Re: [PATCH net-next 1/3] net: mdio: add support for passing a PTP system timestamp to the mii_bus driver

[Vladimir Oltean <olteanv@gmail.com>]

On Mon, 19 Aug 2019 at 16:34, Vladimir Oltean <olteanv@gmail.com> wrote:
>
> Hi Andrew,
>
> On Mon, 19 Aug 2019 at 16:17, Andrew Lunn <andrew@lunn.ch> wrote:
> >
> > On Fri, Aug 16, 2019 at 06:31:55PM +0200, Hubert Feurstein wrote:
> > > In order to improve the synchronisation precision of phc2sys (from
> > > the linuxptp project) for devices like switches which are attached
> > > to the MDIO bus, it is necessary the get the system timestamps as
> > > close as possible to the access which causes the PTP timestamp
> > > register to be snapshotted in the switch hardware. Usually this is
> > > triggered by an MDIO write access, the snapshotted timestamp is then
> > > transferred by several MDIO reads.
> > >
> > > This patch adds the required infrastructure to solve the problem described
> > > above.
> > >
> > > Signed-off-by: Hubert Feurstein <h.feurstein@gmail.com>
> > > ---
> > >  drivers/net/phy/mdio_bus.c | 105 +++++++++++++++++++++++++++++++++++++
> > >  include/linux/mdio.h       |   7 +++
> > >  include/linux/phy.h        |  25 +++++++++
> > >  3 files changed, 137 insertions(+)
> > >
> > > diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
> > > index bd04fe762056..167a21f267fa 100644
> > > --- a/drivers/net/phy/mdio_bus.c
> > > +++ b/drivers/net/phy/mdio_bus.c
> > > @@ -34,6 +34,7 @@
> > >  #include <linux/phy.h>
> > >  #include <linux/io.h>
> > >  #include <linux/uaccess.h>
> > > +#include <linux/ptp_clock_kernel.h>
> > >
> > >  #define CREATE_TRACE_POINTS
> > >  #include <trace/events/mdio.h>
> > > @@ -697,6 +698,110 @@ int mdiobus_write(struct mii_bus *bus, int addr, u32 regnum, u16 val)
> > >  }
> > >  EXPORT_SYMBOL(mdiobus_write);
> > >
> > > +/**
> > > + * __mdiobus_write_sts - Unlocked version of the mdiobus_write_sts function
> > > + * @bus: the mii_bus struct
> > > + * @addr: the phy address
> > > + * @regnum: register number to write
> > > + * @val: value to write to @regnum
> > > + * @sts: the ptp system timestamp
> > > + *
> > > + * Write a MDIO bus register and request the MDIO bus driver to take the
> > > + * system timestamps when sts-pointer is valid. When the bus driver doesn't
> > > + * support this, the timestamps are taken in this function instead.
> > > + *
> > > + * In order to improve the synchronisation precision of phc2sys (from
> > > + * the linuxptp project) for devices like switches which are attached
> > > + * to the MDIO bus, it is necessary the get the system timestamps as
> > > + * close as possible to the access which causes the PTP timestamp
> > > + * register to be snapshotted in the switch hardware. Usually this is
> > > + * triggered by an MDIO write access, the snapshotted timestamp is then
> > > + * transferred by several MDIO reads.
> > > + *
> > > + * Caller must hold the mdio bus lock.
> > > + *
> > > + * NOTE: MUST NOT be called from interrupt context.
> > > + */
> > > +int __mdiobus_write_sts(struct mii_bus *bus, int addr, u32 regnum, u16 val,
> > > +                     struct ptp_system_timestamp *sts)
> > > +{
> > > +     int retval;
> > > +
> > > +     WARN_ON_ONCE(!mutex_is_locked(&bus->mdio_lock));
> > > +
> > > +     if (!bus->ptp_sts_supported)
> > > +             ptp_read_system_prets(sts);
> >
> > How expensive is ptp_read_system_prets()? My original suggestion was
> > to unconditionally call it here, and then let the driver overwrite it
> > if it supports finer grained time stamping. MDIO is slow, so as long
> > as ptp_read_system_prets() is not too expensive, i prefer KISS.
> >
> >    Andrew
>
> While that works for the pre_ts, it doesn't work for the post_ts (the
> MDIO bus core will unconditionally overwrite the system timestamp from
> the driver).
> Unless you're suggesting to keep the pre_ts unconditional and the
> post_ts under the "if" condition, which is a bit odd.
> According to my tests with a scope (measuring the width between SPI
> transfers with and without the ptp_read_system_*ts calls), two calls
> to ktime_get_real_ts64 amount to around 750 ns on a 1200 MHz Cortex A7
> core, or around 90 clock cycles.

900 clock cycles, my bad.

>
> Regards,
> -Vladimir