| Message ID | da6af483-5ee9-45cd-922e-d9d5364674dc@web.de (mailing list archive) |
|---|---|
| State | Handled Elsewhere, archived |
| Headers | show |
| Series | pmdomain: mediatek: Use devm_platform_get_and_ioremap_resource() in init_scp() | expand |
Hi Markus, On Mon, Feb 5, 2024 at 3:23 PM Markus Elfring <Markus.Elfring@web.de> wrote: > From: Markus Elfring <elfring@users.sourceforge.net> > Date: Mon, 5 Feb 2024 15:08:27 +0100 > > A wrapper function is available since the commit 890cc39a879906b63912482dfc41944579df2dc6 > ("drivers: provide devm_platform_get_and_ioremap_resource()"). > Thus reuse existing functionality instead of keeping duplicate source code. > > This issue was detected by using the Coccinelle software. > > Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Thanks for your patch! > --- a/drivers/pmdomain/mediatek/mtk-scpsys.c > +++ b/drivers/pmdomain/mediatek/mtk-scpsys.c > @@ -441,8 +441,7 @@ static struct scp *init_scp(struct platform_device *pdev, > > scp->dev = &pdev->dev; > > - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); > - scp->base = devm_ioremap_resource(&pdev->dev, res); > + scp->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res); Given res is further unused, please use devm_platform_ioremap_resource() instead, and remove the local variable res. > if (IS_ERR(scp->base)) > return ERR_CAST(scp->base); > Gr{oetje,eeting}s, Geert
>> +++ b/drivers/pmdomain/mediatek/mtk-scpsys.c >> @@ -441,8 +441,7 @@ static struct scp *init_scp(struct platform_device *pdev, >> >> scp->dev = &pdev->dev; >> >> - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); >> - scp->base = devm_ioremap_resource(&pdev->dev, res); >> + scp->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res); > > Given res is further unused, please use devm_platform_ioremap_resource() > instead, and remove the local variable res. I got another idea after looking at the implementation of the function “devm_platform_get_and_ioremap_resource” once more. https://elixir.bootlin.com/linux/v6.8-rc3/source/drivers/base/platform.c#L87 It seems that it is supported to pass a null pointer for the last parameter (while this possibility is not mentioned in the interface description so far). How do you think about to benefit from such a design option any more (instead of the determination of a corresponding platform device)? Regards, Markus
Hi Markus, On Mon, Feb 5, 2024 at 6:28 PM Markus Elfring <Markus.Elfring@web.de> wrote: > >> +++ b/drivers/pmdomain/mediatek/mtk-scpsys.c > >> @@ -441,8 +441,7 @@ static struct scp *init_scp(struct platform_device *pdev, > >> > >> scp->dev = &pdev->dev; > >> > >> - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); > >> - scp->base = devm_ioremap_resource(&pdev->dev, res); > >> + scp->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res); > > > > Given res is further unused, please use devm_platform_ioremap_resource() > > instead, and remove the local variable res. > > I got another idea after looking at the implementation of the function > “devm_platform_get_and_ioremap_resource” once more. > https://elixir.bootlin.com/linux/v6.8-rc3/source/drivers/base/platform.c#L87 > > It seems that it is supported to pass a null pointer for the last parameter > (while this possibility is not mentioned in the interface description so far). > How do you think about to benefit from such a design option any more > (instead of the determination of a corresponding platform device)? Yes, you can pass a NULL pointer as the last parameter. And as this is very common, the wrapper devm_platform_ioremap_resource() exists. Gr{oetje,eeting}s, Geert
… >> I got another idea after looking at the implementation of the function >> “devm_platform_get_and_ioremap_resource” once more. >> https://elixir.bootlin.com/linux/v6.8-rc3/source/drivers/base/platform.c#L87 … > Yes, you can pass a NULL pointer as the last parameter. Would you like to support any approaches which can make interface descriptions clearer for such an implementation detail? > And as this is very common, the wrapper devm_platform_ioremap_resource() exists. I find further collateral evolution interesting for the involved parameter reduction. Regards, Markus
Il 05/02/24 17:31, Geert Uytterhoeven ha scritto: > Hi Markus, > > On Mon, Feb 5, 2024 at 3:23 PM Markus Elfring <Markus.Elfring@web.de> wrote: >> From: Markus Elfring <elfring@users.sourceforge.net> >> Date: Mon, 5 Feb 2024 15:08:27 +0100 >> >> A wrapper function is available since the commit 890cc39a879906b63912482dfc41944579df2dc6 >> ("drivers: provide devm_platform_get_and_ioremap_resource()"). >> Thus reuse existing functionality instead of keeping duplicate source code. >> >> This issue was detected by using the Coccinelle software. >> >> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> > > Thanks for your patch! > >> --- a/drivers/pmdomain/mediatek/mtk-scpsys.c >> +++ b/drivers/pmdomain/mediatek/mtk-scpsys.c >> @@ -441,8 +441,7 @@ static struct scp *init_scp(struct platform_device *pdev, >> >> scp->dev = &pdev->dev; >> >> - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); >> - scp->base = devm_ioremap_resource(&pdev->dev, res); >> + scp->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res); > > Given res is further unused, please use devm_platform_ioremap_resource() > instead, and remove the local variable res. > Agreed: res will never be used. Cheers, Angelo
diff --git a/drivers/pmdomain/mediatek/mtk-scpsys.c b/drivers/pmdomain/mediatek/mtk-scpsys.c index b374d01fdac7..9e7f0771e7e4 100644 --- a/drivers/pmdomain/mediatek/mtk-scpsys.c +++ b/drivers/pmdomain/mediatek/mtk-scpsys.c @@ -441,8 +441,7 @@ static struct scp *init_scp(struct platform_device *pdev, scp->dev = &pdev->dev; - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - scp->base = devm_ioremap_resource(&pdev->dev, res); + scp->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res); if (IS_ERR(scp->base)) return ERR_CAST(scp->base);