Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Fri, 4 Sep 2020 10:55:52 +0200
From:   Jacopo Mondi <jacopo@jmondi.org>
To:     Sowjanya Komatineni <skomatineni@nvidia.com>
Cc:     thierry.reding@gmail.com, jonathanh@nvidia.com,
        sakari.ailus@iki.fi, hverkuil@xs4all.nl, jacopo+renesas@jmondi.org,
        luca@lucaceresoli.net, leonl@leopardimaging.com,
        robh+dt@kernel.org, lgirdwood@gmail.com, broonie@kernel.org,
        linux-media@vger.kernel.org, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v5 3/3] media: i2c: imx274: Add IMX274 power on and off
 sequence
Message-ID: <20200904085552.5xddn2kecoktuesg@uno.localdomain>
References: <1599012278-10203-1-git-send-email-skomatineni@nvidia.com>
 <1599012278-10203-4-git-send-email-skomatineni@nvidia.com>
 <20200903144713.fyhmhs2bfcz5br6d@uno.localdomain>
 <094073f1-ef58-a2fd-bed4-7fa3b99dd120@nvidia.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <094073f1-ef58-a2fd-bed4-7fa3b99dd120@nvidia.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hello Sowjanya,

On Thu, Sep 03, 2020 at 09:25:44AM -0700, Sowjanya Komatineni wrote:
>
> On 9/3/20 8:03 AM, Jacopo Mondi wrote:
> > Hello,
> >
> > On Tue, Sep 01, 2020 at 07:04:38PM -0700, Sowjanya Komatineni wrote:
> > > IMX274 has VANA analog 2.8V supply, VDIG digital core 1.8V supply,
> > > and VDDL digital io 1.2V supply which are optional based on camera
> > > module design.
> > >
> > > IMX274 also need external 24Mhz clock and is optional based on
> > > camera module design.
> > >
> > > This patch adds support for IMX274 power on and off to enable and
> > > disable these supplies and external clock.
> > >
> > > Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
> > > Signed-off-by: Sowjanya Komatineni <skomatineni@nvidia.com>
> > > ---
> > >   drivers/media/i2c/imx274.c | 134 ++++++++++++++++++++++++++++++++++++++++++++-
> > >   1 file changed, 131 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/drivers/media/i2c/imx274.c b/drivers/media/i2c/imx274.c
> > > index a4b9dfd..79bfac3c6 100644
> > > --- a/drivers/media/i2c/imx274.c
> > > +++ b/drivers/media/i2c/imx274.c
> > > @@ -18,7 +18,9 @@
> > >   #include <linux/kernel.h>
> > >   #include <linux/module.h>
> > >   #include <linux/of_gpio.h>
> > > +#include <linux/pm_runtime.h>
> > >   #include <linux/regmap.h>
> > > +#include <linux/regulator/consumer.h>
> > >   #include <linux/slab.h>
> > >   #include <linux/v4l2-mediabus.h>
> > >   #include <linux/videodev2.h>
> > > @@ -131,6 +133,15 @@
> > >   #define IMX274_TABLE_WAIT_MS			0
> > >   #define IMX274_TABLE_END			1
> > >
> > > +/* regulator supplies */
> > > +static const char * const imx274_supply_names[] = {
> > > +	"vddl",  /* IF (1.2V) supply */
> > > +	"vdig",  /* Digital Core (1.8V) supply */
> > > +	"vana",  /* Analog (2.8V) supply */
> > > +};
> > > +
> > > +#define IMX274_NUM_SUPPLIES ARRAY_SIZE(imx274_supply_names)
> > > +
> > >   /*
> > >    * imx274 I2C operation related structure
> > >    */
> > > @@ -501,6 +512,8 @@ struct imx274_ctrls {
> > >    * @frame_rate: V4L2 frame rate structure
> > >    * @regmap: Pointer to regmap structure
> > >    * @reset_gpio: Pointer to reset gpio
> > > + * @supplies: List of analog and digital supply regulators
> > > + * @inck: Pointer to sensor input clock
> > >    * @lock: Mutex structure
> > >    * @mode: Parameters for the selected readout mode
> > >    */
> > > @@ -514,6 +527,8 @@ struct stimx274 {
> > >   	struct v4l2_fract frame_interval;
> > >   	struct regmap *regmap;
> > >   	struct gpio_desc *reset_gpio;
> > > +	struct regulator_bulk_data supplies[IMX274_NUM_SUPPLIES];
> > > +	struct clk *inck;
> > >   	struct mutex lock; /* mutex lock for operations */
> > >   	const struct imx274_mode *mode;
> > >   };
> > > @@ -767,6 +782,75 @@ static void imx274_reset(struct stimx274 *priv, int rst)
> > >   	usleep_range(IMX274_RESET_DELAY1, IMX274_RESET_DELAY2);
> > >   }
> > >
> > > +/*
> > > + * imx274_power_on - Function called to power on the sensor
> > > + * @imx274: Pointer to device structure
> > > + */
> > Can I say this does not bring much value ? :)
> > Also the parameter name is wrong
> >
> > > +static int imx274_power_on(struct device *dev)
> > > +{
> > > +	struct i2c_client *client = to_i2c_client(dev);
> > > +	struct v4l2_subdev *sd = i2c_get_clientdata(client);
> > > +	struct stimx274 *imx274 = to_imx274(sd);
> > > +	int ret;
> > > +
> > > +	/* keep sensor in reset before power on */
> > > +	imx274_reset(imx274, 0);
> > > +
> > > +	ret = clk_prepare_enable(imx274->inck);
> > > +	if (ret) {
> > > +		dev_err(&imx274->client->dev,
> > > +			"Failed to enable input clock: %d\n", ret);
> > > +		return ret;
> > > +	}
> > > +
> > > +	ret = regulator_bulk_enable(IMX274_NUM_SUPPLIES, imx274->supplies);
> > > +	if (ret) {
> > > +		dev_err(&imx274->client->dev,
> > > +			"Failed to enable regulators: %d\n", ret);
> > > +		goto fail_reg;
> > > +	}
> > > +
> > > +	usleep_range(1, 2);
> > usleep_range() allows you to provide an interval in which your timeout
> > can be coalesced with others. Giving a [1usec, 2usec] range kind of
> > defeat the purpose. And most than everything, does sleeping for 2usec
> > serve any real purpose ?
>
> Following delay recommendation from DS for power on sequence.
>

