Tested powerpc64le-linux, pushed to trunk.
-- >8 --
This LWG issue was closed as NAD, as it was just a bug in an
implementation, not a defect in the standard. Libstdc++ never had that
bug and always worked for the problem case. Add a test to ensure we
don't regress.
The problem occurs when abs is implemented using a ternary expression:
return d >= d.zero() ? d : -d;
If decltype(-d) is not the same as decltype(d) then this is ambiguous,
because each type can be converted to the other, so there is no common
type.
libstdc++-v3/ChangeLog:
* testsuite/20_util/duration_cast/rounding.cc: Check abs with
non-reduced duration.
---
libstdc++-v3/testsuite/20_util/duration_cast/rounding.cc | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/libstdc++-v3/testsuite/20_util/duration_cast/rounding.cc
b/libstdc++-v3/testsuite/20_util/duration_cast/rounding.cc
index af6e72d9e2e..c5179b6eb6e 100644
--- a/libstdc++-v3/testsuite/20_util/duration_cast/rounding.cc
+++ b/libstdc++-v3/testsuite/20_util/duration_cast/rounding.cc
@@ -58,3 +58,8 @@ static_assert( std::chrono::round<seconds>(2501ms) == 3s );
static_assert( std::chrono::abs(100ms) == 100ms );
static_assert( std::chrono::abs(-100ms) == 100ms );
+
+// LWG 3741. std::chrono::abs(duration) is ill-formed with non-reduced periods
+using D1000 = std::chrono::duration<int, std::ratio<1000, 1000>>;
+static_assert( std::chrono::abs(D1000(-2)) == D1000(2) );
+static_assert( std::is_same_v<decltype(std::chrono::abs(D1000(-2))), D1000> );
--
2.37.2
