[ltp] Weird screen artifacts, but only with vnc

[Alex Deucher]

On 7/27/05, David A. Desrosiers <desrod@gnu-designs.com> wrote:
>=20
>         I use vnc (various versions) quite heavily to manage a bunch
> of headless graphical workstations and development/test machines here
> at gnu-designs, inc. from my laptop (a T42p, 2373-Q1U).
>=20
>         After upgrading to the r300/radeon driver (and getting some
> really nice fps rates), I've noticed some visual "artifacts"
> appearing, but they only seem to affect vnc sessions.
>=20
>         A screenshot of these can be seen here:
>=20
>         http://code.gnu-designs.com/vnc_r300_artifacts.png
>=20
>         Those "dots" are what happens when I move the mouse around
> inside the vnc window. You can see they never go outside the vnc
> window though, onto my "normal" desktop.
>=20
>         I've tried 5 different vnc servers for Windows, various video
> drivers on the Windows side, different configuration optimizations on
> the server and client site (vncviewer, xtightvncviewer, etc.) and
> nothing seems to help. I can move the vnc window around and the dots
> stay in place, until I open another window on the Windows desktop (in
> this case). I can also change the Windows resolution which clears it
> up, until I move the mouse around again.
>=20
>         This only started happening after I switched to the r300
> driver instead of the stock "ati" driver, and it only seems to affect
> vnc sessions. I haven't seen any other application that behaves like
> this at all.
>=20
>         Does anyone have any idea what could cause this behavior?
>=20
>=20

add these lines to your device config:
XaaNoScanlineImageWriteRect
XaaNoScanlineCPUToScreenColorExpandFill

these XAA routines seems to cause problems when the dri is enabled on
some r300 chips.

Alex

>=20
>=20
> p.s. You'll notice that the battery shows 1% in that screenshot. Its
> been plugged in for months, never runs on battery, and the battery has
> slowly dwindled from 100% to 1% and fails to charge back up. This is
> the third battery I've had in it in about 6 months, and they all seem
> to behave the same way. I think the charging circuit is dead.
>=20
>=20
> David A. Desrosiers
> desrod@gnu-designs.com
> http://gnu-designs.com