Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752487AbdCEAeF (ORCPT ); Sat, 4 Mar 2017 19:34:05 -0500 Received: from mx2.suse.de ([195.135.220.15]:33948 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751854AbdCEAeE (ORCPT ); Sat, 4 Mar 2017 19:34:04 -0500 Date: Sun, 5 Mar 2017 01:33:49 +0100 From: Borislav Petkov To: hpa@zytor.com Cc: Logan Gunthorpe , Thomas Gleixner , Ingo Molnar , Tony Luck , Al Viro , the arch/x86 maintainers , Linux Kernel Mailing List Subject: Re: Question Regarding ERMS memcpy Message-ID: <20170305003349.6kgq4ovj7ipezfxu@pd.tnic> References: <20170304224341.zfp4fl37ypt57amg@pd.tnic> <5CCEF10D-5647-4503-A398-0681DF2C8847@zytor.com> <20170305001447.kcxignj3nsq35vci@pd.tnic> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: NeoMutt/20170113 (1.7.2) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 814 Lines: 22 On Sat, Mar 04, 2017 at 04:23:17PM -0800, hpa@zytor.com wrote: > What are the compilation flags? It may be that gcc still does TRT > depending on this call site. I'd check what gcc6 or 7 generates, > though. Well, I don't think that matters: if you're building a kernel on one machine to boot on another machine, the compiler can't know at build time what the best MOVS* variant would be for the target machine. That's why we're doing the alternatives patching at *boot* time. However, if the size is small enough, the CALL/patch overhead would be definitely too much. Hmm, I wish we were able to say, "let gcc decide for small sizes and let us do the patching for larger ones." -- Regards/Gruss, Boris. SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --