Re: [PATCH v2 2/3] media: i2c: ov9282: Switch to using the sub-device state lock

[xiaolei wang]

Hi Sakari,

Thanks for the review!

On 3/2/26 20:49, Sakari Ailus wrote:
> CAUTION: This email comes from a non Wind River email account!
> Do not click links or open attachments unless you recognize the sender and know the content is safe.
>
> Hi Xiaolei,
>
> Thanks for the update. A few comments below...
>
> On Sun, Mar 01, 2026 at 06:48:08PM +0800, Xiaolei Wang wrote:
> > Switch to using the sub-device state lock and properly call
> > v4l2_subdev_init_finalize() / v4l2_subdev_cleanup() on probe() /
> > remove().
> >
> > Signed-off-by: Xiaolei Wang <xiaolei.wang@xxxxxxxxxxxxx>
> > Reviewed-by: Tarang Raval <tarang.raval@xxxxxxxxxxxxxxxxx>
> > ---
> >   drivers/media/i2c/ov9282.c | 50 +++++++++++++++-----------------------
> >   1 file changed, 19 insertions(+), 31 deletions(-)
> >
> > diff --git a/drivers/media/i2c/ov9282.c b/drivers/media/i2c/ov9282.c
> > index 8bfaa3ae4be5..8acbd43838d5 100644
> > --- a/drivers/media/i2c/ov9282.c
> > +++ b/drivers/media/i2c/ov9282.c
> > @@ -221,7 +221,6 @@ struct ov9282 {
> >        bool noncontinuous_clock;
> >        const struct ov9282_mode *cur_mode;
> >        u32 code;
> > -     struct mutex mutex;
> >   };
> >
> >   static const s64 link_freq[] = {
> > @@ -795,8 +794,6 @@ static int ov9282_get_pad_format(struct v4l2_subdev *sd,
> >   {
> >        struct ov9282 *ov9282 = to_ov9282(sd);
> >
> > -     mutex_lock(&ov9282->mutex);
> > -
> >        if (fmt->which == V4L2_SUBDEV_FORMAT_TRY) {
> >                struct v4l2_mbus_framefmt *framefmt;
> >
> > @@ -807,8 +804,6 @@ static int ov9282_get_pad_format(struct v4l2_subdev *sd,
> >                                       fmt);
> >        }
> >
> > -     mutex_unlock(&ov9282->mutex);
> > -
> >        return 0;
> >   }
> >
> > @@ -829,8 +824,6 @@ static int ov9282_set_pad_format(struct v4l2_subdev *sd,
> >        u32 code;
> >        int ret = 0;
> >
> > -     mutex_lock(&ov9282->mutex);
> > -
> >        mode = v4l2_find_nearest_size(supported_modes,
> >                                      ARRAY_SIZE(supported_modes),
> >                                      width, height,
> > @@ -856,8 +849,6 @@ static int ov9282_set_pad_format(struct v4l2_subdev *sd,
> >                }
> >        }
> >
> > -     mutex_unlock(&ov9282->mutex);
> > -
> >        return ret;
> >   }
> >
> > @@ -904,10 +895,8 @@ static int ov9282_get_selection(struct v4l2_subdev *sd,
> >        case V4L2_SEL_TGT_CROP: {
> >                struct ov9282 *ov9282 = to_ov9282(sd);
> >
> > -             mutex_lock(&ov9282->mutex);
> >                sel->r = *__ov9282_get_pad_crop(ov9282, sd_state, sel->pad,
> >                                                sel->which);
> > -             mutex_unlock(&ov9282->mutex);
> >
> >                return 0;
> >        }
> > @@ -1019,9 +1008,10 @@ static int ov9282_stop_streaming(struct ov9282 *ov9282)
> >   static int ov9282_set_stream(struct v4l2_subdev *sd, int enable)
> >   {
> >        struct ov9282 *ov9282 = to_ov9282(sd);
> > +     struct v4l2_subdev_state *state;
> >        int ret;
> >
> > -     mutex_lock(&ov9282->mutex);
> > +     state = v4l2_subdev_lock_and_get_active_state(sd);
> >
> >        if (enable) {
> >                ret = pm_runtime_resume_and_get(ov9282->dev);
> > @@ -1036,14 +1026,14 @@ static int ov9282_set_stream(struct v4l2_subdev *sd, int enable)
> >                pm_runtime_put(ov9282->dev);
> >        }
> >
> > -     mutex_unlock(&ov9282->mutex);
> > +     v4l2_subdev_unlock_state(state);
> >
> >        return 0;
> >
> >   error_power_off:
> >        pm_runtime_put(ov9282->dev);
> >   error_unlock:
> > -     mutex_unlock(&ov9282->mutex);
> > +     v4l2_subdev_unlock_state(state);
> >
> >        return ret;
> >   }
