Lost sliding windows w/ DVI Syncmaster monitor

[Alex Deucher]

On Wed, Apr 1, 2020 at 1:39 PM Paul Rogers <paulgrogers at fastmail.fm> wrote:
>
> > The effect you like is called panning.
>
> Thank you for that.  I couldn't guess what the proper keyword would be for searching, best I could think of was "sliding" and this got lots of hits on vinyl windows! 8-(
>
> >
> > I don't think it has anything to do with the Samsung or the Dell, but
>
> I'm quite sure it does.  I've been doing lots of testing to identify the problem.  If you want the gorey details:  I moved the HD from a box using a Biostar TP43 MoBo w/ Core2-Duo "Conroe" CPU, Nvidia GeForce 6200 w/ NV44 GPU PCI video, HPvs15 VGA monitor, to a Dell XPS-8700, "Haswell" CPU, GeForce GT6351 w/ GK208 GPU PCIe-16 video, Samsung Syncmaster 730E monitor, with NO configuration changes to X.  Panning worked in the first box, not the second.  BUT, it also doesn't work with the same monitor, same software, on an Intel S1200 "Sandybridge" server board with Intel's integrated graphics.  The common denominator is the monitor.  I suspect the EDID is telling X something different than the VGA monitor, that X is taking to mean "don't do panning", that defaulted to panning on all the VGA monitors it has run on before.  I can't find any likely suspects in xorg.conf.
>

Panning has nothing to do with the monitor.  It comes down the the
hardware limitations of the underlying device and what features are
exposed via the driver for that device.  The maximum framebuffer size
is limited by the stride coordinate limits of the display hardware on
the device.  The device also needs to be able to dynamically adjust
the offset of the display frame within the display surface.  When you
slide pan around with in the larger virtual desktop, the driver is
changing the offset of the display controller as you move around.  On
top of the hardware requirements, the X driver also has to expose the
necessary support for panning.  Not all drivers do.  Especially newer
drivers which support randr 1.2 or newer.

Alex

> > rather with however you are trying to configure.
>
> It worked fine before moving the HD.
>
> > How exactly are you configuring?
>
> I attached my X log.
>
> > Also, you're using a very slow generic X driver (VESA)
>
> By choice.  I don't need any advanced functions, so staying with the most basic common driver gives me added flexibility to use it with different video cards without reconfiguratiom.
>
> > that I suspect may not support panning.
>
> Yes, it does.  This same system worked just fine before I changed monitors as described.  It continues to work on other boxen with VGA monitors.
>
> > It's also possible that the problem is your distro's build
>
> My distro, to be sure.  This is an LFS derived "distro" I hand-built, the last of 9 versions since 2.4 kernels.
>
> > of 1.18.4 server or driver is broken WRT panning.
>
> My original complaint said it was working before I upgraded and got this new monitor.  It's not the server unless panning is monitor/EDID dependent.
>
> > ISTR panning was broken in the server for quite some
> > time, depending on configuration method, and the patch to fix for which
> > may not have been included in your quite old 1.18.4 version.
>
> It takes me a long time to build my distro from scratch.  I could upgrade the server, as long as it would retain compatibility with everything else in this version of X-7.2.  I don't want to go into "dependency hell"!
>
> >
> > Please show us your GPU specification, thus:
> >
> >  # inxi -Gxx
> >
>
> [10:32 ~]# inxi -Gxx
> -su: inxi: command not found
> [10:32 ~]#
>
> The GPU ID is in the log, Nvidia GK208.
>
> --
> Paul Rogers
> paulgrogers at fastmail.fm
> Rogers' Second Law: "Everything you do communicates."
> (I do not personally endorse any additions after this line. TANSTAAFL :-)
> _______________________________________________
> xorg at lists.x.org: X.Org support
> Archives: http://lists.freedesktop.org/archives/xorg
> Info: https://lists.x.org/mailman/listinfo/xorg
> Your subscription address: %(user_address)s