Description of Problem: After installing Mike Harris' testing RPMS for XFree4.2.0-0.2, my /var/log/XFree86.0.log keeps filling with this message (II) TRIDENT(0): vgaHWGetIOBase: hwp->IOBase is 0x03d0, hwp->PIOOffset is 0x0000 Version-Release number of selected component (if applicable): XFree86-4.2.0-0.2 RedHat Linux 7.2 Linux xxx 2.4.9-13SGI_XFS_1.0.2 #1 Thu Nov 15 14:28:44 CST 2001 i686 unknown Additional Information: PCI Trident Microsystems Blade 3D PCI video AMD Athlon 1200MHz I will attach lspci, and the first part of the XFree logfile Thanks
Created attachment 43554 [details] First Part of Logfile
Created attachment 43555 [details] lspci -v ; lspci -n
Created attachment 43556 [details] XF86Config-4
How did you manage to use kernel 2.4.9 with XFree86 4.2.0? I thought it required DRM 4.2, which wasn't there.
Trident video hardware isn't supported by DRI at all, so the kernel version wouldn't matter anyway for DRM. The DRM dependancy wasn't added until a later release of my XFree86 package for 4.2.0 anyway, so it wouldn't have prevented installation. One thing I do notice now that I missed before however,is that you're not using the Red Hat Linux kernel. Red Hat does not support using SGI's kernel nor XFS. Can you reproduce this error using a Red Hat kernel?
Hmmm, all my boxes are running XFS now. I did not have this problem with prior versions of XFree, I was using 3.3.6 XF86_SVGA. I updated to 2.4.14 XFS1.0.2, but still have the same messages. Also, this box is routinely crashing for no apparent reason (no logs). That is why I tried the latest XFree, since it would often lock up when a user 'startx', but not every time. This is total ignorance and from left-field, but am I hitting the Athlon kernel bug? cpu family : 6 model : 4 model name : AMD Athlon(tm) Processor stepping : 2 Maybe I can borrow a friends box, and try it there. Thanks for the help!
Hi, I went back and installed the errata XFree for 7.2 (4.1.0-15) and I am no longer getting the error message and X has worked for me so far... I tried to install a newer version of 4.2.0 from rawhide, but that was a dependancy nightmare. I will see how 4.1.0 treats me! Thanks for your time.
Mouse produces tons of vgaHWGetIOBase log messages RED HAT LINUX 8.0 -- XFree86-4.2.0-72 Satellite 2590CDT notebook. RedHat 8.0 installation - fully up2date (as of writing). Running on an old (modified for hibernation) RH kernel 2.4.9-34athena2 [nb. this is probably irrelevent as the logs are produced by XFree and not the kernel] Running XWindows produces a decent logfile on startup, however once it is running moving any attached mouse produces lines upon lines (I've had a log file in and around 200 MB) of: (if rapidly moving mouse rate of generation is around 300KB / 25 seconds - during normal use a few hours usually generate a couple tens of megabytes...) (II) TRIDENT(0): vgaHWGetIOBase: hwp->IOBase is 0x03d0, hwp->PIOOffset is 0x0000 this is interspersed with a few (correctly generated): (II) PM Event received: Power Status Change besides this (and the startup messages) the log is empty. I can produce any additional info you may need, I think it's just a debugging line in the source which hasn't been turned off. [this happens with both builtin mouse and USB logitech 2btn+roller mouse, it seems that displaying the HW mouse cursor produces a log line per display...]
The latest up2date rpms of XWindows (4.2.1-20) solve this problem - However they seem to be compiled with debugging enabled?