[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170320145938.cmj5m33uxunladli@hirez.programming.kicks-ass.net>
Date: Mon, 20 Mar 2017 15:59:38 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Eric Dumazet <eric.dumazet@...il.com>
Cc: Herbert Xu <herbert@...dor.apana.org.au>,
David Miller <davem@...emloft.net>, elena.reshetova@...el.com,
keescook@...omium.org, netdev@...r.kernel.org,
bridge@...ts.linux-foundation.org, linux-kernel@...r.kernel.org,
kuznet@....inr.ac.ru, jmorris@...ei.org, kaber@...sh.net,
stephen@...workplumber.org, ishkamiel@...il.com,
dwindsor@...il.com, akpm@...ux-foundation.org
Subject: Re: [PATCH 07/17] net: convert sock.sk_refcnt from atomic_t to
refcount_t
On Mon, Mar 20, 2017 at 07:51:01AM -0700, Eric Dumazet wrote:
> PowerPC has no efficient atomic_inc() and this definitely shows on
> network intensive workloads involving concurrent cores/threads.
Correct, PPC LL/SC are dreadfully expensive.
> atomic_cmpxchg() on PowerPC is horribly more expensive because of the
> added two SYNC instructions.
Note that refcount_t uses atomic_cmpxchg_release() and
atomic_cmpxchg_relaxed() which avoid most of the painful barriers.
Powered by blists - more mailing lists