[asterisk-dev] IAX2 performance

[Russell Bryant]

Tim Panton wrote:
> Nearer parity I guess.

Probably, so, which is really what I'd like to see.

> Was that on a single or multiprocessor box ?

Dual-core.

> One of the problems with the old code was that it locked the list
> whilst it it searched it, meaning that multi core was not going to be  
> used to
> best effect.

It didn't actually lock the list.  It locked each call individually.  However,
that still meant that all of the worker threads were stepping on each other
checking each call and destroying the cache while they were at it ...

-- 
Russell Bryant
Senior Software Engineer
Open Source Team Lead
Digium, Inc.