Re: [PATCH] net: usb: qmi_wwan: Fix for packets being rejected in the ring buffer used by the xHCI controller.

[Daniele Palmas]

Hi BjÃrn and Paul,

Il giorno dom 8 mar 2020 alle ore 16:28 BjÃrn Mork <bjorn@xxxxxxx> ha scritto:
>
> Paul Gildea <paul.gildea@xxxxxxxxx> writes:
>
> > When MTU of modem is set to less than 1500 and a packet larger than MTU
> > arrives in Linux from a modem, it is discarded with -EOVERFLOW error
> > (Babble error). This is seen on USB3.0 and USB2.0 busses. This is
> > essentially because the MRU (Max Receive Size) is not a separate entity to
> > the MTU (Max Transmit Size) and the received packets can be larger than
> > those transmitted. Following the babble error there were an endless supply
> > of zero-length URBs which are rejected with -EPROTO (increasing the rx
> > input error counter each time). This is only seen on USB3.0. These continue
> > to come ad infinitum until the modem is shutdown, rendering the modem
> > unusable. There is a bug in the core USB handling code in Linux that
> > doesn't deal well with network MTUs smaller than 1500 bytes. By default the
> > dev->hard_mtu (the "real" MTU) is in lockstep with dev->rx_urb_size
> > (essentially an MRU), and it's the latter that is causing trouble. This has
> > nothing to do with the modems; the issue can be reproduced by getting a
> > USB-Ethernet dongle, setting the MTU to 1430, and pinging with size greater
> > than 1406.
> >
> > Signed-off-by: Paul Gildea <Paul.Gildea@xxxxxxxxx>
> > ---
> > drivers/net/usb/qmi_wwan.c | 7 +++++++
> >  1 file changed, 7 insertions(+)
> >
> > diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c
> > index 5754bb6..545c772 100644
> > --- a/drivers/net/usb/qmi_wwan.c
> > +++ b/drivers/net/usb/qmi_wwan.c
> > @@ -815,6 +815,13 @@ static int qmi_wwan_bind(struct usbnet *dev, struct
> > usb_interface *intf)
> >     }
> >     dev->net->netdev_ops = &qmi_wwan_netdev_ops;
> >     dev->net->sysfs_groups[0] = &qmi_wwan_sysfs_attr_group;
> > +    /* LTE Networks don't always respect their own MTU on receive side;
> > +    * e.g. AT&T pushes 1430 MTU but still allows 1500 byte packets from
> > +    * far-end network. Make receive buffer large enough to accommodate
> > +    * them, and add four bytes so MTU does not equal MRU on network
> > +    * with 1500 MTU otherwise usbnet_change_mtu() will change both.
> > +    */
> > +   dev->rx_urb_size = ETH_DATA_LEN + 4;

Isn't this going to break the change MTU workaround for dl data
aggregation when using qmap?

Regards,
Daniele

> >  err:
> >     return status;
> >  }
> > --
> > 1.9.1
>
>
> This is fine as a first step towards saner buffer handling in qmi_wwan.
> If real world devices use asymmetric MTUs, then we should just deal with
> that.
>
> So I was going to add my ack.  But the patch does not apply:
>
>
>  bjorn@miraculix:/usr/local/src/git/linux$ git am /tmp/l
>  Applying: net: usb: qmi_wwan: Fix for packets being rejected in the ring buffer used by the xHCI controller.
>  error: corrupt patch at line 10
>
> and checkpatch says why:
>
>  bjorn@miraculix:/usr/local/src/git/linux$ scripts/checkpatch.pl /tmp/l
>  ERROR: patch seems to be corrupt (line wrapped?)
>  #34: FILE: drivers/net/usb/qmi_wwan.c:814:
>  usb_interface *intf)
>
>
> Could you fix up and resend? You might have to use a different email
> client.  See
> https://www.kernel.org/doc/html/latest/process/email-clients.html#email-clients
>
>
> BjÃrn