[PATCH 00/15] memory: Convert to platform remove callback returning void

[PATCH 00/15] memory: Convert to platform remove callback returning void

[Uwe Kleine-König]

Hello,

this series converts the platform drivers below drivers/memory to make
use of .remove_new. See commit 5c5a7680e67b ("platform: Provide a remove
callback that returns no value") for an extended explanation and the
eventual goal. The TL;DR; is to make it harder for driver authors to
leak resources without noticing.

This is merge window material. All patches are pairwise independent of
each other so they can be applied individually. Still it would be great
to let them go in all together.

Best regards
Uwe

Uwe Kleine-König (15):
  memory: brcmstb_dpfe: Convert to platform remove callback returning void
  memory: brcmstb_memc: Convert to platform remove callback returning void
  memory: emif: Convert to platform remove callback returning void
  memory: fsl-corenet-cf: Convert to platform remove callback returning void
  memory: fsl_ifc: Convert to platform remove callback returning void
  memory: jz4780-nemc: Convert to platform remove callback returning void
  memory: mtk-smi: Convert to platform remove callback returning void
  memory: omap-gpmc: Convert to platform remove callback returning void
  memory: renesas-rpc-if: Convert to platform remove callback returning void
  memory: exynos5422-dmc: Convert to platform remove callback returning void
  memory: stm32-fmc2-ebi: Convert to platform remove callback returning void
  memory: tegra186-emc: Convert to platform remove callback returning void
  memory: tegra210-emc: Convert to platform remove callback returning void
  memory: ti-aemif: Convert to platform remove callback returning void
  memory: ti-emif-pm: Convert to platform remove callback returning void

 drivers/memory/brcmstb_dpfe.c            |  6 ++----
 drivers/memory/brcmstb_memc.c            |  6 ++----
 drivers/memory/emif.c                    |  6 ++----
 drivers/memory/fsl-corenet-cf.c          |  6 ++----
 drivers/memory/fsl_ifc.c                 |  6 ++----
 drivers/memory/jz4780-nemc.c             |  5 ++---
 drivers/memory/mtk-smi.c                 | 10 ++++------
 drivers/memory/omap-gpmc.c               |  6 ++----
 drivers/memory/renesas-rpc-if.c          |  6 ++----
 drivers/memory/samsung/exynos5422-dmc.c  |  6 ++----
 drivers/memory/stm32-fmc2-ebi.c          |  6 ++----
 drivers/memory/tegra/tegra186-emc.c      |  6 ++----
 drivers/memory/tegra/tegra210-emc-core.c |  6 ++----
 drivers/memory/ti-aemif.c                |  5 ++---
 drivers/memory/ti-emif-pm.c              |  6 ++----
 15 files changed, 32 insertions(+), 60 deletions(-)

base-commit: 17cb8a20bde66a520a2ca7aad1063e1ce7382240
-- 
2.42.0

[PATCH 00/15] memory: Convert to platform remove callback returning void

[Uwe Kleine-König]

Hello,

this series converts the platform drivers below drivers/memory to make
use of .remove_new. See commit 5c5a7680e67b ("platform: Provide a remove
callback that returns no value") for an extended explanation and the
eventual goal. The TL;DR; is to make it harder for driver authors to
leak resources without noticing.

This is merge window material. All patches are pairwise independent of
each other so they can be applied individually. Still it would be great
to let them go in all together.

Best regards
Uwe

Uwe Kleine-König (15):
  memory: brcmstb_dpfe: Convert to platform remove callback returning void
  memory: brcmstb_memc: Convert to platform remove callback returning void
  memory: emif: Convert to platform remove callback returning void
  memory: fsl-corenet-cf: Convert to platform remove callback returning void
  memory: fsl_ifc: Convert to platform remove callback returning void
  memory: jz4780-nemc: Convert to platform remove callback returning void
  memory: mtk-smi: Convert to platform remove callback returning void
  memory: omap-gpmc: Convert to platform remove callback returning void
  memory: renesas-rpc-if: Convert to platform remove callback returning void
  memory: exynos5422-dmc: Convert to platform remove callback returning void
  memory: stm32-fmc2-ebi: Convert to platform remove callback returning void
  memory: tegra186-emc: Convert to platform remove callback returning void
  memory: tegra210-emc: Convert to platform remove callback returning void
  memory: ti-aemif: Convert to platform remove callback returning void
  memory: ti-emif-pm: Convert to platform remove callback returning void

 drivers/memory/brcmstb_dpfe.c            |  6 ++----
 drivers/memory/brcmstb_memc.c            |  6 ++----
 drivers/memory/emif.c                    |  6 ++----
 drivers/memory/fsl-corenet-cf.c          |  6 ++----
 drivers/memory/fsl_ifc.c                 |  6 ++----
 drivers/memory/jz4780-nemc.c             |  5 ++---
 drivers/memory/mtk-smi.c                 | 10 ++++------
 drivers/memory/omap-gpmc.c               |  6 ++----
 drivers/memory/renesas-rpc-if.c          |  6 ++----
 drivers/memory/samsung/exynos5422-dmc.c  |  6 ++----
 drivers/memory/stm32-fmc2-ebi.c          |  6 ++----
 drivers/memory/tegra/tegra186-emc.c      |  6 ++----
 drivers/memory/tegra/tegra210-emc-core.c |  6 ++----
 drivers/memory/ti-aemif.c                |  5 ++---
 drivers/memory/ti-emif-pm.c              |  6 ++----
 15 files changed, 32 insertions(+), 60 deletions(-)

