On Tue, 2005-11-29 at 11:31 -0500, Alan Stern wrote: > On Tue, 29 Nov 2005, Steve Bangert wrote: > > > Five days later the printer decides to stop working again, so here's a > > trace when i insmod the printer driver ( printer inop.) > > You know, this suggests strongly that there's something funny going on > with your printer. After all, you've been running the same software for > the last five days, right?
right > The only reason it should stop working now is > if the printer has some weird internal problem. keep in mind that i can reboot into XP and the printer works just fine, consistently. > It's related to the > drivers only in that they aren't able to convince the printer to start > working again once it has stopped. Would a "SOFT_RESET" convince the printer to start working? check out sections 4.2 and 4.2.3 of the usb printer spec at: http://www.usb.org/developers/devclass_docs/usbprint11.pdf > > ... > > cf4f3800 3833320353 C Bi:004:02 0 10 = 0000000a 00007f00 0080 > > > And here's a repost of a working printer when i insmod the printer > > driver: > > ... > > d3f6e500 3212296796 C Bi:003:02 0 0 > > Everything was identical except the two lines shown above. They show a > status message sent back by the non-working printer, so if a SOFT_RESET was implemented when this status message was received maybe it could be cleared, and get the printer working again, does that sound reasonable? > > whereas the working > printer sent an empty message. I have no idea what the message means, > though. Nor do I know which program was responsible for issuing the read > that provoked the message. Do you have hal running on your system? > yes > > > > I tried getting a usbmon trace while doing an insmod of uhci-hcd.ko but > > there's a chicken and egg scenario, /sys/kernel/debug needs to be > > mounted before a usbmon snapshot is taken, but it can't be mounted > > unless uhci-hcd.ko is loaded first. > > Well, that's not _quite_ right. You can mount /sys/kernel/debug any time > you want, and you can modprobe usbmon which will add the .../mon > subdirectory. But the file you need to read won't exist until uhci-hcd is > loaded. > > I suppose you could write a little shell script to read the debugging file > in a loop. When the file doesn't exist the read will fail immediately, so > the script will loop and try again. When the file does exist the read > will succeed and just sit around until you kill it. Then while the script > is running you insmod uhci-hcd.ko. > > Other experiments you might try: If the printer isn't working, see how it > behaves when you do rmmod uhci-hcd followed by modprobe uhci-hcd. Or > leave the printer turned on while you boot, but keep the USB cable > unplugged until the boot-up procedure is finished. > Still no printer output, seems the only think that gets it going is power cycling the printer. > Steve ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
