public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
From: Nelson Chu <nelson.chu@sifive.com>
To: Andrew Burgess <aburgess@redhat.com>
Cc: Binutils <binutils@sourceware.org>
Subject: Re: [PATCH 0/3] disassembler syntax highlighting in objdump (via libopcodes)
Date: Thu, 17 Feb 2022 11:57:34 +0800	[thread overview]
Message-ID: <CAJYME4HD-85A7c9yyPqvoMsR=Et8+dc-bSVcDNpcn7xvP9NY_Q@mail.gmail.com> (raw)
In-Reply-To: <cover.1645043588.git.aburgess@redhat.com>

Hi Andrew,

I get gas/testsuite/gas/riscv/insn testcase fail after applying the
series of patches.  Here is the reduced case,

$ cat tmp.s
.word 0x68c58543
fmadd.s fa0,fa1,fa2,fa3,rne
.insn r  MADD, 0, 0, a0, a1, a2, a3
.insn 0x68c58543

All these are the same instruction encoding - fmadd.s, but the first
".word" will be marked and dumped as data rather than instruction.  We
are used to get the following dump result,

$ /scratch/nelsonc/build-upstream/rv64gc-elf/build-install/bin/riscv64-unknown-elf-objdump
-d tmp.o

0000000000000000 <.text>:
   0:   68c58543                .word   0x68c58543
   4:   68c58543                fmadd.s fa0,fa1,fa2,fa3,rne
   8:   68c58543                fmadd.s fa0,fa1,fa2,fa3,rne
   c:   68c58543                fmadd.s fa0,fa1,fa2,fa3,rne

But now with the patches, I will get segmentation fault when trying to
dump fmadd.s as instruction,

$ /scratch/nelsonc/build-upstream/rv64gc-elf/build-install/bin/riscv64-unknown-elf-objdump
-d tmp.o

0000000000000000 <.text>:
   0:   68c58543                .word   0x68c58543
Segmentation fault (core dumped)

Is it possible if you could help to see what happened?  Thanks!


Nelson

On Thu, Feb 17, 2022 at 4:53 AM Andrew Burgess via Binutils
<binutils@sourceware.org> wrote:
>
> This series is a serious attempt at what I discussed here:
>
>   https://sourceware.org/pipermail/binutils/2021-December/118806.html
>
> This series changes libopcodes so that this disassemblers can supply
> styling information with every piece of disassembly output, e.g. is
> this a register?  an address?  a mnemonic?  etc.
>
> Users of the disassembler can then choose to make use of this
> information to add styling to the disassembler output.
>
> And that is what I do for objdump in this series.  The styling is off
> by default, but can be turned on with a new command line flag:
>     --disassembler-color=off|color|extended-color
>
> I've updated GDB enough to keep it building and running after this
> change, though at this point GDB doesn't make use of the new styling
> information, that will come later.
>
> All feedback would be welcome.
>
> Thanks,
> Andrew
>
> ---
>
> Andrew Burgess (3):
>   objdump/opcodes: add syntax highlighting to disassembler output
>   opcodes/riscv: implement style support in the disassembler
>   opcodes/i386: partially implement disassembler style support
>
>  binutils/NEWS              |   4 +
>  binutils/doc/binutils.texi |  11 ++
>  binutils/objdump.c         | 245 ++++++++++++++++++++++++++++++++-----
>  gdb/disasm.c               |  34 ++++-
>  gdb/disasm.h               |   7 ++
>  include/dis-asm.h          |  62 +++++++++-
>  opcodes/dis-init.c         |   5 +-
>  opcodes/disassemble.c      |  23 +++-
>  opcodes/i386-dis.c         |  71 +++++++----
>  opcodes/riscv-dis.c        | 147 +++++++++++-----------
>  10 files changed, 475 insertions(+), 134 deletions(-)
>
> --
> 2.25.4
>

  parent reply	other threads:[~2022-02-17  3:57 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-16 20:53 Andrew Burgess
2022-02-16 20:53 ` [PATCH 1/3] objdump/opcodes: add syntax highlighting to disassembler output Andrew Burgess
2022-02-28 15:54   ` Tom Tromey
2022-02-16 20:53 ` [PATCH 2/3] opcodes/riscv: implement style support in the disassembler Andrew Burgess
2022-02-19 10:24   ` Andrew Burgess
2022-02-16 20:53 ` [PATCH 3/3] opcodes/i386: partially implement disassembler style support Andrew Burgess
2022-02-17  9:35   ` Jan Beulich
2022-02-17 16:15     ` Andrew Burgess
2022-02-17 16:29       ` Jan Beulich
2022-02-17 22:37         ` Andrew Burgess
2022-02-18  7:14           ` Jan Beulich
2022-02-19 10:54             ` Andrew Burgess
2022-02-21 13:08               ` Jan Beulich
2022-02-21 18:01                 ` Andrew Burgess
2022-02-17  3:57 ` Nelson Chu [this message]
2022-02-17 16:17   ` [PATCH 0/3] disassembler syntax highlighting in objdump (via libopcodes) Andrew Burgess
2022-03-21 14:33 ` [PATCHv2 " Andrew Burgess
2022-03-21 14:33   ` [PATCHv2 1/3] objdump/opcodes: add syntax highlighting to disassembler output Andrew Burgess
2022-03-21 14:33   ` [PATCHv2 2/3] opcodes/riscv: implement style support in the disassembler Andrew Burgess
2022-03-21 14:33   ` [PATCHv2 3/3] opcodes/i386: partially implement disassembler style support Andrew Burgess
2022-03-24 17:08   ` [PATCHv2 0/3] disassembler syntax highlighting in objdump (via libopcodes) Nick Clifton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAJYME4HD-85A7c9yyPqvoMsR=Et8+dc-bSVcDNpcn7xvP9NY_Q@mail.gmail.com' \
    --to=nelson.chu@sifive.com \
    --cc=aburgess@redhat.com \
    --cc=binutils@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).