On Thu, 8 Jul 2004, Rogério Brito wrote: > Well, I think that these are some good news: it printed that the actual > length was 8, as you can see below. > > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > (...) > usb 1-1.1: new full speed USB device using address 3 > usb 1-1.1: new device strings: Mfr=1, Product=4, SerialNumber=0 > usb 1-1.1: default language 0x0409 > usb 1-1.1: Product: USB Mass Storage Device > usb 1-1.1: Manufacturer: Leading Driver Co.,LTD. > usb 1-1.1: hotplug > usb 1-1.1: adding 1-1.1:1.0 (config #1, interface 0) > usb 1-1.1:1.0: hotplug > (...) > scsi0 : SCSI emulation for USB Mass Storage devices > usb-storage: queuecommand called > usb-storage: *** thread awakened. > usb-storage: Command INQUIRY (6 bytes) > usb-storage: 12 00 00 00 24 00 > usb-storage: Bulk Command S 0x43425355 T 0x1 L 36 F 128 Trg 0 LUN 0 CL 6 > usb-storage: usb_stor_bulk_transfer_buf: xfer 31 bytes > hub_irq: actual_length = 8 > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > > I still haven't compiled the kernel with the 8 in the place of the 64, > though.
That will be the real test. But as long as actual_length <= 8, it should work okay. > [Problems with SCSI errors] > > Well, I put some logs of what I got at <http://www.ime.usp.br/~rbrito/usb>, > since they are long and I don't know exactly what is relevant. > > This time, differently from the log that I generated above, I didn't issue > the "echo 4 > /proc/driver/uhci/..." commands, You don't have to do that any more. That was only intended to help track down the problem with the internal hub, which we now understand sufficiently well. > but still usb-storage > generated huge amounts of logs and, as a result, the latter logs had enough > data in them that they don't include parts of the earlier logs. I don't > know if there is any discontinuity in the contents of the logs. > > The logs were taken in single user mode, without hotplug scripts and right > after I loaded the usb-hci module and the usb-storage module. Then, I > plugged the drive, mounted it, unmounted it and unplugged it. > > I hope that their names reflect the order in which I took these actions. > Let me know if you want me to send the information in a different manner in > future e-mails. I see the errors you were worried about. They aren't serious -- for example, you shouldn't have any trouble mounting the disk and reading or writing files. The errors arise because your computer is sending commands that the drive doesn't understand, and the drive responds with an indication that it couldn't carry out the command but no corresponding error code to explain why. Seeing that the command failed but not knowing why it failed, the system prints a message in the log. The particular commands causing your problem are PREVENT-ALLOW MEDIUM REMOVAL; they tell the drive to lock the medium in place or unlock it. Your device probably doesn't have removable media though (that is, you can't remove the storage medium from the drive the way one can with a Zip drive), so naturally it doesn't understand. But those commands get sent because the drive tells the computer that it _does_ have removable media -- probably another bug in the drive. If you want, you can prevent Linux from sending those PREVENT-ALLOW commands, which will prevent the error messages from appearing in your log. Here's how to do it: At any time after the scsi_mod module has been loaded and before you plug in the drive, run this command: echo 'Prolific:USB Flash Disk:0x80400' >/proc/scsi/device_info However this feature was only added very recently, and it probably won't appear in the standard kernels until 2.6.8 comes out. Alan Stern ------------------------------------------------------- This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
