[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20110913074004.GA11397@mtj.dyndns.org>
Date: Tue, 13 Sep 2011 16:40:04 +0900
From: Tejun Heo <tj@...nel.org>
To: Huang Ying <ying.huang@...el.com>
Cc: Eric Dumazet <eric.dumazet@...il.com>,
Christoph Lameter <cl@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: What is the difference between percpu_xxx and this_cpu_xxx?
On Tue, Sep 13, 2011 at 02:14:06PM +0800, Huang Ying wrote:
> Hi, Eric,
>
> On 09/13/2011 02:11 PM, Eric Dumazet wrote:
> > Le mardi 13 septembre 2011 à 12:28 +0800, Huang Ying a écrit :
> >> Hi,
> >>
> >> If my understanding was correct, from the kernel source code, it appears
> >> that percpu_xxx and this_cpu_xxx have the same semantics and
> >> implementation in effect, haven’t them?
> >>
> >> If so, why do we need two API for the same thing?
> >
> > History mostly, and transition.
> >
> > percpu_inc() was an x86 only thing
> >
> > __get_cpu_var(var)++ -> percpu_inc(var) -> this_cpu_inc(var)
Also, we used to need separate accessors for static and dynamic percpu
variables, so we have quite a few duplications in the interface.
> Thanks for clarification. Is there someone working on the transition?
> Is that possible to translate most percpu_xxx to this_cpu_xxx blindly?
It's on my todo list but AFAIK nobody is actually working on it yet.
Thanks.
--
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists