From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail191.messagelabs.com (mail191.messagelabs.com [216.82.242.19]) by kanga.kvack.org (Postfix) with ESMTP id B798F6B0093 for ; Tue, 24 Nov 2009 16:22:18 -0500 (EST) Received: from zps76.corp.google.com (zps76.corp.google.com [172.25.146.76]) by smtp-out.google.com with ESMTP id nAOLMEWV001107 for ; Tue, 24 Nov 2009 21:22:15 GMT Received: from pzk6 (pzk6.prod.google.com [10.243.19.134]) by zps76.corp.google.com with ESMTP id nAOLLNpO007576 for ; Tue, 24 Nov 2009 13:22:10 -0800 Received: by pzk6 with SMTP id 6so4843099pzk.29 for ; Tue, 24 Nov 2009 13:22:10 -0800 (PST) Date: Tue, 24 Nov 2009 13:22:08 -0800 (PST) From: David Rientjes Subject: Re: lockdep complaints in slab allocator In-Reply-To: <1259097150.4531.1822.camel@laptop> Message-ID: References: <84144f020911192249l6c7fa495t1a05294c8f5b6ac8@mail.gmail.com> <1258709153.11284.429.camel@laptop> <84144f020911200238w3d3ecb38k92ca595beee31de5@mail.gmail.com> <1258714328.11284.522.camel@laptop> <4B067816.6070304@cs.helsinki.fi> <1258729748.4104.223.camel@laptop> <1259002800.5630.1.camel@penberg-laptop> <1259003425.17871.328.camel@calx> <4B0ADEF5.9040001@cs.helsinki.fi> <1259080406.4531.1645.camel@laptop> <20091124170032.GC6831@linux.vnet.ibm.com> <1259082756.17871.607.camel@calx> <1259086459.4531.1752.camel@laptop> <1259090615.17871.696.camel@calx> <1259095580.4531.1788.camel@laptop> <1259096004.17871.716.camel@calx> <1259096519.4531.1809.camel@laptop> <1259097150.4531.1822.camel@laptop> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org To: Peter Zijlstra Cc: Matt Mackall , paulmck@linux.vnet.ibm.com, Pekka Enberg , linux-mm@kvack.org, Christoph Lameter , LKML , Nick Piggin List-ID: On Tue, 24 Nov 2009, Peter Zijlstra wrote: > > slqb still has a 5-10% performance regression compared to slab for > > benchmarks such as netperf TCP_RR on machines with high cpu counts, > > forcing that type of regression isn't acceptable. > > Having _4_ slab allocators is equally unacceptable. > So you just advocated to merging slqb so that it gets more testing and development, and then use its inclusion in a statistic to say we should remove others solely because the space is too cluttered? We use slab partially because the regression in slub was too severe for some of our benchmarks, and while CONFIG_SLUB may be the kernel default there are still distros that use slab as the default as well. We cannot simply remove an allocator that is superior to others because it is old or has increased complexity. I'd suggest looking at how widely used slob is and whether it has a significant advantage over slub. We'd then have two allocators for specialized workloads (and slub is much better for diagnostics) and one in development. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org