From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from fmmailgate02.web.de ([217.72.192.227]:54954 "EHLO fmmailgate02.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751668AbZG3JtP (ORCPT ); Thu, 30 Jul 2009 05:49:15 -0400 Received: from smtp06.web.de (fmsmtp06.dlan.cinetic.de [172.20.5.172]) by fmmailgate02.web.de (Postfix) with ESMTP id 921F110EE94E4 for ; Thu, 30 Jul 2009 11:49:14 +0200 (CEST) Received: from [217.228.251.207] (helo=[172.16.99.2]) by smtp06.web.de with asmtp (TLSv1:AES256-SHA:256) (WEB.DE 4.110 #277) id 1MWSG9-0006gm-00 for linux-media@vger.kernel.org; Thu, 30 Jul 2009 11:49:13 +0200 Message-ID: <4A716C96.1080404@magic.ms> Date: Thu, 30 Jul 2009 11:49:10 +0200 From: emagick@magic.ms MIME-Version: 1.0 To: linux-media@vger.kernel.org Subject: Re: Cinergy T2 stopped working with kernel 2.6.30 References: <4A61FD76.8010409@magic.ms> In-Reply-To: <4A61FD76.8010409@magic.ms> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-media-owner@vger.kernel.org List-ID: The more I look into this problem the stranger it becomes. I've compiled the kernel for different CPUs: | mplayer mythtv -----------------+-------------------------------------------- CONFIG_M486 | works works CONFIG_M586 | works cannot tune CONFIG_MCORE2 | cannot tune cannot tune These results are for my Atom N270 board. On my Core2 board, the Cinergy T2 works all the time. By applying -march=i486 and -march=i586 to individual source files, I found out that dvb_frontend.c is the culprit. By editing the assembly output for dvb_frontend.c, I found out that only dvb_frontend_swzigzag_autotune() needs to be compiled with CONFIG_M486 to make the Cinergy T2 work, everything else can be compiled with CONFIG_M586. Both compiled versions of dvb_frontend_swzigzag_autotune() look OK (but I haven't yet strictly verified the assembly code). Anyway, nothing in that code should make a difference on N270 vs. Core, except for timing. Adding NOPs doesn't seem to make a difference. Any ideas? Could this be a CPU bug?