This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: relax jalr $t9 [R_MIPS_JALR symbol] to bal symbol
On Mar 25, 2003, Ralf Baechle <ralf at linux-mips dot org> wrote:
> I can't imagine implementations are intelligent enough to detect
> this special case in their branch prediction so branches may
> occasionally be misspredicted.
I'd think they should. Gee, it's a special register, after all!
Anyway, we could the `likely' version of this instruction, if we find
that the next instruction is a nop. Heck we could even try to copy
the instruction from the branch target and branch to the next
instruction instead, as long as it's not a branch itself. It's just a
matter of programming :-)
> So I'd say go for the bal thing.
Yeah, one step at a time. I just need approval from the MIPS
maintainer now. Hmm... I *think* Eric had pre-approved this. Eric,
can you confirm?
--
Alexandre Oliva Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer aoliva at {redhat dot com, gcc.gnu.org}
CS PhD student at IC-Unicamp oliva at {lsd dot ic dot unicamp dot br, gnu.org}
Free Software Evangelist Professional serial bug killer