Red Hat Bugzilla – Bug 486368
ATI radeon very slow - almost unusable
Last modified: 2009-03-08 06:33:19 EDT
Description of problem:
F11 / rawhide on my system is so slow that it is almost unusable and I must consider giving up trying to test it. To me that is a serious regression in a serious component and I have set severity/priority to indicate that.
I dont' know where the slowness comes from, but it is visible that for example redrawing after switching windows is slow, so I assume it is related to the X driver.
I have seen other reports which might be similar to this one, but I haven't found anything conclusive, so I create this issue to track my issue and my uploads without confusing by mixing them with what other see and report.
Version-Release number of selected component (if applicable):
kernel-PAE-2.6.29-0.124.rc5.fc11.i686 no KMS as mentioned on bug 486327
xorg-x11-drv-ati-6.10.0-3.fc11.i586 no xorg.conf
01:00.0 VGA compatible controller: ATI Technologies Inc RV516 [Radeon X1300/X1550 Series]
.xsession-errors has a lot of other noise, but the only X related is:
Xlib: extension "XFree86-Misc" missing on display ":0.0".
Created attachment 332562 [details]
Created attachment 332563 [details]
Similar to bug 486049 .
Keeping both because of different hardware (RV516 vs R420).
My notebook with a Mobility Radeon 9200 M9+ (R280) certainly feels a lot more sluggish under F11 than it did under F8 or F9 with 2D and no desktop effects (glxgears gets around 1200 fps), this is using kernel-2.6.29-0.137.rc5.git4.fc11.i586 and xorg-x11-drv-ati-6.10.0-3.fc11.i586, with no xorg.conf. Should I report a separate bug for this?
I think you could wait for 6.11.0 and check again. It feels a lot faster on RV350.
I see no improvement with the 6.11.0 driver.
This seems to be indeed a widespread regression. Just to recapitulate from:
related bugs might be:
maybe also in some way bug 242386
There are lot of affected users, adding as F11Blocker.
(In reply to comment #7)
> This seems to be indeed a widespread regression. Just to recapitulate from:
I tried dropping the patches from the SRPM as suggested, but saw no significant performance increase on my Mobility Radeon 9200. Hmm. There are an awful lot of things like
Xorg:2269 freeing invalid memtype ea102000-ea112000
in the syslog after quitting X.
It works fine for me now, both with and without KMS, thanks!
Hmm... the 2D side of X is still sluggish on my M9+ system. At the login screen, for instance, the background doesn't fade in smoothly at all (~1 fps). This is with kernel-2.6.29-0.207.rc7.fc11.i586, xorg-x11-drv-ati-6.11.0-5.fc11.i586 and mesa-dri-drivers-7.3-10.fc11.i586.