On Wed, Jul 31, 2019 at 4:10 PM Rafael J. Wysocki <[email protected]> wrote: > > On Thu, Aug 1, 2019 at 12:59 AM Tri Vo <[email protected]> wrote: > > > > On Wed, Jul 31, 2019 at 3:42 PM Rafael J. Wysocki <[email protected]> > > wrote: > > > > > > On Thursday, August 1, 2019 12:31:16 AM CEST Tri Vo wrote: > > > > On Wed, Jul 31, 2019 at 3:17 PM Rafael J. Wysocki <[email protected]> > > > > wrote: > > > > > > > > > > On Wednesday, July 31, 2019 11:59:32 PM CEST Stephen Boyd wrote: > > > > > > Quoting Tri Vo (2019-07-31 14:55:14) > > > > > > > +/** > > > > > > > + * wakeup_source_sysfs_add - Add wakeup_source attributes to > > > > > > > sysfs. > > > > > > > + * @parent: Device given wakeup source is associated with (or > > > > > > > NULL if virtual). > > > > > > > + * @ws: Wakeup source to be added in sysfs. > > > > > > > + */ > > > > > > > +int wakeup_source_sysfs_add(struct device *parent, struct > > > > > > > wakeup_source *ws) > > > > > > > +{ > > > > > > > + struct device *dev; > > > > > > > + int id; > > > > > > > + > > > > > > > + id = ida_alloc(&wakeup_ida, GFP_KERNEL); > > > > > > > > > > So can anyone remind me why the IDA thing is needed here at all? > > > > > > > > IDA is used to generate the directory name ("ws%d", ID) that is unique > > > > among wakeup_sources. That is what ends up in > > > > /sys/class/wakeup/ws<ID>/* path. > > > > > > That's not my point (see below). > > > > > > > > > > + if (id < 0) > > > > > > > + return id; > > > > > > > + ws->id = id; > > > > > > > + > > > > > > > + dev = device_create_with_groups(wakeup_class, parent, > > > > > > > MKDEV(0, 0), ws, > > > > > > > + wakeup_source_groups, > > > > > > > "ws%d", > > > > > > > > > > > > I thought the name was going to still be 'wakeupN'? > > > > > > > > > > So can't we prefix the wakeup source name with something like > > > > > "wakeup:" or similar here? > > > > > > > > "ws%d" here is the name in the sysfs path rather than the name of the > > > > wakeup source. Wakeup source name is not altered in this patch. > > > > > > > > > > So why wouldn't something like this suffice: > > > > > > dev = device_create_with_groups(wakeup_class, parent, MKDEV(0, 0), ws, > > > wakeup_source_groups, "wakeup:%s", > > > ws->name); > > > > > > ? > > > > ws->name is inherited from the device name. IIUC device names are not > > guaranteed to be unique. So if different devices with the same name > > register wakeup sources, there is an error. > > OK > > So I guess the names are retained for backwards compatibility with > existing user space that may be using them?
Yes, in Android we do rely on the name to aggregate statistics across a fleet of devices. That wouldn't be possible with just the id, as those are generated at dynamically runtime. > > That's kind of fair enough, but having two different identification > schemes for wakeup sources will end up confusing. It's not without precedent though. rtc, input, and other devices have a similar scheme.
