[fpc-other] Multi-threaded programming (was: volatile variables)

[Vinzent Höfler]

On Thu, 30 Jun 2011 11:38:50 +0200, Jonas Maebe
<jonas.maebe at elis.ugent.be> wrote:

> Data races can only be detected without false positives/negatives via  
> dynamic analysis, by intercepting all memory accesses and all  
> synchronization operations.

"That's not entirely ... accurate."(tm)

If you restrict your tasking subset you can at least ensure that no races
or deadlocks can occur. See RavenSPARK.

But that's surely no option for FPC.

> If a program would contain nothing but FPC-compiled code it would in  
> theory be possible to modify the RTL and the compiler to insert all  
> required analysis code, but I would not consider that to be worth the  
> effort.

I agree. It's definitely not worth the trouble considering how many ways
of failing the test there still are.


Vinzent.