From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <5339B015.3080008@xenomai.org> Date: Mon, 31 Mar 2014 20:12:37 +0200 From: Gilles Chanteperdrix MIME-Version: 1.0 References: <20140331174555.GO17765@csclub.uwaterloo.ca> In-Reply-To: <20140331174555.GO17765@csclub.uwaterloo.ca> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai] Porting xenomai 2.6.3 and ipipe to 3.12 kernel List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Lennart Sorensen Cc: xenomai@xenomai.org On 03/31/2014 07:45 PM, Lennart Sorensen wrote: > Does anyone have any hints on how to track this down, or even for how > to properly test that moving to 3.12 is actually working. On what platform do you get this? If on cortex A8, and the user-space is compiled in thumb2 mode, note that an erratum workaround is needed, flushing the BTB at each context switch to avoid problem when ARM code is replaced by THUMB code at the same (virtual) address when switching context. My advice would be to wait, Philippe is currently merging the I-pipe git with 3.14 (you can actually find the 3.12 merge in my git tree, under the branch raw/for-ipipe-3.12.0, it has been lightly tested though, as I wait for 3.14 to be ready to test more thoroughly). If you can not wait, then you can enable CONFIG_DEBUG_USER, and pass user_debug=29 on the kernel command line, to have a complete a dump upon fault (and notably in the case of SIGILL). You can also extend this dump by using a hack done in the FCSE patch to print the equivalent of a dump of /proc/self/maps with markers indicating where the PC, SP, and fault address are when a fault occurs. -- Gilles.