Re: Fw: [cxx-abi-dev] Decimal Floating Point mangling was(Fw: [cxx-abi-dev] C++0x: Mangling of rvalue reference type)s
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Fw: [cxx-abi-dev] Decimal Floating Point mangling was(Fw: [cxx-abi-dev] C++0x: Mangling of rvalue reference type)s



>From: Michael Wong <michaelw@xxxxxxxxxx>
>>>     Decimal128:         "r"
>> 
>> Will that "r" conflict with C99 restrict?

>I think we considered this and C99 restrict cannot exist on its own. It 
>must be a restrict on a type hence it is a qualifier. So the conflict 
>should not exist, in my opinion. I hope we are not wrong. 
>Its a little more work, but not much more.

Do you have the parsing algorithm?
That way everyone can be consistent.  And if any errors in your logic,
everyone can help find them.

It seems like the rule is:
Restrict should only be mangled with pointers and references.
Arrays in function parms degrade to pointers.

And top level qualifiers in template parms are removed??

This would leave:
    U r V K  followed by P or R

So if "r" is followed by V K P or R, it is restrict, else Decimal128?

And we don't have to worry about foo(restrict int) or r then i.