Hello Dmitry, for this patch's Subject my git filter-branch cmdline failed. I guess it should be something like input: i8042-sparcio - Convert to platform remove callback returning void instead. I'll wait until the dust settles. In the meantime feel free to skip this patch or adapt the Subject yourself. In the former case I'll come back to this patch later. Thanks and sorry for the annoyance, Uwe
diff --git a/drivers/input/serio/i8042-sparcio.h b/drivers/input/serio/i8042-sparcio.h index b68793bf05c8..c2fda54dc384 100644 --- a/drivers/input/serio/i8042-sparcio.h +++ b/drivers/input/serio/i8042-sparcio.h @@ -82,11 +82,9 @@ static int sparc_i8042_probe(struct platform_device *op) return 0; } -static int sparc_i8042_remove(struct platform_device *op) +static void sparc_i8042_remove(struct platform_device *op) { of_iounmap(kbd_res, kbd_iobase, 8); - - return 0; } static const struct of_device_id sparc_i8042_match[] = { @@ -103,7 +101,7 @@ static struct platform_driver sparc_i8042_driver = { .of_match_table = sparc_i8042_match, }, .probe = sparc_i8042_probe, - .remove = sparc_i8042_remove, + .remove_new = sparc_i8042_remove, }; static bool i8042_is_mr_coffee(void)
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/input/serio/i8042-sparcio.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)