What's Left for MingW 3.3

Mon Apr 14 12:50:00 GMT 2003

Hi Andrew,
>Thanks. I checked in your strict case patch. Of course I haven't
>tested it, so please make sure you do that.

This checks out okay for the CVS 3.3 sources I checked out on Friday.
What's left:
- GC_DLL in boehm-gc: Tom was supposed to folllow up on this:
 http://gcc.gnu.org/ml/java-patches/2003-q1/msg00669.html
- http://gcc.gnu.org/ml/java-patches/2003-q1/msg00672.html
 needs someone to commit to 3.3
- Ranjit's os.arch patch
 http://gcc.gnu.org/ml/java-patches/2003-q1/msg00445.html
- What do I do about _Jv_select()?:
 http://gcc.gnu.org/ml/java/2003-04/msg00187.html
- Cross-configury (no, I won't stop talking about this :) )
 May I submit a patch or not?
- Other Win32-build-busting stuff in natPlainDatagramSocketImpl.cc:
 http://gcc.gnu.org/ml/java-patches/2003-q1/msg00670.html
 (I have a less-intrusive solution than the one in the above which
 only touches win32.h. Still need to submit this.)
 
- mingwex specs patch (not yet submitted)
 adds -lmingwex to gcc/config/i386/mingw32.h
 Ranjit said that the MingW folks would lobby for their compiler-specific
 patches and I should be patient, but I don't see this happening and
 3.3 seems to be drawing to a close. What's more, I don't see anything
 necessary other than linking in an extra library by default. Without this,
 the MingW native and cross compilers choke when linking anything that
 has opendir() and readdir(), like fastjar.
Like I mentioned before, you might have a hard time assessing the
correctness of proposals like MingW-specific patches like
the mingw32.h and win32.h one, so the question is, do we scare
someone up who fulfills all the CVS requirements you mentioned?
Does such a person exist? Or do you personally just commit
these (perhaps with someone like Ranjit weighing in on these)
knowing that they won't affect the Posix code?
-- Mohan
http://www.thisiscool.com/
http://www.animalsong.org/