Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750716AbVIJJUZ (ORCPT ); Sat, 10 Sep 2005 05:20:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750718AbVIJJUZ (ORCPT ); Sat, 10 Sep 2005 05:20:25 -0400 Received: from cantor2.suse.de ([195.135.220.15]:41157 "EHLO mx2.suse.de") by vger.kernel.org with ESMTP id S1750716AbVIJJUZ (ORCPT ); Sat, 10 Sep 2005 05:20:25 -0400 From: Andi Kleen To: Andrew Morton , torvalds@osdl.org Subject: NUMA mempolicy /proc code in mainline shouldn't have been merged Date: Sat, 10 Sep 2005 11:20:18 +0200 User-Agent: KMail/1.8 Cc: linux-kernel@vger.kernel.org MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200509101120.19236.ak@suse.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1023 Lines: 30 Hi, Just noticed the ugly SGI /proc/*/numa_maps code got merged. I argued several times against it and I very deliberately didn't include a similar facility when I wrote the NUMA policy code because it's a bad idea. - it's a lot of ugly code. - it's basically only a debugging hack right now - it presents lots of kernel internal information and mempolicy internals (like how many people have a page mapped) etc. to userland that shouldn't be exposed to this. - the format is very complicated and the chance of bug free userland parsers of this is near zero. - there is no demonstrated application that needs it (there was a theoretical usecase where it might be needed, but there were better solutions proposed for this) Can the patch please be removed? Thanks, -Andi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/