Re: emacs-current bootstrap fail on FreeBSD 6.2

[Stefan Monnier]

> src/keymap.c  1.330    OK
> src/keymap.c  1.331    NG

OK, so until now it seems we have discovered that the problem appeared with
this change.  Funnily enough, basically 0% of the code introduced by this
change is in the latest revision (1.341).
Also it seems that removing the "-O2" argument makes the problem disappear.

- when it fails, does it always fail at the exact same spot?
- I see one failure at "thai-word" and another at "url-cache", but both have
  pretty much the same C backtrace, ending with mark_maybe_object.  Have all
  the backtraces have this `mark_maybe_object' in common?
- can you run the `bootstrap-emacs' command directly from gdb, rather than
  going through a core file?  Show us the backtraces, of course.
- the backtraces show that some kind of signal thingy happens at
  alloc.c:4244, but that line is just a function call and I can't see how it
  can fail.  Please keep an eye on this.  Try to see why it crashed at that
  point.
- don't use `pp' because it runs C code, so it can change the debugged
  program's behavior.
- try to reproduce the bug with less optimization (so that the debugger's
  info is more reliable): try -O1 and then try to add/remove optimization
  phases one by one (using `man gcc' to see what phases are included in
  -O1 and -O2).
- please check again all the warnings when you compile with -Wall, to see
  if there's some `implicit declaration of function FOO' somewhere.  I guess
  there isn't, but ... just in case.


        Stefan