lkml.org 
[lkml]   [2020]   [Jul]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: kworker/0:3+pm hogging CPU
On Mon 20-07-20 13:48:12, Alan Stern wrote:
> On Mon, Jul 20, 2020 at 07:45:30PM +0200, Michal Hocko wrote:
> > On Mon 20-07-20 13:38:07, Alan Stern wrote:
> > > On Mon, Jul 20, 2020 at 06:33:55PM +0200, Michal Hocko wrote:
> > > > On Mon 20-07-20 11:12:55, Alan Stern wrote:
> > > > [...]
> > > > > sudo echo 'module usbcore =p' >/debug/dynamic_debug/control
> > > > >
> > > > > Then wait long enough for some interesting messages to appear in the
> > > > > kernel log (it should only take a few seconds if the worker thread is as
> > > > > busy as you say) and collect the output from the dmesg command.
> > > > >
> > > > > To turn dynamic debugging back off when you're finished, use the same
> > > > > command with "=p" changed to "-p".
> > > >
> > > > [ 95.395815] hub 2-0:1.0: state 7 ports 6 chg 0006 evt 0000
> > > > [ 95.395824] usb usb2-port1: status 0088, change 0000, 5.0 Gb/s
> > > > [ 95.395838] usb usb2-port2: status 0088, change 0000, 5.0 Gb/s
> > > > [ 95.396016] hub 2-0:1.0: hub_suspend
> > > > [ 95.396023] usb usb2: bus auto-suspend, wakeup 1
> > > > [ 95.396029] usb usb2: bus suspend fail, err -16
> > >
> > > Okay, that's the problem. You may get more specific information about
> > > the suspend failure if you turn on dynamic debugging for the xhci-hcd
> > > module (same command as above but with xhci-hcd instead of usbcore).
> >
> > Would that help to pinpoint the problem down?
>
> It might. I can't say for sure without seeing the output.

echo 'module xhci-hcd =p' >/sys/kernel/debug/dynamic_debug/control

doesn't yield to any output. I have tried both with and without usbcore
debug output enabled. In the former there is no new output and for the
later I get the same I have posted earlier.
--
Michal Hocko
SUSE Labs

\
 
 \ /
  Last update: 2020-07-20 20:17    [W:0.100 / U:0.168 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site