| Message ID | 20200831090813.78841-5-andriy.shevchenko@linux.intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v1,1/8] iio: accel: bma220: Fix returned codes from bma220_init(), bma220_deinit() | expand |
On Mon, 31 Aug 2020 12:08:10 +0300 Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote: > The driver is quite likely used only on ACPI based platforms and > rarely build with CONFIG_ACPI=n. Even though, the few dozens of bytes > is better than ugly ifdeffery and inclusion of heavy header. Given this part is readily available on maker type break out boards I doubt it is mostly ACPI. Rest of the comment is fine though. Jonathan > > As a result, replace acpi.h with mod_devicetable.h. > > Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> > --- > drivers/iio/accel/bma220_spi.c | 6 ++---- > 1 file changed, 2 insertions(+), 4 deletions(-) > > diff --git a/drivers/iio/accel/bma220_spi.c b/drivers/iio/accel/bma220_spi.c > index e55c5f5dd6b4..105c1dece890 100644 > --- a/drivers/iio/accel/bma220_spi.c > +++ b/drivers/iio/accel/bma220_spi.c > @@ -5,8 +5,8 @@ > * Copyright (c) 2016, Intel Corporation. > */ > > -#include <linux/acpi.h> > #include <linux/kernel.h> > +#include <linux/mod_devicetable.h> > #include <linux/module.h> > #include <linux/iio/buffer.h> > #include <linux/iio/iio.h> > @@ -310,20 +310,18 @@ static const struct spi_device_id bma220_spi_id[] = { > {} > }; > > -#ifdef CONFIG_ACPI > static const struct acpi_device_id bma220_acpi_id[] = { > {"BMA0220", 0}, > {} > }; > > MODULE_DEVICE_TABLE(spi, bma220_spi_id); > -#endif > > static struct spi_driver bma220_driver = { > .driver = { > .name = "bma220_spi", > .pm = &bma220_pm_ops, > - .acpi_match_table = ACPI_PTR(bma220_acpi_id), > + .acpi_match_table = bma220_acpi_id, > }, > .probe = bma220_probe, > .remove = bma220_remove,
On Mon, Aug 31, 2020 at 12:19 PM Jonathan Cameron <jic23@kernel.org> wrote: > > On Mon, 31 Aug 2020 12:08:10 +0300 > Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote: > > > The driver is quite likely used only on ACPI based platforms and > > rarely build with CONFIG_ACPI=n. Even though, the few dozens of bytes > > is better than ugly ifdeffery and inclusion of heavy header. > > Given this part is readily available on maker type break out boards > I doubt it is mostly ACPI. Rest of the comment is fine though. In-kernel use seems only ACPI (I grepped for pure SPI driver by name and didn't find anything). I hope we will have DT support for that as well (basically means adding OF ID table).
On Mon, 31 Aug 2020 12:39:10 +0300 Andy Shevchenko <andy.shevchenko@gmail.com> wrote: > On Mon, Aug 31, 2020 at 12:19 PM Jonathan Cameron <jic23@kernel.org> wrote: > > > > On Mon, 31 Aug 2020 12:08:10 +0300 > > Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote: > > > > > The driver is quite likely used only on ACPI based platforms and > > > rarely build with CONFIG_ACPI=n. Even though, the few dozens of bytes > > > is better than ugly ifdeffery and inclusion of heavy header. > > > > Given this part is readily available on maker type break out boards > > I doubt it is mostly ACPI. Rest of the comment is fine though. > > In-kernel use seems only ACPI (I grepped for pure SPI driver by name > and didn't find anything). > I hope we will have DT support for that as well (basically means > adding OF ID table). > Good to add the explicit DT support, but I think the nasty fallback is still in place in which the old style ID table is used if we get a match without manufacturer ID. So should be possible to instantiate it from DT even without the table. Jonathan
On Mon, Aug 31, 2020 at 10:55:04AM +0100, Jonathan Cameron wrote: > On Mon, 31 Aug 2020 12:39:10 +0300 > Andy Shevchenko <andy.shevchenko@gmail.com> wrote: > > On Mon, Aug 31, 2020 at 12:19 PM Jonathan Cameron <jic23@kernel.org> wrote: > > > On Mon, 31 Aug 2020 12:08:10 +0300 > > > Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote: > > > > > > > The driver is quite likely used only on ACPI based platforms and > > > > rarely build with CONFIG_ACPI=n. Even though, the few dozens of bytes > > > > is better than ugly ifdeffery and inclusion of heavy header. > > > > > > Given this part is readily available on maker type break out boards > > > I doubt it is mostly ACPI. Rest of the comment is fine though. > > > > In-kernel use seems only ACPI (I grepped for pure SPI driver by name > > and didn't find anything). > > I hope we will have DT support for that as well (basically means > > adding OF ID table). > > > > Good to add the explicit DT support, but I think the nasty fallback is > still in place in which the old style ID table is used if we get a match > without manufacturer ID. > > So should be possible to instantiate it from DT even without the table. True, but as I said, there is no evidence that there is a single user in kernel.
diff --git a/drivers/iio/accel/bma220_spi.c b/drivers/iio/accel/bma220_spi.c index e55c5f5dd6b4..105c1dece890 100644 --- a/drivers/iio/accel/bma220_spi.c +++ b/drivers/iio/accel/bma220_spi.c @@ -5,8 +5,8 @@ * Copyright (c) 2016, Intel Corporation. */ -#include <linux/acpi.h> #include <linux/kernel.h> +#include <linux/mod_devicetable.h> #include <linux/module.h> #include <linux/iio/buffer.h> #include <linux/iio/iio.h> @@ -310,20 +310,18 @@ static const struct spi_device_id bma220_spi_id[] = { {} }; -#ifdef CONFIG_ACPI static const struct acpi_device_id bma220_acpi_id[] = { {"BMA0220", 0}, {} }; MODULE_DEVICE_TABLE(spi, bma220_spi_id); -#endif static struct spi_driver bma220_driver = { .driver = { .name = "bma220_spi", .pm = &bma220_pm_ops, - .acpi_match_table = ACPI_PTR(bma220_acpi_id), + .acpi_match_table = bma220_acpi_id, }, .probe = bma220_probe, .remove = bma220_remove,
The driver is quite likely used only on ACPI based platforms and rarely build with CONFIG_ACPI=n. Even though, the few dozens of bytes is better than ugly ifdeffery and inclusion of heavy header. As a result, replace acpi.h with mod_devicetable.h. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> --- drivers/iio/accel/bma220_spi.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)