Bug 1067
| Summary: | USB printer is not seen anymore | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Xavier (kernel) |
| Component: | USB | Assignee: | Randy Dunlap (rdunlap) |
| Status: | REJECTED INSUFFICIENT_DATA | ||
| Severity: | high | CC: | nacc |
| Priority: | P2 | ||
| Hardware: | i386 | ||
| OS: | Linux | ||
| Kernel Version: | 2.6-test3 (and 2.5.75) | Subsystem: | |
| Regression: | --- | Bisected commit-id: | |
| Bug Depends on: | |||
| Bug Blocks: | 5089 | ||
| Attachments: |
Output of lspci -v
Output of cat /proc/bus/usb/devices with 2.4.20 Output of cat /proc/bus/usb/devices with 2.6.0-test3 Output of dmesg with 2.4.20 Output of dmesg with 2.6.0-test3 .config files used to compile 2.6.0-test3 |
||
|
Description
Xavier
2003-08-09 15:00:19 UTC
Created attachment 628 [details]
Output of lspci -v
Created attachment 629 [details]
Output of cat /proc/bus/usb/devices with 2.4.20
Created attachment 630 [details]
Output of cat /proc/bus/usb/devices with 2.6.0-test3
Created attachment 631 [details]
Output of dmesg with 2.4.20
Created attachment 632 [details]
Output of dmesg with 2.6.0-test3
Created attachment 633 [details]
.config files used to compile 2.6.0-test3
What are the kernel log messages that occur when you plug your printer in? If you plug it in after booting, does that help? It does not matter whether I plug the printer before or after I boot the machine. I booted the machine with the label printer unplugged and later on plugged it in. The following lines appear in the /proc/bus/usb/devices T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=12 MxCh= 0 D: Ver= 0.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 0 #Cfgs= 0 P: Vendor=0000 ProdID=0000 Rev= 0.00 and here are the lines appended to dmesg hub 1-0:0: port 1, status 101, change 1, 12 Mb/s hub 1-0:0: debounce: port 1: delay 100ms stable 4 status 0x101 hub 1-0:0: new USB device on port 1, assigned address 2 drivers/usb/core/message.c: usb_control/bulk_msg: timeout Is the working 2.4.20 version done with a plain, unchanged 2.4.20, or is it patched? The working install uses a plain vanilla 2.4.20 Linux polly.a.la.maison 2.4.20 #3 SMP Sat Aug 9 22:55:23 CEST 2003 i686 Pentium III (Coppermine) GenuineIntel GNU/Linux Sorry, but haven't been the USB printer support moved to userspace via libusb in current 2.6 kernels? usblp kernel driver is still used for usb printers. not libusb. Xavier, could you test with 2.6.11-rc4? Does the problem still occur? Thanks, Nish I recently spent several days trying to detect my UPS, an MGE Ellipse USB. I found a page (http://linuxfr.org/2003/08/25/13729.html), where a guy said that my problem with the HID module was similar to this one, even if it concerns printers. So that's why I answer to Nishanth, hoping it might help. I tried 2.6.11-rc4, as requested by Nishanth, but it still doesn't work For the moment, the only working kernel for me is still kernel 2.4.20 with USB_TIMEOUT option. One option, I think, may be to increase USB_CTRL_SET_TIMEOUT/USB_CTRL_GET_TIMEOUT directly in the code. This is mentioned in the comment in usb.h: * timeouts, in milliseconds, used for sending/receiving control messages * they typically complete within a few frames (msec) after they're issued * USB identifies 5 second timeouts, maybe more in a few cases, and a few * slow devices (like some MGE Ellipse UPSes) actually push that limit. When the Ellipse UPS was mentioned and I saw the comment, I thought maybe there might be connection. And the fact that you are seeing timeouts. I am not sure what a more reasonable value would be, though. Maybe 10000, that would be 10 seconds, which I think should be more than long enough. Thanks for the advice Nishanth, and sorry for the delay (holidays...). I will modify the given options in the code and re-try with several kernels as soon as possible, and report here if it works. Any update on trying to increase the timeout value? Thanks, Nish Closing this bug due to long delay from last report. If this still shows up on 2.6.13-rc6 or greator, please reopen this. |