[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Problem with resolution on Radeon HD 7459 card (Squeeze)



On Sat, 28 Apr 2012 16:35:11 +0900, Han Soo Chang wrote:

> I might have made a slight advance.
> 
> I added an option ChipID 1002:6779 to designate my card (Radeon HD 7450)
> to be identical with Radeon HD 6450.

Let's don't make guesses ;-) better send/copy your current "/etc/X11/
xorg.conf" file so we can see what you have added and how it looks like 
now.
 
> Then, I got the following /var/log/Xorg.0.log, which says that "DRI
> initialization failed. kernel module (fglrx.ko) may be missing or
> incompatible.
> 2D and 3D acceleration disabled."
> 
> Actually, lsmod | grep fglrx shows nothing.
> 
> aptitude search fglrx-driver shows that the driver is installed.
> 
> Am I trying to use in vain a card that is not supported?

(...)

> [    13.076] (**) fglrx(0): NoAccel = NO 
> [    13.076] (**) fglrx(0): ATI 2D Acceleration Architecture enabled
> [    13.076] (--) fglrx(0): Chipset: "AMD Radeon HD 7400 Series" (Chipset = 0x677b)

2D capabilities are enable without hassle and the chipset seems to be 
properly detected.

> [    13.666] (EE) fglrx(0): atiddxDriScreenInit failed, GPS not been initialized.
> [    13.666] (WW) fglrx(0): *********************************************************** 
> [    13.666] (WW) fglrx(0): * DRI initialization failed                		*
> [    13.666] (WW) fglrx(0): * kernel module (fglrx.ko) may be missing or incompatible *
> [    13.666] (WW) fglrx(0): * 2D and 3D acceleration disabled                		*
> [    13.666] (WW) fglrx(0): ***********************************************************

(...)

> [    13.885] (II) AIGLX: Screen 0 is not DRI capable 
> [    13.885] [glesx] __glESXExtensionInit: No GL ES2.0 capable screen found!

And here it comes the error. The card's accel is disabled and that's 
why you experience a sluggish screen :-/

Just to discard something obvious, do a search for the module 
("locate fglrx.ko") and check if it's indeed there, although I don't 
think that's the problem, as the module should have been compiled 
alongside with the rest of the files just fine. It can be also a bug 
in the driver, I've seen some reports by Googling around, for instance:

http://devgurus.amd.com/thread/158718

Maybe is worth for you to report it in Debian BTS or better yet, first 
ask for further suggestions at "debian-backports" mailing list.

Greetings,

-- 
Camaleón


Reply to: