[Bug target/14344] [3.3.3 Regression] MinGW Bootstrap Failure

["cgf at redhat dot com" <gcc-bugzilla at gcc dot gnu dot org>]

------- Additional Comments From cgf at redhat dot com  2004-02-29 22:29 -------
Subject: Re:  [3.3.3 Regression] MinGW Bootstrap Failure

On Sun, Feb 29, 2004 at 08:54:43PM +0100, Gabriel Dos Reis wrote:
>"stl at caltech dot edu" <gcc-bugzilla@gcc.gnu.org> writes:
>
>| > The immediate conclusion I can draw is that there is no active maintainers
>| > for the MinGW port.
>| 
>| From what I understand, people do actively maintain MinGW (one person you 
>| didn't mention is Danny Smith, who released the 3.3.3 MinGW patch
>| set and has been active in my 3.4 bootstrap failure bug, PR12974),
>| but the critical thing that they don't do is bootstrap MinGW
>| routinely on MSYS, as I do.  Instead they use Cygwin.
>
>OK. The people I mentioned were those listed in $GCC/MAINTAINERS file.
>Danny Smith is listed in the write-after-approval section with no
>particular mention of his focus -- I could not have figured out he
>maintains mingw.
>
>Anyway, if he does actively main MinGW (and that is good news) I would
>suggest he announces so on gcc@gcc.gnu.org so that the SC can
>officially appoint him.
>
>You reported that he produced a 3.3.3 MinGW patch set, then presumably
>he was able to bootstrap?
>
>| Perhaps you could help figure out when and where cygming.h, which
>| 3.3.3 seems to want, disappeared from the 3.3 branch.  I looked at
>| the CVS web interface,  which didn't say anything about cygming.h
>| being removed. 
>
>I don't think that file did disappear in gcc-3.3-branch time frame.
>Rather it *appeared* in gcc-3.4.x with
>
>2003-03-21  Christopher Faylor  <cgf@redhat.com>
>        [...]
>        * config/i386/cygming.h: New file, containing definitions
>        common to mingw32 and cygwin.
>
>
>What I do not understand though, is why it is referenced in the build
>process of gcc-3_3. 

I don't really understand why we're discussing a patched version of gcc.
AFAICT, there are no references to cygming.h anywhere in the
gcc_3_3_3_release branch.  I have closed this bug since it is reported
against the patched mingw sources rather than the pure branch.

I'd suggest that a mingw mailing list would be the best place to get
questions answered about this.  I don't think there is any reason for
further speculation here.

cgf


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14344