Re: h8300-hms-ld segfaults (with a small testcase)

[Alan Modra <amodra at bigpond dot net dot au>]

On Wed, Dec 18, 2002 at 10:42:47AM -0500, Daniel Jacobowitz wrote:
> On Thu, Dec 19, 2002 at 12:51:35AM +1030, Alan Modra wrote:
> > On Mon, Dec 16, 2002 at 04:27:53PM -0000, Max Bowsher wrote:
> > > In this case, a generic hash table is created, then treated as a h8300
> > > extended one.
> > 
> > Would someone please check that the following patch cures the
> > segfault (and doesn't break something)?
> 
> My concern is that this causes .vectors handling to be just skipped
> when linking to non-h8300.  I'd need a second opinion but when I looked
> over this the code generated relocations in the text section to things
> in .vectors, implying that it is necessary...

See the h8300_reloc16_extra_cases patch.  You'll get "cannot handle
R_MEM_INDIRECT reloc when using symbolsrec output" for those relocs.

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre