Re: objdump issues on ppc64?

[Alan Modra <amodra at gmail dot com>]

On Tue, Mar 06, 2018 at 05:37:36PM -0500, Jeffrey Walton wrote:
> On Tue, Mar 6, 2018 at 5:25 PM, Alan Modra <amodra@gmail.com> wrote:
> > On Tue, Mar 06, 2018 at 09:21:01AM -0600, Bill Schmidt wrote:
> >> > On Mar 6, 2018, at 8:25 AM, Jeffrey Walton <noloader@gmail.com> wrote:
> >> > ppc64-be (gcc119.fsffrance.org) (I think this is the same block of
> >> > code, give or take):
> > [snip]
> >> >    100006a4:       38 c9 00 20     cal     r6,32(r9)
> >>
> >> gcc119 is running AIX.  I believe it has its own disassembler.
> >
> > Yes, and likely needs some option to disassemble power8 code.  The
> > "cal" says this isn't objdump.
> 
> I'm working on GCC119 from the compile farm.
> 
> $ /opt/cfarm/binutils-latest/bin/objdump --disassemble sha256-p8.exe |
> /opt/cfarm/binutils-latest/bin/c++filt
> 
> sha256-p8.exe:     file format aixcoff-rs6000

Oh..  Yes, this is objdump.  Sorry for the misinformation.

And the "cal" is indeed possible by disassembling with -Mpwr.  (I
looked at ppc-opc.c and misread
{"addi",	OP(14),		OP_MASK,     PPCCOM,	PPCNONE,	{RT, RA0, SI}},
{"cal",		OP(14),		OP_MASK,     PWRCOM,	PPCNONE,	{RT, D, RA0}},
thinking the PWRCOM was PPCCOM.)

It looks like objdump for 32-bit xcoff accepts but ignores the -M
options, so you're disassembling with the default -Mpwr.  I'll see
about fixing that.

-- 
Alan Modra
Australia Development Lab, IBM