lkml.org 
[lkml]   [2011]   [Jul]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Weird USB over-current messages in recent kernels...
Hello.

Alan Stern wrote:

>> Seen during boot the last few days on my laptop:

>> running 3.0.0-rc4-mmotm0622:
>> Jul 2 13:08:37 turing-police kernel: [ 2.788108] hub 2-0:1.0: over-current condition on port 1
>> Jul 2 13:08:37 turing-police kernel: [ 2.990088] hub 2-0:1.0: over-current condition on port 2
>> Jul 2 13:08:37 turing-police kernel: [ 3.193073] hub 6-0:1.0: over-current condition on port 1
>> Jul 2 13:08:37 turing-police kernel: [ 3.395088] hub 6-0:1.0: over-current condition on port 2

>> running 3.0.0-rc5-mmotm0630:
>> Jul 5 09:43:36 turing-police kernel: [ 203.686082] hub 6-0:1.0: over-current condition on port 1
>> Jul 5 09:43:36 turing-police kernel: [ 203.969179] hub 6-0:1.0: over-current condition on port 2
>> Jul 5 09:43:36 turing-police kernel: [ 204.217077] hub 6-0:1.0: over-current condition on port 1

>> The problem? I don't see any devices that would *cause* the condition:

>> The device 5 and 6 on bus 1 are attached to the docking station, the Jul 2 boot
>> was undocked and those two devices weren't present. Nobody home on bus 2 or 6
>> at any time, as far as I can tell.

>> Any ideas?

> That message was added recently, which may explain why you haven't seen
> it before.

Wasn't there a similar dev_err() message before that patch that added this
message (the previous message was printed on any over-current signal change)?

> For more debugging, please collect a usbmon log for bus 2 or bus 6 (see
> Documentation/usb/usbmon.txt). In fact, you might try doing this for
> an earlier kernel as well.

> It may be that your host controllers claim that an over-current
> condition exists when it really doesn't, or it may be that those ports
> are wired incorrectly and really do have an over-current condition.

If it's EHCI driver, there's module option to suppress over-current
checking, called 'ignore_oc'. It should help with bogus over-current signalling...

> Alan Stern

WBR, Sergei


\
 
 \ /
  Last update: 2011-07-06 17:43    [W:0.096 / U:0.424 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site