[00/12] bus: Convert to platform remove callback returning void

Message ID 20231109202830.4124591-1-u.kleine-koenig@pengutronix.de
Headers
Series bus: Convert to platform remove callback returning void |

Message

Uwe Kleine-König Nov. 9, 2023, 8:28 p.m. UTC
  Hello,

this series converts all drivers below drivers/bus to struct
platform_driver::remove_new(). See commit 5c5a7680e67b ("platform:
Provide a remove callback that returns no value") for an extended
explanation and the eventual goal.

drivers/bus/fsl-mc was already addressed earlier with a separate
series[1].

All conversations are trivial, because all .remove() callbacks returned
zero unconditionally.

Some of the drivers touched here don't have a maintainer and there is no
maintainer for all of drivers/bus. It would be great if someone could pick up
the whole series, maybe Arnd might do that?

Best regards
Uwe

[1] https://lore.kernel.org/lkml/20231103230001.3652259-3-u.kleine-koenig@pengutronix.de

Uwe Kleine-König (12):
  bus: hisi_lpc: Convert to platform remove callback returning void
  bus: omap-ocp2scp: Convert to platform remove callback returning void
  bus: omap_l3_smx: Convert to platform remove callback returning void
  bus: qcom-ssc-block-bus: Convert to platform remove callback returning
    void
  bus: simple-pm-bus: Convert to platform remove callback returning void
  bus: sun50i-de2: Convert to platform remove callback returning void
  bus: sunxi-rsb: Convert to platform remove callback returning void
  bus: tegra-aconnect: Convert to platform remove callback returning
    void
  bus: tegra-gmi: Convert to platform remove callback returning void
  bus: ti-pwmss: Convert to platform remove callback returning void
  bus: ti-sysc: Convert to platform remove callback returning void
  bus: ts-nbus: Convert to platform remove callback returning void

 drivers/bus/hisi_lpc.c           | 6 ++----
 drivers/bus/omap-ocp2scp.c       | 6 ++----
 drivers/bus/omap_l3_smx.c        | 6 ++----
 drivers/bus/qcom-ssc-block-bus.c | 6 ++----
 drivers/bus/simple-pm-bus.c      | 7 +++----
 drivers/bus/sun50i-de2.c         | 5 ++---
 drivers/bus/sunxi-rsb.c          | 6 ++----
 drivers/bus/tegra-aconnect.c     | 6 ++----
 drivers/bus/tegra-gmi.c          | 6 ++----
 drivers/bus/ti-pwmss.c           | 5 ++---
 drivers/bus/ti-sysc.c            | 6 ++----
 drivers/bus/ts-nbus.c            | 6 ++----
 12 files changed, 25 insertions(+), 46 deletions(-)

base-commit: b622d91ca201bf97582e9b09ebbaab005ecee86f
  

Comments

Arnd Bergmann Nov. 10, 2023, 6:32 a.m. UTC | #1
On Thu, Nov 9, 2023, at 21:28, Uwe Kleine-König wrote:
> Hello,
>
> this series converts all drivers below drivers/bus to struct
> platform_driver::remove_new(). See commit 5c5a7680e67b ("platform:
> Provide a remove callback that returns no value") for an extended
> explanation and the eventual goal.
>
> drivers/bus/fsl-mc was already addressed earlier with a separate
> series[1].
>
> All conversations are trivial, because all .remove() callbacks returned
> zero unconditionally.
>
> Some of the drivers touched here don't have a maintainer and there is no
> maintainer for all of drivers/bus. It would be great if someone could pick up
> the whole series, maybe Arnd might do that?

Sure, please send a pull request to soc@kernel.org if there
are no further comments.

