MIPS md_apply_fix()(?) problem.

[cgd@broadcom.com]

"Richard Sandiford" <rsandifo@redhat.com> writes:
> Seems to fix the case and shows no regressions on mips-elf or
> mips-ecoff.  The test case for the original bug was MIPS reloc 4, which
> still passes after this patch.

FYI:

I took out the prev workaround that I had in my sources (the -G 0
patch H.J. mentioned) and put this in.

I compared the results of a build with my old working tools with the
results from building with my newer tools (mostly binutils 2001-10-10
00:00 UTC + your patch + a bunch of others).

Nothing screwy.  Diffs nice and clean even.  8-)

>From where i sit, this looks like a good patch.


chris