Re: sysfs: Add an 'events' class. (was: Re: [RFC][PATCH] perf: sysfstype id)

[Kay Sievers]

On Wed, Nov 10, 2010 at 16:00, Ingo Molnar <mingo@xxxxxxx> wrote:
> * Kay Sievers <kay.sievers@xxxxxxxx> wrote:
>
>> > Same here - calling events a 'bus' is like totally brain-dead IMHO. It implies
>> > something hardware, while many events are not related to any hardware component
>> > but are pure software abstractions: such as context-switches, or syscall
>> > entries, or VM events.
>>
>> Please use a bus, regardless of the name, class is dead and you will need stuff
>> tha can only happen at bus. We'll fix the name later.
>
> When is that 'later' supposed to happen? git annotate says that the
> renaming/unification comments/plans were added:
>
> 46336009    Â(Kay Sievers  Â2007-06-08 13:36:37 -0700    105) Âsymlinks pointing to the unified /sys/devices tree.
>
> ... about 3 and a half years ago.

If all works out as planned, we will do it during the next weeks.
Depends on a bit what Greg and I are dragged into, which is always
kind of unpredictable. It's 'just' a cleanup and stuff that's broken
always comes first, and comes a lot. :)

> Btw., where will thinks like /sys/kernel/slab/ end up? It would be useful (in the
> long run) to have:
>
> Â/sys/kernel/slab/uid_cache/events/
>
> To describe/control events for the uid_cache SLAB cache alone.

It will probably stay as it is, they are not 'struct device' devices,
but raw kobjects. It's a completely unconnected custom tree, which is
not related to the driver core or anything similar.

It's more like its own filesystem. Stuff you put there can not show up
in /sys/class,bus,subsystem unless someone wants convert all of it to
real devices, let them show up in /sys/devices, and assign it a
subsystem (bus).

Kay
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/