From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1DNX34-0001IV-3c for qemu-devel@nongnu.org; Mon, 18 Apr 2005 10:16:26 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1DNX2u-0001C8-02 for qemu-devel@nongnu.org; Mon, 18 Apr 2005 10:16:17 -0400 Received: from [199.232.76.173] (helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1DNX2t-00019d-1e for qemu-devel@nongnu.org; Mon, 18 Apr 2005 10:16:15 -0400 Received: from [64.233.184.195] (helo=wproxy.gmail.com) by monty-python.gnu.org with esmtp (Exim 4.34) id 1DNX0N-0002cp-NQ for qemu-devel@nongnu.org; Mon, 18 Apr 2005 10:13:39 -0400 Received: by wproxy.gmail.com with SMTP id 71so2331194wra for ; Mon, 18 Apr 2005 07:12:05 -0700 (PDT) Message-ID: <46d6db66050418071228139ffd@mail.gmail.com> Date: Mon, 18 Apr 2005 16:12:05 +0200 From: Christian MICHON Subject: Re: [Qemu-devel] Profiling Qemu for speed? In-Reply-To: <95a144a538f61473783743fa426580a0@axiros.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline References: <20050418083542.45378.qmail@web54110.mail.yahoo.com> <42638306.4050200@cs.man.ac.uk> <95a144a538f61473783743fa426580a0@axiros.com> Reply-To: Christian MICHON , qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org I did months ago gcc/FDO with a xp/lite installation as a "repetitive task"= :) I did not improve the timings after all the effort. Christian On 4/18/05, Daniel Egger wrote: > On 18.04.2005, at 11:51, Ian Rogers wrote: >=20 > > I'm not sure if you can get GCC to generate code sequences like this, > > but you probably at least need to use the -fprofile-generate and > > -fprofile-use options > > http://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html >=20 > Feedback optimisation (FDO) will not work for two reasons: > a) qemu itself is something like a realtime compiler so FDO > will only speed up the compiler but not the generated code > b) FDO will only provide speed boosts if the feedback phase > has a chance to analyse a representative work pattern that > is hopefully also repetitive >=20 > After all FDO is mostly about making a tradeoff size/speed > and rearranging code (mostly branches) to avoid branch > mispredictions of the CPU. >=20 > Servus, > Daniel >=20 >=20 > _______________________________________________ > Qemu-devel mailing list > Qemu-devel@nongnu.org > http://lists.nongnu.org/mailman/listinfo/qemu-devel >=20 >=20 >=20 >=20 --=20 Christian