Re: rtems-tld tracing with printk

2014-09-09 Thread Sebastian Huber
On 10/09/14 05:53, Chris Johns wrote: No code in RTEMS was altered to do this. Sounds like you can do pretty nice things with this approach. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09

rtems-tld tracing with printk

2014-09-09 Thread Chris Johns
Hi, This is an update on the RTEMS trace linker being worked on and found in the http://git.rtems.org/chrisj/rtl-host.git repo. The following is a trace using the both_hello example in examples-v2 using the trace configuration file attached: $ sparc-rtems4.11-run both_hello.exe >> _Thread

Re: [PATCH] arinc653 support for virtualpok bsp

2014-09-09 Thread Philipp Eppelt
Hi Janek, I finally came around to review this patch. Can you make your own testsuite/sample folder with the ARINC calls and not patch the hello/init.c? Please fix all trailing whitespace errors in the patch. virtualpok/Makefile.am produces an error at compile time: cat libpart.list | rename 's

[PATCH] sptests/sp39: Convert to sptests/spintrcritical21

2014-09-09 Thread Sebastian Huber
Use interrupt critical section test support. Do not print end of test message in case of failure. --- testsuites/sptests/Makefile.am | 3 +- testsuites/sptests/configure.ac| 2 +- testsuites/sptests/sp39/Makefile.am| 22 --- testsuites/

Re: [PATCH] Adding "eth" linker script section for arm bsp.

2014-09-09 Thread Federico Casares
Hi Sebastian, The config file for the "ETH bsp" (linkcmds.lpc1768_mbed_ahb_ram_eth) exhibit changes in the MEMORY definition, some REGION_ALIAS, and adds the ".eth" section. Consequently, we think this file represents an specific configuration for that BSP, which is different from the others, and

Re: [PATCH] Adding "eth" linker script section for arm bsp.

2014-09-09 Thread Federico Casares
Ok, here is the patch with that change Thanks!. On Tue, Sep 9, 2014 at 10:39 AM, Sebastian Huber < sebastian.hu...@embedded-brains.de> wrote: > On 09/09/14 15:29, Federico Casares wrote: > >> The config file for the "ETH bsp" (linkcmds.lpc1768_mbed_ahb_ram_eth) >> exhibit >> changes in the MEMOR

Re: [PATCH] Adding "eth" linker script section for arm bsp.

2014-09-09 Thread Sebastian Huber
On 09/09/14 15:29, Federico Casares wrote: The config file for the "ETH bsp" (linkcmds.lpc1768_mbed_ahb_ram_eth) exhibit changes in the MEMORY definition, some REGION_ALIAS, and adds the ".eth" section. Consequently, we think this file represents an specific configuration for that BSP, which is

[PATCH] tests: Rework interrupt critical tests

2014-09-09 Thread Sebastian Huber
This avoids test durations of more than one hour on fast targets, since fast targets can count a lot during one clock tick period, so the minor loop iteration count was quite high. Estimate now the test body duration to iterate only through the interesting time window. Add and use interrupt_criti

Re: [PATCH] Add -std=gnu++11 to CXXFLAGS if SMP is enabled

2014-09-09 Thread Nick Withers
On Tue, 2014-09-09 at 16:59 +1000, Nick Withers wrote: > On Tue, 2014-09-09 at 08:52 +0200, Sebastian Huber wrote: > > On 09/09/14 04:00, Nick Withers wrote: > > > On Mon, 2014-09-08 at 11:14 +0200, Sebastian Huber wrote: > > >> >This is necessary to use the header file used for the atomic > > >>

Re: [PATCH] Add -std=gnu++11 to CXXFLAGS if SMP is enabled

2014-09-09 Thread Nick Withers
On Tue, 2014-09-09 at 08:52 +0200, Sebastian Huber wrote: > On 09/09/14 04:00, Nick Withers wrote: > > On Mon, 2014-09-08 at 11:14 +0200, Sebastian Huber wrote: > >> >This is necessary to use the header file used for the atomic > >> >operations. > > "-std=c++11" wouldn't cut it? > > > > I don't k