On Tue, Jun 10, 2025 at 3:08 AM Patrick Palka <ppa...@redhat.com> wrote:
> As with the previous patch, this patch reimplements ranges::sort
> directly instead of incorrectly forwarding to std::sort. In addition to
> the compatibility changes listed in the previous patch we also:
>
> - use ranges::iter_swap instead of std::iter_swap
> - use ranges::move_backward instead of std::move_backward
> - use __bit_floor and __to_unsigned_like instead of __lg
>
> We also need to extend std::__lg to support integer-class types.
>
> PR libstdc++/100795
> PR libstdc++/118209
>
> libstdc++-v3/ChangeLog:
>
> * include/bits/ranges_algo.h (__detail::__move_median_to_first):
> New, based on the stl_algo.h implementation.
> (__detail::__unguarded_liner_insert): Likewise.
> (__detail::__insertion_sort): Likewise.
> (__detail::__sort_threshold): Likewise.
> (__detail::__unguarded_insertion_sort): Likewise.
> (__detail::__final_insertion_sort): Likewise.
> (__detail::__unguarded_partition): Likewise.
> (__detail::__unguarded_partition_pivot): Likewise.
> (__detail::__heap_select): Likewise.
> (__detail::__partial_sort): Likewise.
> (__detail::__introsort_loop): Likewise.
> (__sort_fn::operator()): Reimplement in terms of the above.
> * libstdc++-v3/testsuite/25_algorithms/sort/118209.cc: New test.
> * libstdc++-v3/testsuite/25_algorithms/sort/constrained.cc
> (test03): New test.
> ---
> libstdc++-v3/include/bits/ranges_algo.h | 210 +++++++++++++++++-
> .../testsuite/25_algorithms/sort/118209.cc | 23 ++
> .../25_algorithms/sort/constrained.cc | 31 +++
> 3 files changed, 260 insertions(+), 4 deletions(-)
> create mode 100644 libstdc++-v3/testsuite/25_algorithms/sort/118209.cc
>
> diff --git a/libstdc++-v3/include/bits/ranges_algo.h
> b/libstdc++-v3/include/bits/ranges_algo.h
> index f84e0c42d76a..67bdf4287241 100644
> --- a/libstdc++-v3/include/bits/ranges_algo.h
> +++ b/libstdc++-v3/include/bits/ranges_algo.h
> @@ -32,6 +32,7 @@
>
> #if __cplusplus > 201703L
>
> +#include <bit> // __bit_floor
> #if __cplusplus > 202002L
> #include <optional>
> #endif
> @@ -2166,6 +2167,184 @@ namespace ranges
>
> inline constexpr __is_heap_fn is_heap{};
>
> + namespace __detail
> + {
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __move_median_to_first(_Iter __result, _Iter __a, _Iter __b, _Iter
> __c,
> + _Comp __comp, _Proj __proj)
> + {
> + if (std::__invoke(__comp,
> + std::__invoke(__proj, *__a),
> + std::__invoke(__proj, *__b)))
>
See all this __invoke calls, that are the same in each algo, maybe we could
accept only _Comp here,
and then pass following lambda as comp [&](auto& __a, auto& __b).
}
> + {
> + if (std::__invoke(__comp,
> + std::__invoke(__proj, *__b),
> + std::__invoke(__proj, *__c)))
> + ranges::iter_swap(__result, __b);
> + else if (std::__invoke(__comp,
> + std::__invoke(__proj, *__a),
> + std::__invoke(__proj, *__c)))
> + ranges::iter_swap(__result, __c);
> + else
> + ranges::iter_swap(__result, __a);
> + }
> + else if (std::__invoke(__comp,
> + std::__invoke(__proj, *__a),
> + std::__invoke(__proj, *__c)))
> + ranges::iter_swap(__result, __a);
> + else if (std::__invoke(__comp,
> + std::__invoke(__proj, *__b),
> + std::__invoke(__proj, *__c)))
> + ranges::iter_swap(__result, __c);
> + else
> + ranges::iter_swap(__result, __b);
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __unguarded_linear_insert(_Iter __last, _Comp __comp, _Proj __proj)
> + {
> + iter_value_t<_Iter> __val = ranges::iter_move(__last);
> + _Iter __next = __last;
> + --__next;
> + while (std::__invoke(__comp,
> + std::__invoke(__proj, __val),
> + std::__invoke(__proj, *__next)))
> + {
> + *__last = ranges::iter_move(__next);
> + __last = __next;
> + --__next;
> + }
> + *__last = std::move(__val);
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __insertion_sort(_Iter __first, _Iter __last, _Comp __comp, _Proj
> __proj)
> + {
> + if (__first == __last) return;
> +
> + for (_Iter __i = __first + 1; __i != __last; ++__i)
> + {
> + if (std::__invoke(__comp,
> + std::__invoke(__proj, *__i),
> + std::__invoke(__proj, *__first)))
> + {
> + iter_value_t<_Iter> __val = ranges::iter_move(__i);
