Re: [PATCH-v2 1/3] percpu_ida: Make percpu_ida_alloc + callersaccept task state bitmask

From: Kent Overstreet
Date: Thu Jan 23 2014 - 08:27:15 EST

On Thu, Jan 23, 2014 at 01:47:53PM +0100, Peter Zijlstra wrote:
> On Tue, Jan 21, 2014 at 02:18:52PM -0800, Kent Overstreet wrote:
> > > I do not get the comment near prepare to wait -- why does it matter if
> > > percpu_ida_free() flips a cpus_have_tags bit?
> >
> > Did I write that comment? It is a crappy comment...
> >
> > Ok, in userspace we'd be using condition variables here, but this is the kernel
> > so we need to carefully order putting ourselves on a waitlist, and checking the
> > condition that determines whether we wait, and on the wakeup end changing things
> > that affect that condition and doing the wakeup. steal_tags() is checking the
> > condition that goes with the prepare_to_wait(), that's all.
> How about something like so?

I like it - my only concern is that your patch has the effect of calling
__alloc_global_tag() twice before sleeping on alloc failure - given that
we're also doing a prepare_to_wait() I'm not concerned about touching
the global freelist twice, but we're also calling steal_tags() twice and
that's potentially more expensive.

It should be ok, because I expect when steal_tags() is going to fail
most of the time it'll check the bitmap and not run the loop, but I
think there's enough room for pathological behaviour here to sleep on

pool->lock is also going to be fairly badly contended in the worst case,
and that can get real bad real fast... now that I think about it we
probably want to avoid the __alloc_global_tag() double call just because
of that, pool->lock is going to be quite a bit more contended than the
waitlist lock just because fo the amount of work done under it.

though my old code was also calling prepare_to_wait() with pool->lock
held, which was dumb.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at