2 useconds ? Seems very short :)

> >
> >
> > > +	imx274_reset(imx274, 1);
> > > +
> > > +	return 0;
> > > +
> > > +fail_reg:
> > > +	regulator_bulk_disable(IMX274_NUM_SUPPLIES, imx274->supplies);
> > regulator_bulk_enable() disables all the regulators that were enabled
> > before the one that failed, so I don't think you need this one here
> >
> > > +	clk_disable_unprepare(imx274->inck);
> > > +	return ret;
> > > +}
> > > +
> > > +/*
> > > + * imx274_power_off - Function called to power off the sensor
> > > + * @imx274: Pointer to device structure
> > > + */
> > Same as the above one
> >
> > > +static int imx274_power_off(struct device *dev)
> > > +{
> > > +	struct i2c_client *client = to_i2c_client(dev);
> > > +	struct v4l2_subdev *sd = i2c_get_clientdata(client);
> > > +	struct stimx274 *imx274 = to_imx274(sd);
> > > +
> > > +	imx274_reset(imx274, 0);
> > > +
> > Is reset before power-off necessary ?
>
> Its recommended power off sequence as per data sheet.
>
> Safe to keep sensor in reset before powering down one regulator at a time.
>

Fair enough then!

> >
> > > +	regulator_bulk_disable(IMX274_NUM_SUPPLIES, imx274->supplies);
> > > +
> > > +	clk_disable_unprepare(imx274->inck);
> > > +
> > > +	return 0;
> > > +}
> > > +
> > > +static int imx274_get_regulators(struct device *dev, struct stimx274 *imx274)
> > For symmetry with the regulators API I would call this
> > imx274_regulators_get(). Up to you :)
> >
> > > +{
> > > +	unsigned int i;
> > > +
> > > +	for (i = 0; i < IMX274_NUM_SUPPLIES; i++)
> > > +		imx274->supplies[i].supply = imx274_supply_names[i];
> > > +
> > > +	return devm_regulator_bulk_get(dev, IMX274_NUM_SUPPLIES,
> > > +					imx274->supplies);
> >                                         ^ not sure if it's my email
> >                                         client but you might have a
> >                                         wrong indent here
> >
> > Also, the regulators are optional in the bindings, how do the
> > regulators API cope with that ? I had a look around and they seems to
> > assume regulators are provided. I might be mistaken though
>
> Yes these are optional regulators and based on feedback from sakari changed
> to use regulator_bulk_get() here.
>
> regulator_bulk_get() uses NORMAL_GET and in case if supplies is not found it
> will use dummy regulator.
>

