From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753968AbZCHRXa (ORCPT ); Sun, 8 Mar 2009 13:23:30 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753136AbZCHRXV (ORCPT ); Sun, 8 Mar 2009 13:23:21 -0400 Received: from accolon.hansenpartnership.com ([76.243.235.52]:45659 "EHLO accolon.hansenpartnership.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753011AbZCHRXV (ORCPT ); Sun, 8 Mar 2009 13:23:21 -0400 Subject: Re: [PATCH 01/13] [VOYAGER] x86: add {safe,hard}_smp_processor_id to smp_ops From: James Bottomley To: Jeremy Fitzhardinge Cc: LKML , Thomas Gleixner , "H. Peter Anvin" , Ingo Molnar In-Reply-To: <49B3FD1F.9010000@goop.org> References: <1236530906-7175-1-git-send-email-James.Bottomley@HansenPartnership.com> <1236530906-7175-2-git-send-email-James.Bottomley@HansenPartnership.com> <49B3FD1F.9010000@goop.org> Content-Type: text/plain Date: Sun, 08 Mar 2009 12:23:16 -0500 Message-Id: <1236532996.4861.6.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 (2.22.3.1-1.fc9) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2009-03-08 at 10:15 -0700, Jeremy Fitzhardinge wrote: > James Bottomley wrote: > > Not having apics, Voyager can't use the default apic implementation of > > these, it has to read from a special port in the VIC to get the > > processor ID, so abstract these functions in smp_ops to allow voyager > > to live simultaneously with the apic code. > > > > These aren't performance-sensitive at all, are they? smp_ops is not > subject to patching/inlining optimisations happen to more hotpath pvops. No more than the function pointer indirection of smp_call_function(). Fortunately, the hard version is called in very few places, so the overhead is minimal. > Is safe_smp_processor_id needed at all? It's only got two callers, and > x86-64 just implements it as smp_processor_id(). I can't see a reason, no. If you look at voyager it uses the same function for both, but x86 seems to have extra gunk in the safe path, so I didn't feel entitled to remove it. James