sem issues part 500

[Paolo Mantegazza]

Steven Seeger wrote:
>>That's OK, but you should read my spcs in the documentation of sem_init.
> 
> 
> Missed that. :)
> 
> I have a solution that is actually extremely efficient, and requires no
> dynamic memory allocation. The only problem is that it requires a fixed
> array, and thus you'll have to have a define in the system telling it your
> max number of realtime threads. I am going to default it to 32.
> 
> I spent 4 and 1/2 days trying to find the problem. The first 3 I assumed it
> was me somehow.
> 

It might be a good thing, so let's me have a look at it.

Paolo.