base-commit: 17cb8a20bde66a520a2ca7aad1063e1ce7382240
-- 
2.42.0

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 01/15] memory: brcmstb_dpfe: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/brcmstb_dpfe.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/brcmstb_dpfe.c b/drivers/memory/brcmstb_dpfe.c
index a7ab3d377206..5028467b2dc9 100644
--- a/drivers/memory/brcmstb_dpfe.c
+++ b/drivers/memory/brcmstb_dpfe.c
@@ -909,13 +909,11 @@ static int brcmstb_dpfe_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int brcmstb_dpfe_remove(struct platform_device *pdev)
+static void brcmstb_dpfe_remove(struct platform_device *pdev)
 {
 	struct brcmstb_dpfe_priv *priv = dev_get_drvdata(&pdev->dev);
 
 	sysfs_remove_groups(&pdev->dev.kobj, priv->dpfe_api->sysfs_attrs);
-
-	return 0;
 }
 
 static const struct of_device_id brcmstb_dpfe_of_match[] = {
@@ -936,7 +934,7 @@ static struct platform_driver brcmstb_dpfe_driver = {
 		.of_match_table = brcmstb_dpfe_of_match,
 	},
 	.probe = brcmstb_dpfe_probe,
-	.remove	= brcmstb_dpfe_remove,
+	.remove_new = brcmstb_dpfe_remove,
 	.resume = brcmstb_dpfe_resume,
 };
 
-- 
2.42.0

[PATCH 01/15] memory: brcmstb_dpfe: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/brcmstb_dpfe.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/brcmstb_dpfe.c b/drivers/memory/brcmstb_dpfe.c
index a7ab3d377206..5028467b2dc9 100644
--- a/drivers/memory/brcmstb_dpfe.c
+++ b/drivers/memory/brcmstb_dpfe.c
@@ -909,13 +909,11 @@ static int brcmstb_dpfe_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int brcmstb_dpfe_remove(struct platform_device *pdev)
+static void brcmstb_dpfe_remove(struct platform_device *pdev)
 {
 	struct brcmstb_dpfe_priv *priv = dev_get_drvdata(&pdev->dev);
 
 	sysfs_remove_groups(&pdev->dev.kobj, priv->dpfe_api->sysfs_attrs);
-
-	return 0;
 }
 
 static const struct of_device_id brcmstb_dpfe_of_match[] = {
@@ -936,7 +934,7 @@ static struct platform_driver brcmstb_dpfe_driver = {
 		.of_match_table = brcmstb_dpfe_of_match,
 	},
 	.probe = brcmstb_dpfe_probe,
-	.remove	= brcmstb_dpfe_remove,
+	.remove_new = brcmstb_dpfe_remove,
 	.resume = brcmstb_dpfe_resume,
 };
 
-- 
2.42.0


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 02/15] memory: brcmstb_memc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/brcmstb_memc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/brcmstb_memc.c b/drivers/memory/brcmstb_memc.c
index a6ea51996522..ea9213f7152e 100644
--- a/drivers/memory/brcmstb_memc.c
+++ b/drivers/memory/brcmstb_memc.c
@@ -152,13 +152,11 @@ static int brcmstb_memc_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int brcmstb_memc_remove(struct platform_device *pdev)
+static void brcmstb_memc_remove(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 
 	sysfs_remove_group(&dev->kobj, &dev_attr_group);
-
-	return 0;
 }
 
 enum brcmstb_memc_hwtype {
@@ -284,7 +282,7 @@ static DEFINE_SIMPLE_DEV_PM_OPS(brcmstb_memc_pm_ops, brcmstb_memc_suspend,
 
 static struct platform_driver brcmstb_memc_driver = {
 	.probe = brcmstb_memc_probe,
-	.remove = brcmstb_memc_remove,
+	.remove_new = brcmstb_memc_remove,
 	.driver = {
 		.name		= "brcmstb_memc",
 		.of_match_table	= brcmstb_memc_of_match,
-- 
2.42.0

[PATCH 02/15] memory: brcmstb_memc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/brcmstb_memc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/brcmstb_memc.c b/drivers/memory/brcmstb_memc.c
index a6ea51996522..ea9213f7152e 100644
--- a/drivers/memory/brcmstb_memc.c
+++ b/drivers/memory/brcmstb_memc.c
@@ -152,13 +152,11 @@ static int brcmstb_memc_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int brcmstb_memc_remove(struct platform_device *pdev)
+static void brcmstb_memc_remove(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 
 	sysfs_remove_group(&dev->kobj, &dev_attr_group);
-
-	return 0;
 }
 
 enum brcmstb_memc_hwtype {
@@ -284,7 +282,7 @@ static DEFINE_SIMPLE_DEV_PM_OPS(brcmstb_memc_pm_ops, brcmstb_memc_suspend,
 
 static struct platform_driver brcmstb_memc_driver = {
 	.probe = brcmstb_memc_probe,
-	.remove = brcmstb_memc_remove,
+	.remove_new = brcmstb_memc_remove,
 	.driver = {
 		.name		= "brcmstb_memc",
 		.of_match_table	= brcmstb_memc_of_match,
-- 
2.42.0


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 03/15] memory: emif: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/emif.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index f305643209f0..434982545be6 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -1159,13 +1159,11 @@ static int __init_or_module emif_probe(struct platform_device *pdev)
 	return -ENODEV;
 }
 
-static int __exit emif_remove(struct platform_device *pdev)
+static void __exit emif_remove(struct platform_device *pdev)
 {
 	struct emif_data *emif = platform_get_drvdata(pdev);
 
 	emif_debugfs_exit(emif);
-
-	return 0;
 }
 
 static void emif_shutdown(struct platform_device *pdev)
@@ -1185,7 +1183,7 @@ MODULE_DEVICE_TABLE(of, emif_of_match);
 #endif
 
 static struct platform_driver emif_driver = {
-	.remove		= __exit_p(emif_remove),
+	.remove_new	= __exit_p(emif_remove),
 	.shutdown	= emif_shutdown,
 	.driver = {
 		.name = "emif",
-- 
2.42.0

