[llvm-commits] [llvm] r77706 - in /llvm/trunk: include/llvm/Support/Annotation.h lib/Support/Annotation.cpp

[Nicolas Geoffray]

Hi Dan,

Dan Gohman wrote:
> What if you put a map member in your ModuleProvider subclass?
>   

Yes, that's what I will end up doing. By doing so, I must change lots of 
things in vmkit, as I was tolerating functions of different modules to 
call each other directly. This prevents using the verifier, but works 
fine otherwise!

But that's just me being lazy :)

> That shouldn't require any locking beyond whatever is already
> present to protect the Module itself. 

Indeed. Nice catch.

>   Does vmkit assign contexts
> to Functions when the Module is created? If so, all of the map's
> allocation would happen at the same time, and materializeFunction
> calls wouldn't see any additional allocations.
>   

Not really, functions are created lazily, so new functions may be 
created after creation of the context.

Thanks for having thought about it!

Nicolas

> Dan
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>