From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58639) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZYbAZ-0002Eu-5o for qemu-devel@nongnu.org; Sun, 06 Sep 2015 10:43:47 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZYbAU-00065g-2f for qemu-devel@nongnu.org; Sun, 06 Sep 2015 10:43:46 -0400 Received: from mx1.redhat.com ([209.132.183.28]:60767) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZYbAT-00065Z-Tj for qemu-devel@nongnu.org; Sun, 06 Sep 2015 10:43:42 -0400 References: <1440752053-16961-1-git-send-email-liang.z.li@intel.com> <20150902054009.GF13778@grmbl.mre> From: Paolo Bonzini Message-ID: <55EC5122.4030802@redhat.com> Date: Sun, 6 Sep 2015 16:43:46 +0200 MIME-Version: 1.0 In-Reply-To: <20150902054009.GF13778@grmbl.mre> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 0/2] Optimization with AVX2 instruction List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Amit Shah , Liang Li Cc: yang.z.zhang@intel.com, qemu-devel@nongnu.org, "Dr. David Alan Gilbert" , quintela@redhat.com On 02/09/2015 07:40, Amit Shah wrote: >> The buffer_find_nonzero_offset() will be called to check the zero page >> > during live migration, it's a hot function. buffer_find_nonzero_offset() >> > has already been optimized with SSE2 instructions, for platform that >> > supports AVX2, we can optimize this function with AVX2 instructions and >> > achieve about 25% performance gain. > This should be a good improvement. I recall Dave and I had a chat > about this in the past too. > > I've not yet reviewed the patchset, but I doubt anyone will have > objections. I'll review this shortly. I think we need a better way to enable it than a configure option, however. AVX2 machines are rare, and no one would end up using it except perhaps Gentoo or other source-based distros. Perhaps something like the GCC ifunc attribute? Paolo