The issue was as follows: Every time I tried to start Xorg via the startx command is I would run into a hard freeze where all input was frozen (to my knowledge since I didn't know of sysrq and didn't have it compiled into the kernel prior to then). If the console was allowed to blank normally that it would result in the same thing. All cases were involving the use of the nvidia driver and kernel versions beyond-and possibly including-the gentoo-dev-sources 2.6.9-r4 and greater versions. By not compiling in APM support the system appears to be running smoothly now, although I do have errors dealing with Xorg's config. Reproducible: Sometimes Steps to Reproduce: 1. Compile APM support into kernel. 2. Install kernel and update all conf files as necessary. 3. Reboot. 4.a. Let console blank. 4.b. Type in startx and press enter. 4.c. Type in startx, enter regular Xorg session and exit. Actual Results: Hard crashes. Expected Results: Not crashed. //EMERGE INFO bash-2.05b# emerge info Portage 2.0.51-r3 (default-linux/x86/2004.3, gcc-3.3.4, glibc-2.3.4.20040808- r1, 2.6.9-gentoo-r13 i686) ================================================================= System uname: 2.6.9-gentoo-r13 i686 Intel(R) Pentium(R) 4 CPU 2.40GHz Gentoo Base System version 1.4.16 Autoconf: sys-devel/autoconf-2.59-r5 Automake: sys-devel/automake-1.8.5-r1 Binutils: sys-devel/binutils-2.15.90.0.1.1-r3 Headers: sys-kernel/linux26-headers-2.6.8.1-r1 Libtools: sys-devel/libtool-1.5.2-r7 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-O2 -march=pentium4 -fomit-frame-pointer" CHOST="i686-pc-linux-gnu" COMPILER="" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/share/ config /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" CXXFLAGS="-O2 -march=pentium4 -fomit-frame-pointer" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs autoconfig ccache distlocks sandbox sfperms" GENTOO_MIRRORS="http://gentoo.osuosl.org http://distro.ibiblio.org/pub/Linux/distributions/gentoo" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="x86 X acl alsa anthy apm arts avi berkdb bitmap-fonts cdr cjk crypt cups dvd encode esd fam foomaticdb fortran gdbm gif gpm gtk gtk2 imlib ipv6 jpeg libwww mad mikmod motif mpeg ncurses nls nptl oggvorbis opengl oss pam pdflib perl png python qt quicktime readline sdl slang spell ssl svga tcpd tiff truetype userlocale xml2 xmms xv zlib" //KERNEL MAKEFILE Goto http://students.washington.edu/youshi10/linux/conf/bk.cfg //KERNEL MAKE COMMANDS bash-2.05b# cat /root/mkkernel.sh cd /usr/src/linux; make menuconfig; make; make bzImage modules modules_install; cp System.map /boot/System.map-gentoo-2.6.9-r13; cp arch/i386/boot/bzImage /boot/gentoo-2.6.9-r13 //IMPORTANT EBUILDS INFO bash-2.05b# emerge -p --nodeps nvidia-kernel nvidia-glx xorg-x11 opengl-update These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] media-video/nvidia-kernel-1.0.6111-r3 [ebuild R ] media-video/nvidia-glx-1.0.6111 [ebuild R ] x11-base/xorg-x11-6.8.0-r3 [ebuild R ] x11-base/opengl-update-1.8.2 //Xorg CONF FILE goto http://students.washington.edu/youshi10/linux/conf/xorg.conf
Can you reproduce this with the "nv" driver?
