Mailinglist Archive: radeonhd (307 mails)

[Vladimir Grebenschikov <vova@xxxxxxx>]

On Thu, 2009-02-05 at 11:38 +0100, Matthias Hopf wrote:
> On Feb 03, 09 19:08:28 +0300, Vladimir Grebenschikov wrote:
> > Matthias, what about old bug about "lost of sync" effect on notebook
> > panel ? 
>
> Gimme a hint. Do you have a bug number?

No bug, as far as I remember it was talk on irc or in mail.
Problem - on X start or console switch either to X screen or back to
text console screen looks exactly as CRT monitor which is out of sync
(picture jumping and has no fixed vertical edges).
Switching X-text-X, probably several times, fixes problem.

Problem affects only PANEL output, external monitor is fine.
Problem do not appears (or appears not so often on single-head
configuration).

I can file a bug, but can't provide much details to it, probably short
avi clip ?

> > As well as strange effect with mouse cursor - sometimes mouse cursor
> > starts to shake or blink on some areas of screen, then it passes.
>
> Yes, we still don't have a freaking clue why and when this happens.

Yesterday it crashes system. First mouse cursor grow about 5 times of
normal size and looks as vertically duplicated copy of part of arrow,
then there was hard freeze.

> > And last question: why output naming is different between radeon and
> > radeonhd drivers ? 
>
> Long story short: You can use analog and digital bits of DVI outputs at
> the same time only with this naming scheme.
>
> > It is a bit hard to use one Xorg configuration (dual-head setup) and
> > switch drivers.
>
> Yeah, I know. We had a lengthy discussion on this mailing list about
> whether or not to change, and the result was pretty much 1:1.

Thanks for clarification.

> > Bug google-earth - still does not work (I have FreeBSD, so it tries to
> > run under Linux emulation).
>
> Try the following in your .drirc (Create it if you don't have one):
> see Attachment

> That did it for me. The important line is the vblank line.
> But I don't know whether that fixes it for you (never used any BSD
> variant myself).

Ehh - does not help, actually , after recent Xorg upgrade it just
crashes on start (before it work in non-GL mode)

$ googleearth
unknown chip id 0x7145, can't guess.
libGL warning: 3D driver returned no fbconfigs.
libGL error: InitDriver failed
libGL error: reverting to (slow) indirect rendering
terminate called after throwing an instance of 'QString'
Google Earth has caught signal 6.

Stacktrace from glibc:
  ./googleearth-bin [0x806c3a3]
  ./googleearth-bin [0x806c916]
  [0xbfbfffbb]
  /lib/libc.so.6(abort+0x101) [0x493d0301]
  ./libstdc++.so.6(_ZN9__gnu_cxx27__verbose_terminate_handlerEv+0x179) 
[0x4816a019]
  ./libstdc++.so.6 [0x481679a5]
  ./libstdc++.so.6 [0x481679e2]
  ./libstdc++.so.6 [0x48167b4a]
  ./librender.so(_ZN12RenderWidget6SetApiEPN5earth4evll3APIE+0x34e) [0x49f3b57e]
  ./librender.so(_ZN5earth6render12RenderWindow12createWidgetEv+0xb2) 
[0x49f1cf62]
  
./libgoogleearth_lib.so(_ZN5earth6client12ModuleWidget9showEventEP10QShowEvent+0x8e)
 [0x4931af6e]
...

Probably it is due to outdated Linux packages:
linux-gtk2-2.6.10_1 GTK+ library, version 2.X, Linux binary
linux-xorg-libs-6.8.2_5 Xorg libraries, linux binaries
linux_dri-7.0       Binary Linux DRI libraries for 3D hardware acceleration of 
linux_base-f8-8_11  Base set of packages needed in Linux mode (for i386/amd64)

BSDs libraries:
xorg-libraries-7.4  X.org libraries meta-port
dri-7.3,2           OpenGL hardware acceleration drivers for the DRI

> Matthias

-- 
Vladimir B. Grebenschikov
vova@xxxxxxx
-- 
To unsubscribe, e-mail: radeonhd+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: radeonhd+help@xxxxxxxxxxxx