unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: Pip Cet <pipcet@gmail.com>
To: Michael Heerdegen <michael_heerdegen@web.de>
Cc: Noam Postavsky <npostavs@gmail.com>, 36447@debbugs.gnu.org
Subject: bug#36447: 27.0.50; New "Unknown keyword" errors
Date: Fri, 5 Jul 2019 06:35:03 +0000	[thread overview]
Message-ID: <CAOqdjBe+Sn+u0_PNVePRvimG5tu0EBUVnhc=5_zF0FF_zdhD5Q@mail.gmail.com> (raw)
In-Reply-To: <87lfxdgs1k.fsf@web.de>

On Fri, Jul 5, 2019 at 1:59 AM Michael Heerdegen
<michael_heerdegen@web.de> wrote:
>
> Pip Cet <pipcet@gmail.com> writes:
>
> > As far as I can tell, this is a serious issue that will pop up
> > seemingly at random. We really should fix it.
>
> Yeah, it just happened again, when rebuilding a newer master version.
> And this time, cleaning and make bootstrap didn't fix it.  I still use
> the broken build FWIW as it's not as broken as the last time.

Is it still the same symptom, though?

> Anything I should do or try with it?

I'd like to verify it's indeed the bug I think it is. I think I could
figure that out if I had access to your Emacs binaries (emacs and
emacs.pdmp), but you can also debug things locally.

Can you do the following?

1. Run in gdb, with ASLR disabled
2. Set a breakpoint in print_vectorlike ("b print_vectorlike")
3. evaluate, in Emacs, (aref (aref (symbol-function
#'custom-handle-keyword) 2) 2)
4. wait for the breakpoint to be hit
5. in GDB, disable the breakpoint ("dis")
6. print and prettyprint the hash's next, hash, index, and
key_and_value vectors by typing, in gdb

p XHASH_TABLE(obj)->next
pp XHASH_TABLE(obj)->next
p XHASH_TABLE(obj)->hash
pp XHASH_TABLE(obj)->hash
p XHASH_TABLE(obj)->index
pp XHASH_TABLE(obj)->index
p XHASH_TABLE(obj)->key_and_value
pp XHASH_TABLE(obj)->key_and_value

(maybe you need to go up a stack frame or two so obj isn't optimized away).

7. Set a read/write breakpoint on the index table:

rwatch -l *(long *)&XHASH_TABLE(obj)->index
watch -l *(long *)XHASH_TABLE(obj)->index

8. Rerun emacs with gdb's "r" command, and find out where the
watchpoint is hit. In particular, if I'm right, it's accessed from two
different hash table objects.





  reply	other threads:[~2019-07-05  6:35 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-30 18:23 bug#36447: 27.0.50; New "Unknown keyword" errors Michael Heerdegen
2019-06-30 18:43 ` Eli Zaretskii
2019-06-30 21:44   ` Michael Heerdegen
2019-07-01 12:25     ` Noam Postavsky
2019-07-01 13:20       ` Pip Cet
2019-07-01 22:04       ` Michael Heerdegen
2019-07-02  1:59         ` Stefan Kangas
2019-07-02 14:17           ` Eli Zaretskii
2019-07-02 13:29         ` Pip Cet
2019-07-02 15:35           ` Michael Heerdegen
2019-07-02 16:20             ` Noam Postavsky
2019-07-02 22:50               ` Pip Cet
2019-07-03 11:57                 ` Pip Cet
2019-07-05  1:59                   ` Michael Heerdegen
2019-07-05  6:35                     ` Pip Cet [this message]
2019-07-05  7:50                       ` Eli Zaretskii
2019-07-05  8:12                         ` Pip Cet
2019-07-05  8:25                           ` Eli Zaretskii
2019-07-05  8:36                             ` Pip Cet
2019-07-05  8:41                               ` Eli Zaretskii
2019-07-05  9:09                                 ` Pip Cet
2019-07-05 12:23                                   ` Robert Pluim
2019-07-05 12:33                                   ` Eli Zaretskii
2019-07-05 13:41                                     ` Pip Cet
2019-07-05 18:00                                     ` Stefan Monnier
2019-07-05 18:07                                       ` Eli Zaretskii
2019-07-05 20:16                                         ` Stefan Monnier
2019-07-05 18:57                                       ` Pip Cet
2019-07-05 19:13                                         ` Eli Zaretskii
2019-07-05 20:21                                         ` Stefan Monnier
2019-07-05 21:52                                           ` Pip Cet
2019-07-05 22:10                                             ` Stefan Monnier
2019-07-06  6:45                                               ` Eli Zaretskii
2019-07-06 15:08                                                 ` Pip Cet
2019-07-09 21:05                                                   ` Stefan Monnier
2019-07-10  2:38                                                     ` Eli Zaretskii
2019-07-10  3:19                                                       ` Daniel Colascione
2019-07-10 15:01                                                         ` Pip Cet
2019-07-10 17:16                                                           ` Daniel Colascione
2019-07-10 20:14                                                             ` Pip Cet
2019-07-06 15:32                                             ` Michael Heerdegen
2019-07-08 17:30                                               ` Lars Ingebrigtsen
2019-07-08 17:58                                                 ` Pip Cet
2019-07-08 22:18                                                   ` Lars Ingebrigtsen
2019-07-08 22:25                                                     ` Noam Postavsky
2019-07-09 14:00                                                       ` Pip Cet
2019-07-10  3:01                                                         ` Daniel Colascione
2019-07-14 14:06                                                           ` Noam Postavsky
2019-07-08 23:22                                                     ` Stefan Monnier
2019-07-08 22:23                                                   ` Michael Heerdegen
2019-07-09 15:43                                                     ` Eli Zaretskii
2019-07-09 20:15                                                   ` Stefan Monnier
2019-07-05  7:55                       ` Katsumi Yamaoka

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

  List information: https://www.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAOqdjBe+Sn+u0_PNVePRvimG5tu0EBUVnhc=5_zF0FF_zdhD5Q@mail.gmail.com' \
    --to=pipcet@gmail.com \
    --cc=36447@debbugs.gnu.org \
    --cc=michael_heerdegen@web.de \
    --cc=npostavs@gmail.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.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

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).