[cxx-abi-dev] Proposed ABI changes for new C++0x SFINAE rules

[David Vandevoorde]

On Sep 16, 2010, at 1:10 PM, Jason Merrill wrote:

> On 09/16/2010 09:55 AM, David Vandevoorde wrote:
>>>  I guess that means the compiler needs to track both the full expression and its constant value, and so on through array types and template-ids that use them.
>> 
>> I'm not sure I understand this point.  Mostly (for this purpose), the compiler has to track the "instantiation-dependent" property for the expression.
> 
> Say you have
> 
> template <int I>
> struct A;
> 
> template <> struct A<sizeof(size_t)>
> {
>  typedef size_t type;
> };
> 
> template <class T>
> void f (A<sizeof(sizeof(T))>::type);
> 
> --------------
> 
> The compiler needs to recognize that A<sizeof(sizeof(T))> is not a dependent type, so we need to treat it as A<8> and not require typename to name the nested type.  But for substitution we need to treat it as A<sizeof(sizeof(T))>.

Ah yes, that's correct.

	Daveed