Latest i810 driver

Wang Zhenyu zhenyu.z.wang at intel.com
Sun Jan 21 04:59:37 PST 2007


On 2007.01.19 23:29:20 +0000, Lukas Hejtmanek wrote:
> Hello,
> 
> just tried:
> current:xserver, app, lib, proto, drm, mesa from git (default branch)
> current:xf86-video-intel - modesetting branch.
> 
> Observations:
> EXA - no more image corruption, good. On the other hand, x11perf -shmput500 
>       is about FIVE times slower than XAA.

good to know render result is right, XAA has no hw render accel, our EXA accel
still hasn't been tuned for this, as I know. 

> 
> RANDR - if an application changes resolution and crashes, xrandr -s 0 results in
> 	screen corruption. xrandr -s 1 && xrandr -s 0 fixes the corruption.
> 
> CLONE - works (If I specify MonitorLayout "CRT,LFP") good. It even work as
> 	I would expect - setting the LFP (LVDS) to 1280x800 and the CRT to
> 	1280x1024. However, in this mode, if I do xrandr -s 1, the LVDS goes
> 	blank and screen is only at the CRT (-s 1 corresponds to 1152x870), OK
> 	LVDS is not capable to use this mode. If I do xrandr -s 2 (corresponds
> 	to mode 1024x768), the LVDS is still blank and CRT has the mode,
> 	I guess this is wrong. At this point, I was unable to restore screen
> 	on LVDS (except returning to the text mode).

why randr-1.2? You just need to replace xrandr with randr-1.2 branch.

> 
> DRI - terrible slow (ppracer is unplayable at all), using INTEL_BATCH=1 does
>       not help. (glxinfo:
> 		      name of display: :0.0
> 		      libGL warning: 3D driver claims to not support visual 0x73
> 		      display: :0  screen: 0
> 		      direct rendering: Yes)
> 
> Xserver startup - LVDS is two times turned off and on. After first turning on,
> 		  garbage on screen is displayed, after second turning on,
> 		  correct X screen is displayed.
> 
> Hope, Eric or Keith find this feed back useful.

Thanks for your continuous testing.



More information about the xorg mailing list