> + ranges::move_backward(__first, __i, __i + 1);
> + *__first = std::move(__val);
> + }
> + else
> + __detail::__unguarded_linear_insert(__i, __comp, __proj);
> + }
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __unguarded_insertion_sort(_Iter __first, _Iter __last,
> + _Comp __comp, _Proj __proj)
> + {
> + for (_Iter __i = __first; __i != __last; ++__i)
> + __detail::__unguarded_linear_insert(__i, __comp, __proj);
> + }
> +
> + inline constexpr int __sort_threshold = 16;
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __final_insertion_sort(_Iter __first, _Iter __last,
> + _Comp __comp, _Proj __proj)
> + {
> + if (__last - __first > __sort_threshold)
> + {
> + __detail::__insertion_sort(__first, __first + __sort_threshold,
> + __comp, __proj);
> + __detail::__unguarded_insertion_sort(__first +
> __sort_threshold, __last,
> + __comp, __proj);
> + }
> + else
> + __detail::__insertion_sort(__first, __last, __comp, __proj);
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr _Iter
> + __unguarded_partition(_Iter __first, _Iter __last, _Iter __pivot,
> + _Comp __comp, _Proj __proj)
> + {
> + while (true)
> + {
> + while (std::__invoke(__comp,
> + std::__invoke(__proj, *__first),
> + std::__invoke(__proj, *__pivot)))
> + ++__first;
> + --__last;
> + while (std::__invoke(__comp,
> + std::__invoke(__proj, *__pivot),
> + std::__invoke(__proj, *__last)))
> + --__last;
> + if (!(__first < __last))
> + return __first;
> + ranges::iter_swap(__first, __last);
> + ++__first;
> + }
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr _Iter
> + __unguarded_partition_pivot(_Iter __first, _Iter __last,
> + _Comp __comp, _Proj __proj)
> + {
> + _Iter __mid = __first + (__last - __first) / 2;
> + __detail::__move_median_to_first(__first, __first + 1, __mid,
> __last - 1,
> + __comp, __proj);
> + return __detail::__unguarded_partition(__first + 1, __last,
> __first,
> + __comp, __proj);
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __heap_select(_Iter __first, _Iter __middle, _Iter __last,
> + _Comp __comp, _Proj __proj)
> + {
> + ranges::make_heap(__first, __middle, __comp, __proj);
> + for (_Iter __i = __middle; __i < __last; ++__i)
> + if (std::__invoke(__comp,
> + std::__invoke(__proj, *__i),
> + std::__invoke(__proj, *__first)))
> + __detail::__pop_heap(__first, __middle, __i, __comp, __proj);
> + }
> +
> + template<typename _Iter, typename _Compare, typename _Proj>
> + constexpr void
> + __partial_sort(_Iter __first, _Iter __middle, _Iter __last,
> + _Compare __comp, _Proj __proj)
> + {
> + __detail::__heap_select(__first, __middle, __last, __comp, __proj);
> + ranges::sort_heap(__first, __middle, __comp, __proj);
> + }
> +
> + template<typename _Iter, typename _Comp, typename _Proj>
> + constexpr void
> + __introsort_loop(_Iter __first, _Iter __last, unsigned
> __depth_limit,
> + _Comp __comp, _Proj __proj)
> + {
> + while (__last - __first > __sort_threshold)
> + {
> + if (__depth_limit == 0)
> + {
> + __detail::__partial_sort(__first, __last, __last, __comp,
> __proj);
> + return;
> + }
> + --__depth_limit;
> + _Iter __cut = __detail::__unguarded_partition_pivot(__first,
> __last,
> + __comp,
> __proj);
> + __detail::__introsort_loop(__cut, __last, __depth_limit,
> __comp, __proj);
> + __last = __cut;
> + }
> + }
> + } // namespace __detail
> +
> struct __sort_fn
> {
> template<random_access_iterator _Iter, sentinel_for<_Iter> _Sent,
> @@ -2175,10 +2354,33 @@ namespace ranges
> operator()(_Iter __first, _Sent __last,
> _Comp __comp = {}, _Proj __proj = {}) const
> {
> - auto __lasti = ranges::next(__first, __last);
> - _GLIBCXX_STD_A::sort(std::move(__first), __lasti,
> - __detail::__make_comp_proj(__comp, __proj));
> - return __lasti;
> + if constexpr (!same_as<_Iter, _Sent>)
> + return (*this)(__first, ranges::next(__first, __last),
> + std::move(__comp), std::move(__proj));
> + else
> + {
> + if (__first != __last)
> + {
> + auto __n = __detail::__to_unsigned_like(__last - __first);
> + unsigned __log_n;
> + if constexpr (is_class_v<iter_difference_t<_Iter>>)
> + {
> + // FIXME: <bit> doesn't handle integer-class types.
