Hello,
this series converts all platform drivers below drivers/auxdisplay to
stop using struct platform_driver::remove(). See commit 5c5a7680e67b
("platform: Provide a remove callback that returns no value") for an
extended explanation and the eventual goal.
All conversations are trivial, because the driver's .remove() callbacks
returned zero unconditionally.
There are no interdependencies between these patches, still I'd expect
them to be picked up together. This is merge window material.
Best regards
Uwe
Uwe Kleine-König (3):
auxdisplay: cfag12864bfb: Convert to platform remove callback
returning void
auxdisplay: hd44780: Convert to platform remove callback returning
void
auxdisplay: img-ascii-lcd: Convert to platform remove callback
returning void
drivers/auxdisplay/cfag12864bfb.c | 6 ++----
drivers/auxdisplay/hd44780.c | 5 ++---
drivers/auxdisplay/img-ascii-lcd.c | 5 ++---
3 files changed, 6 insertions(+), 10 deletions(-)
base-commit: 8ffc8b1bbd505e27e2c8439d326b6059c906c9dd
--
2.43.0
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 <[email protected]>
---
drivers/auxdisplay/cfag12864bfb.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/auxdisplay/cfag12864bfb.c b/drivers/auxdisplay/cfag12864bfb.c
index 5ba19c339f08..2b74dabe7e17 100644
--- a/drivers/auxdisplay/cfag12864bfb.c
+++ b/drivers/auxdisplay/cfag12864bfb.c
@@ -96,7 +96,7 @@ static int cfag12864bfb_probe(struct platform_device *device)
return ret;
}
-static int cfag12864bfb_remove(struct platform_device *device)
+static void cfag12864bfb_remove(struct platform_device *device)
{
struct fb_info *info = platform_get_drvdata(device);
@@ -104,13 +104,11 @@ static int cfag12864bfb_remove(struct platform_device *device)
unregister_framebuffer(info);
framebuffer_release(info);
}
-
- return 0;
}
static struct platform_driver cfag12864bfb_driver = {
.probe = cfag12864bfb_probe,
- .remove = cfag12864bfb_remove,
+ .remove_new = cfag12864bfb_remove,
.driver = {
.name = CFAG12864BFB_NAME,
},
--
2.43.0
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 <[email protected]>
---
drivers/auxdisplay/hd44780.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/auxdisplay/hd44780.c b/drivers/auxdisplay/hd44780.c
index d56a5d508ccd..7ac0b1b1d548 100644
--- a/drivers/auxdisplay/hd44780.c
+++ b/drivers/auxdisplay/hd44780.c
@@ -319,7 +319,7 @@ static int hd44780_probe(struct platform_device *pdev)
return ret;
}
-static int hd44780_remove(struct platform_device *pdev)
+static void hd44780_remove(struct platform_device *pdev)
{
struct charlcd *lcd = platform_get_drvdata(pdev);
struct hd44780_common *hdc = lcd->drvdata;
@@ -329,7 +329,6 @@ static int hd44780_remove(struct platform_device *pdev)
kfree(lcd->drvdata);
kfree(lcd);
- return 0;
}
static const struct of_device_id hd44780_of_match[] = {
@@ -340,7 +339,7 @@ MODULE_DEVICE_TABLE(of, hd44780_of_match);
static struct platform_driver hd44780_driver = {
.probe = hd44780_probe,
- .remove = hd44780_remove,
+ .remove_new = hd44780_remove,
.driver = {
.name = "hd44780",
.of_match_table = hd44780_of_match,
--
2.43.0
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 <[email protected]>
---
drivers/auxdisplay/img-ascii-lcd.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/auxdisplay/img-ascii-lcd.c b/drivers/auxdisplay/img-ascii-lcd.c
index 925c4cd101e9..0c91e39389c1 100644
--- a/drivers/auxdisplay/img-ascii-lcd.c
+++ b/drivers/auxdisplay/img-ascii-lcd.c
@@ -279,13 +279,12 @@ static int img_ascii_lcd_probe(struct platform_device *pdev)
*
* Return: 0
*/
-static int img_ascii_lcd_remove(struct platform_device *pdev)
+static void img_ascii_lcd_remove(struct platform_device *pdev)
{
struct img_ascii_lcd_ctx *ctx = platform_get_drvdata(pdev);
sysfs_remove_link(&pdev->dev.kobj, "message");
linedisp_unregister(&ctx->linedisp);
- return 0;
}
static struct platform_driver img_ascii_lcd_driver = {
@@ -294,7 +293,7 @@ static struct platform_driver img_ascii_lcd_driver = {
.of_match_table = img_ascii_lcd_matches,
},
.probe = img_ascii_lcd_probe,
- .remove = img_ascii_lcd_remove,
+ .remove_new = img_ascii_lcd_remove,
};
module_platform_driver(img_ascii_lcd_driver);
--
2.43.0
On Mon, Mar 11, 2024 at 10:59 PM Uwe Kleine-König
<[email protected]> 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 <[email protected]>
Reviewed-by: Geert Uytterhoeven <[email protected]>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
On Mon, Mar 11, 2024 at 10:59 PM Uwe Kleine-König
<[email protected]> 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 <[email protected]>
Reviewed-by: Geert Uytterhoeven <[email protected]>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Uwe,
On Mon, Mar 11, 2024 at 10:59 PM Uwe Kleine-König
<[email protected]> 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 <[email protected]>
Thanks for your patch!
> --- a/drivers/auxdisplay/img-ascii-lcd.c
> +++ b/drivers/auxdisplay/img-ascii-lcd.c
> @@ -279,13 +279,12 @@ static int img_ascii_lcd_probe(struct platform_device *pdev)
> *
> * Return: 0
Please remove this line.
> */
> -static int img_ascii_lcd_remove(struct platform_device *pdev)
> +static void img_ascii_lcd_remove(struct platform_device *pdev)
> {
> struct img_ascii_lcd_ctx *ctx = platform_get_drvdata(pdev);
>
> sysfs_remove_link(&pdev->dev.kobj, "message");
> linedisp_unregister(&ctx->linedisp);
> - return 0;
> }
>
> static struct platform_driver img_ascii_lcd_driver = {
> @@ -294,7 +293,7 @@ static struct platform_driver img_ascii_lcd_driver = {
> .of_match_table = img_ascii_lcd_matches,
> },
> .probe = img_ascii_lcd_probe,
> - .remove = img_ascii_lcd_remove,
> + .remove_new = img_ascii_lcd_remove,
> };
> module_platform_driver(img_ascii_lcd_driver);
With the above fixed:
Reviewed-by: Geert Uytterhoeven <[email protected]>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
On Tue, Mar 12, 2024 at 09:17:54AM +0100, Geert Uytterhoeven wrote:
> Hi Uwe,
>
> On Mon, Mar 11, 2024 at 10:59 PM Uwe Kleine-König
> <[email protected]> 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 <[email protected]>
>
> Thanks for your patch!
>
> > --- a/drivers/auxdisplay/img-ascii-lcd.c
> > +++ b/drivers/auxdisplay/img-ascii-lcd.c
> > @@ -279,13 +279,12 @@ static int img_ascii_lcd_probe(struct platform_device *pdev)
> > *
> > * Return: 0
>
> Please remove this line.
Good catch, thanks.
@Andy: Should I resend this patch? Or the whole series? Or do you want
to fixup while applying?
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | https://www.pengutronix.de/ |
On Mon, Mar 11, 2024 at 10:59 PM Uwe Kleine-König
<[email protected]> 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 <[email protected]>
Acked-by: Miguel Ojeda <[email protected]>
Thanks!
Cheers,
Miguel
On Mon, Mar 11, 2024 at 10:59:21PM +0100, Uwe Kleine-K?nig wrote:
> Hello,
>
> this series converts all platform drivers below drivers/auxdisplay to
> stop using struct platform_driver::remove(). See commit 5c5a7680e67b
> ("platform: Provide a remove callback that returns no value") for an
> extended explanation and the eventual goal.
>
> All conversations are trivial, because the driver's .remove() callbacks
> returned zero unconditionally.
>
> There are no interdependencies between these patches, still I'd expect
> them to be picked up together. This is merge window material.
Pushed to my review and testing queue, thanks!
--
With Best Regards,
Andy Shevchenko
On Tue, Mar 12, 2024 at 11:27:37AM +0100, Uwe Kleine-König wrote:
> On Tue, Mar 12, 2024 at 09:17:54AM +0100, Geert Uytterhoeven wrote:
> > On Mon, Mar 11, 2024 at 10:59 PM Uwe Kleine-König
> > <[email protected]> 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 <[email protected]>
> >
> > Thanks for your patch!
> >
> > > --- a/drivers/auxdisplay/img-ascii-lcd.c
> > > +++ b/drivers/auxdisplay/img-ascii-lcd.c
> > > @@ -279,13 +279,12 @@ static int img_ascii_lcd_probe(struct platform_device *pdev)
> > > *
> > > * Return: 0
> >
> > Please remove this line.
>
> Good catch, thanks.
>
> @Andy: Should I resend this patch? Or the whole series? Or do you want
> to fixup while applying?
Not big deal, I'll fix it whilst applying.
Thank you!
--
With Best Regards,
Andy Shevchenko