Re: [RFC] Do we care about binary compatibility of code produced by cross-compilers?

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

Mark Mitchell wrote:
> Paolo Carlini wrote:
> > As far as I can see, among the 4 libraries only libstdc++-v3 tries to be
> > "smart" and runs it only when link-tests are always safe to do, when
> > native. I find all of this rather puzzling. Shouldn't we run the test
> > only once, at toplevel, propagate the information to the libraries, and,
> > in case, have an infrastructure which allows the libraries to override
> > the toplevel decision? 
> >     
> Yes, we should.
>
> As per our earlier discussions, I think the libstdc++ "smart"-ness is a
> mistake.  We have a good method for probing the C library: link-time
> tests.  We don't use it for libstdc++ only because some people use a
> build procedure in which you can't link C programs by the time you're
> building the C++ library.  I suspect the other library maintainers just
> didn't worry about that scenario.
>   
Ok. To clarify my personal position, I have no personal objections at 
all to moving the GCC_CHECK_TLS call a few lines above, among the checks 
performed unconditionally, for cross builds too. Just want to make sure 
there is consensus about that.

The point of my message of today is also that, if we do that, we can as 
well move the check to the toplevel, in a centralized position, where 
further changes would be clearly result of general agreement between the 
build-system maintainers.  Would it be easy to do? Frankly, on the spot 
I don't know, I'm not an expert in this area, I suspect it would be easy 
indeed...

Paolo.