[Asterisk-Dev] Re: ztdummy? is it necessary?

[Jason DiCioccio]

Eric "ManxPower" Wieling wrote:

> Jason DiCioccio wrote:
>
>> Based on the initial response, I should probably clarify what I'm 
>> asking. I know that some applications, as asterisk is developed now, 
>> require a
>> zaptel timing source.  However, is this requirement necessary?  Would
>> certain platforms, if asterisk was written to accept it, be able to 
>> handle
>> everything fine without the zaptel timing requirement?  My understand of
>> the issue is that older versions of Linux had an inaccurate real-time
>> clock.  I also understand that this has since been fixed?  And that BSD
>> does not have the issue?  So does this extra driver really need to be
>> required on all platforms?  Or just the ones with the broken RTC?
>>
>> Basically, I'm not asking if asterisk as it is today requires 
>> ztdummy. I'm asking if the requirement is necessary.
>
>
> Asterisk requires SOME kind of timing source. It seems silly to have 
> to support each OS's API for RTC in the Asterisk core.  It makes much 
> more sense to me to use a zaptel compatable driver, which has a well 
> known interface, for timing.

It sounds like it's basically a choice between supporting the OS's 
driver APIs (which, in my experience, break more often between OS 
releases) and CPU architecture quirks, versus supporting the RTC APIs 
for each OS (which, from what I'm gathering, isn't standardized).

What makes choice B better than choice A?