Created attachment 54098 [details] Xorg.0.log I'm running Slackware 13.37 x86 with OpenBox 3.4.11.2 as my only desktop, Xorg version 1.9.5 and using xf86-video-ati-6.14.3 for Radeon KMS, on a laptop with an ATI HD3200 chip. I'm running into a few video glitches with the latest ATI driver. 1. Some times text won't appear when I'm typing, as an example, in Firefox 8.x when I'm in gmail it won't appear, also especially if I copy and paste something in the email, it will be blank, then I have to move my mouse around and it appears. I have also had this happen in Xchat, when I type something in the client, then hit enter, nothing appears in the channel window, so I'll move the mouse around where an empty space appears and I get the text to reappear. With xterm, when reopening it will show something that I had opened before, like say I had nano open in a file doing some editing, then closed nano and xterm then opened another xterm, I see for a split second the inside of that file in nano. Also if any of you guys know the pkgtool, which is ncurses, when I have that open in a xterm, then close it, then open another xterm, I'll see that appear again for a slight second. Also I'm noticing a bit more flicker in the Window, not the desktop, then usual when rolling windows up and down.
> I'm running into a few video glitches with the latest ATI driver. Normally, every single issue needs to be tracked in a separate report. Anyway, is any of these issues a regression from 6.14.2 or an older version? If so, can you bisect? Also, can you provide the configure and compiler flags used for building the driver?
Created attachment 54164 [details] Slackware Build Script No this is not a regression from an older version, Slackware comes with xf86-video-ati-6.14.1 and it didn't have these problems, this is something new I've encountered. I've attached the build script I used in Slackware to compile this... THANKS
(In reply to comment #2) > No this is not a regression from an older version, Slackware comes with > xf86-video-ati-6.14.1 and it didn't have these problems, this is something new > I've encountered. That sentence is self-contradictory. :) But it sounds like it actually may be a regression from 6.14.1. Can you try building that from upstream Git, and if it still doesn't have these problems, bisect?
I thought you meant by regression if this was an issuing going back to a previous version I could see the same thing happening... I don't know what you mean to do with bisect? I'll grab from git and see what happens; git://anongit.freedesktop.org/xorg/driver/xf86-video-ati By the way is the git pretty active for the driver? THANKS
By the way what version is the git? I need to know for when I package this, what version I should call it besides something git... :) The only thing I could find was this; Generated by GNU Autoconf 2.68 for xf86-video-ati 6.14.99. Hmm 6.14.99.git? THANKS
(In reply to comment #4) > I don't know what you mean to do with bisect? Google for 'git bisect'. It's a tool for isolating the change that caused a regression as quickly as possible. You can check out the 6.14.1 tree with git checkout xf86-video-ati-6.14.1
I'm here to submit a report and do the best I can but as far as bisect is concerned you're just going to have to tell me exactly how you want me to run it, sorry I don't have time to learn this... As far as xf86-video-ati-6.14.1 is concerned Slackware has it, so do you want me to grab it instead from git or just use the version in Slack? If you want me to grab it from git, what's the complete url? THANKS
(In reply to comment #7) > I'm here to submit a report and do the best I can but as far as bisect is > concerned you're just going to have to tell me exactly how you want me to run > it, sorry I don't have time to learn this... cd xf86-video-ati git checkout xf86-video-ati-6.14.1 [ build and test ] git bisect good # assuming the 6.14.1 snapshot from upstream Git still works git checkout master [ build and test ] git bisect bad # assuming the master snapshot from upstream Git still doesn't work repeat: [ build and test ] git bisect good/bad/skip # according to the current snapshot being good/bad/untestable
Sorry Michel got a little tied up over the holidays, also with the latest kernel, 3.1.x it seems better, so I'm just going to leave it alone... THANKS P.S. Sorry for the bug report...
No need for apologies, but you can't expect bug reports to get magically resolved without your help to narrow down the cause. Note that the Firefox issues could be related to its using OpenGL for acceleration now, it looks like that doesn't work very well yet at least with the Gallium based Radeon drivers.
Mass closure: This bug has been untouched for more than six years, and is not obviously still valid. Please reopen this bug or file a new report if you continue to experience issues with current releases.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.