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

[stl at caltech dot edu]

------- Additional Comments From stl at caltech dot edu  2004-02-29 21:04 -------
[Stephan T. Lavavej]
> When I take the FSF 3.3 sources and patch them with Danny Smith's 3.3 patch
> (sent to a mailing list but never available from mingw.org), the bootstrap
> fails with some message about stdio.h.

[Danny Smith]
> The (complete) patchset and sources (as well as prebuilt binaries)
> are available from www.mingw.org.  Follow the link that says
> 'Candidate: GCC-3.3.3' under News.

Note that there I was talking about 3.3[.0].

I did SOMETHING to bootstrap it with MSYS.  I remember rejoicing when I 
finally got it, since before I had been tediously crossing it over from 
GNU/Linux.  I think maybe I had to put mingw-runtime and w32api in the 
destination directory, whereas that is not necessary for 3.3.1.

I'm currently rebuilding 3.3.1 because I did something bad to my directories. 
After that I'll build FSF 3.3.3 and "real" MinGW 3.3.3, and then I'll see if I 
can get 3.3[.0] to build again, to see if the increase in libstdc++.a size is 
actually occuring, or is caused by some change between my forgotten old build 
process and the one I'm using right now.

Actually, looking at it, the error I just got with 3.3[.0] seems to have been 
identical to the FSF 3.3.2 error I mentioned above.

What do the MinGW-local patches "do", anyways?  If I only care about gcc/g++ 
and not other languages, do they do anything for me?  It's all very confusing.

-- 


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