[PATCH 04/15] memory: fsl-corenet-cf: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/fsl-corenet-cf.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/fsl-corenet-cf.c b/drivers/memory/fsl-corenet-cf.c
index 8096c4f33303..f47d05f7c5c5 100644
--- a/drivers/memory/fsl-corenet-cf.c
+++ b/drivers/memory/fsl-corenet-cf.c
@@ -223,7 +223,7 @@ static int ccf_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int ccf_remove(struct platform_device *pdev)
+static void ccf_remove(struct platform_device *pdev)
 {
 	struct ccf_private *ccf = dev_get_drvdata(&pdev->dev);
 
@@ -241,8 +241,6 @@ static int ccf_remove(struct platform_device *pdev)
 		iowrite32be(0, &ccf->err_regs->errinten);
 		break;
 	}
-
-	return 0;
 }
 
 static struct platform_driver ccf_driver = {
@@ -251,7 +249,7 @@ static struct platform_driver ccf_driver = {
 		.of_match_table = ccf_matches,
 	},
 	.probe = ccf_probe,
-	.remove = ccf_remove,
+	.remove_new = ccf_remove,
 };
 
 module_platform_driver(ccf_driver);
-- 
2.42.0

[PATCH 05/15] memory: fsl_ifc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/fsl_ifc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/fsl_ifc.c b/drivers/memory/fsl_ifc.c
index 2509e5152036..15e919c24f81 100644
--- a/drivers/memory/fsl_ifc.c
+++ b/drivers/memory/fsl_ifc.c
@@ -84,7 +84,7 @@ static int fsl_ifc_ctrl_init(struct fsl_ifc_ctrl *ctrl)
 	return 0;
 }
 
-static int fsl_ifc_ctrl_remove(struct platform_device *dev)
+static void fsl_ifc_ctrl_remove(struct platform_device *dev)
 {
 	struct fsl_ifc_ctrl *ctrl = dev_get_drvdata(&dev->dev);
 
@@ -98,8 +98,6 @@ static int fsl_ifc_ctrl_remove(struct platform_device *dev)
 	iounmap(ctrl->gregs);
 
 	dev_set_drvdata(&dev->dev, NULL);
-
-	return 0;
 }
 
 /*
@@ -318,7 +316,7 @@ static struct platform_driver fsl_ifc_ctrl_driver = {
 		.of_match_table = fsl_ifc_match,
 	},
 	.probe       = fsl_ifc_ctrl_probe,
-	.remove      = fsl_ifc_ctrl_remove,
+	.remove_new  = fsl_ifc_ctrl_remove,
 };
 
 static int __init fsl_ifc_init(void)
-- 
2.42.0

[PATCH 06/15] memory: jz4780-nemc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/jz4780-nemc.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/memory/jz4780-nemc.c b/drivers/memory/jz4780-nemc.c
index e5a93e7da15f..fb6db2ffe71b 100644
--- a/drivers/memory/jz4780-nemc.c
+++ b/drivers/memory/jz4780-nemc.c
@@ -384,12 +384,11 @@ static int jz4780_nemc_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int jz4780_nemc_remove(struct platform_device *pdev)
+static void jz4780_nemc_remove(struct platform_device *pdev)
 {
 	struct jz4780_nemc *nemc = platform_get_drvdata(pdev);
 
 	clk_disable_unprepare(nemc->clk);
-	return 0;
 }
 
 static const struct jz_soc_info jz4740_soc_info = {
@@ -408,7 +407,7 @@ static const struct of_device_id jz4780_nemc_dt_match[] = {
 
 static struct platform_driver jz4780_nemc_driver = {
 	.probe		= jz4780_nemc_probe,
-	.remove		= jz4780_nemc_remove,
+	.remove_new	= jz4780_nemc_remove,
 	.driver	= {
 		.name	= "jz4780-nemc",
 		.of_match_table = of_match_ptr(jz4780_nemc_dt_match),
-- 
2.42.0

[PATCH 07/15] memory: mtk-smi: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/mtk-smi.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c
index 6523cb510518..572c7fbdcfd3 100644
--- a/drivers/memory/mtk-smi.c
+++ b/drivers/memory/mtk-smi.c
@@ -566,14 +566,13 @@ static int mtk_smi_larb_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int mtk_smi_larb_remove(struct platform_device *pdev)
+static void mtk_smi_larb_remove(struct platform_device *pdev)
 {
 	struct mtk_smi_larb *larb = platform_get_drvdata(pdev);
 
 	device_link_remove(&pdev->dev, larb->smi_common_dev);
 	pm_runtime_disable(&pdev->dev);
 	component_del(&pdev->dev, &mtk_smi_larb_component_ops);
-	return 0;
 }
 
 static int __maybe_unused mtk_smi_larb_resume(struct device *dev)
@@ -616,7 +615,7 @@ static const struct dev_pm_ops smi_larb_pm_ops = {
 
 static struct platform_driver mtk_smi_larb_driver = {
 	.probe	= mtk_smi_larb_probe,
-	.remove	= mtk_smi_larb_remove,
+	.remove_new = mtk_smi_larb_remove,
 	.driver	= {
 		.name = "mtk-smi-larb",
 		.of_match_table = mtk_smi_larb_of_ids,
@@ -795,14 +794,13 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int mtk_smi_common_remove(struct platform_device *pdev)
+static void mtk_smi_common_remove(struct platform_device *pdev)
 {
 	struct mtk_smi *common = dev_get_drvdata(&pdev->dev);
 
 	if (common->plat->type == MTK_SMI_GEN2_SUB_COMM)
 		device_link_remove(&pdev->dev, common->smi_common_dev);
 	pm_runtime_disable(&pdev->dev);
-	return 0;
 }
 
 static int __maybe_unused mtk_smi_common_resume(struct device *dev)
@@ -842,7 +840,7 @@ static const struct dev_pm_ops smi_common_pm_ops = {
 
 static struct platform_driver mtk_smi_common_driver = {
 	.probe	= mtk_smi_common_probe,
-	.remove = mtk_smi_common_remove,
+	.remove_new = mtk_smi_common_remove,
 	.driver	= {
 		.name = "mtk-smi-common",
 		.of_match_table = mtk_smi_common_of_ids,
-- 
2.42.0

