sched/core: Separate out io_schedule_prepare() and io_schedule_finish()
authorTejun Heo <[email protected]>
Fri, 28 Oct 2016 16:58:10 +0000 (12:58 -0400)
committerIngo Molnar <[email protected]>
Sat, 14 Jan 2017 10:30:04 +0000 (11:30 +0100)
commit10ab56434f2f633a51e432ee8b7c29e12438e163
tree12407831c0c8d7001ed3220510eff75e9d77979e
parente33a9bba85a869b85fd0a7f16e21a5ec8977e325
sched/core: Separate out io_schedule_prepare() and io_schedule_finish()

Now that IO schedule accounting is done inside __schedule(),
io_schedule() can be split into three steps - prep, schedule, and
finish - where the schedule part doesn't need any special annotation.
This allows marking a sleep as iowait by simply wrapping an existing
blocking function with io_schedule_prepare() and io_schedule_finish().

Because task_struct->in_iowait is single bit, the caller of
io_schedule_prepare() needs to record and the pass its state to
io_schedule_finish() to be safe regarding nesting.  While this isn't
the prettiest, these functions are mostly gonna be used by core
functions and we don't want to use more space for ->in_iowait.

While at it, as it's simple to do now, reimplement io_schedule()
without unnecessarily going through io_schedule_timeout().

Signed-off-by: Tejun Heo <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
include/linux/sched.h
kernel/sched/core.c