On Wed, Jun 5, 2013 at 4:15 PM, Eli Friedman <[email protected]> wrote:

> How is the change to TypeOfExprType::desugar related to the other changes?
>

I'll delete it.  I don't need it anymore.

At one point I was breaking this code in test/Sema/function-redecl.c:
void *h0(unsigned a0,     ...);
extern __typeof (h0) h1 __attribute__((__sentinel__));
extern __typeof (h1) h1 __attribute__((__sentinel__));

Code in SemaExpr.cpp makes the __typeof(h1) expression become 'void
(*)(...)'.

I changed something about the way I retreive the function type before
adjusting parameters.  I don't remember what it was before, but now I'm
using:
  const FunctionType *FT = R->castAs<FunctionType>();

I added a test for the case where I need to rebuild the type because one of
the parameters needs to be adjusted and everything seems to work fine.

+          // Get the type as written.  It will be adusted later in
>
> *adjusted
>
> Otherwise, it looks good.
>

Thanks!  Committing.


> -Eli
>
>
> On Mon, Jun 3, 2013 at 6:36 AM, Reid Kleckner <[email protected]> wrote:
>
>>
>>   Ping
>>
>> http://llvm-reviews.chandlerc.com/D883
>> _______________________________________________
>> cfe-commits mailing list
>> [email protected]
>> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>>
>
>
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to