[PATCH 07/15] memory: mtk-smi: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/mtk-smi.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c
index 6523cb510518..572c7fbdcfd3 100644
--- a/drivers/memory/mtk-smi.c
+++ b/drivers/memory/mtk-smi.c
@@ -566,14 +566,13 @@ static int mtk_smi_larb_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int mtk_smi_larb_remove(struct platform_device *pdev)
+static void mtk_smi_larb_remove(struct platform_device *pdev)
 {
 	struct mtk_smi_larb *larb = platform_get_drvdata(pdev);
 
 	device_link_remove(&pdev->dev, larb->smi_common_dev);
 	pm_runtime_disable(&pdev->dev);
 	component_del(&pdev->dev, &mtk_smi_larb_component_ops);
-	return 0;
 }
 
 static int __maybe_unused mtk_smi_larb_resume(struct device *dev)
@@ -616,7 +615,7 @@ static const struct dev_pm_ops smi_larb_pm_ops = {
 
 static struct platform_driver mtk_smi_larb_driver = {
 	.probe	= mtk_smi_larb_probe,
-	.remove	= mtk_smi_larb_remove,
+	.remove_new = mtk_smi_larb_remove,
 	.driver	= {
 		.name = "mtk-smi-larb",
 		.of_match_table = mtk_smi_larb_of_ids,
@@ -795,14 +794,13 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int mtk_smi_common_remove(struct platform_device *pdev)
+static void mtk_smi_common_remove(struct platform_device *pdev)
 {
 	struct mtk_smi *common = dev_get_drvdata(&pdev->dev);
 
 	if (common->plat->type == MTK_SMI_GEN2_SUB_COMM)
 		device_link_remove(&pdev->dev, common->smi_common_dev);
 	pm_runtime_disable(&pdev->dev);
-	return 0;
 }
 
 static int __maybe_unused mtk_smi_common_resume(struct device *dev)
@@ -842,7 +840,7 @@ static const struct dev_pm_ops smi_common_pm_ops = {
 
 static struct platform_driver mtk_smi_common_driver = {
 	.probe	= mtk_smi_common_probe,
-	.remove = mtk_smi_common_remove,
+	.remove_new = mtk_smi_common_remove,
 	.driver	= {
 		.name = "mtk-smi-common",
 		.of_match_table = mtk_smi_common_of_ids,
-- 
2.42.0


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 08/15] memory: omap-gpmc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/omap-gpmc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/omap-gpmc.c b/drivers/memory/omap-gpmc.c
index d78f73db37c8..80d038884207 100644
--- a/drivers/memory/omap-gpmc.c
+++ b/drivers/memory/omap-gpmc.c
@@ -2690,7 +2690,7 @@ static int gpmc_probe(struct platform_device *pdev)
 	return rc;
 }
 
-static int gpmc_remove(struct platform_device *pdev)
+static void gpmc_remove(struct platform_device *pdev)
 {
 	int i;
 	struct gpmc_device *gpmc = platform_get_drvdata(pdev);
@@ -2702,8 +2702,6 @@ static int gpmc_remove(struct platform_device *pdev)
 	gpmc_mem_exit();
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
-
-	return 0;
 }
 
 #ifdef CONFIG_PM_SLEEP
@@ -2747,7 +2745,7 @@ MODULE_DEVICE_TABLE(of, gpmc_dt_ids);
 
 static struct platform_driver gpmc_driver = {
 	.probe		= gpmc_probe,
-	.remove		= gpmc_remove,
+	.remove_new	= gpmc_remove,
 	.driver		= {
 		.name	= DEVICE_NAME,
 		.of_match_table = of_match_ptr(gpmc_dt_ids),
-- 
2.42.0

[PATCH 09/15] memory: renesas-rpc-if: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/renesas-rpc-if.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/renesas-rpc-if.c b/drivers/memory/renesas-rpc-if.c
index 9695b2d3ae59..3167826b236a 100644
--- a/drivers/memory/renesas-rpc-if.c
+++ b/drivers/memory/renesas-rpc-if.c
@@ -777,13 +777,11 @@ static int rpcif_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int rpcif_remove(struct platform_device *pdev)
+static void rpcif_remove(struct platform_device *pdev)
 {
 	struct rpcif_priv *rpc = platform_get_drvdata(pdev);
 
 	platform_device_unregister(rpc->vdev);
-
-	return 0;
 }
 
 static const struct of_device_id rpcif_of_match[] = {
@@ -797,7 +795,7 @@ MODULE_DEVICE_TABLE(of, rpcif_of_match);
 
 static struct platform_driver rpcif_driver = {
 	.probe	= rpcif_probe,
-	.remove	= rpcif_remove,
+	.remove_new = rpcif_remove,
 	.driver = {
 		.name =	"rpc-if",
 		.of_match_table = rpcif_of_match,
-- 
2.42.0

