From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rusty Russell Subject: Re: [patch 9/11] net: dst_entry.refcount, use, lastuse to use alloc_percpu Date: Wed, 14 Sep 2005 17:21:02 +1000 Message-ID: <1126682462.7896.103.camel@localhost.localdomain> References: <20050913220737.GA6249@localhost.localdomain> <20050913.151216.48124942.davem@davemloft.net> <20050913231717.GC6249@localhost.localdomain> <20050913.162748.86496945.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: kiran@scalex86.org, akpm@osdl.org, linux-kernel@vger.kernel.org, dipankar@in.ibm.com, bharata@in.ibm.com, shai@scalex86.org, netdev@vger.kernel.org Return-path: To: "David S. Miller" In-Reply-To: <20050913.162748.86496945.davem@davemloft.net> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Tue, 2005-09-13 at 16:27 -0700, David S. Miller wrote: > From: Ravikiran G Thirumalai > Date: Tue, 13 Sep 2005 16:17:17 -0700 > > > But even 1 Million dst cache entries would be 16+4 MB additional for > > a 4 cpu box....is that too much? > > Absolutely. > > Per-cpu counters are great for things like single instance > statistics et al. But once you start doing them per-object > that's out of control bloat as far as I'm concerned. This is why my original per-cpu allocator patch was damn slow, and GFP_KERNEL only. I wasn't convinced that high-churn objects are a good fit for spreading across cpus. I thought that net devices and modules (which uses a primitive hard-coded "bigref" currently) were a fair uses for bigrefs, though I'd like to see some stats. Cheers, Rusty. -- A bad analogy is like a leaky screwdriver -- Richard Braakman