public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Gareth Rees <grees@undo.io>
To: Andrew Burgess <aburgess@redhat.com>
Cc: Eli Zaretskii <eliz@gnu.org>, gdb-patches@sourceware.org
Subject: Re: [PATCH v5] [PR mi/29554] New PRINT-VALUES option '--scalar-values'.
Date: Thu, 16 Mar 2023 12:28:01 +0000	[thread overview]
Message-ID: <CAA8DSpGQ4terD-cEv+z0vtyF=PEyqci-D_a+fu8QCXs4TmrSqw@mail.gmail.com> (raw)
In-Reply-To: <87v8j4fu48.fsf@redhat.com>

Andrew Burgess wrote:
> If we really feel that the old behaviour is worth saving then adding a
> new flag isn't the end of the world.  I'd just want to make sure we
> really have considered the alternatives first.

I agree with Andrew's analysis. Another consideration is that it is
likely that existing GDB/MI clients that use --simple-values are subject
to latent bugs or performance issues that could be fixed by changing the
behaviour of the option.

When implementing a GDB/MI client for an IDE, you typically want to
update the displayed call stack every time the debuggee stops, and the
most convenient way to do this is to use the -stack-list-arguments
command. Since it is helpful for the user to be able to see the type of
each argument, it is natural to use the --simple-values option, as
otherwise you have to loop over the arguments calling -var-create.

If you use -stack-list-arguments with --simple-values, it can be hard to
spot that you are introducing bugs or performance issues when debugging
C++ programs with functions taking references to large data structures,
and it is easy to omit this case in testing. Visual Studio Code's
initial implementation of their GDB/MI client had exactly such an issue.

Searching GitHub, I found a couple of other bugs that might take this
form, in particular [1] and [2] are hangs in gdbgui whose descriptions
suggest that they might be related to the handling of reference types by
the --simple-values option.

[1] https://github.com/cs01/gdbgui/issues/205
[2] https://github.com/cs01/gdbgui/issues/206

  reply	other threads:[~2023-03-16 12:28 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-08  7:18 [PATCH] [gdb/mi] Don't treat references to compound values as "simple" Gareth Rees
2022-09-08 10:27 ` Andrew Burgess
2022-09-08 11:02   ` [PATCH v2] " Gareth Rees
2022-09-08 13:30     ` Eli Zaretskii
2022-09-08 13:58       ` Gareth Rees
2022-09-08 14:07         ` Eli Zaretskii
2022-09-09  8:01       ` [PATCH v3] [PR mi/29554] New PRINT-VALUES option '--scalar-values' Gareth Rees
2022-09-15  9:06         ` [PING] " Gareth Rees
2022-09-25  8:15         ` Gareth Rees
2022-09-25  8:25           ` Eli Zaretskii
2022-09-25  9:00             ` Gareth Rees
2022-09-25 10:16               ` Eli Zaretskii
2022-09-26 12:48                 ` Gareth Rees
2022-09-25 10:16           ` Eli Zaretskii
2022-09-26 12:46         ` [PATCH v4] " Gareth Rees
2022-10-04  9:08           ` [PING] " Gareth Rees
2022-10-18 11:59             ` Gareth Rees
2022-10-12 16:38           ` Andrew Burgess
2022-10-20 17:47             ` [PATCH v5] " Gareth Rees
2022-10-20 18:00               ` Eli Zaretskii
2022-11-03 16:20               ` [PING] " Gareth Rees
2022-11-14  9:25                 ` Gareth Rees
2022-12-01 13:41                 ` Gareth Rees
2022-12-14  8:50                 ` Gareth Rees
2023-02-01 10:00                 ` Gareth Rees
2023-02-16 10:08                 ` Gareth Rees
2023-03-06  9:52                 ` Gareth Rees
2023-03-08 12:35               ` Andrew Burgess
2023-03-10 11:04                 ` Gareth Rees
2023-03-10 12:05                   ` Eli Zaretskii
2023-03-10 12:58                     ` Gareth Rees
2023-03-13 17:17                     ` Andrew Burgess
2023-03-16 12:28                       ` Gareth Rees [this message]
2023-03-11 11:58                   ` Gareth Rees
2023-04-11 13:15                     ` Pedro Alves
2023-03-11 11:49               ` [PATCH v6] [gdb/mi] Don't treat references to compound values as "simple" Gareth Rees
2023-03-21  9:50                 ` [PING] " Gareth Rees
2023-03-26  9:56                   ` Gareth Rees
2023-04-03  9:22                     ` Gareth Rees
2023-05-04 15:08                       ` Tom Tromey
2023-04-18  9:23                   ` Gareth Rees
2023-04-24  9:53                   ` Gareth Rees
2023-05-02  9:13                   ` Gareth Rees
2023-03-27 14:34                 ` Tom Tromey
2023-03-29  9:14                   ` Gareth Rees
2023-04-06 17:18                   ` Gareth Rees
2022-10-20 17:58             ` [PATCH v4] [PR mi/29554] New PRINT-VALUES option '--scalar-values' Gareth Rees
2022-09-09  8:04       ` [PATCH v2] [gdb/mi] Don't treat references to compound values as "simple" Gareth Rees
2022-09-08 11:09   ` [PATCH] " Gareth Rees

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='CAA8DSpGQ4terD-cEv+z0vtyF=PEyqci-D_a+fu8QCXs4TmrSqw@mail.gmail.com' \
    --to=grees@undo.io \
    --cc=aburgess@redhat.com \
    --cc=eliz@gnu.org \
    --cc=gdb-patches@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).