[LITMUS^RT] request for review: patches for arbitrary deadlines

[Björn Brandenburg]

On Jul 17, 2012, at 8:20 AM, Christopher Kenna wrote:

> Why don't we just make a get_release(t) macro? We have one for almost
> all the other parameters.
> 
> Also related to these macros, would it be possible to make them more
> consistent? We have a mix of get_X(t) and get_rt_X(t) macros.
> Initially I thought that get_rt were for task-related properties, but
> the get_exec_cost(t) is does not use _rt. Maybe I'm missing something.

Agreed, please send a patch. The inconsistent naming is a leftover from the very first LITMUS^RT. From my point of view, we should either use consistent macros (maybe even prefixed with a 'litmus_' or 'lt_') or no macros at all (tsk->rt_param.XXX is not much worse than get_rt_XXX(tsk)).

I guess the macros have the advantage that we can move stuff around more easily (we might want to move the parameters out of task_struct eventually; it's getting largish).

- Björn