Re: [PATCH v4 05/12] can: slcan: use CAN network device driver API

[Marc Kleine-Budde]

On 28.06.2022 11:48:48, Dario Binacchi wrote:
> Hi Marc,
> 
> On Tue, Jun 28, 2022 at 11:28 AM Marc Kleine-Budde <mkl@xxxxxxxxxxxxxx> wrote:
> >
> > On 14.06.2022 14:28:14, Dario Binacchi wrote:
> > > As suggested by commit [1], now the driver uses the functions and the
> > > data structures provided by the CAN network device driver interface.
> > >
> > > Currently the driver doesn't implement a way to set bitrate for SLCAN
> > > based devices via ip tool, so you'll have to do this by slcand or
> > > slcan_attach invocation through the -sX parameter:
> > >
> > > - slcan_attach -f -s6 -o /dev/ttyACM0
> > > - slcand -f -s8 -o /dev/ttyUSB0
> > >
> > > where -s6 in will set adapter's bitrate to 500 Kbit/s and -s8 to
> > > 1Mbit/s.
> > > See the table below for further CAN bitrates:
> > > - s0 ->   10 Kbit/s
> > > - s1 ->   20 Kbit/s
> > > - s2 ->   50 Kbit/s
> > > - s3 ->  100 Kbit/s
> > > - s4 ->  125 Kbit/s
> > > - s5 ->  250 Kbit/s
> > > - s6 ->  500 Kbit/s
> > > - s7 ->  800 Kbit/s
> > > - s8 -> 1000 Kbit/s
> > >
> > > In doing so, the struct can_priv::bittiming.bitrate of the driver is not
> > > set and since the open_candev() checks that the bitrate has been set, it
> > > must be a non-zero value, the bitrate is set to a fake value (-1U)
> > > before it is called.
> > >
> > > The patch also changes the slcan_devs locking from rtnl to spin_lock. The
> > > change was tested with a kernel with the CONFIG_PROVE_LOCKING option
> > > enabled that did not show any errors.
> >
> > You're not allowed to call alloc_candev() with a spin_lock held. See
> > today's kernel test robot mail:
> >
> > | https://lore.kernel.org/all/YrpqO5jepAvv4zkf@xsang-OptiPlex-9020
> >
> > I think it's best to keep the rtnl for now.
> 
> The rtnl_lock() uses a mutex while I used a spin_lock.
> 
> static DEFINE_MUTEX(rtnl_mutex);
> 
> void rtnl_lock(void)
> {
> mutex_lock(&rtnl_mutex);
> }
> EXPORT_SYMBOL(rtnl_lock);
> 
> So might it be worth trying with a mutex instead of rtnl_lock(), or do
> you think it is
> safer to return to rtn_lock () anyway?

As Max pointed out the whole static dev array is not needed at all. Just
keep the rtnl, until removing the array altogether.

regards,
Marc

-- 
Pengutronix e.K.                 | Marc Kleine-Budde           |
Embedded Linux                   | https://www.pengutronix.de  |
Vertretung West/Dortmund         | Phone: +49-231-2826-924     |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-5555 |

Attachment: signature.asc
Description: PGP signature