From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?B?SsO2cm4=?= Engel Subject: Re: [00/41] Large Blocksize Support V7 (adds memmap support) Date: Sun, 23 Sep 2007 19:44:03 +0200 Message-ID: <20070923174402.GA15525@lazybastard.org> References: <20070911121225.GE13132@lazybastard.org> <20070915014449.4f9cdb51.akpm@linux-foundation.org> <87ir6c3z2l.fsf@informatik.uni-tuebingen.de> <20070915155100.GA21861@v2.random> <87tzpvy9cb.fsf@informatik.uni-tuebingen.de> <20070915223032.GA6708@v2.random> <20070916174657.GA2393@lazybastard.org> <20070916182136.GC2393@lazybastard.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: =?utf-8?B?SsO2cm4=?= Engel , Andrea Arcangeli , Goswin von Brederlow , Andrew Morton , Nick Piggin , Christoph Lameter , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Christoph Hellwig , Mel Gorman , William Lee Irwin III , David Chinner , Jens Axboe , Badari Pulavarty , Maxim Levitsky , Fengguang Wu , swin wang , totty.lu@gmail.com, hugh@veritas.com To: Linus Torvalds Return-path: Received: from lazybastard.de ([212.112.238.170]:35801 "EHLO longford.lazybastard.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753644AbXIWRsv (ORCPT ); Sun, 23 Sep 2007 13:48:51 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Sun, 16 September 2007 11:44:09 -0700, Linus Torvalds wrote: > On Sun, 16 Sep 2007, J=C3=B6rn Engel wrote: > >=20 > > My approach is to have one for mount points and ramfs/tmpfs/sysfs/e= tc. > > which are pinned for their entire lifetime and another for regular > > files/inodes. One could take a three-way approach and have > > always-pinned, often-pinned and rarely-pinned. > >=20 > > We won't get never-pinned that way. >=20 > That sounds pretty good. The problem, of course, is that most of the = time,=20 > the actual dentry allocation itself is done before you really know wh= ich=20 > case the dentry will be in, and the natural place for actually giving= the=20 > dentry lifetime hint is *not* at "d_alloc()", but when we "instantiat= e"=20 > it with d_add() or d_instantiate(). >=20 > [...] >=20 > And yes, you'd end up with the reallocation overhead quite often, but= at=20 > least it would now happen only when filling in a dentry, not in the=20 > (*much* more critical) cached lookup path. There may be another approach. We could create a never-pinned cache, without trying hard to keep it full. Instead of moving a hot dentry at dput() time, we move a cold one from the end of lru. And if the lru list is short, we just chicken out. Our definition of "short lru list" can either be based on a ratio of pinned to unpinned dentries or on a metric of cache hits vs. cache misses. I tend to dislike the cache hit metric, because updatedb would cause tons of misses and result in the same mess we have right now. With this double cache, we have a source of slabs to cheaply reap under memory pressure, but still have a performance advantage (memcpy beats disk io by orders of magnitude). J=C3=B6rn --=20 The story so far: In the beginning the Universe was created. This has made a lot of people very angry and been widely regarded as a bad move. -- Douglas Adams - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel= " in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html