> > @@ -1285,9 +1275,6 @@ static int ov9282_init_controls(struct ov9282 *ov9282)
> >        if (ret)
> >                return ret;
> >
> > -     /* Serialize controls with sensor device */
> > -     ctrl_hdlr->lock = &ov9282->mutex;
> > -
> >        /* Initialize exposure and gain */
> >        lpfr = mode->vblank + mode->height;
> >        ov9282->exp_ctrl = v4l2_ctrl_new_std(ctrl_hdlr,
> > @@ -1409,13 +1396,10 @@ static int ov9282_probe(struct i2c_client *client)
> >                return dev_err_probe(ov9282->dev, PTR_ERR(ov9282->regmap),
> >                                     "Failed to init CCI\n");
> >
> > -     mutex_init(&ov9282->mutex);
> > -
> >        ret = ov9282_power_on(ov9282->dev);
> > -     if (ret) {
> > -             dev_err(ov9282->dev, "failed to power-on the sensor");
> > -             goto error_mutex_destroy;
> > -     }
> > +     if (ret)
> > +             return dev_err_probe(ov9282->dev, ret,
> > +                                  "failed to power-on the sensor");
> >
> >        /* Check module identity */
> >        ret = ov9282_detect(ov9282);
> > @@ -1448,10 +1432,10 @@ static int ov9282_probe(struct i2c_client *client)
> >                goto error_handler_free;
> >        }
> >
> > -     ret = v4l2_async_register_subdev_sensor(&ov9282->sd);
> > +     ov9282->sd.state_lock = ov9282->ctrl_handler.lock;
> > +     ret = v4l2_subdev_init_finalize(&ov9282->sd);
> >        if (ret < 0) {
> > -             dev_err(ov9282->dev,
> > -                     "failed to register async subdev: %d", ret);
> > +             ret = dev_err_probe(ov9282->dev, ret, "failed to init subdev\n");
> Assigning ret here won't do anything, will it?
You're right. I'll remove the assignment and just call dev_err_probe() 
directly.
> >                goto error_media_entity;
> >        }
> >
> > @@ -1459,16 +1443,22 @@ static int ov9282_probe(struct i2c_client *client)
> >        pm_runtime_enable(ov9282->dev);
> >        pm_runtime_idle(ov9282->dev);
> The sensor may be powered down here...
>
> > +     ret = v4l2_async_register_subdev_sensor(&ov9282->sd);
> > +     if (ret < 0)
> > +             goto v4l2_subdev_cleanup;
> > +
> >        return 0;
> >
> > +v4l2_subdev_cleanup:
> > +     v4l2_subdev_cleanup(&ov9282->sd);
> > +     pm_runtime_disable(ov9282->dev);
> > +     pm_runtime_set_suspended(ov9282->dev);
> >   error_media_entity:
> >        media_entity_cleanup(&ov9282->sd.entity);
> >   error_handler_free:
> >        v4l2_ctrl_handler_free(ov9282->sd.ctrl_handler);
> >   error_power_off:
> >        ov9282_power_off(ov9282->dev);
> and again here.
>
> I'd suggest to call pm_runtime_idle() after registering the async
> sub-device to avoid doing this twice.
Thanks for catching this. I'll move pm_runtime_idle() after
v4l2_async_register_subdev_sensor() to avoid the potential double
power-off in the error path.

I'll send v2 with these fixes.

Best regards,
Xiaolei

> > -error_mutex_destroy:
> > -     mutex_destroy(&ov9282->mutex);
> >
> >        return ret;
> >   }
> > @@ -1482,9 +1472,9 @@ static int ov9282_probe(struct i2c_client *client)
> >   static void ov9282_remove(struct i2c_client *client)
> >   {
> >        struct v4l2_subdev *sd = i2c_get_clientdata(client);
> > -     struct ov9282 *ov9282 = to_ov9282(sd);
> >
> >        v4l2_async_unregister_subdev(sd);
> > +     v4l2_subdev_cleanup(sd);
> >        media_entity_cleanup(&sd->entity);
> >        v4l2_ctrl_handler_free(sd->ctrl_handler);
> >
> > @@ -1492,8 +1482,6 @@ static void ov9282_remove(struct i2c_client *client)
> >        if (!pm_runtime_status_suspended(&client->dev))
> >                ov9282_power_off(&client->dev);
> >        pm_runtime_set_suspended(&client->dev);
> > -
> > -     mutex_destroy(&ov9282->mutex);
> >   }
> >
> >   static const struct dev_pm_ops ov9282_pm_ops = {
> --
> Kind regards,
>
> Sakari Ailus