Re: [PATCH] Handle undefined extern vars in output_in_order

[Jeff Law <law at redhat dot com>]

On 07/14/2016 12:33 PM, Alexander Monakov wrote:
> On Thu, 14 Jul 2016, Jeff Law wrote:
> > Is the point here to be able to deduce what symbols are external & undefined
> > and emit some kind of directive to the assembler in that case?
>
> Yes, PTX assembly requires that properly typed declarations are emitted for
> external references.  Today, the NVPTX backend in GCC performs that internally
> for function symbols, and relies on the middle-end to do that for data symbols,
> but when the port was added that was implemented only when -ftoplevel-reorder is
> in effect (the default).
Right.  I remember noting that in my brief review of the PTX assembly 
specs and figuring it wouldn't be a big deal since we had a functional 
hack in place for the PA/SOM which has similar requirements.

> *nod*, although as I mentioned above today it's handled in a generic way only
> for undefined data symbols, not undefined function symbols.
Understood.  Quickly reviewing the PA code, the other bit of magic is we 
avoided importing a symbol which was declared as extern, but never 
actually used.  IIRC that caused the HP/SOM linker to core dump. 
Hopefully the PTX assembler/linker/finalizer handles that case more 
gracefully...

Jeff