From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ravikiran G Thirumalai Subject: Re: [patch 3/4] net: Percpufy frequently used variables -- proto.sockets_allocated Date: Fri, 27 Jan 2006 16:40:27 -0800 Message-ID: <20060128004027.GE3565@localhost.localdomain> References: <20060126185649.GB3651@localhost.localdomain> <20060126190357.GE3651@localhost.localdomain> <43D9DFA1.9070802@cosmosbay.com> <20060127195227.GA3565@localhost.localdomain> <20060127121602.18bc3f25.akpm@osdl.org> <43DA9EFF.1020200@cosmosbay.com> <20060127225036.GC3565@localhost.localdomain> <43DAAAE3.2030107@cosmosbay.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Andrew Morton , davem@davemloft.net, linux-kernel@vger.kernel.org, shai@scalex86.org, netdev@vger.kernel.org, pravins@calsoftinc.com Return-path: To: Eric Dumazet Content-Disposition: inline In-Reply-To: <43DAAAE3.2030107@cosmosbay.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Sat, Jan 28, 2006 at 12:21:07AM +0100, Eric Dumazet wrote: > Ravikiran G Thirumalai a =E9crit : > >On Fri, Jan 27, 2006 at 11:30:23PM +0100, Eric Dumazet wrote: >=20 > Why not use a boot time allocated percpu area (as done today in=20 > setup_per_cpu_areas()), but instead of reserving extra space for modu= le's=20 > percpu data, being able to serve alloc_percpu() from this reserved ar= ea (ie=20 > no kmalloced data anymore), and keeping your At that time ia64 placed a limit on the max size of per-CPU area=20 (PERCPU_ENOUGH_ROOM). I think that the limit is still there, But hopef= ully 64K per-CPU should be enough for static + dynamic + modules? Let me do a allyesconfig on my box and verify. Thanks, Kiran