RTEMS available tasks

Gedare Bloom gedare at rtems.org
Mon Jan 27 14:07:45 UTC 2014


On Mon, Jan 27, 2014 at 7:30 AM, Andre Marques
<andre.lousa.marques at gmail.com> wrote:
> On 01/27/14 08:00, Sebastian Huber wrote:
>>
>> Hello Andre,
>>
>> On 2014-01-26 01:31, Andre Marques wrote:
>>>
>>> Hello,
>>>
>>> I'm a CS undergraduate student and i've decided to do something RTEMS
>>> related
>>> for my final project.
>>
>>
>> this is nice.
>>
>>>
>>> I already have a RTEMS build (4.11) working on my Gentoo system and
>>> played a
>>> bit with the EDF and RM schedulers (just launching some tasks and
>>> counting the
>>> deadline misses).
>>
>>
>> Which BSP do you use?  If you want to work with a simulator BSP, then you
>> should have a look at the arm/realview_pbx_a9_qemu BSP since it offers a
>> NULL pointer read and write protection.  Qemu is also a fast simulator which
>> provides support for GDB watch points.  This is quite handy for development
>> and test suite runs.
>>
> I'm using the pc386 BSP using Qemu with the pc386 script provided on
> rtems_testing. Only today I tested GDB with Qemu, which I'm finding a bit
> tricky with the pc386 script because as soon as the program finishes It
> closes Qemu, so I can't connect to the gdb server. To solve that I used the
> instructions on the Wiki about Qemu (which seems to be a little outdated, so
> I needed to change some steps) and I can now load a program and connect to
> the gdb server, but I don't have the symbol table for any program. I can't
> recall enabling debugging during the rtems configure stage, but will that
> generate symbol tables for every compiled program? Can't I just generate a
> symbol table for a specific program?
>
>
If you don't mind could you update the wiki page with your notes? you
might also like to include sebastian's notes about using the realview,
if you go that route..

>>>
>>> As of right now i'm between semesters, and looking to fill this gap with
>>> some
>>> coding fun. In [1] there is a task for making the implementation of
>>> rename()
>>> POSIX conformant. If this task is still open I would like to start
>>> working on it.
>>
>>
>> Yes, this task is still open and it is a good start point.
>>
>> I would start with a new test in
>>
>> http://git.rtems.org/rtems/tree/testsuites/fstests/
>>
>> which covers the POSIX rename() specification
>>
>> http://pubs.opengroup.org/onlinepubs/9699919799/functions/rename.html
>>
> In that case that's where I'm going to start. Any questions that I might
> have should I post them on the users list or somewhere else?
>
>>>
>>> I also went through the Google Code-In open tasks and found this bug
>>> [2][3],
>>
>>
>> Fixing this bug is a very boring task.  You have to install also the tool
>> chain for each RTEMS target and build all BSPs.
>>
>>> which I could also work on, but if there's anything more interesting
>>> (possibly
>>> some small task related with scheduling) please let me know.
>>>
>>> [1] -
>>>
>>> https://wiki.rtems.org/wiki/index.php/Open_Projects#API_Layers_.28POSIX.2C_Classic.2C_SAPI.29
>>>
>>> [2] -
>>> http://www.google-melange.com/gci/task/view/google/gci2013/5227914052239360
>>> [3] - https://www.rtems.org/bugzilla/show_bug.cgi?id=1279
>>>
>>> Thanks for your time,
>>> Andre Marques.
>>> _______________________________________________
>>> rtems-users mailing list
>>> rtems-users at rtems.org
>>> http://www.rtems.org/mailman/listinfo/rtems-users
>>
>>
>>
>
> _______________________________________________
> rtems-users mailing list
> rtems-users at rtems.org
> http://www.rtems.org/mailman/listinfo/rtems-users



More information about the users mailing list