On Tue, Apr 18, 2017 at 5:12 PM, Simon Glass <s...@chromium.org> wrote:
> On 17 April 2017 at 13:00, Maxim Sloyko <max...@google.com> wrote:
> > Add Device Model based I2C driver for ast2500/ast2400 SoCs.
> > The driver is very limited, it only supports master mode and
> > synchronous byte-by-byte reads/writes, no DMA or Pool Buffers.
> >
> > Signed-off-by: Maxim Sloyko <max...@google.com>
> >
> > ---
> >
> > Changes in v1:
> > - Style fixes
> >
> >
> > ---
> > drivers/i2c/Kconfig | 9 ++
> > drivers/i2c/Makefile | 1 +
> > drivers/i2c/ast_i2c.c | 357 ++++++++++++++++++++++++++++++
> ++++++++++++++++++++
> > drivers/i2c/ast_i2c.h | 132 +++++++++++++++++++
> > 4 files changed, 499 insertions(+)
> > create mode 100644 drivers/i2c/ast_i2c.c
> > create mode 100644 drivers/i2c/ast_i2c.h
>
> Reviewed-by: Simon Glass <s...@chromium.org>
>
> nit below
>
> [..]
> > +static int ast_i2c_ofdata_to_platdata(struct udevice *dev)
> > +{
> > + struct ast_i2c_priv *priv = dev_get_priv(dev);
> > + int ret;
> > +
> > + priv->regs = dev_get_addr_ptr(dev);
> > + if (IS_ERR(priv->regs))
> > + return PTR_ERR(priv->regs);
>
> Should be
>
> if (!priv->regs)
>
> I think
>
Looks like dev_get_addr_ptr returns FDT_ADDR_T_NONE (cast to void*) in case
of error. FDT_ADDR_T_NONE is -1, so simple !priv->regs check would be
incorrect, as far as I understand.
>
> > +
> > + ret = clk_get_by_index(dev, 0, &priv->clk);
> > + if (ret < 0) {
> > + debug("%s: Can't get clock for %s: %d\n", __func__,
> dev->name,
> > + ret);
> > + return ret;
>
> Regards,
> Simon
>
--
*M*axim *S*loyko
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
https://lists.denx.de/listinfo/u-boot
