Re: [syzbot] INFO: rcu detected stall in tx

[Guido Kiener <Guido.Kiener@rohde-schwarz.com>]

Sorry for the messing "RE: Re: ....". One should be enough.
Next mails I'll send from my private account again, since I can't change the company rules.
This thread continues from my last mail.
Alan, please see below and add other recipient if required.

> > > Otherwise I see only the other approach to change hundred drivers
> > > and add the cases EPROTO, EILSEQ and ETIME in each callback handler.
> > > The usbtmc driver already respects the EILSEQ and ETIME, and only
> > > EPROTO is missing.
> > > The rest should be more a management task.
> > > BTW do you assume it is only a problem for INT pipes or is it also a
> > > problem for isochronous and bulk transfers?
> >
> > All of them.  Control too.
> >
> > > > Will you be able to test patches?
> > >
> > > I only can test the USBTMC function in some different PCs. I do not
> > > have automated regression tests for USB drivers or Linux kernels.
> > > Maybe there is company who could do that.
> >
> > Well then, if I do find time to write a patch, I'll ask you to try it
> > out with the usbtmc driver.
> 
> You mean that you will do a patch in urb.c or a host driver? Or just add a line in
> usbtmc.c?
> Anyhow there is no hurry. On May 20 I will send you a mail if I'm able to provoke one
> of these hardware errors EPROTO, EILSQ, or ETIME. Otherwise it doesn't make
> sense to test it.
> 
> -Guido

I've tested to pull the cable and I can easy produce the -EPROTO (-71) error in the usbtmc.c driver.
The error also happens when I have no traffic and just switch off the USBTMC device (which I can do from home office).
When there is traffic, the IN/OUT pipe will also show some -EPROTO errors from the flying urbs.

On my test PC the amount of error messages vary from 1 to ca. 10 error message issued by the INT pipe (see below).
Nevertheless I do not see the endless loop, since the USB disconnect stops the loop.
(Tested with USB 2.0, older kernel 4.15, but latest modified usbtmc.c driver using ehci-pci)
So I can test fixes (of course with latest kernel) and also setup a machine with USB 3.0.

1. Example
[1616527.017863] usbtmc 1-1.4:1.0: usbtmc_write_bulk_cb - nonzero write bulk status received: -71
[1616527.032489] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.053998] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.076145] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.098143] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.120244] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.142244] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.164440] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.186517] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.208175] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.230272] usbtmc 1-1.4:1.0: unknown status received: -71
[1616527.241982] usb 1-1.4: USB disconnect, device number 3

2. Example
[1622454.299131] usbtmc 1-1.4:1.0: unknown status received: -71
[1622454.305815] usbtmc 1-1.4:1.0: usbtmc_write_bulk_cb - nonzero write bulk status received: -71
[1622454.309921] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -71
[1622454.319121] usbtmc 1-1.4:1.0: unknown status received: -71
[1622454.341199] usbtmc 1-1.4:1.0: unknown status received: -71
[1622454.363336] usbtmc 1-1.4:1.0: unknown status received: -71
[1622454.385466] usbtmc 1-1.4:1.0: unknown status received: -71
[1622454.407576] usbtmc 1-1.4:1.0: unknown status received: -71
[1622454.422870] usb 1-1.4: USB disconnect, device number 7

3. Example
[1623429.713784] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -71
[1623429.714037] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.714279] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.714528] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.714778] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.715029] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.715278] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.715528] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.715777] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.716028] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.716277] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.718916] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.719168] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.722404] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.725666] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.725918] usbtmc 1-1.4:1.0: usbtmc_read_bulk_cb - nonzero read bulk status received: -2
[1623429.727837] usbtmc 1-1.4:1.0: unknown status received: -71
[1623429.749949] usbtmc 1-1.4:1.0: unknown status received: -71
[1623429.772094] usbtmc 1-1.4:1.0: unknown status received: -71
[1623429.794184] usbtmc 1-1.4:1.0: unknown status received: -71
[1623429.816297] usbtmc 1-1.4:1.0: unknown status received: -71
[1623429.838479] usbtmc 1-1.4:1.0: unknown status received: -71
[1623429.838795] usb 1-1.4: USB disconnect, device number 8