Hello,
this series convert all platform drivers below drivers/net/mdio to
use remove_new. The motivation is to get rid of an integer return code
that is (mostly) ignored by the platform driver core and error prone on
the driver side.
See commit 5c5a7680e67b ("platform: Provide a remove callback that
returns no value") for an extended explanation and the eventual goal.
There are no interdependencies between the patches. As there are still
quite a few drivers to convert, I'm happy about every patch that makes
it in. So even if there is a merge conflict with one patch until you
apply, please apply the remainder of this series anyhow.
Best regards
Uwe
Uwe Kleine-König (19):
net: mdio: aspeed: Convert to platform remove callback returning void
net: mdio: bcm-iproc: Convert to platform remove callback returning
void
net: mdio: bcm-unimac: Convert to platform remove callback returning
void
net: mdio: gpio: Convert to platform remove callback returning void
net: mdio: hisi-femac: Convert to platform remove callback returning
void
net: mdio: ipq4019: Convert to platform remove callback returning void
net: mdio: ipq8064: Convert to platform remove callback returning void
net: mdio: moxart: Convert to platform remove callback returning void
net: mdio: mscc-miim: Convert to platform remove callback returning
void
net: mdio: mux-bcm-iproc: Convert to platform remove callback
returning void
net: mdio: mux-bcm6368: Convert to platform remove callback returning
void
net: mdio: mux-gpio: Convert to platform remove callback returning
void
net: mdio: mux-meson-g12a: Convert to platform remove callback
returning void
net: mdio: mux-meson-gxl: Convert to platform remove callback
returning void
net: mdio: mux-mmioreg: Convert to platform remove callback returning
void
net: mdio: mux-multiplexer: Convert to platform remove callback
returning void
net: mdio: octeon: Convert to platform remove callback returning void
net: mdio: sun4i: Convert to platform remove callback returning void
net: mdio: xgene: Convert to platform remove callback returning void
drivers/net/mdio/mdio-aspeed.c | 6 ++----
drivers/net/mdio/mdio-bcm-iproc.c | 6 ++----
drivers/net/mdio/mdio-bcm-unimac.c | 6 ++----
drivers/net/mdio/mdio-gpio.c | 6 ++----
drivers/net/mdio/mdio-hisi-femac.c | 6 ++----
drivers/net/mdio/mdio-ipq4019.c | 6 ++----
drivers/net/mdio/mdio-ipq8064.c | 7 ++-----
drivers/net/mdio/mdio-moxart.c | 6 ++----
drivers/net/mdio/mdio-mscc-miim.c | 6 ++----
drivers/net/mdio/mdio-mux-bcm-iproc.c | 6 ++----
drivers/net/mdio/mdio-mux-bcm6368.c | 6 ++----
drivers/net/mdio/mdio-mux-gpio.c | 5 ++---
drivers/net/mdio/mdio-mux-meson-g12a.c | 6 ++----
drivers/net/mdio/mdio-mux-meson-gxl.c | 6 ++----
drivers/net/mdio/mdio-mux-mmioreg.c | 6 ++----
drivers/net/mdio/mdio-mux-multiplexer.c | 6 ++----
drivers/net/mdio/mdio-octeon.c | 5 ++---
drivers/net/mdio/mdio-sun4i.c | 6 ++----
drivers/net/mdio/mdio-xgene.c | 6 ++----
19 files changed, 38 insertions(+), 75 deletions(-)
base-commit: 7fc7222d9680366edeecc219c21ca96310bdbc10
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-mux-meson-g12a.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mdio/mdio-mux-meson-g12a.c b/drivers/net/mdio/mdio-mux-meson-g12a.c
index 910e5cf74e89..754b0f2cf15b 100644
--- a/drivers/net/mdio/mdio-mux-meson-g12a.c
+++ b/drivers/net/mdio/mdio-mux-meson-g12a.c
@@ -336,7 +336,7 @@ static int g12a_mdio_mux_probe(struct platform_device *pdev)
return ret;
}
-static int g12a_mdio_mux_remove(struct platform_device *pdev)
+static void g12a_mdio_mux_remove(struct platform_device *pdev)
{
struct g12a_mdio_mux *priv = platform_get_drvdata(pdev);
@@ -344,13 +344,11 @@ static int g12a_mdio_mux_remove(struct platform_device *pdev)
