[asterisk-dev] conflicting timer modules

Tzafrir Cohen tzafrir.cohen at xorcom.com
Sun Feb 15 08:24:19 CST 2009


On Sun, Feb 15, 2009 at 12:18:13PM +0200, Tzafrir Cohen wrote:
> Hi
> 
> In trunk (1.6.2) I see that the separate res_timing_* module conflict
> with each other. In 1.6.1 I don't see it.
> 
> Why is this conflict? Is it merely a run-time issue or also a build-time
> issue? Is it enforced on build-time or on run-time?

Just built 1.6.1 with live_ast on a system that has both (as non-root
user, and hence without the ability to ue dahdi devices). The module
dahdi is actually loaded, and hence the device files exist.

Both modules seems to have been loaded. 

  sweetmorn*CLI> timing test
  Attempting to test a timer with 50 ticks per second ...
  It has been 1000 milliseconds, and we got 50 timer ticks

I can't tell from the message (or generally from the CLI interface) what
timing source is used. But when I explicitly disable loading
res_timing_pthread I get:

  sweetmorn*CLI> timing test
  Attempting to test a timer with 50 ticks per second ...
  Failed to open timing fd
  Command 'timing test ' failed.

Looks good to me. 

BTW: it would probably help to add a "name" function to the timing
interface so I could tell which timing source is in use.

-- 
               Tzafrir Cohen
icq#16849755              jabber:tzafrir.cohen at xorcom.com
+972-50-7952406           mailto:tzafrir.cohen at xorcom.com
http://www.xorcom.com  iax:guest at local.xorcom.com/tzafrir



More information about the asterisk-dev mailing list