From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 30 Nov 2001 13:51:21 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 30 Nov 2001 13:50:26 -0500 Received: from e1.ny.us.ibm.com ([32.97.182.101]:40090 "EHLO e1.ny.us.ibm.com") by vger.kernel.org with ESMTP id ; Fri, 30 Nov 2001 13:49:18 -0500 Date: Sat, 1 Dec 2001 00:24:11 +0530 From: Dipankar Sarma To: linux-kernel@vger.kernel.org Subject: SLAB_HWCACHE_ALIGN question Message-ID: <20011201002411.A12832@in.ibm.com> Reply-To: dipankar@in.ibm.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Greetings. Why does SLAB_HWCACHE_ALIGN forces alignment on L1_CACHE_BYTES and not SMP_CACHE_BYTES ? Does this have anything to do with fitting cache objects in a single L1 cache line even when it is uniprocessor and SMP_CACHE_BYTES is no equal to L1_CACHE_BYTES ? Thanks Dipankar -- Dipankar Sarma http://lse.sourceforge.net Linux Technology Center, IBM Software Lab, Bangalore, India.