On Thu, Mar 2, 2023 at 1:53 PM Keguang Zhang <keguang.zhang@xxxxxxxxx> wrote: > > This patch adds DT support for Loongson-1 GPIO driver. > > Signed-off-by: Keguang Zhang <keguang.zhang@xxxxxxxxx> > --- > V1 -> V2: Let gpiolib parse ngpios property > Remove unnecessary alias id parsing > Remove superfluous initialization done by bgpio_init() > Add MODULE_DEVICE_TABLE() > Other minor fixes > --- > drivers/gpio/gpio-loongson1.c | 19 ++++++++++++++++--- > 1 file changed, 16 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpio/gpio-loongson1.c b/drivers/gpio/gpio-loongson1.c > index 3ac9e49e7efb..94ac0ccb450f 100644 > --- a/drivers/gpio/gpio-loongson1.c > +++ b/drivers/gpio/gpio-loongson1.c > @@ -68,25 +68,38 @@ static int ls1x_gpio_probe(struct platform_device *pdev) > ls1x_gc->gc.owner = THIS_MODULE; > ls1x_gc->gc.request = ls1x_gpio_request; > ls1x_gc->gc.free = ls1x_gpio_free; > - ls1x_gc->gc.base = pdev->id * 32; > + /* > + * Clear ngpio to let gpiolib get the correct number > + * by reading ngpios property > + */ > + ls1x_gc->gc.ngpio = 0; > Who could have set it before and why would this information need to be unconditionally discarded? Bart > ret = devm_gpiochip_add_data(dev, &ls1x_gc->gc, ls1x_gc); > if (ret) > goto err; > > platform_set_drvdata(pdev, ls1x_gc); > - dev_info(dev, "Loongson1 GPIO driver registered\n"); > + > + dev_info(dev, "GPIO controller registered with %d pins\n", > + ls1x_gc->gc.ngpio); > > return 0; > err: > - dev_err(dev, "failed to register GPIO device\n"); > + dev_err(dev, "failed to register GPIO controller\n"); > return ret; > } > > +static const struct of_device_id ls1x_gpio_dt_ids[] = { > + { .compatible = "loongson,ls1x-gpio" }, > + { /* sentinel */ } > +}; > +MODULE_DEVICE_TABLE(of, ls1x_gpio_dt_ids); > + > static struct platform_driver ls1x_gpio_driver = { > .probe = ls1x_gpio_probe, > .driver = { > .name = "ls1x-gpio", > + .of_match_table = ls1x_gpio_dt_ids, > }, > }; > > -- > 2.34.1 >