Re: [PATCH][libstdc++-v3 parallel mode] Enable user-defined swap

[François Dumont <francois dot cppdevs at free dot fr>]

Hi

    Sorry if this is trivial but I don't really understand the rational 
of such modifications.

    It is supposed to have libstdc++ use user defined overloads of swap 
when available. But if so shouldn't it be overloaded in the std 
namespace ? I remember having read something like that but I am not sure 
it was in the C++ Standard or it was about the swap algo.

    If we open libstdc++ to use swap defined in any namespace shouln't 
we do the same for all other algos used internally. For instance this 
patch use std::iter_swap, shouldn't it use:

using std::iter_swap;
iter_swap(...);

Regards

On 02/25/2011 03:09 PM, Johannes Singler wrote:
> On 01/24/2011 07:12 PM, Benjamin Kosnik wrote:
> > > Also commit to gcc-4_5-branch?
> >
> > Ok.
> >
> > -benjamin
>
> The following patch enables user-defined swap for parallel sorting and 
> the like, which can improve performance.
>
> Tested x86_64-unknown-linux-gnu: No regressions
>
> Committed.
>
> 2011-02-25  Johannes Singler <singler@kit.edu>
>
>         PR libstdc++/47433
>         * include/parallel/losertree.h
>         (_LoserTree<>::__delete_min_insert):
>         Do not qualify swap with std:: for value type,
>         but include a using directive instead.
>         (_LoserTreeUnguarded<>::__delete_min_insert): Likewise.
>         * include/parallel/balanced_quicksort.h (__qsb_divide):
>         Use std::iter_swap instead of std::swap.
>         (__qsb_local_sort_with_helping): Likewise.
>         * include/parallel/partition.h (__parallel_partition):
>         Likewise. (__parallel_nth_element): Likewise.
>
> Johannes