Home Original page

x86-64 large data sections updated

Jan Hubicka jh@suse.cz
Wed Jun 15 10:39:00 GMT 2005 
> On Wed, Jun 15, 2005 at 01:43:31AM +0200, Jan Hubicka wrote:
> > Hmm, right.  I guess elf_backend_additional_program_headers is the right
> > place for this.  Now when lbss comes just after .bss, we will likely
> > need 1 segment for ldata, 1 segment for lrodata and lbss doesn't count,
> > or am I counting total nonsense?
> 
> Well, one for (ldata | lbss), one for lrodata.
OK
(however I was under impression that ordering bss next to lbss would
result in merging the segments when possible)
> 
> > OK, so dropping here the lcommon test sounds sane?
> 
> The entire function is not sane.  The .largecomm assembler directive
> should set the section to SHN_X86_64_LARGE_COMMON, and that should be
> the end of it.  Once COMMON symbols get mapped into .lcommon/.lbss
> sections, they don't return.

I just duplicated here what mips and m32r is doing but this function
ineed look pretty redundant.

I will do some more testing by GCC medium model bootstrap and send
updated patch later today.

Thanks,
Honza
> 
> 
> r~
More information about the Binutils mailing list