hotplug/usb.agent consumes all CPU

hotplug/usb.agent consumes all CPU

[Simon Wu]

Hello,

I am using udev with hotplug on kernel 2.6.5 and 2.6.9 (Fedora2 and 3). When I
plug one single cdc_acm device (Nokia wireless cellphones/modem) into
the USB port, dozens of hotplug usb.agent processes invoked and took
away all CPU for 7 to 8 seconds. During this time, the computer
doesn't even accept keyboard input.

If I plug N Nokia phones in the USB ports at the same time, there seem
to have N times of usb.agent processes and Nx7 seconds delay.

Any way to improve it?

Thanks,


Simon


-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: hotplug/usb.agent consumes all CPU

[Kay Sievers]

On Tue, 2005-02-01 at 16:22 -0500, Simon Wu wrote:
> Hello,
> 
> I am using udev with hotplug on kernel 2.6.5 and 2.6.9 (Fedora2 and 3). When I
> plug one single cdc_acm device (Nokia wireless cellphones/modem) into
> the USB port, dozens of hotplug usb.agent processes invoked and took
> away all CPU for 7 to 8 seconds. During this time, the computer
> doesn't even accept keyboard input.
> 
> If I plug N Nokia phones in the USB ports at the same time, there seem
> to have N times of usb.agent processes and Nx7 seconds delay.
> 
> Any way to improve it?

How can you be sure, that the hotplug processes are taking your
cpu and that it's not the kernel with the cdc_acm driver that blocks?

Are you using other USB devices on the system? Any chance to try maybe
a storage device on the same port, if that works as expected?

It would be nice to know if you get the same behavior with a 2.6.10
kernel, if you can get one for your system.

Thanks,
Kay




-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: hotplug/usb.agent consumes all CPU

[Simon Wu]

I was monitoring the processes with top. I disabled (by using exit as
first line on script) usb.agent, and the problem is gone.  The cdc_acm
devices is OK with usb.agent running. So, this solves my problem but
it would be nice to know which program called by usb.agent  was
consuming all the CPU.

By the way, I also noticed the similar behaviour with Redhat 9's
usb.agent script, in which "/sibn/devlabel restart" was run and
consumes all the CPU. When I comment this line out, everything is OK.

I will try kernel 2.6.10 later. But the above 

Simon


On Tue, 01 Feb 2005 23:33:39 +0100, Kay Sievers <kay.sievers@vrfy.org> wrote:
> On Tue, 2005-02-01 at 16:22 -0500, Simon Wu wrote:
> > Hello,
> >
> > I am using udev with hotplug on kernel 2.6.5 and 2.6.9 (Fedora2 and 3). When I
> > plug one single cdc_acm device (Nokia wireless cellphones/modem) into
> > the USB port, dozens of hotplug usb.agent processes invoked and took
> > away all CPU for 7 to 8 seconds. During this time, the computer
> > doesn't even accept keyboard input.
> >
> > If I plug N Nokia phones in the USB ports at the same time, there seem
> > to have N times of usb.agent processes and Nx7 seconds delay.
> >
> > Any way to improve it?
> 
> How can you be sure, that the hotplug processes are taking your
> cpu and that it's not the kernel with the cdc_acm driver that blocks?
> 
> Are you using other USB devices on the system? Any chance to try maybe
> a storage device on the same port, if that works as expected?
> 
> It would be nice to know if you get the same behavior with a 2.6.10
> kernel, if you can get one for your system.
> 
> Thanks,
> Kay
> 
>


-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel