Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262436AbTIOFG2 (ORCPT ); Mon, 15 Sep 2003 01:06:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262438AbTIOFG1 (ORCPT ); Mon, 15 Sep 2003 01:06:27 -0400 Received: from e31.co.us.ibm.com ([32.97.110.129]:39672 "EHLO e31.co.us.ibm.com") by vger.kernel.org with ESMTP id S262436AbTIOFG1 (ORCPT ); Mon, 15 Sep 2003 01:06:27 -0400 Date: Mon, 15 Sep 2003 10:43:11 +0530 From: Ravikiran G Thirumalai To: Dipankar Sarma Cc: Manfred Spraul , akpm@osdl.org, linux-kernel@vger.kernel.org, Robert Love , arjanv@redhat.com Subject: Re: [patch] Make slab allocator work with SLAB_MUST_HWCACHE_ALIGN Message-ID: <20030915051308.GA1159@llm08.in.ibm.com> References: <20030910081654.GA1129@llm08.in.ibm.com> <1063208464.700.35.camel@localhost> <20030911055428.GA1140@llm08.in.ibm.com> <20030911110853.GB3700@llm08.in.ibm.com> <3F60A08A.7040504@colorfullife.com> <20030912085921.GB1128@llm08.in.ibm.com> <3F6378B0.8040606@colorfullife.com> <20030914080942.GA9302@in.ibm.com> <20030914130037.GA1781@in.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030914130037.GA1781@in.ibm.com> User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1157 Lines: 24 On Sun, Sep 14, 2003 at 06:30:37PM +0530, Dipankar Sarma wrote: > > While we are at it, we should also probably look up the cache line > size for a cpu family from a table, store it in some per-cpu data > (cpuinfo_x86 ?) and provide an l1_cache_bytes() api to > return it at run time. If we are going to solve the cache line size mismatch due to compile time arch versus run time arch (compiling on a PIII and running on a P4), we might end up with kernel code which sometimes refer to the static line size and sometimes to the run time size, which might cause correctness issues. So maybe it is good idea to have just one macro for l1 size? We can't do away with the static one, so maybe we should keep it and expect users to compile with the target cpu properly set (otherwise they lose out on performance -- correctness won't be a problem). I could be wrong...just thinking out loud... Thanks, Kiran - 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/