Description of Problem: I have a fairly predictable (insofar as I know it will happen, just not exactly when) problem with the system completely locking up when using the SMP kernel under X. If I leave the computer for long enough and the screensaver kicks I'm assured a lockup after some period of time (within a couple of hours). The lockup is a complete one: I can't CTRL-ALT-BACKSPACE, I can't ping the system, I can't CTRL-ALT-DEL, and I can't even ALT-SysRq-K (yes, I've enabled SysRq to try to deal with this problem). So far in my tests I have not had any problems with this if I use the uniproc kernel. I've just disabled DRI to see if that helps. Version-Release number of selected component (if applicable): Null-re0816.3 (previous Milan and Limbo releases as well, however) [crunge@mothership data]$ rpm -q kernel-smp kernel-smp-2.4.18-11 [crunge@mothership data]$ rpm -q kernel kernel-2.4.18-11 [crunge@mothership data]$ rpm -q XFree86 XFree86-4.2.0-65 How Reproducible: 100% thus far Steps to Reproduce: 1. install Workstation class system 2. startx 3. leave computer for several hours Actual Results: Expected Results: Additional Information: graphics card is Radeon 7500 sound card is SB Audigy (problem happened with SB Live! as well, however) ServerWorks III HE-SL chipset 2xP3 1GHz 512 MB RAM
some additional information based on some tests I've done: card kernel DRI locks-up? ---------------------------------------------------- radeon 7500 SMP y y radeon 7500 SMP n n radeon 7500 uni y n matrox g450 SMP y n basically, everything except the radeon 7500 running on an SMP kernel with DRI works -- perhaps there is a race condition in the radeon driver? fwiw, I have the same problems with the radeon/smp/dri config under 7.3 as well
still an issue with Null
still an issue with 8.0 final I've tried removing all PCI cards and fiddling with the BIOS settings yet I can still reproduce this. If it isn't an issue in the radeon code then perhaps there is a problem with the serverworks agpgart code when run on SMP, but since the latter chipset is only used on SMP systems I find this hard to believe. Any ideas?
Does this occur with 8.0 plus all released erratum?
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/