> + __log_n = 0;
> + while (__n != 1)
> + {
> + __n >>= 1;
> + ++__log_n;
> + }
>
I would add overload for std::__bit_width for __max_size_type
in max_size_type.h,
as it should be simple to implement:
if (_M_msb)
return numeric_limits
<https://en.cppreference.com/w/cpp/types/numeric_limits.html><__rep>::digits +
1;
else
return std::__bit_width(_M_val);
> + }
> + else
> + __log_n = std::__bit_floor(__n);
> + __detail::__introsort_loop(__first, __last, __log_n * 2,
> + __comp, __proj);
> + __detail::__final_insertion_sort(__first, __last, __comp,
> __proj);
> + }
> + return __last;
> + }
> }
>
> template<random_access_range _Range,
> diff --git a/libstdc++-v3/testsuite/25_algorithms/sort/118209.cc
> b/libstdc++-v3/testsuite/25_algorithms/sort/118209.cc
> new file mode 100644
> index 000000000000..6d8fa5fd5966
> --- /dev/null
> +++ b/libstdc++-v3/testsuite/25_algorithms/sort/118209.cc
> @@ -0,0 +1,23 @@
> +// PR libstdc++ - ranges::sort doesn't use iter_move, cannot sort zip of
> move-only type
> +// { dg-do compile { target c++23 } }
> +
> +#include <algorithm>
> +#include <ranges>
> +#include <vector>
> +
> +struct F {
> + int a = -1;
> + explicit F(int d) : a(d) { }
> + F(const F&) = delete;
> + F(F&& o) : a(o.a) { }
> + void operator=(const F&) = delete;
> + F& operator=(F&& o) { return *this; }
> + auto operator<=>(const F&) const = default;
> +};
> +
> +int main () {
> + int a[] = {3,2,1};
> + std::vector<F> v(a, a+3);
> + std::ranges::sort(v); // OK
> + std::ranges::sort(std::views::zip(v, v)); // didn't compile
> +}
> diff --git a/libstdc++-v3/testsuite/25_algorithms/sort/constrained.cc
> b/libstdc++-v3/testsuite/25_algorithms/sort/constrained.cc
> index 82754af96430..930dbd77a376 100644
> --- a/libstdc++-v3/testsuite/25_algorithms/sort/constrained.cc
> +++ b/libstdc++-v3/testsuite/25_algorithms/sort/constrained.cc
> @@ -20,6 +20,7 @@
>
> #include <algorithm>
> #include <random>
> +#include <ranges>
> #include <vector>
> #include <testsuite_hooks.h>
> #include <testsuite_iterators.h>
> @@ -72,9 +73,39 @@ test02()
> && ranges::equal(x, y, {}, &X::i, &X::i));
> }
>
> +constexpr bool
> +test03()
> +{
> + // PR libstdc++/100795 - ranges::sort should not use std::sort directly
> +#if __SIZEOF_INT128__
> + auto v = std::views::iota(__int128(0), __int128(20));
> +#else
> + auto v = std::views::iota(0ll, 20ll);
> +#endif
> +
> + int storage[20] = {2,5,4,3,1,6,7,9,10,8,11,14,12,13,15,16,18,0,19,17};
> + auto w = v | std::views::transform([&](auto i) -> int& { return
> storage[i]; });
> + using type = decltype(w);
> + using cat =
> std::iterator_traits<std::ranges::iterator_t<type>>::iterator_category;
> + static_assert( std::same_as<cat, std::output_iterator_tag> );
> + static_assert( std::ranges::random_access_range<type> );
> +
> + ranges::sort(w);
> + VERIFY( ranges::equal(w, v) );
> +
> + ranges::sort(w, std::ranges::greater{});
> + VERIFY( ranges::equal(w, v | std::views::reverse) );
> +
> + ranges::sort(w, std::ranges::greater{}, std::negate{});
> + VERIFY( ranges::equal(w, v) );
> +
> + return true;
> +}
> +
> int
> main()
> {
> test01();
> static_assert(test02());
> + static_assert(test03());
> }
> --
> 2.50.0.rc1.89.g8db3019401
>
>
