Re: emacs-current bootstrap fail on FreeBSD 6.2

[Richard Stallman]

    While compiler bugs are always a possibility, C has "unspecified" so many
    parts of its behavior, that many bugs in your programs can stay hidden until
    some compiler's optimization bumps into them.
    So it's actually more probable that it's a bug in our code, which happens to
    be uncovered by one of the -O2 optimizations.

I am not sure this is more probable, but it is a possibility.  The way
to look for these problems is to study the cpp output carefully after
refreshing your memory about what sort of thing is undefined.

Can hashiz provide the cpp output for that function?