      Arnd
  
Andre Przywara Nov. 10, 2023, 10:20 a.m. UTC | #2
On Thu,  9 Nov 2023 21:28:37 +0100
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> wrote:

> The .remove() callback for a platform driver returns an int which makes
> many driver authors wrongly assume it's possible to do error handling by
> returning an error code. However the value returned is ignored (apart
> from emitting a warning) and this typically results in resource leaks.
> 
> To improve here there is a quest to make the remove callback return
> void. In the first step of this quest all drivers are converted to
> .remove_new(), which already returns void. Eventually after all drivers
> are converted, .remove_new() will be renamed to .remove().
> 
> Trivially convert this driver from always returning zero in the remove
> callback to the void returning variant.
> 
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

Reviewed-by: Andre Przywara <andre.przywara@arm.com>

Thanks,
Andre

> ---
>  drivers/bus/sunxi-rsb.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/bus/sunxi-rsb.c b/drivers/bus/sunxi-rsb.c
> index db0ed4e5d315..fd3e9d82340a 100644
> --- a/drivers/bus/sunxi-rsb.c
> +++ b/drivers/bus/sunxi-rsb.c
> @@ -817,15 +817,13 @@ static int sunxi_rsb_probe(struct platform_device *pdev)
>  	return 0;
>  }
>  
> -static int sunxi_rsb_remove(struct platform_device *pdev)
> +static void sunxi_rsb_remove(struct platform_device *pdev)
>  {
>  	struct sunxi_rsb *rsb = platform_get_drvdata(pdev);
>  
>  	device_for_each_child(rsb->dev, NULL, sunxi_rsb_remove_devices);
>  	pm_runtime_disable(&pdev->dev);
>  	sunxi_rsb_hw_exit(rsb);
> -
> -	return 0;
>  }
>  
>  static const struct dev_pm_ops sunxi_rsb_dev_pm_ops = {
> @@ -842,7 +840,7 @@ MODULE_DEVICE_TABLE(of, sunxi_rsb_of_match_table);
>  
>  static struct platform_driver sunxi_rsb_driver = {
>  	.probe = sunxi_rsb_probe,
> -	.remove	= sunxi_rsb_remove,
> +	.remove_new = sunxi_rsb_remove,
>  	.driver	= {
>  		.name = RSB_CTRL_NAME,
>  		.of_match_table = sunxi_rsb_of_match_table,
  
Jernej Škrabec Nov. 18, 2023, 9:51 p.m. UTC | #3
Dne četrtek, 09. november 2023 ob 21:28:37 CET je Uwe Kleine-König napisal(a):
> The .remove() callback for a platform driver returns an int which makes
> many driver authors wrongly assume it's possible to do error handling by
> returning an error code. However the value returned is ignored (apart
> from emitting a warning) and this typically results in resource leaks.
> 
> To improve here there is a quest to make the remove callback return
> void. In the first step of this quest all drivers are converted to
> .remove_new(), which already returns void. Eventually after all drivers
> are converted, .remove_new() will be renamed to .remove().
> 
> Trivially convert this driver from always returning zero in the remove
> callback to the void returning variant.
> 
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>

Best regards,
Jernej

> ---
>  drivers/bus/sunxi-rsb.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/bus/sunxi-rsb.c b/drivers/bus/sunxi-rsb.c
> index db0ed4e5d315..fd3e9d82340a 100644
> --- a/drivers/bus/sunxi-rsb.c
> +++ b/drivers/bus/sunxi-rsb.c
> @@ -817,15 +817,13 @@ static int sunxi_rsb_probe(struct platform_device *pdev)
>  	return 0;
>  }
>  
> -static int sunxi_rsb_remove(struct platform_device *pdev)
> +static void sunxi_rsb_remove(struct platform_device *pdev)
>  {
>  	struct sunxi_rsb *rsb = platform_get_drvdata(pdev);
>  
>  	device_for_each_child(rsb->dev, NULL, sunxi_rsb_remove_devices);
>  	pm_runtime_disable(&pdev->dev);
>  	sunxi_rsb_hw_exit(rsb);
> -
> -	return 0;
>  }
>  
>  static const struct dev_pm_ops sunxi_rsb_dev_pm_ops = {
> @@ -842,7 +840,7 @@ MODULE_DEVICE_TABLE(of, sunxi_rsb_of_match_table);
>  
>  static struct platform_driver sunxi_rsb_driver = {
>  	.probe = sunxi_rsb_probe,
> -	.remove	= sunxi_rsb_remove,
> +	.remove_new = sunxi_rsb_remove,
>  	.driver	= {
>  		.name = RSB_CTRL_NAME,
>  		.of_match_table = sunxi_rsb_of_match_table,
>