BUG: USB3.0 µPD720200 complains about over-current

BUG: USB3.0 µPD720200 complains about over-current

[Tõnu Samuel]

Hi!

I know, this is place where only Gods speak and others not. But I am
sure there is something odd I found and someone here can take care of
it.

In situation when nothing is connected to any USB ports Ubuntu 10.04.1
stock kernel complains:
spider@spider:~$ zcat /var/log/kern.log.2.gz | grep -i current | head
Sep 29 23:30:12 spider kernel: [    9.048388] hub 8-0:1.0: over-current
change on port 1
Sep 29 23:30:12 spider kernel: [    9.152036] hub 8-0:1.0: over-current
change on port 2
Sep 29 23:30:12 spider kernel: [    9.256038] hub 8-0:1.0: over-current
change on port 3
Sep 29 23:30:12 spider kernel: [    9.360037] hub 8-0:1.0: over-current
change on port 1
Sep 29 23:30:12 spider kernel: [    9.464036] hub 8-0:1.0: over-current
change on port 2
Sep 29 23:30:12 spider kernel: [    9.570621] hub 8-0:1.0: over-current
change on port 3
Sep 29 23:30:12 spider kernel: [    9.672037] hub 8-0:1.0: over-current
change on port 1
Sep 29 23:30:12 spider kernel: [    9.776038] hub 8-0:1.0: over-current
change on port 2
Sep 29 23:30:12 spider kernel: [    9.880049] hub 8-0:1.0: over-current
change on port 3
Sep 29 23:30:13 spider kernel: [    9.984041] hub 8-0:1.0: over-current
change on port 1
spider@spider:~$ 

This runs forever and makes computer busy forever just running those
messages. I verified same behavior exists with vanilla 2.6.35.7.
Disabling USB3.0 in BIOS turns off problem and all USB2 ports remain to
work. So probably this is xHCI driver who complains. I looked bit around
in source and see it is copied from USB2 code and possibly USB3 has
something different. I've read USB3 can supply more power, so possibly
something here has changed.

My hardware:

Motherboard ASUS M4A88T-I Deluxe
http://www.asus.com/product.aspx?P_ID=jGAoIFEziW5sPYy7

it has USB3 chip "NEC D720200F1" on it:
http://digi.physic.ut.ee/mw/images/d/de/20100926_digi_007.png

NEC is joined Renesas, so this is his link for datasheet etc:
http://www2.renesas.com/usb/en/product/upd720200.html

If you propose patch, I can give him test run.

   Tõnu

Re: BUG: USB3.0 µPD720200 complains about over-current

[Tõnu Samuel]

On Wed, 2010-10-13 at 19:01 +0300, Tõnu Samuel wrote:
> Sep 29 23:30:12 spider kernel: [    9.048388] hub 8-0:1.0: over-current
> change on port 1

I forgot to add dmesg output. Because it is 840 lines I put it here
http://pastebin.com/xEfiKYW5

  Tõnu

Re: BUG: USB3.0 ?PD720200 complains about over-current

[Greg KH]

On Wed, Oct 13, 2010 at 07:01:09PM +0300, T?nu Samuel wrote:
> Hi!
> 
> I know, this is place where only Gods speak and others not. But I am
> sure there is something odd I found and someone here can take care of
> it.

For usb specific stuff like this, try using the
linux-usb@vger.kernel.org mailing list.

> In situation when nothing is connected to any USB ports Ubuntu 10.04.1
> stock kernel complains:
> spider@spider:~$ zcat /var/log/kern.log.2.gz | grep -i current | head
> Sep 29 23:30:12 spider kernel: [    9.048388] hub 8-0:1.0: over-current
> change on port 1
> Sep 29 23:30:12 spider kernel: [    9.152036] hub 8-0:1.0: over-current
> change on port 2
> Sep 29 23:30:12 spider kernel: [    9.256038] hub 8-0:1.0: over-current
> change on port 3
> Sep 29 23:30:12 spider kernel: [    9.360037] hub 8-0:1.0: over-current
> change on port 1
> Sep 29 23:30:12 spider kernel: [    9.464036] hub 8-0:1.0: over-current
> change on port 2
> Sep 29 23:30:12 spider kernel: [    9.570621] hub 8-0:1.0: over-current
> change on port 3
> Sep 29 23:30:12 spider kernel: [    9.672037] hub 8-0:1.0: over-current
> change on port 1
> Sep 29 23:30:12 spider kernel: [    9.776038] hub 8-0:1.0: over-current
> change on port 2
> Sep 29 23:30:12 spider kernel: [    9.880049] hub 8-0:1.0: over-current
> change on port 3
> Sep 29 23:30:13 spider kernel: [    9.984041] hub 8-0:1.0: over-current
> change on port 1
> spider@spider:~$ 
> 
> This runs forever and makes computer busy forever just running those
> messages. I verified same behavior exists with vanilla 2.6.35.7.
> Disabling USB3.0 in BIOS turns off problem and all USB2 ports remain to
> work. So probably this is xHCI driver who complains. I looked bit around
> in source and see it is copied from USB2 code and possibly USB3 has
> something different. I've read USB3 can supply more power, so possibly
> something here has changed.

Do you have the xhci driver loaded for the USB 3.0 controller?

> My hardware:
> 
> Motherboard ASUS M4A88T-I Deluxe
> http://www.asus.com/product.aspx?P_ID=jGAoIFEziW5sPYy7
> 
> it has USB3 chip "NEC D720200F1" on it:
> http://digi.physic.ut.ee/mw/images/d/de/20100926_digi_007.png
> 
> NEC is joined Renesas, so this is his link for datasheet etc:
> http://www2.renesas.com/usb/en/product/upd720200.html
> 
> If you propose patch, I can give him test run.

Any xhci developers have an idea?

thanks,

greg k-h