From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32385 invoked by alias); 2 Sep 2002 13:12:57 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 32359 invoked from network); 2 Sep 2002 13:12:54 -0000 Received: from unknown (HELO dns.debis.hu) (195.228.20.2) by sources.redhat.com with SMTP; 2 Sep 2002 13:12:54 -0000 Received: from exchange.uns.t-systems.tss ([172.16.1.5]) by dns.debis.hu (8.12.1/8.12.1/Debian -5) with ESMTP id g82EAwdK025413; Mon, 2 Sep 2002 16:11:10 +0200 X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Subject: [GCC 3.x] Performance testing for QA Date: Mon, 02 Sep 2002 06:12:00 -0000 Message-ID: <71EE24368CCFB940A79BD7002F14D76014BC21@exchange.uns.t-systems.tss> X-MS-Has-Attach: X-MS-TNEF-Correlator: From: =?iso-8859-1?Q?Sasi_P=E9ter?= To: Cc: X-SW-Source: 2002-09/txt/msg00026.txt.bz2 Dear GCC people, Allow me the humble question to ask you if it would possible to employ a si= mple yet efficient testing method from real life as opposed to the SPEC* (a= rtificial) test? Specifically I was thinking of running a few encoding/decoding/manipulation= runs with mplayer on sample multimedia streams. Main reasons I could think of are: - Easy to test: all you need (sources, sample files) is available free at w= ww.mplayerhq.hu (for Linux(intel, alpha), Solaris, Irix, etc.) - Important for masses: it is one of the most widely used free software in = the area gaining probably the most momentum nowadays (amongst performace hu= ngry software) - There is a lot of space for improvements: even the 2.95 series perform be= tter than the ongoing 3.x series, but if going for the long term Intel's co= mpiler currently wins hands down I know it is not a very scientific measurement, still it does apply to a lo= t of people, andd is a reason why a lot of people still stick to the older = versions. New features are really an important thing to have, still they should not m= ake it instead of the good performance, but exploiting it. Summarizing the above: - Do you think such measurements are feasible/worth and can supply usable i= nput for the development? - Do you lack the interest, the resources, or what else? - Ifyou are not interested, is there a reasoning behind that? (point me to = RTFM if applicable) - In case of positive answers do you think profiling, code analysis or what= else could be the way to improve the current situation? (get 3.x to the pe= rf. level of 2.95.x) Please CC my personal address in your reply, since I am not subscribed to a= ny of your mailing lists. Thank you for your precious time. Best regards, Peter -- Sasi P=E9ter E-mail: peter.sasi@t-systems.co.hu=20 Sr. Systems Engineer Internet: =20 T-Systems Unisoftware Ltd. Phone: +36 (1) 456 54 13=20 Hungary Fax: +36 (1) 456 54 99=20 Postal address: H-1453 Budapest 92, Pf.28. Mobile: +36 (30) 280 34 31=20 Address: H-1097 Budapest, K=F6nyves K=E1lm=E1n krt. 12-14. 3.em., Lurdy-H= =E1z=20