Re: [PATCH 1/2] usb: gadget: u_ether: Do not make UDC parent of the net device

[Paul Cercueil]

Hi Sascha,

Le jeudi 09 février 2023 à 11:18 +0100, Sascha Hauer a écrit :
> Hi Paul,
> 
> On Wed, Feb 01, 2023 at 01:32:51PM +0000, Paul Cercueil wrote:
> > Hi Sascha, Greg,
> > 
> > I have a breakage in 6.2-rc* that I eventually bisected to this
> > commit,
> > on a Ingenic SoC (using the jz4740 musb driver) with ECM or RNDIS
> > configured through gadgetfs.
> > 
> > When plugging the board to my PC, the USB network interface is
> > recognized, but 'ip link' sees it as 'NO-CARRIER'. With this commit
> > reverted on v6.2-rc5, everything works fine.
> 
> I don't have this hardware available. I just tried with a i.MX
> hardware
> and it works as expected. I have no idea where the jz4740 musb could
> behave differently.
> 
> Here's exactly what I did:
> 
> mkdir -p /sys/kernel/config/usb_gadget/mygadget
> cd /sys/kernel/config/usb_gadget/mygadget
> mkdir -p configs/c.1/strings/0x409
> echo "C1:Composite Device" > configs/c.1/strings/0x409/configuration
> mkdir -p functions/ecm.usb0
> ln -s functions/ecm.usb0 configs/c.1/
> echo "ci_hdrc.0" > UDC
> 
> Did you do something differently apart from the "ci_hdrc.0" of
> course?

Nothing very different, no.

I do:

cd /sys/kernel/config/usb_gadget
mkdir mtp \
  mtp/strings/0x409 \
  mtp/configs/c.1 \
  mtp/configs/c.1/strings/0x409 \
  mtp/functions/ffs.mtp \
  mtp/functions/ecm.net \
  mtp/functions/rndis.net

echo 0x80 > mtp/configs/c.1/bmAttributes
echo 500 > mtp/configs/c.1/MaxPower

echo 0x049f > mtp/idVendor
echo 0x505a > mtp/idProduct
echo cdc > mtp/configs/c.1/strings/0x409/configuration
ln -s mtp/functions/ecm.net mtp/configs/c.1/ecm.net

echo ci_hdrc.0 > mtp/UDC

> BTW when you say 'NO-CARRIER' is it on the PC side, board side, or
> both?

PC side. I don't know what it says on the board side, I can't
telnet/SSH.

> It would be great if we could sort this out, but if not I am fine
> with
> reverting this patch. I guess this topic will come back to my desk
> sooner or later then

Considering that the clock is ticking, let's revert it for now; that
will give me some time to debug the issue, and then we can work on a
revised patch.

Cheers,
-Paul