On Wed, 22 Jun 2022 at 07:35, Alexandre Oliva via Libstdc++ <libstd...@gcc.gnu.org> wrote: > > > The last_write_time functions are defined in ways that are useful, or > that fail immediately, depending on various macros. When they fail > immediately, the filesystem last_write_time.cc tests fail noisily, but > the fail is entirely expected. > > Define HAVE_LWT in the last_write_time.cc tests, according to the > macros that select implementations of last_write_time, and use it to > skip tests that are expected to fail. > > Regstrapped on x86_64-linux-gnu, also tested with a cross to > aarch64-rtems6. Ok to install?
OK > PS: I realize _GLIBCXX_HAVE_SYS_STAT_H is tested for in two different > ways in the #if expressions added to the tests. This mirrors the > different uses in the do_stat template body, and in > fs::last_write_time(const path&, file_time_type, error_code&). Perhaps > they should all be using either value or definedness, but I didn't want > to go there, at least not at first, so I retained the apparent > inconsistency. Yes, they should be tested consistently in the libstdc++ sources. And again, this could be a macro defined in testsuite_fs.h
