This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RFA/RFT] libffi reorg
- From: Tom Tromey <tromey at redhat dot com>
- To: "Paolo Bonzini" <bonzini at gnu dot org>
- Cc: <gcc-patches at gcc dot gnu dot org>
- Date: 19 Sep 2003 10:16:47 -0600
- Subject: Re: [RFA/RFT] libffi reorg
- References: <1058195670.3331.25.camel@escape><003401c37e9b$de5fb230$e314623e@philo>
- Reply-to: tromey at redhat dot com
>>>>> "Paolo" == Paolo Bonzini <bonzini@gnu.org> writes:
>> The intent here is to move the bulk of the runtime configury into per-cpu
>> header files instead of ifdefs in ffi.h. It also cleans up some namespace
>> issues, such that nothing that doesn't start with FFI_ or ffi_ is seen in
>> the header files.
This seems like a nice idea.
However, I've also wanted a way for libffi to tell client programs
about its local configury. For instance, right now in
libjava/configure.host we have code that knows whether a given
platform supports the closure API (or libffi at all). But why should
we do that, when libffi already knows this? I wish we could just
unconditionally include some libffi header and use some exported
define to decide what to do. Any thoughts on that?
Tom