[cfe-commits] [libcxx] r135893 - /libcxx/trunk/include/string

Benjamin Kramer benny.kra at googlemail.com
Mon Jul 25 11:21:12 PDT 2011


On Sun, Jul 24, 2011 at 14:45, Howard Hinnant <hhinnant at apple.com> wrote:
> Author: hhinnant
> Date: Sun Jul 24 16:45:06 2011
> New Revision: 135893
>
> URL: http://llvm.org/viewvc/llvm-project?rev=135893&view=rev
> Log:
> Optimization of string::operator< by M.E. O'Neill.  Discussion in http://llvm.org/bugs/show_bug.cgi?id=10461
>
> Modified:
>    libcxx/trunk/include/string
>
> Modified: libcxx/trunk/include/string
> URL: http://llvm.org/viewvc/llvm-project/libcxx/trunk/include/string?rev=135893&r1=135892&r2=135893&view=diff
> ==============================================================================
> --- libcxx/trunk/include/string (original)
> +++ libcxx/trunk/include/string Sun Jul 24 16:45:06 2011
> @@ -3392,7 +3392,17 @@
>  int
>  basic_string<_CharT, _Traits, _Allocator>::compare(const basic_string& __str) const _NOEXCEPT
>  {
> -    return compare(0, npos, __str.data(), __str.size());
> +    size_t __lhs_sz = size();
> +    size_t __rhs_sz = __str.size();
> +    int __result = traits_type::compare(data(), __str.data(),
> +                                        _VSTD::min(__lhs_sz, __rhs_sz));
> +    if (__result != 0)
> +        return __result;
> +    if (__lhs_sz < __rhs_sz)
> +        return -1;
> +    if (__lhs_sz > __rhs_sz)
> +        return 1;
> +    return 0;
>  }
>
>  template <class _CharT, class _Traits, class _Allocator>

Wouldn't it be faster to compare sizes first and do the data
comparison only if sizes are equal? I suppose traits_type::compare is
the expensive part of this function.

- Ben




More information about the cfe-commits mailing list