The device is changing it's serial number -- likely it's trying to use it's USB bus address in place of a serial number.
In other words, the device is busted. 2.5/6 treat things differently, so this isn't a problem. Matt On Sun, Oct 26, 2003 at 03:12:00AM +0200, Vadim wrote: Content-Description: signed data > This is happening right now and is 100% reproducible on two > computers. > > First computer is a dual Athlon MP 2000+, running the Gentoo > 2.4.20-gentoo-r7 kernel. The second one is a Toshiba 470CDT > laptop running the 2.4.22 vanilla kernel. > > The laptop currently doesn't have a working network card so I'm > using a PQI TravelFlash card reader to exchange data. > > The problem is that every time I remove the card reader and > insert it again it gets a new device. The old ones remain in > /proc/partitions and get set to 1GB size if I try to access > them. Otherwise they remain showing the 64MB of the removed card > reader. > > For example, on the Athlon, /proc/partitions looks like this > after inserting and removing the card several times: > > 8 0 1048575 scsi/host1/bus0/target0/lun0/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 16 61440 scsi/host1/bus0/target0/lun1/disc 160 6 174 > 484 23 4012 4035 26798 0 2866 27282 > 8 17 61375 scsi/host1/bus0/target0/lun1/part1 159 3 > 166 478 23 4012 4035 26798 0 2860 27276 > 8 32 1048575 scsi/host1/bus0/target0/lun2/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 48 1048575 scsi/host1/bus0/target0/lun3/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 64 1048575 scsi/host2/bus0/target0/lun0/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 80 61440 scsi/host2/bus0/target0/lun1/disc 1 3 8 6 0 > 0 0 0 0 6 6 > 8 81 61375 scsi/host2/bus0/target0/lun1/part1 0 0 0 0 > 0 0 0 0 0 0 0 > 8 96 1048575 scsi/host2/bus0/target0/lun2/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 112 1048575 scsi/host2/bus0/target0/lun3/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 128 1048575 scsi/host3/bus0/target0/lun0/disc 2 6 16 0 > 0 0 0 0 0 0 0 > 8 144 61440 scsi/host3/bus0/target0/lun1/disc 1 3 8 6 0 > 0 0 0 0 6 6 > 8 145 61375 scsi/host3/bus0/target0/lun1/part1 0 0 0 0 > 0 0 0 0 0 0 0 > > I have also noticed that in /proc/scsi/ there is a usb-storage-N > directory created every time I insert the card reader, with the > file inside having a content like: > > Host scsi2: usb-storage > Vendor: Generic > Product: USB Storage Device > Serial Number: 0AEC305000001A006 > Protocol: Transparent SCSI > Transport: Bulk > GUID: 0aec5010aec305000001a006 > Attached: No > > Only the last one has 'Attached' set to 'Yes': > > Host scsi4: usb-storage > Vendor: Generic > Product: USB Storage Device > Serial Number: 0AEC305000001A008 > Protocol: Transparent SCSI > Transport: Bulk > GUID: 0aec5010aec305000001a008 > Attached: Yes > > I'm guessing the kernel is keeping it there for some reason, but > I can't figure out why. I guess it might be trying to allow to > reinsert a device later, but if so, it's definitely not working. > > Here are the USB lines from .config for the laptop. > CONFIG_USB=y > CONFIG_USB_DEBUG=y > CONFIG_USB_DEVICEFS=y > CONFIG_USB_OHCI=y > CONFIG_USB_STORAGE=y > CONFIG_USB_STORAGE_DATAFAB=y > CONFIG_USB_STORAGE_FREECOM=y > CONFIG_USB_STORAGE_ISD200=y > CONFIG_USB_STORAGE_DPCM=y > CONFIG_USB_STORAGE_SDDR09=y > CONFIG_USB_STORAGE_SDDR55=y > CONFIG_USB_STORAGE_JUMPSHOT=y > CONFIG_USB_PWC=y > CONFIG_USB_USBNET=y > > > Perhaps I'm missing something obvious here, but I googled for > this and didn't find anything. Other people seem to have had > this problem, but I haven't found any answers. > > I've also just tried the 2.6.0-test8 kernel, and verified that > the same device is being used every time. > > Thanks in advance > -- > Use the PGP path finder to find a trust path to my key: > http://the.earth.li/~noodles/pathfind.html -- Matthew Dharm Home: [EMAIL PROTECTED] Maintainer, Linux USB Mass Storage Driver It was a new hope. -- Dust Puppy User Friendly, 12/25/1998
pgp00000.pgp
Description: PGP signature
