3.3 problem of -fzero-initialized-in-bss

[Nathanael Nerode]

>What I see above all is that we said that we were commited to making 
>sure that GCC produces a working Emacs and we failed.

We did produce a working emacs.  It has the binary name "temacs".

>From the "GNU Emacs Lisp Reference Manual",
"GNU Emacs Internals",
"Building Emacs":
>  Some operating systems don't support dumping.  On those systems, you
>must start Emacs with the `temacs -l loadup' command each time you use
>it.

The emacs people recognized that the bug in unexec was exactly that -- 
their bug.  Shipping 'bug-compatible' versions of GCC has never been a 
release goal, as far as I know.

We don't turn off features of GCC by default to allow buggy 
versions of the Linux kernel to compile, and we shouldn't do it for 
emacs either.  The release criteria recognize this explicitly when 
talking about the Linux kernel, and should recognize this for emacs as 
well.

-- 
Nathanael Nerode  <neroden at gcc.gnu.org>
http://home.twcny.rr.com/nerode/neroden/fdl.html