public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Luis Machado <lgustavo@codesourcery.com>
To: Simon Marchi <simon.marchi@ericsson.com>, <gdb-patches@sourceware.org>
Subject: Re: [PATCH 2/2] arm-tdep.c: Refactor displaced stepping relocation functions
Date: Fri, 26 Feb 2016 16:50:00 -0000	[thread overview]
Message-ID: <56D08256.80502@codesourcery.com> (raw)
In-Reply-To: <56D07C04.9060000@ericsson.com>

On 02/26/2016 01:23 PM, Simon Marchi wrote:
>>> diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
>>> index 43b61c2..ef48a90 100644
>>> --- a/gdb/arm-tdep.c
>>> +++ b/gdb/arm-tdep.c
>>> @@ -7111,9 +7111,8 @@ thumb_copy_pop_pc_16bit (uint16_t insn1, struct arm_insn_reloc_data *data)
>>>      return 0;
>>>    }
>>>
>>> -static void
>>> -thumb_process_displaced_16bit_insn (uint16_t insn1,
>>> -				    struct arm_insn_reloc_data *data)
>>> +static int
>>> +arm_relocate_insn_thumb_16bit (uint16_t insn1, struct arm_insn_reloc_data *data)
>>>    {
>>>      unsigned short op_bit_12_15 = bits (insn1, 12, 15);
>>>      unsigned short op_bit_10_11 = bits (insn1, 10, 11);
>>> @@ -7202,9 +7201,7 @@ thumb_process_displaced_16bit_insn (uint16_t insn1,
>>>          err = 1;
>>>        }
>>>
>>> -  if (err)
>>> -    internal_error (__FILE__, __LINE__,
>>> -		    _("thumb_process_displaced_16bit_insn: Instruction decode error"));
>>> +  return err;
>>
>> Should we keep this internal error message under a different context
>> instead of exporting just an error code? Maybe the error code should
>> trigger this internal error for GDB?
>
> I am not sure I understand your comment. Given this call tree:
>
> - arm_process_displaced_insn
>    - arm_relocate_insn_arm
>      ...
>    - arm_relocate_insn_thumb_32bit
>      ...
>    - arm_relocate_insn_thumb_16bit
>      ...
>
> my patch makes it so that the arm_relocate_insn* functions return an error code, and
> arm_process_displaced_insn calls internal_error if an error is returned.  Do you suggest
> putting the internal_error calls in the arm_relocate_insn_* functions directly?
>

No. If we want them shared, i don't think we'd want them to throw 
internal errors.

I was just pointing out the fact that we're losing the function name 
information from arm_relocate_insn_thumb_16bit's and 
thumb_process_displaced_32bit_insn's internal error messages. This 
information may make debugging easier. We would need to throw errors 
with custom messages from within arm_process_displaced_insn in order to 
maintain those names. For example:

"thumb_process_displaced_16bit_insn: Instruction decode error"
"thumb_process_displaced_32bit_insn: Instruction decode error"

Instead of:

"arm_process_displaced_insn: Instruction decode error"

I see the regular non-thumb functions don't throw internal errors 
themselves though. So i'm fine if others think the more specific error 
message is not needed.

Luis

  reply	other threads:[~2016-02-26 16:50 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-25 15:47 [PATCH 0/2] Decouple ARM instruction decoding/relocating from GDB Simon Marchi
2016-02-25 15:47 ` [PATCH 2/2] arm-tdep.c: Refactor displaced stepping relocation functions Simon Marchi
2016-02-26 16:11   ` Luis Machado
2016-02-26 16:23     ` Simon Marchi
2016-02-26 16:50       ` Luis Machado [this message]
2016-02-25 15:47 ` [PATCH 1/2] arm-tdep.c: Replace arguments to relocation functions by a structure Simon Marchi
2016-02-26 16:12   ` Luis Machado
2016-02-26 16:25     ` Simon Marchi
2016-02-26 16:26       ` Simon Marchi
2016-03-04 12:03 ` [PATCH 0/2] Decouple ARM instruction decoding/relocating from GDB Yao Qi
2016-03-04 18:42   ` Simon Marchi

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=56D08256.80502@codesourcery.com \
    --to=lgustavo@codesourcery.com \
    --cc=gdb-patches@sourceware.org \
    --cc=simon.marchi@ericsson.com \
    /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).