Re: alignof(int[]) (was Re: [C++ Patch] PR 51316)

2011-12-30 Thread Jason Merrill
On 12/30/2011 12:38 PM, Joseph S. Myers wrote: In the absence of a resolution of a DR against C11 that allows alignof in this case, I think we should continue to disallow it for C. Ah, I didn't realize that C11 had alignof as well. OK, then. Jason

Re: alignof(int[]) (was Re: [C++ Patch] PR 51316)

2011-12-30 Thread Joseph S. Myers
On Fri, 30 Dec 2011, Jason Merrill wrote: > On 12/28/2011 07:07 PM, Paolo Carlini wrote: > > I think the resolution of core/930 and C++11 itself are pretty clear: > > alignof of an array of unknown bound is fine, provided the element type > > is complete of course. > > I think it makes sense for

Re: alignof(int[]) (was Re: [C++ Patch] PR 51316)

2011-12-30 Thread Paolo Carlini
Hi, On 12/30/2011 06:27 PM, Jason Merrill wrote: On 12/28/2011 07:07 PM, Paolo Carlini wrote: I think the resolution of core/930 and C++11 itself are pretty clear: alignof of an array of unknown bound is fine, provided the element type is complete of course. I think it makes sense for this chan

alignof(int[]) (was Re: [C++ Patch] PR 51316)

2011-12-30 Thread Jason Merrill
On 12/28/2011 07:07 PM, Paolo Carlini wrote: I think the resolution of core/930 and C++11 itself are pretty clear: alignof of an array of unknown bound is fine, provided the element type is complete of course. I think it makes sense for this change to apply to C as well, on the same principle

[C++ Patch] PR 51316

2011-12-28 Thread Paolo Carlini
Hi, I think the resolution of core/930 and C++11 itself are pretty clear: alignof of an array of unknown bound is fine, provided the element type is complete of course. Tested x86_64-linux. Thanks, Paolo. // /c-family 2011-12-29 Paolo Carlini PR c++/51316