This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [patch] testsuite file fails concept-checks
On Mon, May 17, 2004 at 08:08:08PM -0500, Benjamin Kosnik wrote:
>
> >I'll rerun the whole testsuite later on x86_64/linux where there's less
> >noise from other FAILs and make sure I've not missed anything else.
>
> list/vector/deque maybe.
No, they play nicely with concept-checks. It was just the associative
containers that didn't define a StrictWeakOrdering for their keys.
> Do you know how the debug mode and concept checking get along? I'm
> curious. (You'd run the testsuite with -D_GLIBCXX_DEBUG set in
> scripts/testsuite_flags --cxxflags.)
That's the next thing I was going to check :-)
There _might_ be a problem with pch and concept-checks - I have both
enabled in my build, and keep getting compilation FAILs from the file
17_intro/headers_cstdio.cc, that go away if I remake the precompiled
headers. I've not yet had time to investigate properly, and I was
going to look at the debug mode first.
The noise I was referring to includes 46 27_io FAILs on x86/FreeBSD,
that I haven't identified, and half a dozen fails on x86_64/Linux from
22_locale - but I think they're due to missing locale defs on my box.
I also get a couple of XPASS results on x86_64/Linux from 27_io/fpos.
I'll test again tonight, but for now I have a day job to do...
jon
--
"Only two things are infinite: the universe and human stupidity,
and I'm not sure about the former."
- Albert Einstein