Ah thanks, I had a look at the regulator_get() implementation and
missed that. So we're safe here!

> > > +}
> > > +
> > >   /**
> > >    * imx274_s_ctrl - This is used to set the imx274 V4L2 controls
> > >    * @ctrl: V4L2 control to be set
> > > @@ -781,6 +865,9 @@ static int imx274_s_ctrl(struct v4l2_ctrl *ctrl)
> > >   	struct stimx274 *imx274 = to_imx274(sd);
> > >   	int ret = -EINVAL;
> > >
> > > +	if (!pm_runtime_get_if_in_use(&imx274->client->dev))
> > > +		return 0;
> > > +
> > Right, but then you should call __v4l2_ctrl_handler_setup() in the
> > s_stream(1) call path to have controls updated (after
> > pm_runtime_get_sync() call for power on). I had a look at it seems
> > only exposure is updated.
>
> Existing driver does v4l2_ctrl_handler_setup() in probe(). So, sensor power
> on happens prior to that in probe() and then powers down during idle.
>

mmm, my point is that with this patch if a control is set while the
sensor is powered off it does not get applied, as you return 0 here.

This mean the newly set control values have to be applied as soon as
it possible, or at least before starting to stream. If you have a look
at the imx219 driver, in the s_stream() call path there's a call to
v4l2_ctrl_handler_setup() (in imx219_start_streaming()).

I think you should do the same unless I mis-interpreted your reply.

> >
> > >   	dev_dbg(&imx274->client->dev,
> > >   		"%s : s_ctrl: %s, value: %d\n", __func__,
> > >   		ctrl->name, ctrl->val);
> > > @@ -811,6 +898,8 @@ static int imx274_s_ctrl(struct v4l2_ctrl *ctrl)
> > >   		break;
> > >   	}
> > >
> > > +	pm_runtime_put(&imx274->client->dev);
> > > +
> > >   	return ret;
> > >   }
> > >
> > > @@ -1327,6 +1416,13 @@ static int imx274_s_stream(struct v4l2_subdev *sd, int on)
> > >   	mutex_lock(&imx274->lock);
> > >
> > >   	if (on) {
> > > +		ret = pm_runtime_get_sync(&imx274->client->dev);
> > > +		if (ret < 0) {
> > > +			pm_runtime_put_noidle(&imx274->client->dev);
> > > +			mutex_unlock(&imx274->lock);
> > > +			return ret;
> > > +		}
> > > +
> > >   		/* load mode registers */
> > >   		ret = imx274_mode_regs(imx274);
> > >   		if (ret)
> > > @@ -1362,6 +1458,7 @@ static int imx274_s_stream(struct v4l2_subdev *sd, int on)
> > >   		ret = imx274_write_table(imx274, imx274_stop);
> > >   		if (ret)
> > >   			goto fail;
> > > +		pm_runtime_put(&imx274->client->dev);
> > >   	}
> > >
> > >   	mutex_unlock(&imx274->lock);
> > > @@ -1369,6 +1466,7 @@ static int imx274_s_stream(struct v4l2_subdev *sd, int on)
> > >   	return 0;
> > >
> > >   fail:
> > > +	pm_runtime_put(&imx274->client->dev);
> > >   	mutex_unlock(&imx274->lock);
> > >   	dev_err(&imx274->client->dev, "s_stream failed\n");
> > >   	return ret;
> > > @@ -1834,6 +1932,14 @@ static int imx274_probe(struct i2c_client *client)
> > >
> > >   	mutex_init(&imx274->lock);
> > >
> > > +	imx274->inck = devm_clk_get_optional(&client->dev, "inck");
> > clk_get_optional() might return error. I would check this with IS_ERR
> >
> > > +	ret = imx274_get_regulators(&client->dev, imx274);
> > > +	if (ret) {
> > > +		dev_err(&client->dev,
> > > +			"Failed to get power regulators, err: %d\n", ret);
> > > +		return ret;
> > > +	}
> > > +
> > >   	/* initialize format */
> > >   	imx274->mode = &imx274_modes[IMX274_DEFAULT_BINNING];
> > >   	imx274->crop.width = IMX274_MAX_WIDTH;
> > > @@ -1881,15 +1987,23 @@ static int imx274_probe(struct i2c_client *client)
> > >   		goto err_me;
> > >   	}
> > >
> > > -	/* pull sensor out of reset */
> > > -	imx274_reset(imx274, 1);
> > > +	/* power on the sensor */
> > > +	ret = imx274_power_on(&client->dev);
> > > +	if (ret < 0) {
> > > +		dev_err(&client->dev,
> > > +			"%s : imx274 power on failed\n", __func__);
> > > +		goto err_me;
> > > +	}
> > Doesn't pm_runtime_get calls the poweron function for you ?
> >
> > But anyway, I don't see the device being probed for, in example,
> > querying it's VID/PID for identification during the driver's probe
> > routine. Do you need to power on ?
>
> existing driver does v4l2_ctrl handler setup and loads sensor default
> control values during probe.

