| Summary: | modprobe sis-agp fails with 2.6.13-gentoo | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Volker Hemmann <volkerarmin> |
| Component: | [OLD] Core system | Assignee: | Gentoo's Team for Core System packages <base-system> |
| Status: | RESOLVED TEST-REQUEST | ||
| Severity: | minor | CC: | kernel, rusty |
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | x86 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
|
Description
Volker Hemmann
2005-09-02 16:56:25 UTC
Hi,
ok, I found it, it is not the kernel!
good news everybody, it is modprobe.
modprobe sis-agp --> agpgart is loaded, sis-agp not
insmod /lib/modules/2.6.13-gentoo/kernel/drivers/char/agp/sis-agp.ko --->
sis-agp&agpgart are loaded, agp works.
but why? I made a depmod -a after the first problems, and it solved not the problem.
Plus I am using this hardware setup/kernel-config combination for a long time and
this was the first time that this happened
So I change it to minor - it is 2:45 in the morning and I panic'ed ;)
Hi, one more point: after sis-agp did not modprobe, I first rebooted, without changes, than build a vanilla kernel.org 2.6.13, which after the reboot did not modprobe sis-agp too. After that I rebooted into 2.6.13-gentoo again, did checked the logs, did a lot of modprobe than the insmod and bingo, it worked. My main problem is: Please post the output of "modprobe -v sis-agp" and the messages which appear in dmesg after you run that. Hi, modprobe -v sis-agp insmod /lib/modules/2.6.13-gentoo/kernel/drivers/char/agp/agpgart.ko dmesg part of that 'event': [17179693.704000] Linux agpgart interface v0.101 (c) Dave Jones and when I rmmod agpgart and insmod /lib/modules/2.6.13-gentoo/kernel/drivers/char/agp/sis-agp.ko 17179739.208000] Linux agpgart interface v0.101 (c) Dave Jones [17179757.632000] agpgart: Detected SiS 746 chipset [17179757.656000] agpgart: AGP aperture is 256M @ 0xd0000000 [17179785.332000] ACPI: PCI Interrupt 0000:01:00.0[A] -> GSI 16 (level, low) -> IRQ 21 [17179785.332000] NVRM: loading NVIDIA Linux x86 NVIDIA Kernel Module 1.0-7676 Fri Jul 29 12:58:54 PDT 2005 [17179785.644000] agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0. [17179785.644000] agpgart: Putting AGP V3 device at 0000:00:00.0 into 8x mode [17179785.644000] agpgart: SiS delay workaround: giving bridge time to recover. [17179785.660000] agpgart: Putting AGP V3 device at 0000:01:00.0 into 8x mode [17179785.812000] agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0. [17179785.812000] agpgart: Putting AGP V3 device at 0000:00:00.0 into 8x mode [17179785.812000] agpgart: SiS delay workaround: giving bridge time to recover. [17179785.828000] agpgart: Putting AGP V3 device at 0000:01:00.0 into 8x mode oh, em, to clear it up a little bit: 17179739.208000] Linux agpgart interface v0.101 (c) Dave Jones [17179757.632000] agpgart: Detected SiS 746 chipset [17179757.656000] agpgart: AGP aperture is 256M @ 0xd0000000 [17179785.332000] ACPI: PCI Interrupt 0000:01:00.0[A] -> GSI 16 (level, low) -> IRQ 21 that was after insmod'ing sis-agp.ko and 'standard' output I got with all older kernels. This stuff is just starting X and also totally normal: [17179785.332000] NVRM: loading NVIDIA Linux x86 NVIDIA Kernel Module 1.0-7676 Fri Jul 29 12:58:54 PDT 2005 [17179785.644000] agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0. [17179785.644000] agpgart: Putting AGP V3 device at 0000:00:00.0 into 8x mode [17179785.644000] agpgart: SiS delay workaround: giving bridge time to recover. [17179785.660000] agpgart: Putting AGP V3 device at 0000:01:00.0 into 8x mode [17179785.812000] agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0. [17179785.812000] agpgart: Putting AGP V3 device at 0000:00:00.0 into 8x mode [17179785.812000] agpgart: SiS delay workaround: giving bridge time to recover. [17179785.828000] agpgart: Putting AGP V3 device at 0000:01:00.0 into 8x mode Which version of module-init-tools is in use? Hi,
emerge -s:
sys-apps/module-init-tools
Latest version available: 3.2_pre7-r1
Latest version installed: 3.2_pre7-r1
Looks like a module-init-tools bug, apparently it's not even _trying_ to load sis-agp when you run "modprobe sis-agp"... Hi, hmm, if it is a module-init-tool bug, should it not have shown itself with earlier kernels? Because AFAIR 2.6.12-X were working fine with the same module-init-tools version. Hi, I have to correct mmyself - just installed 2.6.12-gentoo-r9 to make sure - and modprobe sis-agp does not work there too. does it work with older module-init-tools ? Hi, I tried module-init-tools 3.1-r1 and sis-agp fails there too. if you have a minute, think you could take a look at this issue for us rusty ? This seems very strange; I suspect the kernel is trying to load modules itself, confusing the issue. I suggest creating a simple script in /tmp/modprobe, like so: #! /bin/sh echo ARGS: "$@" >> /tmp/modprobe.log exec /sbin/modprobe "$@" Then chmod a+x /tmp/modprobe, and echo /tmp/modprobe > /proc/sys/kernel/modrpobe. Now, you'll get a message in /tmp/modprobe.log if the kernel is trying to load things. If you do a modprobe -v sis-agp, what happens now? Puzzled, Rusty. Hi, I did and there was no modprobe.log in /tmp. when I did modprobe -v sis-agp If I do it I get this: modprobe -v sis-agp insmod /lib/modules/2.6.13-gentoo/kernel/drivers/char/agp/agpgart.ko but no /tmp/modprobe.log BUT when I start X, I have suddenly the modprobe.log, which has this: cat /tmp/modprobe.log ARGS: nvidia so at least it works - for nvidia&co but not for sis-agp Hi, I just updated to 2.6.13-gentoo-r1 and upon reboot, sis-agp.ko was modprobed correctly. Fails for me with 2.6.13-gentoo-r3,r2 (and vanilla 2.6.14-rc3), and "an indeterminate number of" previous gentoo kernels. Specific insmod works as listed in the other reports in this thread. Mobo is ugly. cheap K7S5A. Stale bug. Reopen if you can reproduce this w/ current module-init-tools and uptodate kernels. Thanks. I don't have the hardware anymore, so testing is impossible ... I am fine with closing it. If anybody still using sis-boards would have had hit this bug, he would have left a comment... so I am pretty sure, that this bug died a long time ago (and I completly forgot its existence). |
