This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Patch to fix non-sun targets pretending to be sun-solaris2


"Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> writes:

> As discussed previously,
> http://gcc.gnu.org/ml/gcc-patches/2003-11/msg00195.html
>
> several platforms are reusing config/sol2.h and this causes them to
> erroneously define `sun' and misidentify themselves.  This causes
> problems in user code when one checks "#ifdef sun" and makes any
> assumptions based on that.
>
> Perhaps the 'real' fix is to separate out the common elements of
> sol2.h into yet another file, then disentangle the entire mess.  I
> felt that was too invasive for 3.4 in stage3, but I still think we
> should fix the macro problem.  The solution I propose is to undefine
> sun in the right places.  These are easier to identify and edit rather
> than pulling apart the config stuff.

This is probably what we'll have to do for 3.4 but I would like to see
what your 'real' fix would look like, before making a decision.  If
nothing else, that can then be put on tree-ssa in anticipation of 3.5.

zw


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]