Re: [patch v2 0/5] percpu_counter: bug fix and enhancement

[Eric Dumazet <eric.dumazet@gmail.com>]

Le vendredi 13 mai 2011 à 12:37 +0800, Shaohua Li a écrit :
> On Thu, 2011-05-12 at 17:05 +0800, Tejun Heo wrote:
> > Hello,
> > 
> > On Thu, May 12, 2011 at 11:02:15AM +0200, Eric Dumazet wrote:
> > > > I don't think @maxfuzzy is necessary there.  I wrote this before but
> > > > why can't we track the actual deviation instead of the number of
> > > > deviation events?
> > > 
> > > Thats roughly same thing (BATCH multiplicator factor apart)
> > > 
> > > Most percpu_counter users for a given percpu_counter object use a given
> > > BATCH, dont they ?
> > 
> > Well, @maxfuzzy is much harder than @batch.  It's way less intuitive.
> > Although I haven't really thought about it that much, I think it might
> > be possible to eliminate it.  Maybe I'm confused.  I'll take another
> > look later but if someone can think of something, please jump right
> > in.
> Hmm, looks Eric's approach doesn't work. because we want to remove lock
> in _add, checking seq in _sum still races with _add.
> 

Why ?

I'll code a patch, I believe it should work.

A seqcount is not a 'lock'.

The thing is we want _add to be real fast, so it must not hit a lock set
in _sum()

[Think about a machine with 4096 cpus]