From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ricardo Wurmus Subject: Re: [PATCH 2/2] gnu: zita-convolver: Drop the '-march=native' optimization. Date: Thu, 07 Jan 2016 09:37:31 +0100 Message-ID: <87ziwh4wfo.fsf@elephly.net> References: <1452095330-14147-1-git-send-email-iyzsong@gmail.com> <874meq5s30.fsf@elephly.net> <87y4c2qbv9.fsf@member.fsf.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:55329) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aH64k-0005y8-EJ for guix-devel@gnu.org; Thu, 07 Jan 2016 03:37:43 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aH64h-0001as-EQ for guix-devel@gnu.org; Thu, 07 Jan 2016 03:37:42 -0500 Received: from sender163-mail.zoho.com ([74.201.84.163]:25376) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aH64h-0001Zp-6T for guix-devel@gnu.org; Thu, 07 Jan 2016 03:37:39 -0500 In-reply-to: <87y4c2qbv9.fsf@member.fsf.org> List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org To: =?utf-8?B?5a6L5paH5q2m?= Cc: guix-devel@gnu.org 宋文武 writes: > Ricardo Wurmus writes: > >>> + (snippet >>> + ;; Don't optimize for a specific processor architecture. >>> + '(substitute* "libs/Makefile" >>> + (("^CXXFLAGS \\+= -march=native") ""))) >>> + (modules '((guix build utils))) >> >> Is this to avoid that packages are optimised for the CPU of the build >> slave? > Yes, using that from hydra will crash guitarix for me. > Same as: . Guitarix also crashes for me on i686 — and I thought it was something to do with my machine or the broken build of eigen. >> If so, could we instead pass different optimisation flags for >> different architectures? For a convolver I’d prefer to have *some* >> optimisation, even if it’s just the greatest common divisor. > I don't familiar with optimization flags, but according to: > > I think '-march=native' should be dropped definitely, it enable all > the CPU specified instruction set when building on the slave. > And it's not clear to me what *some* flags are, suppose we don't want > any CPU specified (-msse3, -msse4, etc.) thing. Okay. I think removing “-march=native” is worth doing anyway, so let’s ignore adding optimisation flags for now. Thank you!