Re: [PATCH] Performance counter opcodes for MIPS R1[02]000

[cgd at sibyte dot com]

"Thiemo Seufer" <ica2_ts@csv.ica.uni-stuttgart.de> writes:
> [snip]
> > The selection of t0 for that register name, for a binary unmarked with
> > ABI, is a change, and I don't understand why it's correct.
> 
> Ah, now I understand Your concern. I wasn't aware of the fact that
> there are unmarked binaries. I thought the use of ABI-flags was
> mandatory in ELF.

Ahh, no.  See also Daniel J.'s response.


There was a discussion on the gcc-patches list last August about MIPS
ABI selection and ELF flags, under the subject of "RFA: Recording MIPS
ABI selection in binaries" -- you might go have a look at parts of it.
Of particular interest is:

	http://gcc.gnu.org/ml/gcc-patches/2000-08/msg00418.html

which details the e_flags values that the SGI compilers apparently
use.

That, combined with the current flags values in the GNU tree should
give you a mostly-complete idea of what you're up against.


> 2001-06-21  Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
> 
> 	/gas/testsuite/ChangeLog
> 	* mips/lb.d: Reflect disassembler output fixes.
> 	* mips/mips32.d: Likewise.
> 	* mips/mips64.d: Likewise. Typo.
> 	* mips/sb.d: Likewise.
> 	* mips/trunc.d: Likewise.
> 
> 	/opcodes/ChangeLog
> 	* mips-dis.c (print_insn_arg): Don't use software integer registers
> 	for coprocessor registers.
> 	(get_mips_isa): Removed.
> 	(_print_insn_mips): Get distinction between old ABI and new ABI right.

This looks reasonable to me.

(thanks for noticing the mips64.d typo... d'oh!  8-)



chris