Printable View
searching about it I found this. I'm wondering if and how is it possible to correct the uncorrect line and make usb 2.0 working? Is it possible to change while I'm using the SO or the unique option is to edit the ISO image (how?) and burn another CD?
Thank you very much
I think it is too late to "modprobe ehci-hcd" once the system is running, because usb-storage module will be loaded by then and
it will probably cause a kernel crash.
Therefore suggest to make a new CD.
To make the CD, you can follow the "Knoppix remastering guide" - it takes a while and requires a fair bit of disk space, but it is an educational experience!
Before building the final ISO image, use this procedure to edit the linuxrc file:
cd /mnt/drive/knoppix-tmp # assuming that is where you are building the release
cp ./master/boot/isolinux/minirt.gz .
gunzip minirt.gz
mkdir /rt
mount -o loop minirt /rt
vi /rt/linuxrc # i.e. change USB2="ehci-hcd.ko" to USB2="ehci-hcd"
umount /rt
gzip minirt
cp minirt.gz ./master/boot/isolinux/minirt.gz
Hope this helps,
eh? crash? why? it works in more than 90% of cases.
Code:sudo modprobe ehci
If manually loading the driver later works, then fine. It's a quick and easy solution.
It crashed 2 out of 2 attempts for me (manually loading USB1.1 ohci-hcd after usb-storage).
Anyway, I have another problem with Knoppix 4.0.2: It freezes at the line "Checking for for USB" (that's a typo in linuxrc).
It seems that:
1. It only occurs if a USB 1.1 multi-chip flash reader is installed, and only about 50% of the time.
2. It only occurs with Knoppix 4.0.2: kernel 2.6.12, not with Knoppix 3.6 kernel 2.4.27 or 2.6.7, and not wtih Ubuntu 5.10 kernel 2.6.12.10
Could it be that an interrupt is raised by this type of hardware soon after the driver is loaded, and maybe at a critical time before the usb-storage module is ready to handle it?
Any tips on debugging this problem would be most appreciated.
Hardware: Asus A7N266: onboard ohci USB; 4 in 1 chip reader permanently connected.
I have encountered the same problem. The board specs are almost identical excepting: Asus A7V8-X. Wondering what to do, this Deb. incarnation uses my modem, all others fail.
Fogcat
Hi Fogcat,
do you have a multi-chip reader permanently connected?
If so, does removing the reader fix the problem?
I think this is worth a new discussion thread, to find out how to debug this problem. It stops me from using Knoppix, which I like. Ubuntu doesn't have the problem, which is hard to fathom because the kernel is almost identical.
I have spent a few days playing around with the startup file linuxrc, and there doesn't seem to be anything wrong with it, apart from a very minor issue I noticed - there is a "nousb2" option that is effectively nullified by the fact that the text matches "nousb" and therefore disables all USB. This is not the problem here, though.
cheers,
I have the same reader (4 in 1) permanently. This is a feature of Asus I think. No work around in the BIOS however; only a legacy USB enable/disable.
I am at a loss here.
I think I misunderstood you. I thought you ment the VIA 4 in 1 chipset on the board. If this is incorrect I apologize for the false lead.
Sorry to post so many. I found this info :http://groups.google.com/group/linux...&start=0&num=2
I think this si pertinent since these are VIA UHCI.
It turns out the problem with freeze ("oops") during startup with OHCI (USB 1.1) is actually a kernel bug, reported here
http://bugzilla.kernel.org/show_bug.cgi?id=5350
It's current status is that patches are available, or I understand it will be fixed in kernel 2.6.15.
For completeness, here is a description of the problem:
With Knoppix 4.0.2, kernel 2.6.12, the startup process may freeze about 50% - 70% of the time, if the motherboard has
USB 1.1 e.g. the Asus A7N266 or similar motherboard, AND there are USB devices attached, such as a flash card reader.
The last messages on screen mention USB devices being detected.
If you install Knoppix to hard disk, there are some more informative messages and a stack trace.
This information led me to the bugzilla problem report.
I was puzzled why Ubuntu with kernel 2.6.12-10 does not exhibit this problem, but did not get to the bottom of this.
One possibility is that the patch is applied to its "-10" kernel. Another possibility is that it appears Ubuntu compiles the USB drivers into the kernel, and does not load them as modules.
Maybe the problem is only evident when loading modules. (pure speculation here)