if (__clk_is_enabled(priv->pll))
clk_disable_unprepare(priv->pll);
-
- return 0;
}
static struct platform_driver g12a_mdio_mux_driver = {
.probe = g12a_mdio_mux_probe,
- .remove = g12a_mdio_mux_remove,
+ .remove_new = g12a_mdio_mux_remove,
.driver = {
.name = "g12a-mdio_mux",
.of_match_table = g12a_mdio_mux_match,
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-mux-gpio.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/net/mdio/mdio-mux-gpio.c b/drivers/net/mdio/mdio-mux-gpio.c
index 3c7f16f06b45..38fb031f8979 100644
--- a/drivers/net/mdio/mdio-mux-gpio.c
+++ b/drivers/net/mdio/mdio-mux-gpio.c
@@ -62,11 +62,10 @@ static int mdio_mux_gpio_probe(struct platform_device *pdev)
return 0;
}
-static int mdio_mux_gpio_remove(struct platform_device *pdev)
+static void mdio_mux_gpio_remove(struct platform_device *pdev)
{
struct mdio_mux_gpio_state *s = dev_get_platdata(&pdev->dev);
mdio_mux_uninit(s->mux_handle);
- return 0;
}
static const struct of_device_id mdio_mux_gpio_match[] = {
@@ -87,7 +86,7 @@ static struct platform_driver mdio_mux_gpio_driver = {
.of_match_table = mdio_mux_gpio_match,
},
.probe = mdio_mux_gpio_probe,
- .remove = mdio_mux_gpio_remove,
+ .remove_new = mdio_mux_gpio_remove,
};
module_platform_driver(mdio_mux_gpio_driver);
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-mux-bcm6368.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mdio/mdio-mux-bcm6368.c b/drivers/net/mdio/mdio-mux-bcm6368.c
index 8b444a8eb6b5..1b77e0e3e6e1 100644
--- a/drivers/net/mdio/mdio-mux-bcm6368.c
+++ b/drivers/net/mdio/mdio-mux-bcm6368.c
@@ -153,14 +153,12 @@ static int bcm6368_mdiomux_probe(struct platform_device *pdev)
return rc;
}
-static int bcm6368_mdiomux_remove(struct platform_device *pdev)
+static void bcm6368_mdiomux_remove(struct platform_device *pdev)
{
struct bcm6368_mdiomux_desc *md = platform_get_drvdata(pdev);
mdio_mux_uninit(md->mux_handle);
mdiobus_unregister(md->mii_bus);
-
- return 0;
}
static const struct of_device_id bcm6368_mdiomux_ids[] = {
@@ -175,7 +173,7 @@ static struct platform_driver bcm6368_mdiomux_driver = {
.of_match_table = bcm6368_mdiomux_ids,
},
.probe = bcm6368_mdiomux_probe,
- .remove = bcm6368_mdiomux_remove,
+ .remove_new = bcm6368_mdiomux_remove,
};
module_platform_driver(bcm6368_mdiomux_driver);
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-ipq8064.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/net/mdio/mdio-ipq8064.c b/drivers/net/mdio/mdio-ipq8064.c
index fd9716960106..f71b6e1c66e4 100644
--- a/drivers/net/mdio/mdio-ipq8064.c
+++ b/drivers/net/mdio/mdio-ipq8064.c
@@ -147,14 +147,11 @@ ipq8064_mdio_probe(struct platform_device *pdev)
return 0;
}
-static int
-ipq8064_mdio_remove(struct platform_device *pdev)
+static void ipq8064_mdio_remove(struct platform_device *pdev)
{
struct mii_bus *bus = platform_get_drvdata(pdev);
mdiobus_unregister(bus);
-
- return 0;
}
static const struct of_device_id ipq8064_mdio_dt_ids[] = {
@@ -165,7 +162,7 @@ MODULE_DEVICE_TABLE(of, ipq8064_mdio_dt_ids);
static struct platform_driver ipq8064_mdio_driver = {
.probe = ipq8064_mdio_probe,
- .remove = ipq8064_mdio_remove,
+ .remove_new = ipq8064_mdio_remove,
.driver = {
.name = "ipq8064-mdio",
.of_match_table = ipq8064_mdio_dt_ids,
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-bcm-iproc.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mdio/mdio-bcm-iproc.c b/drivers/net/mdio/mdio-bcm-iproc.c
index 77fc970cdfde..5a2d26c6afdc 100644
--- a/drivers/net/mdio/mdio-bcm-iproc.c
+++ b/drivers/net/mdio/mdio-bcm-iproc.c
@@ -168,14 +168,12 @@ static int iproc_mdio_probe(struct platform_device *pdev)
return rc;
}
-static int iproc_mdio_remove(struct platform_device *pdev)
