Bug in emacs-20.3.9.1 with egs-1.1.2 (IRIX6.5/mips4)

[Richard Stallman]

It seems that GCC is generating legitimate code, but that unexec fails
to handle it properly.  So this is probably an unexec bug rather than
a GCC bug.  (Does anyone disagree?)

If so, the next thing to do is to examine the dumped emacs executable.
See what it has to correspond with the pointer location .LC5, and try
to figure out whether the executable is simply incorrect.

I hope that someone from SGI will offer to help figure this out.