Re: [Libmesh-users] parallel.h bug report

[John P. <jwp...@gm...>]

On Mon, Oct 13, 2008 at 1:58 PM, Roy Stogner <roy...@ic...> wrote:
>
>
> On Mon, 13 Oct 2008, John Peterson wrote:
>
>> I checked out GCC's implementation of std::complex<T> and I see the
>> class contains two T's rather than e.g. a length=2 array such as: "T
>> array[2];".  So in the (purely academic) case where someone had a
>> vector<complex<char>> I would be a little worried about the struct
>> padding in doing what we are trying to do, but with a float or double
>> I feel pretty confident everything is fine...
>
> I think that there's no padding for either a struct of 2 chars or an
> array of char...  but there's nothing in the standard to prevent
> compilers from adding padding to the former, I don't think.

Okay.  For some reason I thought tiny structs like that might get
padded out to a 4 or 8-byte boundary in memory.

-- 
John