From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <548B481E.3050103@nta-inc.net> Date: Fri, 12 Dec 2014 13:55:10 -0600 From: Jeff Webb MIME-Version: 1.0 References: <5489D81D.7070003@nta-inc.net> <548AAB8A.7060406@xenomai.org> In-Reply-To: <548AAB8A.7060406@xenomai.org> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai] switchtest in Xenomai-3.0-rc2 List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Philippe Gerum , Xenomai On 12/12/2014 02:47 AM, Philippe Gerum wrote: > On 12/11/2014 06:45 PM, Jeff Webb wrote: >> In Xenomai-3.0-rc2, doc/asciidoc/TROUBLESHOOTING.COBALT.adoc still says >> this: >> >> >> === Issues when running the _switchtest_ program >> >> ==== pthread_create: Resource temporarily unavailable >> >> The switchtest test creates many kernel threads, this means that the >> options +CONFIG_XENO_OPT_SYS_HEAPSZ+ and >> +CONFIG_XENO_OPT_SYS_STACKPOOLSZ+, in your kernel configuration, >> should be configured to large enough values. Try increasing them, >> rebuilding the kernel. >> >> >> but, it looks like CONFIG_XENO_OPT_SYS_STACKPOOLSZ has been removed in >> Xenomai-3. Is there another an equivalent variable that I might need to >> increase, or is increasing CONFIG_XENO_OPT_SYS_HEAPSZ sufficient to get >> rid of the error below? >> > > It is sufficient to increase SYS_HEAPSZ, there is no private stack pool > anymore, since Xenomai kthreads are now fully mapped over regular > kthreads. i.e. we don't have to allocate stacks for our kernel threads, > we simply inherit the kernel stack underlying the new regular kthread. > > The doc was fixed accordingly. Thanks for the heads up. > That worked. Thanks! -Jeff