From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: ext4 performance falloff Date: Mon, 07 Apr 2014 09:40:28 -0700 Message-ID: <87bnwdjdoj.fsf@tassilo.jf.intel.com> References: <533EE547.3030504@numascale.com> <20140404205604.GC10275@thunk.org> <533F7851.30803@numascale.com> <20140407141935.GA22171@quack.suse.cz> Mime-Version: 1.0 Content-Type: text/plain Cc: Daniel J Blueman , Theodore Ts'o , linux-ext4@vger.kernel.org, LKML , Steffen Persvold , Andreas Dilger To: Jan Kara Return-path: In-Reply-To: <20140407141935.GA22171@quack.suse.cz> (Jan Kara's message of "Mon, 7 Apr 2014 16:19:35 +0200") Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org Jan Kara writes: > > What we really need is a counter where we can better estimate counts > accumulated in the percpu part of it. As the counter approaches zero, it's > CPU overhead will have to become that of a single locked variable but when > the value of counter is relatively high, we want it to be fast as the > percpu one. Possibly, each CPU could "reserve" part of the value in the > counter (by just decrementing the total value; how large that part should > be really needs to depend to the total value of the counter and number of > CPUs - in this regard we really differ from classical percpu couters) and > allocate/free using that part. If CPU cannot reserve what it is asked for > anymore, it would go and steal from parts other CPUs have accumulated, > returning them to global pool until it can satisfy the allocation. That's a percpu_counter() isn't it? (or cookie jar) The MM uses similar techniques. -Andi -- ak@linux.intel.com -- Speaking for myself only