xf86-video-intel 2.2.1 release testing

[Alexander E. Patrakov]

2008/1/22, Sergio Monteiro Basto <sergio at sergiomb.no-ip.org>:
> On Tue, 2008-01-22 at 12:57 +0500, Alexander E. Patrakov wrote:
> > Also in this configuration it fails to restore the VGA mode (gives
> > blank screen)
> > when I zap the X server with Ctrl+Alt+Backspace.
>
> I think, not guaranty ,   I saw this problem when I had xerver 1.3.0,
> and correct by upgrade to xserver 1.4+

With my LFS hat on, now I say that the "blank screen" bug (which, BTW,
can be reproduced with Ctrl+Alt+F1, too) still exists on my computer
with the following software versions:

Mesa 557b0d9
xorg-server b6d4cdf
xf86-video-intel c35ab31

However, while doing some GL testing, I did something that made this
bug go away. Here are the errors from Xorg log:

(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): Unable to write to SDVOCTRL_E for SDVOB Slave 0x70.
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): First SDVO output reported failure to sync
(EE) intel(0): First SDVO output reported failure to sync

(and if I switch to the first VT and back, there appear two new
"failure to sync" messages).

The other bug ("drmMap of framebuffer failed") no longer exists.

Another observation:

If I run the "glplanet" demo from xlockmore and then run glxgears in
parallel to glplanet, glplanet's rendering slows down very much.

-- 
Alexander E. Patrakov