julian Coleman wrote:
>=20
> > The next thing I noticed was that the keyboard focus wasn't following
> > the mouse cursor as it normally should with ctwm.
>=20
> I have noticed both these on sparc and sparc64 when using Xsun.  When
> using XFree86, these problems aren't there.  (This bit me again recently
> on sparc because we have X -> Xsun by default).
>=20
> So, I guess the simplest fix is to add support for the other framebuffers
> to XFree.

That's simply not true.  It might be the most _desirable_ fix (for
some), at least until Xorg came along, but it's not likely the simplest
fix by any definition of "simplest".  I'm not even convinced yet that
it's the "right" fix in terms of being the most elegant and secure and
such (i.e. that using the XFree86 Xserver on sparc hardware is the
"right" fix).

The "old" Xsun code used to work just fine w.r.t. focus follows mouse,
even on MUCH slower machines (I was happy once upon a time with an SS-1
on my desktop, using the exact same framebuffer card).  That was way
back with NetBSD-1.3 or before.

It would seem that somebody somewhere broke something in the new code.
(I just can't believe that XFree86's Xsun code could be that much older
and thus not have some fix that made it work in the old code.)

It's the same kind of breakage that may also be causing the screen saver
to kick in even while event have occured within microseconds of the
screeen blanking -- that never happened before, not even with the very
ancient Xserver I used to run on SunOS-4 on the same hardware.  Here too
it would seem as though some more recent change, in the xfree code that
was most certainly branced from the working code, must be responsible.

I have no idea where to look or I'd have looked for more recent changes
in the xfree code that might be responsible for the breakage.

My frustration comes from having "upgraded" to something that's far more
broken than all of what came before it.

--=20
						Greg A. Woods

H:+1 416 218-0098 W:+1 416 489-5852 x122 VE3TCP RoboHack <woods@robohack.ca>
Planix, Inc. <woods@planix.com>       Secrets of the Weird <woods@weird.com>

--pgp-sign-Multipart_Wed_Dec__6_12:05:33_2006-1
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: AnLfaQtYAsCu90dz2NuhEOToEkTG7SLL

iQA/AwUBRXb4YmJ7XxTCWceFEQK+QwCeOhp6k3JqOxDhnzWDgFarGtIFbOkAmgJK
8V8JJ33jVIBX4ivhHPskW1cm
=8GDn
-----END PGP SIGNATURE-----

--pgp-sign-Multipart_Wed_Dec__6_12:05:33_2006-1--