[openal] Announcing MojoAL

Chris Robinson chris.kcat at gmail.com
Sat Mar 24 07:14:07 EDT 2018


On 03/23/2018 10:15 PM, Ryan C. Gordon wrote:
> I've started working on a new OpenAL implementation, so I thought I'd 
> mention it here.

Cool. Nice to see other implementations spring up. :)

> - I was aiming for the implementation to be 100% lock-free, so there's 
> no chance that you might call alSourcef() at the wrong moment and have 
> it hang for a long time while the mixer thread runs (or alternately: you 
> call alSourcef at the wrong moment and the mixer thread doesn't complete 
> in time). This might be foolish, I haven't decided yet.

If it's foolish, you've got company. I put in a lot of effort to get 
OpenAL Soft mostly lock-free on the mixer, to avoid similar issues with 
either the audio mix or the function call getting delayed from 
contention (the original code was not very async-friendly). There's 
still a few places it can happen, but I have managed to get a 
significant portion to not block with the mixer.

> The expectation is that this will be useful in the near future for 
> embedded and game console developers that need a small, drop-in, 
> license-friendly implementation and don't need any advanced features or 
> extensions.
> 
> I'm currently calling it "MojoAL" and the source code is here:
> 
> https://hg.icculus.org/icculus/mojoAL

I may have to give this a try when I get some time, and see how it works 
with various apps.


More information about the openal mailing list