Re: [PATCH v6 08/11] i2c: designware: Convert driver to using regmap API

From: Serge Semin
Date: Sun May 31 2020 - 13:13:23 EST


On Sat, May 30, 2020 at 11:05:54PM +0200, Wolfram Sang wrote:
> On Sat, May 30, 2020 at 01:09:30PM +0200, Wolfram Sang wrote:
> > On Thu, May 28, 2020 at 12:33:18PM +0300, Serge Semin wrote:
> > > Seeing the DW I2C driver is using flags-based accessors with two
> > > conditional clauses it would be better to replace them with the regmap
> > > API IO methods and to initialize the regmap object with read/write
> > > callbacks specific to the controller registers map implementation. This
> > > will be also handy for the drivers with non-standard registers mapping
> > > (like an embedded into the Baikal-T1 System Controller DW I2C block, which
> > > glue-driver is a part of this series).
> > >
> > > As before the driver tries to detect the mapping setup at probe stage and
> > > creates a regmap object accordingly, which will be used by the rest of the
> > > code to correctly access the controller registers. In two places it was
> > > appropriate to convert the hand-written read-modify-write and
> > > read-poll-loop design patterns to the corresponding regmap API
> > > ready-to-use methods.
> > >
> > > Note the regmap IO methods return value is checked only at the probe
> > > stage. The rest of the code won't do this because basically we have
> > > MMIO-based regmap so non of the read/write methods can fail (this also
> > > won't be needed for the Baikal-T1-specific I2C controller).
> > >
> > > Suggested-by: Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
> > > Signed-off-by: Serge Semin <Sergey.Semin@xxxxxxxxxxxxxxxxxxxx>
> > > Tested-by: Jarkko Nikula <jarkko.nikula@xxxxxxxxxxxxxxx>
> > > Acked-by: Jarkko Nikula <jarkko.nikula@xxxxxxxxxxxxxxx>
> > > Cc: Alexey Malahov <Alexey.Malahov@xxxxxxxxxxxxxxxxxxxx>
> > > Cc: Thomas Bogendoerfer <tsbogend@xxxxxxxxxxxxxxxx>
> > > Cc: Rob Herring <robh+dt@xxxxxxxxxx>
> > > Cc: devicetree@xxxxxxxxxxxxxxx
> > > Cc: linux-mips@xxxxxxxxxxxxxxx
> >
> > My codecheckers found these, rightfully I'd say:
> >
> > SPARSE
> > drivers/i2c/busses/i2c-designware-master.c:427:53: warning: incorrect type in argument 3 (different signedness)
> > drivers/i2c/busses/i2c-designware-master.c:427:53: expected unsigned int *val
> > drivers/i2c/busses/i2c-designware-master.c:427:53: got int *
> >
> > CC
> > drivers/i2c/busses/i2c-designware-common.c:127: warning: Excess function parameter 'base' description in 'i2c_dw_init_regmap'
> >
> > ^ means there is an argument documented which does not exist in the
> > function declaration
>
> Well, I fixed these two minor things for your now...
>

Great! Sorry for me not doing it on time. I was away from my laptop and internet
last day to do that. I saw your message about the sparse detecting the problems
while I was on my way to outskirts and was going to fix them today either in a
new version of the patchset or sending follow-up patches. But you already fixed
the problems. Thank you very much for doing that for me.

-Sergey