Mirror of the gdb mailing list
 help / color / mirror / Atom feed
From: Nick Roberts <nickrob@snap.net.nz>
To: "André Pönitz" <apoenitz@trolltech.com>
Cc: gdb@sourceware.org
Subject: Re: Questions about gdb/mi support on the Mac
Date: Tue, 28 Aug 2007 08:56:00 -0000	[thread overview]
Message-ID: <18131.58128.433536.819432@kahikatea.snap.net.nz> (raw)
In-Reply-To: <200708281029.27514.apoenitz@trolltech.com>

 > >     Are there other known differences between the Mac and other platforms?
 > > (I'm using gdb 6.3.5 on the Mac).
 > 
 > One thing I am aware of is the gdb/Mac sometimes produces results
 > that to not fit into the MI grammar. E.g. when listing locals
 > it returns:   9^done,locals={{name="a"},{name="w"}}

This is old style MI (mi0) output.

 > instead of:   9^done,locals=[{name="a"},{name="w"}]

Looking at Apple's mi-cmd-stack.c:

  cleanup_list = make_cleanup_ui_out_list_begin_end (uiout, locals ? "locals" : "args");

I think it will output the latter now.

 > IIRC (haven't looked at it for a while) also the output for -break-insert
 > is different.

Apple have documented some of their own commands in their version of the
GDB manual.

 > And (also IIRC) Mac gdb automatically creates MI variables when
 > running -stack-list-locals.

It can do with the "--make-varobjs" option but this doesn't seem to be
documented.

 > In case you need details (or want to be sure that I remember correctly ;-))
 > I can dig through my project history...

It's probably enough to say that you will get unreliable behaviour trying 
to use Apple GDB from a frontend that is expecting the MI output of FSF GDB.
If your frontend must run on a Mac then you either need to compile FSF GDB
for the Mac (if that's possible) or read the Apple GDB source/manual and
adapt your frontend accordingly.

-- 
Nick                                           http://www.inet.net.nz/~nickrob


  reply	other threads:[~2007-08-28  8:56 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-08-24 17:51 Gordon Prieur
2007-08-24 18:02 ` Daniel Jacobowitz
2007-08-24 18:30   ` Jim Ingham
2007-08-24 18:43   ` Bob Rossi
2007-08-25  9:56     ` Eli Zaretskii
2007-08-25 10:19       ` Bob Rossi
2007-08-28  8:49     ` André Pönitz
2007-08-28  8:29 ` André Pönitz
2007-08-28  8:56   ` Nick Roberts [this message]
2007-08-28 10:55     ` Daniel Jacobowitz
2007-08-28 11:24       ` Bob Rossi
2007-08-28 11:33         ` Daniel Jacobowitz
2007-08-28 12:53           ` Bob Rossi
2007-08-28 13:13             ` Daniel Jacobowitz
2007-08-28 13:37               ` Bob Rossi
2007-08-28 13:45                 ` Daniel Jacobowitz
2007-08-28 13:54                   ` Bob Rossi
2007-08-28 15:05               ` Gordon Prieur
2007-08-28 11:54         ` Nick Roberts
2007-08-28 12:40           ` Bob Rossi
2007-08-29  5:03             ` Nick Roberts
2007-08-30 21:10           ` Mark Kettenis

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=18131.58128.433536.819432@kahikatea.snap.net.nz \
    --to=nickrob@snap.net.nz \
    --cc=apoenitz@trolltech.com \
    --cc=gdb@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