Created attachment 47897 [details] Photo of Kernel Panic Hmm, I've been experiencing very similar symptoms with my Dell Inspiron 9100 (laptop). Interestingly, I do not have APM compiled into my kernel (2.6.9-r13, though r6 exhibits the same problem); I do, however, have ACPI compiled in. Removing ACPI from the kernel also removes the problem. I wonder if this is the same problem, or a different one altogether. Both technologies are used for power management -- do they share some code? I was lucky enough to switch to a virtual terminal (Ctrl-Alt-F1) just after logging in, but before the lock up, and caught the output of the kernel panic in the terminal. I didn't know how to dump the output to a text file (sorry), since my computer had locked up, but I did take a picture which I've attached. Try switching to a terminal to see if you get the same output. Let me know, so that we can determine if this problem is the same. If it is the same, I have a few more symptoms that may be of some help in solving the problem. Thanks, all, for your help; Kris Kerwin kkerwin@insi__REMOVE_ME__ghtbb.com
Please reopen when you respond to comment #1.
Using nv with Xorg 6.8.0 still hard crashed the system once I finished my session using the newer version of the kernel, gentoo-dev-2.6.10-r4. This only occurs when I compile in Power Management support. Therefore the issue is most likely not one dealing with the nvidia drivers, but rather a kernel/Xorg issue. HOWEVER, considering that APM console blanking also hard crashes my tty when I allow it to blank, it sounds primarily like an issue with the kernel and possibly some other driver.
Quoting Garrett Cooper: "it sounds primarily like an issue with the kernel and possibly some other driver." This rather intrigues me. I have since found out that my problem is instead one with the ndiswrapper out-of-tree kernel module that allows many proprietary wireless network cards to work under Linux. Dell is a large reseller of such cards in many of their machines. The corresponding bug report may be found at http://bugs.gentoo.org/show_bug.cgi?id=73747 For myself, I noticed that ndiswrapper was dependent on ACPI and APM, so removing either one disabled ndiswrapper, preventing it from crashing the system. This allowed my situation to at least mimic yours, if it is not indeed the same. I'm curious, since we both have Dell machines, and since our problems are indeed so similar, if by some long shot you do use the ndiswrapper module? Thanks for your help. Kris Kerwin kkerwin@insi__REMOVE_ME__ghtbb.com
Hmmm.. I actually run a Dell Dimension 8200 and I do not have a wireless card. I should have listed all my specs... Dell Dimension 8200 Intel Pentium 4 2.4 GHz 533 MHz FSB 512 Mb 1033 MHz RD Ram nVidia GeForce4 Ti 4200 Realtek 8139 10/100 MBit Networking Unknown maker ieee1394 card Turtle Beach Santa Cruz card M$ Intellimouse 3.0 Hauppage TV card (PCI without radio) I'll be sure to post my new kernel config sometime soon.
Ok, can you confirm that this issue exists without using any binary drivers (e.g. ndiswrapper), without the nvidia module even being *loaded* (check with lsmod), using the nv driver under Xorg Ideally you should test this under 2.6.11-rc1 too, if possible.
Please reopen when you reply to comment #7
As you can see, I don't have ndiswrapper bash-2.05b# cd /usr/src/linux; cat .config | grep -i NDIS bash-2.05b# bash-2.05b# emerge -C ndiswrapper --- Couldn't find ndiswrapper to unmerge. >>> unmerge: No packages selected for removal. I will try 2.6.11 soon. I have just been busy with schoolwork and don't want to deal with reconfig'ing right away.
I tried development-sources-2.6.11-r3 to see if that caused an issue, but once again X crashed. I also had emerged the latest RC of Xorg in an attempt to see if that was an issue, but it wasn't. So I've narrowed the issue down to a few things with APM: [] Enable APM at boot [] Enable console blanking [] Use real mode APM BIOS call to power off
One thing that I've been considering recently is that it may be an issue with permissions dealing with /dev/apm_bios and udev. Could someone post a valid set of permissions for apm_bios for me to test?
Looks like APM is at fault then Please file a bug for this at http://bugzilla.kernel.org
This particular set of symptoms have been resolved in further versions of the kernel / Xorg (last operating version of the Linux kernel was 2.6.18 and X.org was 7.0 latest, as of 2006.12.16). No longer running Linux though so I cannot comment upon the issue further really or provide debugging information. Resolving on my own in the upstream ticket.