From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34641) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eeV6L-0004vP-L2 for qemu-devel@nongnu.org; Wed, 24 Jan 2018 19:09:10 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eeV6H-00044i-By for qemu-devel@nongnu.org; Wed, 24 Jan 2018 19:09:09 -0500 Received: from mail-pf0-x233.google.com ([2607:f8b0:400e:c00::233]:46543) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1eeV6H-00044P-55 for qemu-devel@nongnu.org; Wed, 24 Jan 2018 19:09:05 -0500 Received: by mail-pf0-x233.google.com with SMTP id y5so4427245pff.13 for ; Wed, 24 Jan 2018 16:09:05 -0800 (PST) References: From: Richard Henderson Message-ID: <9cc3c0d8-26f4-5626-e0f1-e449068eaac1@linaro.org> Date: Wed, 24 Jan 2018 16:09:00 -0800 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v3 00/22] re-factor softfloat and add fp16 functions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Howard Spoelstra , =?UTF-8?Q?Alex_Benn=c3=a9e?= Cc: qemu-devel qemu-devel On 01/24/2018 01:49 PM, Howard Spoelstra wrote: > Hi, > > I built qemu-system-ppc for OSX and Windows from > https://github.com/stsquad/qemu/tree/softfloat-refactor-and-fp16-v3 > and noticed a considerable drop in floating point performance on both > hosts. > Running Mac OS 9.2 in OSX, using MacBench 3.0, the score for the > floating point performance dropped from ~60 to ~42. > > Recent tcg optimisations had improved processor and floating point > performance considerably, but that gain seems to be more than lost for > the floating point performance. > > Any idea what is causing this? Is this a 64-bit or a 32-bit build? We made a conscious choice to share code and use all 64-bit operations, under the assumption that x86_64 is the most common host. Otherwise we have not really done extensive benchmarking. r~