[rtems commit] posix: Set pthread attribute values in detach and setschedparm methods.

[Joel Sherrill]

On 2/20/2014 2:45 AM, Sebastian Huber wrote:
> On 2014-02-19 17:33, Jennifer Averett wrote:
>> Module:    rtems
>> Branch:    master
>> Commit:    52eecbaa089523a4afaef1024225c7bf5181495e
>> Changeset:http://git.rtems.org/rtems/commit/?id=52eecbaa089523a4afaef1024225c7bf5181495e
>>
>> Author:    Jennifer Averett<jennifer.averett at oarcorp.com>
>> Date:      Wed Feb 19 10:36:01 2014 -0600
>>
>> posix: Set pthread attribute values in detach and setschedparm methods.
> Why do we need the values in two different places in the same structure?  I see 
> no test suite changes, do we miss a test case?
>
>
We don't need them in two places. This is historical and needs to be
eliminated. But
when we added pthread_get_attr_np(), we noticed that they were
inconsistent. We
never had an external API to test this with.

Let this go in so the new API call and associated tests can go in. Then
we can eliminate
the duplication that has been there for at least a decade.

FYI Jennifer has a large series of patches to add the Classic and POSIX
affinity APIs and
support for storing the requested affinity in the Thread_Control,
validating it, and
keeping it consistent across the Classic and POSIX APIs. This includes
tests and
documentation. If there is some clean up that can be postponed until all
that is
merged, it would be appreciated.  This all should be additive and not
impact the
schedulers at all. It is just APIs and the underlying data store.

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill at OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985