Ouch, they're pretty evident and I've missed that. Although I think
you can call pm_runtime_get_sync() to have resume executed, but this
makes no difference in practice I guess.

Thanks for the clarifications, there's just a few items left to
address in my opinion.

Thanks
  j

>
> So doing sensor power_on here prior to setup. Power off happens during idle.
>
> >
> > > +
> > > +	pm_runtime_set_active(&client->dev);
> > > +	pm_runtime_enable(&client->dev);
> > >
> > >   	/* initialize controls */
> > >   	ret = v4l2_ctrl_handler_init(&imx274->ctrls.handler, 4);
> > >   	if (ret < 0) {
> > >   		dev_err(&client->dev,
> > >   			"%s : ctrl handler init Failed\n", __func__);
> > > -		goto err_me;
> > > +		goto err_disable_rpm;
> > >   	}
> > >
> > >   	imx274->ctrls.handler.lock = &imx274->lock;
> > > @@ -1951,11 +2065,16 @@ static int imx274_probe(struct i2c_client *client)
> > >   		goto err_ctrls;
> > >   	}
> > >
> > > +	pm_runtime_idle(&client->dev);
> > > +
> > >   	dev_info(&client->dev, "imx274 : imx274 probe success !\n");
> > >   	return 0;
> > >
> > >   err_ctrls:
> > >   	v4l2_ctrl_handler_free(&imx274->ctrls.handler);
> > > +err_disable_rpm:
> > > +	pm_runtime_disable(&client->dev);
> > > +	pm_runtime_set_suspended(&client->dev);
> > >   err_me:
> > >   	media_entity_cleanup(&sd->entity);
> > >   err_regmap:
> > > @@ -1973,14 +2092,23 @@ static int imx274_remove(struct i2c_client *client)
> > >
> > >   	v4l2_async_unregister_subdev(sd);
> > >   	v4l2_ctrl_handler_free(&imx274->ctrls.handler);
> > > +
> > > +	pm_runtime_disable(&client->dev);
> > > +	pm_runtime_set_suspended(&client->dev);
> > > +
> > >   	media_entity_cleanup(&sd->entity);
> > >   	mutex_destroy(&imx274->lock);
> > >   	return 0;
> > >   }
> > >
> > > +static const struct dev_pm_ops imx274_pm_ops = {
> > > +	SET_RUNTIME_PM_OPS(imx274_power_off, imx274_power_on, NULL)
> > > +};
> > > +
> > >   static struct i2c_driver imx274_i2c_driver = {
> > >   	.driver = {
> > >   		.name	= DRIVER_NAME,
> > > +		.pm = &imx274_pm_ops,
> > >   		.of_match_table	= imx274_of_id_table,
> > >   	},
> > >   	.probe_new	= imx274_probe,
> > > --
> > > 2.7.4
> > >