Re: [patch] convert aio event reap to use atomic-op instead of spin_lock

[Andrew Morton <akpm@linux-foundation.org>]

On Wed, 11 Apr 2007 11:00:38 -0700
Zach Brown <zach.brown@oracle.com> wrote:

> > -	/* Compensate for the ring buffer's head/tail overlap entry */
> > -	nr_events += 2;	/* 1 is required, 2 for good luck */
> > +	/* round nr_event to next power of 2 */
> > +	nr_events = roundup_pow_of_two(nr_events);
> 
> Is that buggy?  How will the code tell if head == tail means a full ring
> or an empty ring?  The old code added that extra event to let it tell
> the ring was full before head == tail and it would think it's empty
> again, I think.  I'm guessing roundup(nr_events + 1) is needed.

Ken uses the other (superior!) way of implementing ringbuffers: the head
and tail pointers (the naming of which AIO appears to have reversed) are
not constrained to the ringsize - they are simply allowed to wrap through
0xfffffff.  Consequently:

ring full:		(head-tail == size)
ring empty:		head==tail
numer-of-items-in-ring:	head-tail
add to ring:		ring[head++]=item
remove from ring: 	item=ring[tail++]

(adjust the above for AIO naming assbackwardness)
(requires that size be a power of 2)

Many net drivers do it this way for their DMA rings.