Re: [PATCH] PM: runtime: Add kerneldoc comments to multiple helpers
From: Sakari Ailus <sakari.ailus@linux.intel.com>
Date: 2020-08-03 08:54:24
Also in:
lkml
Hi Rafael, Thanks for the patch. On Fri, Jul 31, 2020 at 07:03:26PM +0200, Rafael J. Wysocki wrote:
quoted hunk ↗ jump to hunk
From: Rafael J. Wysocki <redacted> Add kerneldoc comments to multiple PM-runtime helper functions defined as static inline wrappers around lower-level routines to provide quick reference decumentation of their behavior. Some of them are similar to each other with subtle differences only and the behavior of some of them may appear as counter-intuitive, so clarify all that to avoid confusion. Signed-off-by: Rafael J. Wysocki <redacted> --- include/linux/pm_runtime.h | 246 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 246 insertions(+) Index: linux-pm/include/linux/pm_runtime.h ===================================================================--- linux-pm.orig/include/linux/pm_runtime.h +++ linux-pm/include/linux/pm_runtime.h@@ -60,58 +60,151 @@ extern void pm_runtime_put_suppliers(str extern void pm_runtime_new_link(struct device *dev); extern void pm_runtime_drop_link(struct device *dev); +/** + * pm_runtime_get_if_in_use - Conditionally bump up runtime PM usage counter. + * @dev: Target device. + * + * Increment the runtime PM usage counter of @dev if its runtime PM status is + * %RPM_ACTIVE and its runtime PM usage counter is greater than 0.
The implementation of the non-runtime PM variants (used when CONFIG_PM is disabled) isn't here but I think it'd be nice if their behaviour was also documented here. pm_runtime_get_if_in_use() returns -EINVAL if CONFIG_PM is disabled, for instance. pm_runtime_disable() is defined here but the documentation in corresponding pm_runtime_enable() in drivers/base/power/runtime.c is rather terse. It'd be nice to improve that now (or separately). -- Kind regards, Sakari Ailus