When I startx with my PCMCIA network card plugged in and enabled, X takes a few minutes to start up. The computer is now very slow overall and the network card will not function. When I look in /var/log/messages (I'm using syslog-ng if that makes any difference) the log is being flooded with the following two messages: Nov 28 15:17:50 localhost eth0: Host error, FIFO diagnostic register 0000. Nov 28 15:17:50 localhost eth0: PCI bus error, bus status 80000020. I cannot stop the network card via '/etc/init.d/pcmcia stop' or '/etc/init.d/net.eth0 stop'. If I pull the PCMCIA card out, my computer gets a kernel panic passed to it from syslog-ng. I can avoid this problem by using startx without my network card plugged. Once X is running, I plug my network card in, it is recognized, and I receive no errors. My network card itself is configured by pcmcia-cs-3.2.4 and uses the drivers provided by it. Reproducible: Always Steps to Reproduce: 1.Plug in PCMCIA network card 2.startx Actual Results: The computer's speed is signifigantly diminished. The logger is getting flooded with: Nov 28 15:17:50 localhost eth0: Host error, FIFO diagnostic register 0000. Nov 28 15:17:50 localhost eth0: PCI bus error, bus status 80000020. The network card also stops working. Expected Results: Their should be no errors or conflicts between X and my network card. Portage 2.0.49-r15 (default-x86-1.4, gcc-3.2.3, glibc-2.3.2-r3, 2.4.20-gentoo-r9) ================================================================= System uname: 2.4.20-gentoo-r9 i686 Pentium III (Coppermine) Gentoo Base System version 1.4.3.10 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-O3 -march=pentium3 -fprefetch-loop-arrays -funroll-loops" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /var/qmail/control /usr/kde/2/share/config /usr/kde/3/share/config /usr/X11R6/lib/X11/xkb /usr/kde/3.1/share/config /usr/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" CXXFLAGS="-O3 -march=pentium3 -fprefetch-loop-arrays -funroll-loops" DISTDIR="/usr/portage/distfiles" FEATURES="sandbox ccache autoaddcvs" GENTOO_MIRRORS="http://gentoo.oregonstate.edu 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.us.gentoo.org/gentoo-portage" USE="x86 apm avi crypt cups encode foomaticdb gif gpm imlib kde libg++ libwww mad motif mpeg ncurses pdflib png quicktime spell truetype xml2 xv zlib gdbm berkdb slang readline arts svga java X sdl tcpd pam ssl perl python oggvorbis gtk qt opengl -gnome -esd gtk2 mozilla pcmcia pnp -nls dvd oss xmms jpeg sse mmx -mikmod fbcon directfb tiff" dmesg | grep eth0 eth0: 3Com 3CCFE575CT Tornado CardBus at 0x280, 00:00:86:59:80:f6, irq 9 eth0: Setting full-duplex based on MII #0 link partner capability of 01e1. Kernel line in grub: kernel (hd0,0)/boot/bzImage root=/dev/hda3 pci=usepirqmask vga=792
Can you please try vanilla-sources * 2.4.20 * to confirm whether this is a Gentoo bug?
The same problem occurred with the vanilla sources. I emerged vanilla source 2.4.20, setup the /usr/src/linux symlink to it, and built my kernel as close as I could to gentoo-sources-2.4.20-r9. After the kernel booted, I needed to reemerge pcmcia-cs to get my pcmcia card working with the the new kernel. After I did that, I rebooted and started X with the same result happening.
I'm not too sure if we can do that much if this is a problem in the underlying vanilla core. Can I suggest that you try a recent 2.4 series kernel, like a 2.4.23 vanilla or a 'ck' kernel? If this still persists in the latest vanilla kernel, then this should be filed upstream.
The problem is still persisting with the later vanillas.
We can't really do much here as this is a problem in the underlying vanilla sources. Can I recommend that you try the 2.6 sources and see if the problem persists - if it does, please file a bug at <http://bugme.osdl.org> or send mail to the LKML <http://www.lkml.org>. Thanks a lot.
I haven't found a way of getting my NIC to work with the 2.6 kernel. Because of that issue, testing it with the 2.6 kernel has been an impossiblity.