| Message ID | 1399483554-8824-3-git-send-email-boris.brezillon@free-electrons.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, May 07, 2014 at 07:25:49PM +0200, Boris BREZILLON wrote: > The current implementation uses sunxi_reset_init function for both early > init and platform device probe. > > The sunxi_reset_init function uses DT to retrieve device resources, which > will be an issue if reset controllers are registered from an MFD device > that define resources from mfd_cell definition. > > Moreover, we can make of devm functions when we're in the probe context. > > Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com> > --- > drivers/reset/reset-sunxi.c | 21 ++++++++++++++++++--- > 1 file changed, 18 insertions(+), 3 deletions(-) > > diff --git a/drivers/reset/reset-sunxi.c b/drivers/reset/reset-sunxi.c > index 695bd34..1b5fea6 100644 > --- a/drivers/reset/reset-sunxi.c > +++ b/drivers/reset/reset-sunxi.c > @@ -145,7 +145,24 @@ MODULE_DEVICE_TABLE(of, sunxi_reset_dt_ids); > > static int sunxi_reset_probe(struct platform_device *pdev) > { > - return sunxi_reset_init(pdev->dev.of_node); > + struct sunxi_reset_data *data; > + struct resource *res; > + > + data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL); > + if (!data) > + return -ENOMEM; > + > + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); > + data->membase = devm_request_and_ioremap(&pdev->dev, res); > + if (!data->membase) > + return -ENOMEM; You'd probably be better off using devm_ioremap_resource so that you get a meaningful error code. Apart from this, you have my Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com> Thanks! Maxime
On 08/05/2014 05:07, Maxime Ripard wrote: > On Wed, May 07, 2014 at 07:25:49PM +0200, Boris BREZILLON wrote: >> The current implementation uses sunxi_reset_init function for both early >> init and platform device probe. >> >> The sunxi_reset_init function uses DT to retrieve device resources, which >> will be an issue if reset controllers are registered from an MFD device >> that define resources from mfd_cell definition. >> >> Moreover, we can make of devm functions when we're in the probe context. >> >> Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com> >> --- >> drivers/reset/reset-sunxi.c | 21 ++++++++++++++++++--- >> 1 file changed, 18 insertions(+), 3 deletions(-) >> >> diff --git a/drivers/reset/reset-sunxi.c b/drivers/reset/reset-sunxi.c >> index 695bd34..1b5fea6 100644 >> --- a/drivers/reset/reset-sunxi.c >> +++ b/drivers/reset/reset-sunxi.c >> @@ -145,7 +145,24 @@ MODULE_DEVICE_TABLE(of, sunxi_reset_dt_ids); >> >> static int sunxi_reset_probe(struct platform_device *pdev) >> { >> - return sunxi_reset_init(pdev->dev.of_node); >> + struct sunxi_reset_data *data; >> + struct resource *res; >> + >> + data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL); >> + if (!data) >> + return -ENOMEM; >> + >> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); >> + data->membase = devm_request_and_ioremap(&pdev->dev, res); >> + if (!data->membase) >> + return -ENOMEM; > You'd probably be better off using devm_ioremap_resource so that you > get a meaningful error code. Oh, I missed this one. I'll fix it for the next version. > Apart from this, you have my > Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com> > > Thanks! > Maxime >
diff --git a/drivers/reset/reset-sunxi.c b/drivers/reset/reset-sunxi.c index 695bd34..1b5fea6 100644 --- a/drivers/reset/reset-sunxi.c +++ b/drivers/reset/reset-sunxi.c @@ -145,7 +145,24 @@ MODULE_DEVICE_TABLE(of, sunxi_reset_dt_ids); static int sunxi_reset_probe(struct platform_device *pdev) { - return sunxi_reset_init(pdev->dev.of_node); + struct sunxi_reset_data *data; + struct resource *res; + + data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + data->membase = devm_request_and_ioremap(&pdev->dev, res); + if (!data->membase) + return -ENOMEM; + + data->rcdev.owner = THIS_MODULE; + data->rcdev.nr_resets = resource_size(res) * 32; + data->rcdev.ops = &sunxi_reset_ops; + data->rcdev.of_node = pdev->dev.of_node; + + return reset_controller_register(&data->rcdev); } static int sunxi_reset_remove(struct platform_device *pdev) @@ -153,8 +170,6 @@ static int sunxi_reset_remove(struct platform_device *pdev) struct sunxi_reset_data *data = platform_get_drvdata(pdev); reset_controller_unregister(&data->rcdev); - iounmap(data->membase); - kfree(data); return 0; }
The current implementation uses sunxi_reset_init function for both early init and platform device probe. The sunxi_reset_init function uses DT to retrieve device resources, which will be an issue if reset controllers are registered from an MFD device that define resources from mfd_cell definition. Moreover, we can make of devm functions when we're in the probe context. Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com> --- drivers/reset/reset-sunxi.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-)