Bug#727621: armel: std::future appears to be broken?
Hi,
2013-10-24 23:51, Bastian Blank:
> On Fri, Oct 25, 2013 at 12:12:59AM +0300, Eugene V. Lyubimkin wrote:
> > 2013-10-24 21:45, Bastian Blank:
> > > Please fix cupt to show the complete command-line. The build-logs are
> > > near to useless.
> > In general, I tend to disagree, for noise-to-signal and readability
> > ratio.
>
> Log from buildds are usually only inspected in the case of errors or by
> other automatic means, so there is no noise. As the errors sometimes
> are not really reproducible, it needs to provide all information needed
> to solve the problem.
I agree that sometimes that helps. But still package is built not only
by buildds.
> Well. Did you already try to reproduce it on the developer machine?
Not back then, but I got to it now:
-------------------------------8<-------------------------------
(jessie_armel-dchroot)jackyf@abel:~$ g++ -std=c++11 future.cpp
future.cpp: In function 'int main()':
future.cpp:5:20: error: variable 'std::future<bool> fb' has initializer but incomplete type
std::future<bool> fb = std::async([]{ return false; });
^
future.cpp:5:55: error: invalid use of incomplete type 'class std::future<bool>'
std::future<bool> fb = std::async([]{ return false; });
^
In file included from future.cpp:1:0:
/usr/include/c++/4.8/future:113:11: error: declaration of 'class std::future<bool>'
class future;
^
/usr/include/c++/4.8/future: At global scope:
/usr/include/c++/4.8/future:175:5: error: 'std::future<typename std::result_of<_Functor(_ArgTypes ...)>::type> std::async(_Fn&&, _Args&& ...) [with _Fn = main()::__lambda0; _Args = {}; typename std::result_of<_Functor(_ArgTypes ...)>::type = bool]', declared using local type 'main()::__lambda0', is used but never defined [-fpermissive]
async(_Fn&& __fn, _Args&&... __args);
^
(jessie_armel-dchroot)jackyf@abel:~$ cat future.cpp
#include <future>
int main()
{
std::future<bool> fb = std::async([]{ return false; });
return 0;
}
------------------------------->8-------------------------------
The log is from abel.debian.org. The same command works on my amd64 machine.
Please let me know if something more if needed to help with an investigation.
--
Eugene V. Lyubimkin aka JackYF, JID: jackyf.devel(maildog)gmail.com
C++ GNU/Linux userspace developer, Debian Developer
Reply to: