RFC: weakref GCC attribute and .weakref assembly directive

[Aaron W. LaFramboise]

Alexandre Oliva wrote:
> On Oct 13, 2005, Daniel Jacobowitz <drow@false.org> wrote:
 >
>>The difference is that ".weak sym1 = sym2" resolves to sym1 (if
>>available) else sym2; but ".weakref sym1, sym2" resolves to sym2 (if
>>available) else zero.  Also sym1 does not become an external, only a
>>local alias, IIRC.
> 
> Yep.  In `.weak sym1 = sym2', sym1 is a weak alias, which I actually
> contrast with a weakref in the spec text I posted.

Well, that syntax is specific to the PE port as far as I know, and the 
semantics don't quite match weak aliases, which I understand to be the 
equivilent of:

	.set	sym1, sym2
	.weak	sym1

For instance, in PECOFF 'weak externals' (probably a rather misleading 
name),

 > A weak alias cannot reference an undefined symbol, weak or strong.

this statement is not true.  This is because PECOFF's 'weak externals' 
are actually a unique storage class, and so it can do special stuff.

But thank you guys for your help; it answers my question.


Aaron W. LaFramboise