From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <49E45473.4050409@domain.hid> Date: Tue, 14 Apr 2009 11:16:35 +0200 From: Jan Kiszka MIME-Version: 1.0 References: <49C91B24.2020304@domain.hid> <49C95A76.4020802@domain.hid> <49D9E000.4070009@domain.hid> <1239099594.9869.65.camel@domain.hid> <49DB3717.4090402@domain.hid> <49DC6612.40802@domain.hid> <1239696760.5780.12.camel@domain.hid> In-Reply-To: <1239696760.5780.12.camel@domain.hid> Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai-help] Xenomai LTTng support List-Id: Help regarding installation and common use of Xenomai List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Vikesh Rambaran Cc: xenomai-help Vikesh Rambaran wrote: > Hi > Hope everybody had a good Easter weekend :) Yeah, sunny and relaxing, just a little bit hacking. 8) > > On Wed, 2009-04-08 at 10:53 +0200, Jan Kiszka wrote: >> vikesh rambaran wrote: >>> Hi >>> >>> On Tue, Apr 7, 2009 at 1:20 PM, Jan Kiszka wrote: >>> >>>> Hmm, maybe we have issues with MTRR again. Can you also try X without >>>> CONFIG_MTRR set in the kernel? >>> >>> Tried the following combinations on intel and amd platforms >>> >>> 1. No MTRR on 2.6.28.7 + Xenomai 2.4.7 for Intel core 2 >>> 2. No MTRR on 2.6.28.7 + Xenomai 2.4.7 + LTTng for Amd Athlon >>> >>> Lockup still persists (: >> OK, something to rule out. So I'm hoping for serial/netconsole dumps now >> (don't forget to switch off klogd for this, otherwise the information >> may get lost between crashing ipipe/xenomai and klogd. >> > > Unfortunately, no last words on serial console (: > > Stopped klogd (using /etc/init.d/klogd stop) and set dmesg -n 8 prior to > logging into gnome desktop. > > Serial console setup with console=115200n81 with minicom on the monitor > PC. Startup messages and other events (eg USB attach/detach) visible > > For info, > > 1) Lockup seems to occur especially when applications are opened or when > trying to access disks > > 2) a fellow colleague is using Xenomai 2.4.7 with the 2.6.27 kernel > (without LTTng) with similar lockups. > > > If there are any other tests that i can do, please let me know ... NMI watchdog, see Xenomai machine kconfig menu, use nmi_watchdog=2 to finally enable it at the kernel command line. > >>> Going to compare the vanilla installation message log with that of the >>> xenomai patched installation >>> ("Unable to set hardware clock" is something i don't remember seeing in the >>> vanilla logs - will check again) >> The RTC issue is most probably unrelated, should rather have something >> to do with conflicting old/new RTC config switches (sorry, don't recall >> details ATM). >> > Yes, agreed. Built the RTC into the kernel and the warning goes away. > Ah, that was the trick. Jan -- Siemens AG, Corporate Technology, CT SE 2 Corporate Competence Center Embedded Linux