> > Either hsh->target_section->owner or hsh->bh_root.string is NULL. =C2=A0C=
> ould
> > you put a break on this line and print the struct hsh in gdb?
>
> bfd/elf32-hppa.c:
> 3568 /* Make sure we can reach the stub. */
> 3569 if (max_branch_offset !=3D 0
> 3570 && value + addend + max_branch_offset >=3D 2*max_branch_offset)
> 3571 {
> 3572 (*_bfd_error_handler)
> 3573 (_("%B(%A+0x%lx): cannot reach %s, recompile with
> -ffunction-sections"),
> 3574 input_bfd,
> 3575 input_section,
> 3576 offset,
> 3577 hsh->bh_root.string);
> 3578 bfd_set_error (bfd_error_bad_value);
> 3579 return bfd_reloc_notsupported;
> 3580 }
>
> (gdb) step
> 3572 (*_bfd_error_handler)
...
> (gdb) print hsh->bh_root.string
> No symbol "hsh" in current context.
> I do not understand why there is no such symbols.
As soon as you step into the error handler, you change the function
context. So, there is no local symbol hsh there. It's also possible
the symbol hsh is optimized away, particularly because of the way it's
derived fro the argument bh.
If hsh is not available, try bh->bh_root. hsh is derived from
bh as follows:
hsh = (struct bfd_hash_entry *) bh;
It's always tricky to debug optimized code.
Dave
--
J. David Anglin dave.anglin@xxxxxxxxxxxxxx
National Research Council of Canada (613) 990-0752 (FAX: 952-6602)
--
To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
