This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


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

Re: AIX I/O failures


>>>>> "benjamin" == benjamin kosnik <bkoz@nabi.net> writes:

    benjamin> That's why I'm for this, if it doesn't break other

Great.

    benjamin> FWIW, I think g++ needs a FAQ entry or explict
    benjamin> user-level documentation on how to deal with static
    benjamin> initialization portably across platforms.

Yes, that would be a good thing indeeed.

Also, some details on the various limitations in G++ when using it
with platforms that do not support weak symbols.

There are, by the way, ways to deal with the no-weak-symbols problem
relatively portably, and that only affect the darkest corners of
standards conformance.  EDG and its customers, for example, has
successfully targeted their compiler front-end and template prelinker
(it is these two components, not the back-end, that handle the
details) at just about every plaform on earth.  

It's just that nobody has implemented these approaches in G++.  I'd
love to get a chance to do so at some point; you can actually win for
platforms that do have weak symbols by doing this, too, in some
cases...

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com

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