Hi, On Mon, Mar 10, 2003 at 09:45:11PM +0100, Paul Stravers wrote: > Every time I use my Nikon Coolscan IV LS-40 USB scanner the usb driver > crashes after scanning a small number of frames with the message:
What do you mean by crashing? Do you get a kernel oops or does the kernel freeze? > Mar 10 03:13:48 kletskop kernel: usb_control/bulk_msg: timeout > Mar 10 03:13:48 kletskop kernel: scanner.c: write_scanner: NAK received. Do you get this message after 60 seconds of "nothing happens"? The timeout for writing is 60 seconds. Do you know if there is any bulk write that may take longer with this scanner? You can change the timeout in the source code (scanner.c) but I don't think that will help. Search for the function write_scanner and on this function for result = usb_bulk_msg. Change the last argument from 60*HZ to e.g. 120*HZ. > This is with linux-2.4.20 and in fact it has been a problem with all > stable linux kernels that I have tried. Kernels <= 2.4.18 would crash > within a few seconds, while kernels >= 2.4.19 do a decent job until they > crash after some 10-15 minutes scan time. There was no change in the kernel USB scanner driver between 2.4.18 and 2.4.19. There were some changes for 2.4.20 but I don't think they could have any influence on the behaviour you noticed. You could try 2.4.21-pre5 which has the latest scanner driver with quite some updates. Please also enable debuging in drivers/usb/scanner.h (remove the "//2 from #define DEBUG). Please send me the debugging messages that occure during a failed scan. > I am using vuescan which is also available for windows; when booting my > computer in Windows the exact same hardware runs vuescan without > problems, so I am assuming it is a problem with the linux usb driver. To > be sure I have bought a new usb pci card (uhci instead of the > motherboard ohci) but this did not help. So the problem is with both the > usb-uhci and usb-ohci module (or something that they share). If it's not the scanner driver: USB core or hub driver? Or it's a problem with vuescan. There may be small timing differences between Windows and Linux or other hardware oddities that lead to different behaviour. Is the scanner connected to the root hub directly or do you use any external hubs or a USB extension cable? > I saw a message on this mailing list by Martin Thornquist dated > 2002-05-30 where he described the same problem. In his description the first error occurs in read_scanner not write_scanner. That error was a "funky result", not a timeout. > Could anybody out there > suggest a remedy? If you need more details, please let me know. Which USB transfers cause this problem? Are there any problems with the read transfers before write? Does SANE work instead of vuescan? Or do you get the same timeouts? I don't know if that's the same scanner but a "Nikon LS 40 ED" is listed as supported device. You can use current versions of sane with libusb instead of the kernel scanner driver. If it works with libusb but not the kernel driver we know who is to blame :-) Bye, Henning ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
