Re: [v3] c++0x <mutex> and <condition_variable>

[Paolo Carlini <paolo dot carlini at oracle dot com>]

Hi Chris,
> One thing that's unclear to me is how to determine the targets in the
> testcase's dg-do run option. I'm just using what I think Benjamin (or
> whoever did the initial commit) was using. I'm not sure how to go
> about determining whether its still correct or not.
>   
We should be very careful indeed. I did these issues a bit of thought. 
For anything using threads I think we need the same dg-* directives we 
have in the shared_ptr tests, for instance and known to be safe:

// { dg-do run { target *-*-freebsd* *-*-netbsd* *-*-linux* *-*-solaris* 
*-*-cygwin *-*-darwin* alpha*-*-osf* mips-sgi-irix6* } }
// { dg-options "-pthread -std=gnu++0x" { target *-*-freebsd* 
*-*-netbsd* *-*-linux* alpha*-*-osf* mips-sgi-irix6* } }
// { dg-options "-pthreads -std=gnu++0x" { target *-*-solaris* } }
// { dg-options " -std=gnu++0x " { target *-*-cygwin *-*-darwin* } }

+ in case the test relies on the new gthread support, we need a new 
dg-require very similar to the one I added for the atomic builtins (look 
for check_v3_target_atomic_builtins) checking that <bits/c++config.h> 
defines _GLIBCXX_HAS_GTHREADS. Likewise for any other feature required 
for the correct build and run of the test (__GTHREAD_HAS_COND?!?). 
Really rather mechanical additions.

That way the dg-* directives should ok. Maybe a little redundant in case 
of dg-require for gthreads, because likely many of the above targets 
currently do not use posix threads, but safe, anyway and ready in case 
further threading models besides posix define the C++0x functions.

Paolo.