[PATCH 10/15] memory: exynos5422-dmc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/samsung/exynos5422-dmc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/samsung/exynos5422-dmc.c b/drivers/memory/samsung/exynos5422-dmc.c
index 6d019dbd721c..da7ecd921c72 100644
--- a/drivers/memory/samsung/exynos5422-dmc.c
+++ b/drivers/memory/samsung/exynos5422-dmc.c
@@ -1558,7 +1558,7 @@ static int exynos5_dmc_probe(struct platform_device *pdev)
  * clean the device's resources. It just calls explicitly disable function for
  * the performance counters.
  */
-static int exynos5_dmc_remove(struct platform_device *pdev)
+static void exynos5_dmc_remove(struct platform_device *pdev)
 {
 	struct exynos5_dmc *dmc = dev_get_drvdata(&pdev->dev);
 
@@ -1569,8 +1569,6 @@ static int exynos5_dmc_remove(struct platform_device *pdev)
 
 	clk_disable_unprepare(dmc->mout_bpll);
 	clk_disable_unprepare(dmc->fout_bpll);
-
-	return 0;
 }
 
 static const struct of_device_id exynos5_dmc_of_match[] = {
@@ -1581,7 +1579,7 @@ MODULE_DEVICE_TABLE(of, exynos5_dmc_of_match);
 
 static struct platform_driver exynos5_dmc_platdrv = {
 	.probe	= exynos5_dmc_probe,
-	.remove = exynos5_dmc_remove,
+	.remove_new = exynos5_dmc_remove,
 	.driver = {
 		.name	= "exynos5-dmc",
 		.of_match_table = exynos5_dmc_of_match,
-- 
2.42.0

[PATCH 10/15] memory: exynos5422-dmc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/samsung/exynos5422-dmc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/samsung/exynos5422-dmc.c b/drivers/memory/samsung/exynos5422-dmc.c
index 6d019dbd721c..da7ecd921c72 100644
--- a/drivers/memory/samsung/exynos5422-dmc.c
+++ b/drivers/memory/samsung/exynos5422-dmc.c
@@ -1558,7 +1558,7 @@ static int exynos5_dmc_probe(struct platform_device *pdev)
  * clean the device's resources. It just calls explicitly disable function for
  * the performance counters.
  */
-static int exynos5_dmc_remove(struct platform_device *pdev)
+static void exynos5_dmc_remove(struct platform_device *pdev)
 {
 	struct exynos5_dmc *dmc = dev_get_drvdata(&pdev->dev);
 
@@ -1569,8 +1569,6 @@ static int exynos5_dmc_remove(struct platform_device *pdev)
 
 	clk_disable_unprepare(dmc->mout_bpll);
 	clk_disable_unprepare(dmc->fout_bpll);
-
-	return 0;
 }
 
 static const struct of_device_id exynos5_dmc_of_match[] = {
@@ -1581,7 +1579,7 @@ MODULE_DEVICE_TABLE(of, exynos5_dmc_of_match);
 
 static struct platform_driver exynos5_dmc_platdrv = {
 	.probe	= exynos5_dmc_probe,
-	.remove = exynos5_dmc_remove,
+	.remove_new = exynos5_dmc_remove,
 	.driver = {
 		.name	= "exynos5-dmc",
 		.of_match_table = exynos5_dmc_of_match,
-- 
2.42.0


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 11/15] memory: stm32-fmc2-ebi: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/stm32-fmc2-ebi.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/stm32-fmc2-ebi.c b/drivers/memory/stm32-fmc2-ebi.c
index 9015e8277dc8..47d0ea5f1616 100644
--- a/drivers/memory/stm32-fmc2-ebi.c
+++ b/drivers/memory/stm32-fmc2-ebi.c
@@ -1146,7 +1146,7 @@ static int stm32_fmc2_ebi_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int stm32_fmc2_ebi_remove(struct platform_device *pdev)
+static void stm32_fmc2_ebi_remove(struct platform_device *pdev)
 {
 	struct stm32_fmc2_ebi *ebi = platform_get_drvdata(pdev);
 
@@ -1154,8 +1154,6 @@ static int stm32_fmc2_ebi_remove(struct platform_device *pdev)
 	stm32_fmc2_ebi_disable_banks(ebi);
 	stm32_fmc2_ebi_disable(ebi);
 	clk_disable_unprepare(ebi->clk);
-
-	return 0;
 }
 
 static int __maybe_unused stm32_fmc2_ebi_suspend(struct device *dev)
@@ -1197,7 +1195,7 @@ MODULE_DEVICE_TABLE(of, stm32_fmc2_ebi_match);
 
 static struct platform_driver stm32_fmc2_ebi_driver = {
 	.probe	= stm32_fmc2_ebi_probe,
-	.remove	= stm32_fmc2_ebi_remove,
+	.remove_new = stm32_fmc2_ebi_remove,
 	.driver	= {
 		.name = "stm32_fmc2_ebi",
 		.of_match_table = stm32_fmc2_ebi_match,
-- 
2.42.0

[PATCH 11/15] memory: stm32-fmc2-ebi: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/stm32-fmc2-ebi.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/stm32-fmc2-ebi.c b/drivers/memory/stm32-fmc2-ebi.c
index 9015e8277dc8..47d0ea5f1616 100644
--- a/drivers/memory/stm32-fmc2-ebi.c
+++ b/drivers/memory/stm32-fmc2-ebi.c
@@ -1146,7 +1146,7 @@ static int stm32_fmc2_ebi_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int stm32_fmc2_ebi_remove(struct platform_device *pdev)
+static void stm32_fmc2_ebi_remove(struct platform_device *pdev)
 {
 	struct stm32_fmc2_ebi *ebi = platform_get_drvdata(pdev);
 
@@ -1154,8 +1154,6 @@ static int stm32_fmc2_ebi_remove(struct platform_device *pdev)
 	stm32_fmc2_ebi_disable_banks(ebi);
 	stm32_fmc2_ebi_disable(ebi);
 	clk_disable_unprepare(ebi->clk);
-
-	return 0;
 }
 
 static int __maybe_unused stm32_fmc2_ebi_suspend(struct device *dev)
@@ -1197,7 +1195,7 @@ MODULE_DEVICE_TABLE(of, stm32_fmc2_ebi_match);
 
 static struct platform_driver stm32_fmc2_ebi_driver = {
 	.probe	= stm32_fmc2_ebi_probe,
-	.remove	= stm32_fmc2_ebi_remove,
+	.remove_new = stm32_fmc2_ebi_remove,
 	.driver	= {
 		.name = "stm32_fmc2_ebi",
 		.of_match_table = stm32_fmc2_ebi_match,
-- 
2.42.0


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 12/15] memory: tegra186-emc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/tegra/tegra186-emc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/tegra/tegra186-emc.c b/drivers/memory/tegra/tegra186-emc.c
index 4007f4e16d74..fcd4aea48bda 100644
--- a/drivers/memory/tegra/tegra186-emc.c
+++ b/drivers/memory/tegra/tegra186-emc.c
@@ -378,7 +378,7 @@ static int tegra186_emc_probe(struct platform_device *pdev)
 	return err;
 }
 
-static int tegra186_emc_remove(struct platform_device *pdev)
+static void tegra186_emc_remove(struct platform_device *pdev)
 {
 	struct tegra_mc *mc = dev_get_drvdata(pdev->dev.parent);
 	struct tegra186_emc *emc = platform_get_drvdata(pdev);
@@ -387,8 +387,6 @@ static int tegra186_emc_remove(struct platform_device *pdev)
 
 	mc->bpmp = NULL;
 	tegra_bpmp_put(emc->bpmp);
-
-	return 0;
 }
 
 static const struct of_device_id tegra186_emc_of_match[] = {
@@ -413,7 +411,7 @@ static struct platform_driver tegra186_emc_driver = {
 		.sync_state = icc_sync_state,
 	},
 	.probe = tegra186_emc_probe,
-	.remove = tegra186_emc_remove,
+	.remove_new = tegra186_emc_remove,
 };
 module_platform_driver(tegra186_emc_driver);
 
-- 
2.42.0

[PATCH 13/15] memory: tegra210-emc: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/tegra/tegra210-emc-core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/tegra/tegra210-emc-core.c b/drivers/memory/tegra/tegra210-emc-core.c
index 3300bde47c13..78ca1d6c0977 100644
--- a/drivers/memory/tegra/tegra210-emc-core.c
+++ b/drivers/memory/tegra/tegra210-emc-core.c
@@ -1985,15 +1985,13 @@ static int tegra210_emc_probe(struct platform_device *pdev)
 	return err;
 }
 
-static int tegra210_emc_remove(struct platform_device *pdev)
+static void tegra210_emc_remove(struct platform_device *pdev)
 {
 	struct tegra210_emc *emc = platform_get_drvdata(pdev);
 
 	debugfs_remove_recursive(emc->debugfs.root);
 	tegra210_clk_emc_detach(emc->clk);
 	of_reserved_mem_device_release(emc->dev);
-
-	return 0;
 }
 
 static int __maybe_unused tegra210_emc_suspend(struct device *dev)
@@ -2053,7 +2051,7 @@ static struct platform_driver tegra210_emc_driver = {
 		.pm = &tegra210_emc_pm_ops,
 	},
 	.probe = tegra210_emc_probe,
-	.remove = tegra210_emc_remove,
+	.remove_new = tegra210_emc_remove,
 };
 
 module_platform_driver(tegra210_emc_driver);
-- 
2.42.0

[PATCH 14/15] memory: ti-aemif: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/ti-aemif.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/memory/ti-aemif.c b/drivers/memory/ti-aemif.c
index f81e7df8798a..e192db9e0e4b 100644
--- a/drivers/memory/ti-aemif.c
+++ b/drivers/memory/ti-aemif.c
@@ -427,17 +427,16 @@ static int aemif_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int aemif_remove(struct platform_device *pdev)
+static void aemif_remove(struct platform_device *pdev)
 {
 	struct aemif_device *aemif = platform_get_drvdata(pdev);
 
 	clk_disable_unprepare(aemif->clk);
-	return 0;
 }
 
 static struct platform_driver aemif_driver = {
 	.probe = aemif_probe,
-	.remove = aemif_remove,
+	.remove_new = aemif_remove,
 	.driver = {
 		.name = "ti-aemif",
 		.of_match_table = of_match_ptr(aemif_of_match),
-- 
2.42.0

[PATCH 15/15] memory: ti-emif-pm: Convert to platform remove callback returning void

[Uwe Kleine-König]

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>
---
 drivers/memory/ti-emif-pm.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/memory/ti-emif-pm.c b/drivers/memory/ti-emif-pm.c
index cef0d3beb63b..592f70e9c8e5 100644
--- a/drivers/memory/ti-emif-pm.c
+++ b/drivers/memory/ti-emif-pm.c
@@ -315,15 +315,13 @@ static int ti_emif_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int ti_emif_remove(struct platform_device *pdev)
+static void ti_emif_remove(struct platform_device *pdev)
 {
 	struct ti_emif_data *emif_data = emif_instance;
 
 	emif_instance = NULL;
 
 	ti_emif_free_sram(emif_data);
-
-	return 0;
 }
 
 static const struct dev_pm_ops ti_emif_pm_ops = {
@@ -332,7 +330,7 @@ static const struct dev_pm_ops ti_emif_pm_ops = {
 
 static struct platform_driver ti_emif_driver = {
 	.probe = ti_emif_probe,
-	.remove = ti_emif_remove,
+	.remove_new = ti_emif_remove,
 	.driver = {
 		.name = KBUILD_MODNAME,
 		.of_match_table = ti_emif_of_match,
-- 
2.42.0

Re: [PATCH 06/15] memory: jz4780-nemc: Convert to platform remove callback returning void

