| Message ID | 20180920183151.2933-1-ricardo.ribalda@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | None | expand |
On Thu 2018-09-20 20:31:51, Ricardo Ribalda Delgado wrote: > Since kernel 4.16, i2c devices with DT compatible tag are modprobed > using their DT modalias. > Without this patch, if this driver is build as module it would never > be autoprobed. > > Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com> 1,4: Acked-by: Pavel Machek <pavel@ucw.cz>
Hi Ricardo, Thank you for the patch. On Thursday, 20 September 2018 21:31:51 EEST Ricardo Ribalda Delgado wrote: > Since kernel 4.16, i2c devices with DT compatible tag are modprobed > using their DT modalias. > Without this patch, if this driver is build as module it would never > be autoprobed. > > Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com> > --- > drivers/media/i2c/ad5820.c | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/drivers/media/i2c/ad5820.c b/drivers/media/i2c/ad5820.c > index 20931217e3b1..75b9b8aa5533 100644 > --- a/drivers/media/i2c/ad5820.c > +++ b/drivers/media/i2c/ad5820.c > @@ -375,12 +375,19 @@ static const struct i2c_device_id ad5820_id_table[] = > { }; > MODULE_DEVICE_TABLE(i2c, ad5820_id_table); > > +static const struct of_device_id ad5820_of_table[] = { > + { .compatible = "adi,"AD5820_NAME }, I'd spell this out explicitly, to make it easier to grep for the compatible string. > + { } > +}; > +MODULE_DEVICE_TABLE(of, ad5820_of_table); > + > static SIMPLE_DEV_PM_OPS(ad5820_pm, ad5820_suspend, ad5820_resume); > > static struct i2c_driver ad5820_i2c_driver = { > .driver = { > .name = AD5820_NAME, > .pm = &ad5820_pm, > + .of_match_table = ad5820_of_table, As the driver doesn't depend on CONFIG_OF, would it make sense to use of_config_ptr() (and to compile the of table conditionally on CONFIG_OF) ? > }, > .probe = ad5820_probe, > .remove = ad5820_remove,
Hi Laurent, On Thu, Sep 20, 2018 at 11:10:23PM +0300, Laurent Pinchart wrote: > > +MODULE_DEVICE_TABLE(of, ad5820_of_table); > > + > > static SIMPLE_DEV_PM_OPS(ad5820_pm, ad5820_suspend, ad5820_resume); > > > > static struct i2c_driver ad5820_i2c_driver = { > > .driver = { > > .name = AD5820_NAME, > > .pm = &ad5820_pm, > > + .of_match_table = ad5820_of_table, > > As the driver doesn't depend on CONFIG_OF, would it make sense to use > of_config_ptr() (and to compile the of table conditionally on CONFIG_OF) ? You get ACPI support as a bonus if you don't use of_config_ptr(). :-) Other changes could be needed but this enables probing the driver for a device. In this case the probability of anyone using this device on an ACPI system could be pretty low though.
diff --git a/drivers/media/i2c/ad5820.c b/drivers/media/i2c/ad5820.c index 20931217e3b1..75b9b8aa5533 100644 --- a/drivers/media/i2c/ad5820.c +++ b/drivers/media/i2c/ad5820.c @@ -375,12 +375,19 @@ static const struct i2c_device_id ad5820_id_table[] = { }; MODULE_DEVICE_TABLE(i2c, ad5820_id_table); +static const struct of_device_id ad5820_of_table[] = { + { .compatible = "adi,"AD5820_NAME }, + { } +}; +MODULE_DEVICE_TABLE(of, ad5820_of_table); + static SIMPLE_DEV_PM_OPS(ad5820_pm, ad5820_suspend, ad5820_resume); static struct i2c_driver ad5820_i2c_driver = { .driver = { .name = AD5820_NAME, .pm = &ad5820_pm, + .of_match_table = ad5820_of_table, }, .probe = ad5820_probe, .remove = ad5820_remove,
Since kernel 4.16, i2c devices with DT compatible tag are modprobed using their DT modalias. Without this patch, if this driver is build as module it would never be autoprobed. Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com> --- drivers/media/i2c/ad5820.c | 7 +++++++ 1 file changed, 7 insertions(+)