Mirror of the gdb mailing list
 help / color / mirror / Atom feed
From: Kyle Huey <me@kylehuey.com>
To: gdb@sourceware.org
Cc: rr-dev <rr-dev@mozilla.org>, Pedro Alves <palves@redhat.com>
Subject: [regression] gdb 7.11 breaks rr
Date: Mon, 16 May 2016 22:06:00 -0000	[thread overview]
Message-ID: <CAP045Ao2usw07e0iUJ7jfYp8EHCBKBTw6UvGZFVDrD-0Xq54JA@mail.gmail.com> (raw)

(Resending because the gdb list rejects HTML mail, sorry for duplicates)

Steps to reproduce:

1. Clone rr from https://github.com/mozilla/rr.git
2. Build rr rev 8566e3038892
3. ./bin/rr record ./bin/reverse_continue_process_signal
4. ./bin/rr replay -d /path/to/gdb-7.11
5. c
6. Once, the program has reached the SIGKILL, b 22
7. rc

Instead of reversing to the fork on line 22, rr will stop again at the
SIGKILL.

Beginning with [0] gdb seems to forget which direction execution is heading
in when a signal is reported by the gdb server.  When gdb resumes execution
after deciding to ignore the signal it sends a vCont instead of a bc to the
gdb server, causing rr to execute forwards again.

This was fixed after the release of gdb 7.11 by [1]. I verified that
cherry-picking this commit on top of the gdb-7.11-release tag also fixes
rr.  Could we get this commit backported for a 7.11.1?  rr's reverse
execution features are largely unusable with complex programs (e.g.
Firefox) because of this bug.

Thanks,

- Kyle

[0]
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commit;h=0b333c5e7d6c3fc65d37ffa11bd21ba52c4adb25
[1]
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commit;h=3a00c80277a54abe0b286a6e8babc8fe50120205


             reply	other threads:[~2016-05-16 22:06 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-16 22:06 Kyle Huey [this message]
2016-05-17 16:27 ` Pedro Alves
2016-05-17 18:02   ` Kyle Huey

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=CAP045Ao2usw07e0iUJ7jfYp8EHCBKBTw6UvGZFVDrD-0Xq54JA@mail.gmail.com \
    --to=me@kylehuey.com \
    --cc=gdb@sourceware.org \
    --cc=palves@redhat.com \
    --cc=rr-dev@mozilla.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