This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: objdump issues on ppc64?
- From: Bill Schmidt <wschmidt at linux dot vnet dot ibm dot com>
- To: noloader at gmail dot com
- Cc: Binutils <binutils at sourceware dot org>
- Date: Tue, 6 Mar 2018 09:21:01 -0600
- Subject: Re: objdump issues on ppc64?
- Authentication-results: sourceware.org; auth=none
- References: <CAH8yC8kSHPkavdMMArh7=-VExP5m4q9-amuUs_JtF4xVfrDohw@mail.gmail.com>
> On Mar 6, 2018, at 8:25 AM, Jeffrey Walton <noloader@gmail.com> wrote:
>
> Hi Everyone,
>
> I'm working on a Power8 implementation of SHA using built-ins. I'm
> seeing some unusual results. For example:
>
> ppc64-le (gcc112.fsffrance.org) (notice the unused v0 in 'vperm v0,v0,v0,v17'):
>
> ...
> 10000b0c: a6 03 09 7d mtctr r8
> 10000b10: 57 02 00 f0 xxswapd vs32,vs32
> 10000b14: 6b 04 00 10 vperm v0,v0,v0,v17
> 10000b18: 57 02 00 f0 xxswapd vs32,vs32
> 10000b1c: 99 57 00 7c stxvd2x vs32,0,r10
> 10000b20: 99 26 0c 7c lxvd2x vs32,r12,r4
> 10000b24: 57 02 00 f0 xxswapd vs32,vs32
> 10000b28: 6b 04 00 10 vperm v0,v0,v0,v17
> 10000b2c: 57 02 00 f0 xxswapd vs32,vs32
> 10000b30: 99 67 0a 7c stxvd2x vs32,r10,r12
> 10000b34: 99 26 0b 7c lxvd2x vs32,r11,r4
> 10000b38: 57 02 00 f0 xxswapd vs32,vs32
vs32 is an alias for v0. Is that what's bothering you? The VMX registers
v0-v31 equate to VSX registers vs32-vs63. So 100000b10 feeds 100000b14
which feeds 10000b18, etc.
>
> ppc64-be (gcc119.fsffrance.org) (I think this is the same block of
> code, give or take):
>
> 10000680 <.bb>:
> 10000680: 7d 40 26 18 .long 0x7d402618
> 10000684: 7d 09 03 a6 mtctr r8
> 10000688: 7d 40 57 18 .long 0x7d405718
> 1000068c: 7d 6a ff 18 .long 0x7d6aff18
> 10000690: 7d 8a 67 18 .long 0x7d8a6718
> 10000694: 7c 0a 07 18 .long 0x7c0a0718
> 10000698: 60 00 00 00 oril r0,r0,0
>
> 1000069c <.bb>:
> 1000069c: 7c 0a 4e 19 .long 0x7c0a4e19
>
> 100006a0 <.bb>:
> 100006a0: 38 a9 ff fc cal r5,-4(r9)
>
> 100006a4 <.bb>:
> 100006a4: 38 c9 00 20 cal r6,32(r9)
gcc119 is running AIX. I believe it has its own disassembler.
Thanks,
Bill
>
> I'm guessing there are some outstanding issues with objdump.
>
> Is anyone interested in working on them with me?
>
> Jeff
>