| Message ID | 20230919133207.1400430-10-u.kleine-koenig@pengutronix.de (mailing list archive) |
|---|---|
| State | Handled Elsewhere |
| Headers | show |
| Series | dma: Convert to platform remove callback returning void | expand |
Hi Uwe, Le mardi 19 septembre 2023 à 15:31 +0200, 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() is renamed to .remove(). "is renamed" -> "will be renamed"? > > Trivially convert this driver from always returning zero in the > remove > callback to the void returning variant. > > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Reviewed-by: Paul Cercueil <paul@crapouillou.net> Cheers, -Paul > --- > drivers/dma/dma-jz4780.c | 6 ++---- > 1 file changed, 2 insertions(+), 4 deletions(-) > > diff --git a/drivers/dma/dma-jz4780.c b/drivers/dma/dma-jz4780.c > index adbd47bd6adf..c9cfa341db51 100644 > --- a/drivers/dma/dma-jz4780.c > +++ b/drivers/dma/dma-jz4780.c > @@ -1008,7 +1008,7 @@ static int jz4780_dma_probe(struct > platform_device *pdev) > return ret; > } > > -static int jz4780_dma_remove(struct platform_device *pdev) > +static void jz4780_dma_remove(struct platform_device *pdev) > { > struct jz4780_dma_dev *jzdma = platform_get_drvdata(pdev); > int i; > @@ -1020,8 +1020,6 @@ static int jz4780_dma_remove(struct > platform_device *pdev) > > for (i = 0; i < jzdma->soc_data->nb_channels; i++) > tasklet_kill(&jzdma->chan[i].vchan.task); > - > - return 0; > } > > static const struct jz4780_dma_soc_data jz4740_dma_soc_data = { > @@ -1124,7 +1122,7 @@ MODULE_DEVICE_TABLE(of, jz4780_dma_dt_match); > > static struct platform_driver jz4780_dma_driver = { > .probe = jz4780_dma_probe, > - .remove = jz4780_dma_remove, > + .remove_new = jz4780_dma_remove, > .driver = { > .name = "jz4780-dma", > .of_match_table = jz4780_dma_dt_match,
On Tue, Sep 19, 2023 at 03:39:10PM +0200, Paul Cercueil wrote: > Hi Uwe, > > Le mardi 19 septembre 2023 à 15:31 +0200, 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() is renamed to .remove(). > > "is renamed" -> "will be renamed"? I guess you're right here. I fixed that in my template, but I won't resend this series (and the others with the same issue) for now. Thanks Uwe
On 19/9/23 15:31, 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 <u.kleine-koenig@pengutronix.de> > --- > drivers/dma/dma-jz4780.c | 6 ++---- > 1 file changed, 2 insertions(+), 4 deletions(-) Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
diff --git a/drivers/dma/dma-jz4780.c b/drivers/dma/dma-jz4780.c index adbd47bd6adf..c9cfa341db51 100644 --- a/drivers/dma/dma-jz4780.c +++ b/drivers/dma/dma-jz4780.c @@ -1008,7 +1008,7 @@ static int jz4780_dma_probe(struct platform_device *pdev) return ret; } -static int jz4780_dma_remove(struct platform_device *pdev) +static void jz4780_dma_remove(struct platform_device *pdev) { struct jz4780_dma_dev *jzdma = platform_get_drvdata(pdev); int i; @@ -1020,8 +1020,6 @@ static int jz4780_dma_remove(struct platform_device *pdev) for (i = 0; i < jzdma->soc_data->nb_channels; i++) tasklet_kill(&jzdma->chan[i].vchan.task); - - return 0; } static const struct jz4780_dma_soc_data jz4740_dma_soc_data = { @@ -1124,7 +1122,7 @@ MODULE_DEVICE_TABLE(of, jz4780_dma_dt_match); static struct platform_driver jz4780_dma_driver = { .probe = jz4780_dma_probe, - .remove = jz4780_dma_remove, + .remove_new = jz4780_dma_remove, .driver = { .name = "jz4780-dma", .of_match_table = jz4780_dma_dt_match,
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 <u.kleine-koenig@pengutronix.de> --- drivers/dma/dma-jz4780.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)