Re: failure of cross-compilation detection on BlueGene/L

[Steven G. Johnson]

Ralf Wildenhues wrote:
> One simple but a bit ugly workaround would be to
>   ./configure cross_compiling=yes

This doesn't work.  The configure script ignores any pre-existing value 
of cross_compiling.

> How effective and efficient would such a test be?  IOW, are we sure
> writing to a file is sufficient for all current (and maybe the future)
> BG versions and other cross compilation setups to detect them as such?
> OTOH, can we be reasonably sure that this works with all emulators we
> might care about not to be seen as cross setups?  (wine seems to be fine
> with this.)

It seems impossible to guarantee that a test will detect cross-compiling 
on all future hardware and system software.  I don't see how one can 
find any test that makes such a guarantee.

However, as I said, we are *already* assuming that this works on all 
existing non-cross-compile setups, because fopen/fclose is assumed by 
the runtime version of AC_COMPUTE_INT (see AC_LANG_INT_SAVE).

So, the test I proposed seems strictly better than what we have now: it 
should not give any false positives, and should reduce the rate of false 
negatives.

> It matters not as the host alias you pass to --host, but definitely for
> macros like those from Libtool which decide things based on $host.  In
> the other thread, powerpc-bgp-linux seemed like a sensible thing.  But

This doesn't make sense to me, because the compute nodes (as far as I 
understand from IBM's papers online) are not running any version of a 
Linux kernel.  See 
http://www.mcs.anl.gov/~beckman/bluegene/SSW-Utah-2005/BGL-SSW03-CNK-CIOD.pdf

Steven