[PATCH 10/12] smp: Optimize Simple SMP scheduler

Gedare Bloom gedare at rtems.org
Wed Aug 14 19:36:07 UTC 2013


On Wed, Aug 14, 2013 at 2:52 PM, Chris Johns <chrisj at rtems.org> wrote:
> Gedare Bloom wrote:
>>
>> On Tue, Aug 13, 2013 at 9:42 AM, Sebastian Huber
>> <sebastian.huber at embedded-brains.de>  wrote:
>>>
>>>
>>>     /**
>>> +   * @brief This field is true if the thread is in the air.
>>> +   *
>>> +   * A thread is in the air if it is in a transient state.  The extract
>>> +   * operation on a scheduled thread will produce threads in the air.
>>> Such
>>> +   * threads are no longer on the scheduled chain, but are still
>>> allocated to a
>>> +   * processor.  The next enqueue or schedule operation will decide what
>>> to do.
>>> +   */
>>> +  bool                                  is_in_the_air;
>>> +
>>
>> At first I did not like this term "in the air" but I cannot think of a
>> better term right now.
>>
>
> What about 'is_bound_and_transient' ?
>
the term we use now is "allocated" for a thread that has been assigned
to a specific core for execution. So it would be
"is_allocated_and_transient" to be precise, or change the "allocated"
to "bound" to mean a thread on a core...

> Chris



More information about the devel mailing list