Fedora Account System
Red Hat Associate
Red Hat Customer
Description of problem: Both kde and gnome grawl to a halt (login becomes impossible) after update from xorg-x11-drv-ati-6.9.0-61.fc10.i386 to xorg-x11-drv-ati-6.9.0-62.fc10.i386. Version-Release number of selected component (if applicable): xorg-x11-drv-ati-6.9.0-62.fc10.i386 How reproducible: kdm greet screen does not complete and everything halts Steps to Reproduce: 1. update xorg-x11-drv-ati to 6.9.0-62 2. telinit 3, then telinit 5 Actual results: screen freezes or slows down to an almost halted state Additional info: after replacing /usr/lib/xorg/modules/drivers/radeon_drv.so with the corresponding file from the 6.9.0.61 package everything works as before the update.
exactly the same bug since I upgraded today to 6.9.0-62 I've had to move to the vesa driver on 2 computers with 2 different ATI cards. The card on this computer is ATI Technologies Inc RV350 AP [Radeon 9600]. I don't understand how this update came out from the testing repository ...
Created attachment 327367 [details] Xorg.0.log with the final error It ends with (EE) RADEON(0): FIFO timed out, resetting engine...
(In reply to comment #1) > exactly the same bug since I upgraded today to 6.9.0-62 > I've had to move to the vesa driver on 2 computers with 2 different ATI cards. > The card on this computer is ATI Technologies Inc RV350 AP [Radeon 9600]. > I don't understand how this update came out from the testing repository ... If it helps I placed the previous /usr/lib/xorg/modules/drivers/radeon_drv.so in http://dmf.unicatt.it/~paolini/radeon/ By chance I didn't upgrade also the computer at work, so I still had the old version there :-) Actually this was a *very annoying* problem!
IBM T30 loses all fonts at run level 5. Can't even read the login screen though blind-typing lets me log in and do stuff. Run Level 3 and grub booting fine. Here's the chipset info from the Xorg.0.log: RADEON(0): Chipset: "ATI Radeon Mobility M7 LW (AGP)" (ChipID = 0x4c57) Reverting to -61 fixed the issue.
The -62 driver causes the display to go weird on my Dell D600 with a Mobility Radeon 9000 chipset as well, especially text is jumbled beyond recognition. Downgrading to -61 fixed it, it's mostly gone from the mirrors but can be found here: http://koji.fedoraproject.org/koji/buildinfo?buildID=72791
As mentioned, the ATI Radeon 9600 experiences the same symptoms after upgrading to xorg-x11-drv-ati-6.9.0-62.fc10.i386. In addition to the timeout listed in Xorg.0.log, an error is logged in messages, excerpted below: Dec 20 13:11:55 localhost kernel: Linux agpgart interface v0.103 Dec 20 13:11:55 localhost kernel: agpgart-intel 0000:00:00.0: Intel 865 Chipset Dec 20 13:11:55 localhost kernel: agpgart-intel 0000:00:00.0: AGP aperture is 64M @ 0xac000000 ... Dec 20 13:11:55 localhost kernel: [drm] Initialized drm 1.1.0 20060810 Dec 20 13:11:55 localhost kernel: pci 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 Dec 20 13:11:55 localhost kernel: [drm] Forcing AGP to PCI mode Dec 20 13:11:55 localhost kernel: [drm] Initialized radeon 1.29.0 20080528 on minor 0 ... Dec 20 13:12:08 localhost kernel: agpgart-intel 0000:00:00.0: AGP 3.0 bridge Dec 20 13:12:08 localhost kernel: agpgart-intel 0000:00:00.0: putting AGP V3 device into 8x mode Dec 20 13:12:08 localhost kernel: pci 0000:01:00.0: putting AGP V3 device into 8x mode Dec 20 13:12:09 localhost kernel: [drm:radeon_do_init_cp] *ERROR* PCI GART memory not allocated! When using the proprietary ATI driver, the symptoms do not arise; however, this is not a feasible workaround, as several applications, including OpenOffice, do not seem to function.
please test with the xorg-x11-drv-ati-6.9.0-63 in koji. I see two problems here, one is already fixed in a later kernel update, without that fix the fallback position was broken in -62. So -63 should fix the fallback position.
(In reply to comment #7) > please test with the xorg-x11-drv-ati-6.9.0-63 in koji. > > I see two problems here, one is already fixed in a later kernel update, without > that fix the fallback position was broken in -62. So -63 should fix the > fallback position. Tested. All seems fine now, thank you very much!
works fine here too. Thanks.
Great! The new build works fine here.
Works fine here, thanks!