From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265113AbUD3Ia0 (ORCPT ); Fri, 30 Apr 2004 04:30:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265116AbUD3IaZ (ORCPT ); Fri, 30 Apr 2004 04:30:25 -0400 Received: from holomorphy.com ([207.189.100.168]:16000 "EHLO holomorphy.com") by vger.kernel.org with ESMTP id S265113AbUD3IaW (ORCPT ); Fri, 30 Apr 2004 04:30:22 -0400 Date: Fri, 30 Apr 2004 01:30:17 -0700 From: William Lee Irwin III To: Ulrich Drepper Cc: Linux Kernel Subject: Re: NUMA API Message-ID: <20040430083017.GB1298@holomorphy.com> Mail-Followup-To: William Lee Irwin III , Ulrich Drepper , Linux Kernel References: <409201BE.9000909@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <409201BE.9000909@redhat.com> User-Agent: Mutt/1.5.5.1+cvs20040105i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Apr 30, 2004 at 12:35:26AM -0700, Ulrich Drepper wrote: > In the last weeks I have been working on designing a new API for a NUMA > support library. I am aware of the code in libnuma by ak but this code > has many shortcomings: > ~ inadequate topology discovery > ~ fixed cpu set size > ~ no clear separation of memory nodes > ~ no inclusion of SMT/multicore in the cpu hierarchy > ~ awkward (at best) memory allocation interface > ~ etc etc > and last but not least > ~ a completely unacceptable library interface (e.g., global variables as > part of the API, WTF?) Regardless of issues addressed, Andi's been working with everyone for something on the order of 12+ months and this is out of the blue. I very very strongly suggest that you take up each of these issues with him so that they can be addressed as individual incremental improvements to the API everyone's been working with for all that time as opposed to screwing the world (esp. now that commodity NUMA boxen are becoming more prevalent) with transparent and deliberate distro-competition motivated API skew. -- wli