On Fri, 17 Jan 2003, Richard Henderson wrote: > On Thu, Jan 16, 2003 at 09:18:20AM -0600, Brian Ford wrote: > > I am still consulting the DWARF2 spec to see if gcc and gas are > > correct in generating VMA addresses. If so, I guess I have to fix the > > dwarf parsing code in bfd and gdb to subtract the section base VMA. > > No, dwarf specifies a section-relative address. The issue is that, > with the exception of IA-64, no target has section-relative relocations. > So instead we force the VMA to zero such that the VMA address and > the section offset are identical. > > If you cannot do this with PE, then you have to add new relocs. > There's no other way around it.
Wait a minute, are you saying IA-64 implements section-relative relocations? Is that for any kind of sections, or just the dwarf's ones? Hope the former, because in that's exactly what I need for AROS. Fabio Alemagna -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/