[asterisk-dev] [Code Review] Various updates to the timing API

[Kevin Fleming]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard.digium.com/r/162/#review438
-----------------------------------------------------------

Ship it!


This looks fine, but will require the heap stuff to be added to 1.6.1 without the scheduler changes, as discussed on IRC.


- Kevin


On 2009-02-16 10:00:03, Russell Bryant wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviewboard.digium.com/r/162/
> -----------------------------------------------------------
> 
> (Updated 2009-02-16 10:00:03)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Summary
> -------
> 
> Now that we have multiple timing modules, only being to load one module is causing some usability issues.  So, that inspired this patch.  I am proposing that this patch go in both trunk and 1.6.1.
> 
> 1) Add module use count handling so that timing modules can be unloaded.
> 
> 2) Allow multiple timing modules to be loaded, and use the one with the highest priority value.
> 
> 3) Report which timing module is being use in the "timing test" CLI command, at the request of Tzafrir on the -dev list.
> 
> Note: This patch requires the heap implementation from /r/160/.
> 
> 
> Diffs
> -----
> 
>   /trunk/include/asterisk/timing.h 176034 
>   /trunk/main/Makefile 176034 
>   /trunk/main/channel.c 176034 
>   /trunk/main/timing.c 176034 
>   /trunk/res/res_timing_dahdi.c 176034 
>   /trunk/res/res_timing_pthread.c 176034 
>   /trunk/res/res_timing_timerfd.c 176034 
> 
> Diff: http://reviewboard.digium.com/r/162/diff
> 
> 
> Testing
> -------
> 
> It compiles.  "timing test" still works and reports the module I would expect to be in use.
> 
> 
> Thanks,
> 
> Russell
> 
>