+static void iproc_mdio_remove(struct platform_device *pdev)
{
struct iproc_mdio_priv *priv = platform_get_drvdata(pdev);
mdiobus_unregister(priv->mii_bus);
mdiobus_free(priv->mii_bus);
-
- return 0;
}
#ifdef CONFIG_PM_SLEEP
@@ -210,7 +208,7 @@ static struct platform_driver iproc_mdio_driver = {
#endif
},
.probe = iproc_mdio_probe,
- .remove = iproc_mdio_remove,
+ .remove_new = iproc_mdio_remove,
};
module_platform_driver(iproc_mdio_driver);
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-mux-meson-gxl.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mdio/mdio-mux-meson-gxl.c b/drivers/net/mdio/mdio-mux-meson-gxl.c
index 76188575ca1f..89554021b5cc 100644
--- a/drivers/net/mdio/mdio-mux-meson-gxl.c
+++ b/drivers/net/mdio/mdio-mux-meson-gxl.c
@@ -140,18 +140,16 @@ static int gxl_mdio_mux_probe(struct platform_device *pdev)
return ret;
}
-static int gxl_mdio_mux_remove(struct platform_device *pdev)
+static void gxl_mdio_mux_remove(struct platform_device *pdev)
{
struct gxl_mdio_mux *priv = platform_get_drvdata(pdev);
mdio_mux_uninit(priv->mux_handle);
-
- return 0;
}
static struct platform_driver gxl_mdio_mux_driver = {
.probe = gxl_mdio_mux_probe,
- .remove = gxl_mdio_mux_remove,
+ .remove_new = gxl_mdio_mux_remove,
.driver = {
.name = "gxl-mdio-mux",
.of_match_table = gxl_mdio_mux_match,
--
2.40.1
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() is 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 <[email protected]>
---
drivers/net/mdio/mdio-mux-multiplexer.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/mdio/mdio-mux-multiplexer.c b/drivers/net/mdio/mdio-mux-multiplexer.c
index bfa5af577b0a..569b13383191 100644
--- a/drivers/net/mdio/mdio-mux-multiplexer.c
+++ b/drivers/net/mdio/mdio-mux-multiplexer.c
@@ -85,7 +85,7 @@ static int mdio_mux_multiplexer_probe(struct platform_device *pdev)
return ret;
}
-static int mdio_mux_multiplexer_remove(struct platform_device *pdev)
+static void mdio_mux_multiplexer_remove(struct platform_device *pdev)
{
struct mdio_mux_multiplexer_state *s = platform_get_drvdata(pdev);
@@ -93,8 +93,6 @@ static int mdio_mux_multiplexer_remove(struct platform_device *pdev)
if (s->do_deselect)
mux_control_deselect(s->muxc);
-
- return 0;
}
static const struct of_device_id mdio_mux_multiplexer_match[] = {
@@ -109,7 +107,7 @@ static struct platform_driver mdio_mux_multiplexer_driver = {
.of_match_table = mdio_mux_multiplexer_match,
},
.probe = mdio_mux_multiplexer_probe,
- .remove = mdio_mux_multiplexer_remove,
+ .remove_new = mdio_mux_multiplexer_remove,
};
module_platform_driver(mdio_mux_multiplexer_driver);
--
2.40.1
On 9/18/23 12:50, Uwe Kleine-König 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() is 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 <[email protected]>
Reviewed-by: Florian Fainelli <[email protected]>
--
Florian
On Mon, Sep 18, 2023 at 09:50:43PM +0200, Uwe Kleine-König wrote:
> Hello,
>
> this series convert all platform drivers below drivers/net/mdio to
> use remove_new. The motivation is to get rid of an integer return code
> that is (mostly) ignored by the platform driver core and error prone on
> the driver side.
>
> See commit 5c5a7680e67b ("platform: Provide a remove callback that
> returns no value") for an extended explanation and the eventual goal.
>
> There are no interdependencies between the patches. As there are still
> quite a few drivers to convert, I'm happy about every patch that makes
> it in. So even if there is a merge conflict with one patch until you
> apply, please apply the remainder of this series anyhow.
>
> Best regards
> Uwe
For series,
Reviewed-by: Simon Horman <[email protected]>