[Paul Cercueil]

Hi,

Le dimanche 17 décembre 2023 à 15:29 +0100, Uwe Kleine-König a écrit :
> 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>

Acked-by: Paul Cercueil <paul@crapouillou.net>

Cheers,
-Paul

> ---
>  drivers/memory/jz4780-nemc.c | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/memory/jz4780-nemc.c b/drivers/memory/jz4780-
> nemc.c
> index e5a93e7da15f..fb6db2ffe71b 100644
> --- a/drivers/memory/jz4780-nemc.c
> +++ b/drivers/memory/jz4780-nemc.c
> @@ -384,12 +384,11 @@ static int jz4780_nemc_probe(struct
> platform_device *pdev)
>  	return 0;
>  }
>  
> -static int jz4780_nemc_remove(struct platform_device *pdev)
> +static void jz4780_nemc_remove(struct platform_device *pdev)
>  {
>  	struct jz4780_nemc *nemc = platform_get_drvdata(pdev);
>  
>  	clk_disable_unprepare(nemc->clk);
> -	return 0;
>  }
>  
>  static const struct jz_soc_info jz4740_soc_info = {
> @@ -408,7 +407,7 @@ static const struct of_device_id
> jz4780_nemc_dt_match[] = {
>  
>  static struct platform_driver jz4780_nemc_driver = {
>  	.probe		= jz4780_nemc_probe,
> -	.remove		= jz4780_nemc_remove,
> +	.remove_new	= jz4780_nemc_remove,
>  	.driver	= {
>  		.name	= "jz4780-nemc",
>  		.of_match_table =
> of_match_ptr(jz4780_nemc_dt_match),

Re: [PATCH 01/15] memory: brcmstb_dpfe: Convert to platform remove callback returning void

[Markus Mayer]

On Sun, 17 Dec 2023 at 06:30, 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>
> ---
>  drivers/memory/brcmstb_dpfe.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/memory/brcmstb_dpfe.c b/drivers/memory/brcmstb_dpfe.c
> index a7ab3d377206..5028467b2dc9 100644
> --- a/drivers/memory/brcmstb_dpfe.c
> +++ b/drivers/memory/brcmstb_dpfe.c
> @@ -909,13 +909,11 @@ static int brcmstb_dpfe_probe(struct platform_device *pdev)
>         return ret;
>  }
>
> -static int brcmstb_dpfe_remove(struct platform_device *pdev)
> +static void brcmstb_dpfe_remove(struct platform_device *pdev)
>  {
>         struct brcmstb_dpfe_priv *priv = dev_get_drvdata(&pdev->dev);
>
>         sysfs_remove_groups(&pdev->dev.kobj, priv->dpfe_api->sysfs_attrs);
> -
> -       return 0;
>  }
>
>  static const struct of_device_id brcmstb_dpfe_of_match[] = {
> @@ -936,7 +934,7 @@ static struct platform_driver brcmstb_dpfe_driver = {
>                 .of_match_table = brcmstb_dpfe_of_match,
>         },
>         .probe = brcmstb_dpfe_probe,
> -       .remove = brcmstb_dpfe_remove,
> +       .remove_new = brcmstb_dpfe_remove,
>         .resume = brcmstb_dpfe_resume,
>  };

Acked-by: Markus Mayer <mmayer@broadcom.com>

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 01/15] memory: brcmstb_dpfe: Convert to platform remove callback returning void

[Markus Mayer]

On Sun, 17 Dec 2023 at 06:30, 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>
> ---
>  drivers/memory/brcmstb_dpfe.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/memory/brcmstb_dpfe.c b/drivers/memory/brcmstb_dpfe.c
> index a7ab3d377206..5028467b2dc9 100644
> --- a/drivers/memory/brcmstb_dpfe.c
> +++ b/drivers/memory/brcmstb_dpfe.c
> @@ -909,13 +909,11 @@ static int brcmstb_dpfe_probe(struct platform_device *pdev)
>         return ret;
>  }
>
> -static int brcmstb_dpfe_remove(struct platform_device *pdev)
> +static void brcmstb_dpfe_remove(struct platform_device *pdev)
>  {
>         struct brcmstb_dpfe_priv *priv = dev_get_drvdata(&pdev->dev);
>
>         sysfs_remove_groups(&pdev->dev.kobj, priv->dpfe_api->sysfs_attrs);
> -
> -       return 0;
>  }
>
>  static const struct of_device_id brcmstb_dpfe_of_match[] = {
> @@ -936,7 +934,7 @@ static struct platform_driver brcmstb_dpfe_driver = {
>                 .of_match_table = brcmstb_dpfe_of_match,
>         },
>         .probe = brcmstb_dpfe_probe,
> -       .remove = brcmstb_dpfe_remove,
> +       .remove_new = brcmstb_dpfe_remove,
>         .resume = brcmstb_dpfe_resume,
>  };

Acked-by: Markus Mayer <mmayer@broadcom.com>

Re: [PATCH 08/15] memory: omap-gpmc: Convert to platform remove callback returning void

[Tony Lindgren]

* Uwe Kleine-König <u.kleine-koenig@pengutronix.de> [231217 14:30]:
> Trivially convert this driver from always returning zero in the remove
> callback to the void returning variant.

Reviewed-by: Tony Lindgren <tony@atomide.com>

Re: [PATCH 00/15] memory: Convert to platform remove callback returning void

[Krzysztof Kozlowski]

On Sun, 17 Dec 2023 15:29:26 +0100, Uwe Kleine-König wrote:
> this series converts the platform drivers below drivers/memory to make
> use of .remove_new. See commit 5c5a7680e67b ("platform: Provide a remove
> callback that returns no value") for an extended explanation and the
> eventual goal. The TL;DR; is to make it harder for driver authors to
> leak resources without noticing.
> 
> This is merge window material. All patches are pairwise independent of
> each other so they can be applied individually. Still it would be great
> to let them go in all together.
> 
> [...]

