Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Sat, 2 Apr 2022 16:26:33 +0100
From:   Jonathan Cameron <jic23@kernel.org>
To:     Gwendal Grignou <gwendal@chromium.org>
Cc:     Stephen Boyd <swboyd@chromium.org>,
        Lars-Peter Clausen <lars@metafoo.de>,
        linux-kernel@vger.kernel.org, linux-iio@vger.kernel.org
Subject: Re: [PATCH] iio:proximity:sx_common: Fix device property parsing on
 DT systems
Message-ID: <20220402162633.0050d289@jic23-huawei>
In-Reply-To: <CAPUE2uvpBaE+YLkg6RXq4CCEFBr5ZFhjS1dXkbse9nyH8C=2Jg@mail.gmail.com>
References: <20220331210425.3908278-1-swboyd@chromium.org>
        <CAPUE2uvpBaE+YLkg6RXq4CCEFBr5ZFhjS1dXkbse9nyH8C=2Jg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Fri, 1 Apr 2022 13:54:51 -0700
Gwendal Grignou <gwendal@chromium.org> wrote:

> On Thu, Mar 31, 2022 at 2:04 PM Stephen Boyd <swboyd@chromium.org> wrote:
> >
> > After commit 7a3605bef878 ("iio: sx9310: Support ACPI property") we
> > started using the 'indio_dev->dev' to extract device properties for
> > various register settings in sx9310_get_default_reg(). This broke DT
> > based systems because dev_fwnode() used in the device_property*() APIs
> > can't find an 'of_node'. That's because the 'indio_dev->dev.of_node'
> > pointer isn't set until iio_device_register() is called. Set the pointer
> > earlier, next to where the ACPI companion is set, so that the device
> > property APIs work on DT systems.
> >
> > Cc: Gwendal Grignou <gwendal@chromium.org>
> > Fixes: 7a3605bef878 ("iio: sx9310: Support ACPI property")
> > Signed-off-by: Stephen Boyd <swboyd@chromium.org>  
> Reviewed-by: Gwendal Grignou <gwendal@chromium.org>

Applied to the fixes-togreg branch of iio.git though I'll not push
that out until rc1 is out in a few days time and I can rebase appropriately.

Thinking on the cause of this problem it seems like we should really be setting
the indio_dev->dev.of_node in iio_device_alloc() and then let a driver
override it if desired rather than this approach of fill it in if not set.
Such a change only affects a few drivers based on a dumb bit of grepping.
I'd treat such a change as a cleanup though, so having this fix in place
makes sense anyway. 

Jonathan

> > ---
> >  drivers/iio/proximity/sx_common.c | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/iio/proximity/sx_common.c b/drivers/iio/proximity/sx_common.c
> > index a7c07316a0a9..8ad814d96b7e 100644
> > --- a/drivers/iio/proximity/sx_common.c
> > +++ b/drivers/iio/proximity/sx_common.c
> > @@ -521,6 +521,7 @@ int sx_common_probe(struct i2c_client *client,
> >                 return dev_err_probe(dev, ret, "error reading WHOAMI\n");
> >
> >         ACPI_COMPANION_SET(&indio_dev->dev, ACPI_COMPANION(dev));
> > +       indio_dev->dev.of_node = client->dev.of_node;
> >         indio_dev->modes = INDIO_DIRECT_MODE;
> >
> >         indio_dev->channels =  data->chip_info->iio_channels;
> > --
> > https://chromeos.dev
> >