From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2b) Gecko/20021102 Description of problem: WHen using a Motif app (e.g., xpdf 2.00) right clicking over the window will shut down the attached monitor(s) when using a Radeon 7500. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.Get system with Radeon 7500 2.Get xpdf-2.00-1 3.start with a PDF fileloaded 4.right click Actual Results: Monitor(s) shut down. The video card stopped all signals. Expected Results: Monitor(s) stay(s) up. Additional info: After the mouse button is released all is normal. So there is no actual data loss but the monitors do suffer.
Created attachment 84244 [details] XFree86 log
Created attachment 84245 [details] XFree86 config
If you disable accel for troubleshooting purposes by using: Option "noaccel" in the Device section of the config file, does this problem go away? If so, there are other options we can experiment with to narrow it down.
Sorry for the delay. This is my main machine which has the problem and I hardly ever reboot it. But I did now, to get the 2.4.18-18.8.0 kernel. The result: I cannot reproduce the problem. A few things changed: - the kernel: 2.4.18-18.8.0 instead of 2.4.18-17.8.0 - the system's uptime (and use of X and its resources) - the file I viewed Anyway, I'll keep an eye on it.
I just tried it with 2.4.18-18.8.0smp and the problem is still there. I don't know why I haven't seen it the last time. Are there more recent XF binaries than 4.2.0-72 available for RHL8?
There is 4.2.1-something on my mharris/testing/8.0 dirs on people.redhat.com but it most likely doesn't fix this. You can test it though nonetheless. The packages should be as stable or more stable than what is in 8.0, and haven't been released as erratum more or less because we don't have the resources to process the erratum currently, but there shouldn't be any problems with the packages. I just went through all XFree86 bug reports and classified them all for drivers. I noticed along the way about 3-4 bugs almost identical to this one, so there is a chance it is not driver specific, but a bug in XAA itself perhaps. I should be able to debug this sometime though.
Uli: Does this issue still exist in Fedora Core 3 or rawhide? Please let me know so we can schedule FC4 development time to troubleshoot the issue if it's still present. TIA
I don't use that equipment anymore.
Ok. The "radeon" driver has went through several fairly major revisions since this was initially reported. I've asked if anyone can reproduce this in #fedora-devel and #freedesktop using current generation xorg-x11 6.8.2. So far only recieved "it works fine for me" type responses. Since nobody else seems to have reported the bug here, nor in freedesktop bugzilla, we're going to assume the problem is resolved now. Should anyone experience the problem again however, please reopen and we will reinvestigate. Setting status to CURRENTRELEASE.