mn10300 assembler fix for error in debugging info

[Alexandre Oliva]

On Oct  3, 2002, Alan Modra <amodra@bigpond.net.au> wrote:

> Indeed, the problem was introduced with my gas reloc changes.  I
> made a suggestion to Alex on a possible fix a few minutes before
> receiving your email, but on looking a little more closely, I
> think my suggestion was probably wrong.

> Instead, I think you want

> #define TC_VALIDATE_FIX_SUB(FIX) 1

> in tc-mn10300.h to disable the write.c check.  (Difference
> expressions will be checked later in tc-mn10300.c:tg_gen_reloc)

This fix works, indeed, and it produces the right results.  Even
though I'm not entirely happy about it, I'm checking in the first
patch below, at least as a stop-gap.

The second patch was an attempt to extend one of the
TC_FORCE_RELOCATION_SUB_ macros such that we could implement the plan
detailed in the comments I added before TC_LINKRELAX_FIXUP in the
first patch.  Does it sound like a reasonable plan?  Should the same
change be applied to other such macros so that we could implement it?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: gas-mn10300-diff-validate.patch
Type: text/x-patch
Size: 1786 bytes
Desc: not available
URL: <https://sourceware.org/pipermail/binutils/attachments/20021211/e2ea1e57/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gas-extend-force-reloc-sub-same.patch
Type: text/x-patch
Size: 6780 bytes
Desc: not available
URL: <https://sourceware.org/pipermail/binutils/attachments/20021211/e2ea1e57/attachment-0001.bin>
-------------- next part --------------

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                 aoliva@{redhat.com, gcc.gnu.org}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer