Change the mechanism in `iterator_operations.h` to pass around a generic
policy tag indicating whether an internal function is being invoked from
a "classic" STL algorithm or a ranges algorithm. `IterOps` is now
a template class specialized on the policy tag.
The advantage is that this mechanism is more generic and allows defining
arbitrary conditions in a clean manner.
Also add a few more iterator functions to `IterOps`.
Differential Revision: https://reviews.llvm.org/D129390
- create the headers (but not include them from `<algorithm>`);
- define the niebloid and its member functions with the right signatures
(as no-ops);
- make sure all the right headers are included that are required by each
algorithm's signature;
- update `CMakeLists.txt` and the module map;
- create the test files with the appropriate synopses.
The synopsis in `<algorithm>` is deliberately not updated because that
could be taken as a readiness signal. The new headers aren't included
from `<algorithm>` for the same reason.
Differential Revision: https://reviews.llvm.org/D129549
implement `std::ranges::set_intersection` by reusing the classic `std::set_intersenction`
added unit tests
Differential Revision: https://reviews.llvm.org/D129233
Also, improve the test for nasty macros to define min and max, so this
will be caught in the future.
Differential Revision: https://reviews.llvm.org/D128655
This patch also adds a new optimization to `std::move`. It unwraps three `reverse_iterator`s if the wrapped iterator is a `contiguous_iterator` and the iterated type is trivially_movable. This allows us to simplify `ranges::move_backward` to a forward to `std::move` without any pessimization.
Reviewed By: var-const, #libc
Spies: libcxx-commits, mgorny
Differential Revision: https://reviews.llvm.org/D126616
Simplify the implementation of `std::copy` and `std::move` by using `__unwrap_iter` and `__rewrap_iter` to unwrap and rewrap `reverse_iterator<reverse_iterator<Iter>>` instead of specializing `__copy_impl` and `__move_impl`.
Reviewed By: ldionne, #libc
Spies: wenlei, libcxx-commits
Differential Revision: https://reviews.llvm.org/D127049
The flexibility around extern template instantiation declarations in
libc++ result in a very complicated model, especially when support for
slightly different configurations (like the debug mode or assertions
in the dylib) are taken into account. That results in unexpected bugs
like http://llvm.org/PR50534 (and there have been multiple similar
bugs in the past, notably around the debug mode).
This patch gets rid of the _LIBCPP_DISABLE_EXTERN_TEMPLATE knob, which
I don't think is fundamental. Indeed, the motivation for that knob was to
avoid taking a dependency on the library, however that can be done better
by linking against the static library instead. And in fact, some parts of
the headers will always depend on things defined in the library, which
defeats the original goal of _LIBCPP_DISABLE_EXTERN_TEMPLATE.
Differential Revision: https://reviews.llvm.org/D103960
Get rid of the __is_trivially_copy_assignable_unwrapped helper, which
is only used in one place, and use __iter_value_type instead of
iterator_traits<T>::value_type.
Differential Revision: https://reviews.llvm.org/D127230
The debug mode has been broken pretty much ever since it was shipped
because it was possible to enable the debug mode in user code without
actually enabling it in the dylib, leading to ODR violations that
caused various kinds of failures.
This commit makes the debug mode a knob that is configured when
building the library and which can't be changed afterwards. This is
less flexible for users, however it will actually work as intended
and it will allow us, in the future, to add various kinds of checks
that do not assume the same ABI as the normal library. Furthermore,
this will make the debug mode more robust, which means that vendors
might be more tempted to support it properly, which hasn't been the
case with the current debug mode.
This patch shouldn't break any user code, except folks who are building
against a library that doesn't have the debug mode enabled and who try
to enable the debug mode in their code. Such users will get a compile-time
error explaining that this configuration isn't supported anymore.
In the future, we should further increase the granularity of the debug
mode checks so that we can cherry-pick which checks to enable, like we
do for unspecified behavior randomization.
Differential Revision: https://reviews.llvm.org/D122941
In D122982 I accidentally disabled the memmove optimization. This re-enables it and adds more cases where copy forwards to memmove.
Fixes https://github.com/llvm/llvm-project/issues/33687
Reviewed By: var-const, #libc, ldionne
Spies: pkasting, ayzhao, dcheng, xbolva00, libcxx-commits
Differential Revision: https://reviews.llvm.org/D124328
While looking at D113413 I noticed that __log2i could perhaps be
improved to be both slightly smaller and faster.
projects/libcxx/benchmarks/sort.libcxx.out --benchmark_filter=BM_Sort_uint32_QuickSortAdversary*
suggests this is performance neutral, but it shaves a few bytes off the
benchmark binary, and even more off a Chromium build.
Differential revision: https://reviews.llvm.org/D125958
This is to simplify the changes made in D122780. This diff was generated using the command:
```
clang-format include/__algorithm/sort.h -i
```
Reviewed By: #libc, ldionne
Differential Revision: https://reviews.llvm.org/D122858
We are introducing branchless variants for sort3, sort4 and sort5.
These sorting functions have been generated using Reinforcement
Learning and aim to replace __sort3, __sort4 and __sort5 variants
for integral types.
The libc++ benchmarks were run on isolated machines for Skylake, ARM and
AMD architectures and achieve statistically significant improvement in
sorting random integers on test cases from sort1 to sort262144 for
uint32 and uint64.
A full performance overview for Intel Skylake, AMD and Arm can be
found here: https://bit.ly/3AtesYf
Reviewed By: ldionne, #libc, philnik
Spies: daniel.mankowitz, mgrang, Quuxplusone, andreamichi, philnik, libcxx-commits, nilayvaish, kristof.beyls
Differential Revision: https://reviews.llvm.org/D118029
