From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 2 Apr 2004 17:45:59 +0200 From: Andrea Arcangeli Subject: Re: [RFC][PATCH 1/3] radix priority search tree - objrmap complexity fix Message-ID: <20040402154559.GE21341@dualathlon.random> References: <20040402011627.GK18585@dualathlon.random> <20040401173649.22f734cd.akpm@osdl.org> <20040402020022.GN18585@dualathlon.random> <20040401180802.219ece99.akpm@osdl.org> <20040402022233.GQ18585@dualathlon.random> <20040402070525.A31581@infradead.org> <20040402152240.GA21341@dualathlon.random> <20040402162709.A4312@infradead.org> <20040402153801.GD21341@dualathlon.random> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040402153801.GD21341@dualathlon.random> Sender: owner-linux-mm@kvack.org Return-Path: To: Christoph Hellwig , Andrew Morton , hugh@veritas.com, vrajesh@umich.edu, linux-kernel@vger.kernel.org, linux-mm@kvack.org List-ID: On Fri, Apr 02, 2004 at 05:38:01PM +0200, Andrea Arcangeli wrote: > 100 kernel developers, who cares about saving some cycles in 100 > machines? Get real. just to avoid any misunderstanding, I want to optimize it _everywhere_, I mean that optimizing it in only 100 machines and an embedded niche is worthless. I'm not saying it's worthless to optimize it everywhere (though I doubt it's a measurable slowdown given the order > 0 is unlikely in the first place). if you check my first emails about the compound thing I wasn't very happy about it. The only single reason I had to keep it on by default is that currently I feel unsafe about optimizing it away turning it off by default, since the big testing (on weird drivers too) has happened so far with compound on by default, and disabling it everywhere would risk to trigger bugs, and this clearly shows you how unreliable it is to return different things from alloc_pages in function of an unrelated hugetlbfs option, and this is a basic problem I'm fixing. -- 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: aart@kvack.org