Return-Path: Received: from mail.kernel.org ([198.145.29.99]:35540 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726842AbgEZGdi (ORCPT ); Tue, 26 May 2020 02:33:38 -0400 Date: Tue, 26 May 2020 08:33:34 +0200 From: Greg Kroah-Hartman Subject: Re: [PATCH 1/8] driver core: Add helper for accessing Power Management callbacs Message-ID: <20200526063334.GB2578492@kroah.com> References: <20200525182608.1823735-1-kw@linux.com> <20200525182608.1823735-2-kw@linux.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20200525182608.1823735-2-kw@linux.com> Sender: linux-s390-owner@vger.kernel.org List-ID: To: Krzysztof =?utf-8?Q?Wilczy=C5=84ski?= Cc: Dan Carpenter , "Rafael J. Wysocki" , Len Brown , Kevin Hilman , Ulf Hansson , Pavel Machek , Johan Hovold , Alex Elder , Bjorn Helgaas , "James E.J. Bottomley" , "Martin K. Petersen" , Felipe Balbi , Julian Wiedmann , Karsten Graul , Ursula Braun , Jakub Kicinski , Bjorn Andersson , John Stultz , "David S. Miller" , greybus-dev@lists.linaro.org, netdev@vger.kernel.org, linux-acpi@vger.kernel.org, linux-pci@vger.kernel.org, linux-pm@vger.kernel.org, linux-s390@vger.kernel.org, linux-scsi@vger.kernel.org, linux-usb@vger.kernel.org On Mon, May 25, 2020 at 06:26:01PM +0000, Krzysztof Wilczyński wrote: > Add driver_to_pm() helper allowing for accessing the Power Management > callbacs for a particular device. Access to the callbacs (struct > dev_pm_ops) is normally done through using the pm pointer that is > embedded within the device_driver struct. > > Helper allows for the code required to reference the pm pointer and > access Power Management callbas to be simplified. Changing the > following: > > struct device_driver *drv = dev->driver; > if (dev->driver && dev->driver->pm && dev->driver->pm->prepare) { > int ret = dev->driver->pm->prepare(dev); > > To: > > const struct dev_pm_ops *pm = driver_to_pm(dev->driver); > if (pm && pm->prepare) { > int ret = pm->prepare(dev); > > Or, changing the following: > > const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; > > To: > const struct dev_pm_ops *pm = driver_to_pm(dev->driver); > > Signed-off-by: Krzysztof Wilczyński > --- > include/linux/device/driver.h | 15 +++++++++++++++ > 1 file changed, 15 insertions(+) > > diff --git a/include/linux/device/driver.h b/include/linux/device/driver.h > index ee7ba5b5417e..ccd0b315fd93 100644 > --- a/include/linux/device/driver.h > +++ b/include/linux/device/driver.h > @@ -236,6 +236,21 @@ driver_find_device_by_acpi_dev(struct device_driver *drv, const void *adev) > } > #endif > > +/** > + * driver_to_pm - Return Power Management callbacs (struct dev_pm_ops) for > + * a particular device. > + * @drv: Pointer to a device (struct device_driver) for which you want to access > + * the Power Management callbacks. > + * > + * Returns a pointer to the struct dev_pm_ops embedded within the device (struct > + * device_driver), or returns NULL if Power Management is not present and the > + * pointer is not valid. > + */ > +static inline const struct dev_pm_ops *driver_to_pm(struct device_driver *drv) > +{ > + return drv && drv->pm ? drv->pm : NULL; I hate ? : lines with a passion, as they break normal pattern mattching in my brain. Please just spell this all out: if (drv && drv->pm) return drv->pm; return NULL; Much easier to read, and the compiler will do the exact same thing. Only place ? : are ok to use in my opinion, are as function arguments. thanks, greg k-h