public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Jan Vrany <jan.vrany@labware.com>
To: gdb-patches@sourceware.org
Subject: Re: [PATCH v3 0/1] PR20684, preserve user selected context when invoking MI commands
Date: Mon, 07 Feb 2022 11:56:04 +0000	[thread overview]
Message-ID: <b9994552adfd82df10c1a00b62a9400999ad78c0.camel@labware.com> (raw)
In-Reply-To: <20220127145011.1153142-1-jan.vrany@labware.com>

Polite ping. 

Jan

On Thu, 2022-01-27 at 14:50 +0000, Jan Vrany wrote:
> This is a next version of the patch fixing some regression
> 
> Changes since previous iteration [1]:
> 
> * updates to accomodate new C++ification of MI commands pushed in
> December last year.
> * updates to handle cases pointed out by Simon
> * add testcase
> 
> Changes since v1:
> 
> * address Simon's comments [2]
> * add more tests for -stack-select-frame
> 
> Changes since v2:
> 
> Fix regression in mi-nonstop.exp and mi-nsthrexec.exp. This
> was caused by extra user selected thread notification being sent
> when resuming a thread like:
> 
>  -exec-continue --thread 2
> 
> The culprit was this part of mi_excecute_command ():
> 
>          else if (inferior_ptid != null_ptid)
>            {
>              struct thread_info *ti = inferior_thread ();
> 
>              report_change = (ti->global_num != command->thread);
>            }
> 
> This "else" is only executed when command specifies --thread.
> 
> Imagine currently selected thread is thread 1 and -exec-continue --thread 2
> is issued.
> 
> BEFORE applying this patch, no change would be reported because mi_cmd_execute ()
> would switch the thread to thread 2 so `ti->global_num != command->thread` would
> be false.
> 
> AFTER appling this patch, mi_cmd_execute () would switch the thread *and then restore*
> back so `ti->global_num != command->thread` would be true.
> 
> Therefore I removed this whole `else if` branch: since mi_cmd_execute () now restores
> the thread if --thread is passed (except few limited commands that notify themselves),
> this condition would hold each time thread passed to --thread is different than current
> thread. Removing it preserves the behavior.
> 
> Removing that "else if" branch allowed me to merge two nested ifs into one to (hopefully)
> make the code simpler.
> 
> Thanks!
> 
> Jan
> 
> 
> [1]: https://sourceware.org/pipermail/gdb-patches/2021-February/175897.html
> 
> 
> Jan Vrany (1):
>   gdb/mi: PR20684, preserve user selected thread and frame when invoking
>     MI commands
> 
>  gdb/mi/mi-cmds.h                             |  12 ++
>  gdb/mi/mi-main.c                             |  54 ++++---
>  gdb/testsuite/gdb.mi/mi-cmd-user-context.exp | 155 +++++++++++++++++++
>  3 files changed, 198 insertions(+), 23 deletions(-)
>  create mode 100644 gdb/testsuite/gdb.mi/mi-cmd-user-context.exp
> 



  reply	other threads:[~2022-02-07 11:56 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-25 14:51 [PATCH " Jan Vrany
2022-01-25 14:51 ` [PATCH 1/1] gdb/mi: PR20684, preserve user selected thread and frame " Jan Vrany
2022-01-25 18:59   ` Simon Marchi
2022-01-26 13:21     ` [PATCH v2 0/1] PR20684, preserve user selected context " Jan Vrany
2022-01-26 13:21     ` [PATCH v2 1/1] gdb/mi: PR20684, preserve user selected thread and frame " Jan Vrany
2022-01-26 15:01       ` Simon Marchi
2022-01-26 17:08         ` Jan Vrany
2022-01-27 14:50         ` [PATCH v3 0/1] PR20684, preserve user selected context " Jan Vrany
2022-02-07 11:56           ` Jan Vrany [this message]
2022-02-17 16:07             ` [PING] " Jan Vrany
2022-01-27 14:50         ` [PATCH v3 1/1] gdb/mi: PR20684, preserve user selected thread and frame " Jan Vrany
2022-02-17 22:24           ` Andrew Burgess
2022-02-18 17:45             ` Jan Vrany
2022-03-01 11:59               ` [PING] " Jan Vrany
2022-03-02 10:00               ` Andrew Burgess
2022-03-02 13:23                 ` [PATCH v4] " Jan Vrany
2022-03-08 16:58                   ` Andrew Burgess

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=b9994552adfd82df10c1a00b62a9400999ad78c0.camel@labware.com \
    --to=jan.vrany@labware.com \
    --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).