Re: HP/SGI g++ 2.95.2 valarray bug

[llewelly at dbritsch dot dsl dot xmission dot com]

[snip]
> Description of the problem
>   The g++ 2.95.2 valarray class seems to violate part of
>   the specifications for one of its constructors:
> 
>     explicit valarray(size_t);
> 
>   1 The  array created by this constructor has a length equal to the value
>     of the argument.  The elements of the array are constructed using the
>     default constructor for the instantiating type T.
> 
>   ( The text quoted above is from the Dec 1996 working paper;
>     Straustrup::The_C++_Programming_Language( edition = 3, page = 662 )
>     has the same specifications ).
> 
>   The g++ 2.95.2 implementation of this constructor does not
>   seem to initialize the elements with the default value,
>   as demonstrated by the example source code below. The problem
>   shows up second iteration onwards.
>
[snip]

Thank you for your bug report.

This bug is fixed in libstdc++-v3, which will probably be the standard
  c++ library for the next major release of gcc .