Re: [RFC] sched: make callers check lock contention for cond_resched_lock()

[Takuya Yoshikawa <takuya.yoshikawa@gmail.com>]

On Fri, 18 May 2012 09:26:05 +0200
Ingo Molnar <mingo@kernel.org> wrote:

> I'm not sure we had a usable spin_is_contended() back then, nor 
> was the !PREEMPT case in my mind really.

The fact that both spin_needbreak() and spin_is_contended() can be
used outside of sched is a bit confusing.

For example, in mm/compaction.c we are using spin_is_contended(), but
in mm/memory.c spin_needbreak().

BTW, the actual users of spin_is_contended() look to be only:
	mm/compaction.c
	security/keys/gc.c

> ( The patch looks ugly though, in 99% of the lines it just does
>   something that cond_resched_lock() itself could do. )

Please ignore the patch.  I have already found a way to solve my
problem without cond_resched().

Thanks,
	Takuya