Applied, thanks!

[01/15] memory: brcmstb_dpfe: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/431187eadbc7b0f2650d4e55111b3fff4720f867
[02/15] memory: brcmstb_memc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/f7754712ad6094de5be18674777b265ed4db2f45
[03/15] memory: emif: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/c8a53461990cb697ca494d6671fab9e196d20ce4
[04/15] memory: fsl-corenet-cf: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/021d044b0f9c9a09aa2f778e876e467a8810fb4a
[05/15] memory: fsl_ifc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/f17130855d51f24563a24cd957add769ad59eee9
[06/15] memory: jz4780-nemc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/9024fbbd77b4d73279bbbe2c748a4e4b414d50cc
[07/15] memory: mtk-smi: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/08c1aeaa45ce0fd18912e92c6705586c8aa5240f
[08/15] memory: omap-gpmc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/6a4edb1a4f61e28cc127cd06c470ce3599ee0d9c
[09/15] memory: renesas-rpc-if: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/961abc9f7d6771e8f13db1f4d8b0ffff3f0f41a4
[10/15] memory: exynos5422-dmc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/8013408e4912fb7e469bb8b14fd3a5c956257eec
[11/15] memory: stm32-fmc2-ebi: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/1455b6b0c83132960826d0e527a79a355e096a80
[12/15] memory: tegra186-emc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/dcefa0368458e9e20642dbd2608adae6b22e6464
[13/15] memory: tegra210-emc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/622fa819a2f0f3e6d8322a0b6d3177302ae937b6
[14/15] memory: ti-aemif: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/7852eb8c8ac7e0164b43cc5f8d8245cc3a037620
[15/15] memory: ti-emif-pm: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/365fcc03b6321f36eb7cbda8baa737238c387907

Best regards,
-- 
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

Re: [PATCH 00/15] memory: Convert to platform remove callback returning void

[Krzysztof Kozlowski]

On Sun, 17 Dec 2023 15:29:26 +0100, Uwe Kleine-König wrote:
> this series converts the platform drivers below drivers/memory to make
> use of .remove_new. See commit 5c5a7680e67b ("platform: Provide a remove
> callback that returns no value") for an extended explanation and the
> eventual goal. The TL;DR; is to make it harder for driver authors to
> leak resources without noticing.
> 
> This is merge window material. All patches are pairwise independent of
> each other so they can be applied individually. Still it would be great
> to let them go in all together.
> 
> [...]

Applied, thanks!

[01/15] memory: brcmstb_dpfe: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/431187eadbc7b0f2650d4e55111b3fff4720f867
[02/15] memory: brcmstb_memc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/f7754712ad6094de5be18674777b265ed4db2f45
[03/15] memory: emif: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/c8a53461990cb697ca494d6671fab9e196d20ce4
[04/15] memory: fsl-corenet-cf: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/021d044b0f9c9a09aa2f778e876e467a8810fb4a
[05/15] memory: fsl_ifc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/f17130855d51f24563a24cd957add769ad59eee9
[06/15] memory: jz4780-nemc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/9024fbbd77b4d73279bbbe2c748a4e4b414d50cc
[07/15] memory: mtk-smi: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/08c1aeaa45ce0fd18912e92c6705586c8aa5240f
[08/15] memory: omap-gpmc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/6a4edb1a4f61e28cc127cd06c470ce3599ee0d9c
[09/15] memory: renesas-rpc-if: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/961abc9f7d6771e8f13db1f4d8b0ffff3f0f41a4
[10/15] memory: exynos5422-dmc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/8013408e4912fb7e469bb8b14fd3a5c956257eec
[11/15] memory: stm32-fmc2-ebi: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/1455b6b0c83132960826d0e527a79a355e096a80
[12/15] memory: tegra186-emc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/dcefa0368458e9e20642dbd2608adae6b22e6464
[13/15] memory: tegra210-emc: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/622fa819a2f0f3e6d8322a0b6d3177302ae937b6
[14/15] memory: ti-aemif: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/7852eb8c8ac7e0164b43cc5f8d8245cc3a037620
[15/15] memory: ti-emif-pm: Convert to platform remove callback returning void
        https://git.kernel.org/krzk/linux-mem-ctrl/c/365fcc03b6321f36eb7cbda8baa737238c387907

Best regards,
-- 
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 00/15] memory: Convert to platform remove callback returning void

[Krzysztof Kozlowski]

On 17/12/2023 15:29, Uwe Kleine-König wrote:
> Hello,
> 
> this series converts the platform drivers below drivers/memory to make
> use of .remove_new. See commit 5c5a7680e67b ("platform: Provide a remove
> callback that returns no value") for an extended explanation and the
> eventual goal. The TL;DR; is to make it harder for driver authors to
> leak resources without noticing.

Thanks, applied.
It is however very late in the cycle, so there is a chance this will
miss the merge window. If this happens, I will keep it for the next
cycle (no need for resending).

Best regards,
Krzysztof

Re: [PATCH 00/15] memory: Convert to platform remove callback returning void

[Krzysztof Kozlowski]

On 17/12/2023 15:29, Uwe Kleine-König wrote:
> Hello,
> 
> this series converts the platform drivers below drivers/memory to make
> use of .remove_new. See commit 5c5a7680e67b ("platform: Provide a remove
> callback that returns no value") for an extended explanation and the
> eventual goal. The TL;DR; is to make it harder for driver authors to
> leak resources without noticing.

Thanks, applied.
It is however very late in the cycle, so there is a chance this will
miss the merge window. If this happens, I will keep it for the next
cycle (no need for resending).

Best regards,
Krzysztof


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel