Re: [PATCH RFC] PM: runtime: Drop device usage only after remove is done

[Wysocki, Rafael J]

CC-ing linux-pm would have been nice (done now).

On 4/2/2023 12:39 PM, Uwe Kleine-König wrote:
> Many device drivers call one of the variants of pm_runtime_resume() in
> their remove callback. So calling pm_runtime_put_sync() just before that
> is ineffective as the suspend callback might just be called just to
> resume the device directly afterwards again.
>
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
> ---
> Hello,
>
> this is a theoretical issue that I noticed while doing some research how
> pm-runtime works. Not sure it makes sense, so I marked it as RFC.
>
> Best regards
> Uwe
>
>   drivers/base/dd.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> index 8def2ba08a82..6beac141d3d2 100644
> --- a/drivers/base/dd.c
> +++ b/drivers/base/dd.c
> @@ -1235,10 +1235,10 @@ static void __device_release_driver(struct device *dev, struct device *parent)
>   
>   		bus_notify(dev, BUS_NOTIFY_UNBIND_DRIVER);
>   
> -		pm_runtime_put_sync(dev);
> -
>   		device_remove(dev);
>   
> +		pm_runtime_put_sync(dev);
> +
>   		if (dev->bus && dev->bus->dma_cleanup)
>   			dev->bus->dma_cleanup(dev);
>   
>
> base-commit: fe15c26ee26efa11741a7b632e9f23b01aca4cc6