From: Eli Zaretskii <eliz@gnu.org>
To: 31312@debbugs.gnu.org, npostavs@gmail.com, rudalics@gmx.at
Cc: cyrus.and@gmail.com
Subject: bug#31312: Segmentation fault with doom-emacs, NeoTree and Zoom
Date: Wed, 02 May 2018 15:45:31 +0300 [thread overview]
Message-ID: <0DFE6C67-38C5-45EE-B658-8BEF80B595FF@gnu.org> (raw)
In-Reply-To: <87fu3ant8k.fsf@gmail.com>
On May 2, 2018 3:21:15 PM GMT+03:00, Noam Postavsky <npostavs@gmail.com> wrote:
> martin rudalics <rudalics@gmx.at> writes:
>
> > and the only ways this could cause w != XWINDOW(selected_window)
> > before the last call is either !WINDOWP (selected_window) or that
> > clear_garbaged_frames or prepare_menu_bars change selected_window.
> > Can you find the responsible?
>
> Yes, found it, seems to be a combination of zoom
> window-size-change-functions called via prepare_menu_bars and doom
> advice on balance-windows. I got 12 changes of selected_window.
> First
> hit summarized inline, the rest of the backtraces are attached below.
>
> #0 select_window_1 (window=XIL(0x1601c35), inhibit_point_swap=false)
> at ../../src/window.c:562
> #1 0x00000000004aff15 in select_window (window=XIL(0x1601c35),
> norecord=XIL(0xc090), inhibit_point_swap=false) at
> ../../src/window.c:520
> #2 0x00000000004b0183 in Fselect_window (window=XIL(0x1601c35),
> norecord=XIL(0xc090)) at ../../src/window.c:590
> #3 0x00000000004beaa7 in Fdelete_window_internal
> (window=XIL(0x1696c35)) at ../../src/window.c:4667
>
> #79 0x0000000000441b1a in safe_call1 (fn=XIL(0x4a9e6d0),
> arg=XIL(0x1600c35)) at ../../src/xdisp.c:2644
> #80 0x00000000004bb27a in run_window_size_change_functions
> (frame=XIL(0x1600c35)) at ../../src/window.c:3457
> #81 0x0000000000460409 in prepare_menu_bars () at
> ../../src/xdisp.c:12065
> #82 0x0000000000465141 in redisplay_internal () at
> ../../src/xdisp.c:13937
> #83 0x0000000000463c16 in redisplay () at ../../src/xdisp.c:13518
>
> Lisp Backtrace:
> [...]
> "delete-window" (0xffff9ab0)
> [...]
> "doom/popup-close" (0xffffa068)
> [...]
> "doom*popups-save" (0xffffaa00)
> [...]
> "ad-Advice-balance-windows" (0xffffb530)
> "apply" (0xffffb528)
> "balance-windows" (0xffffb960)
> "let" (0xffffbcc0)
> "zoom--update" (0xffffbe10)
> [...]
> "zoom--handler" (0xffffc928)
> "redisplay_internal (C function)" (0x0)
>
> (defun zoom--on ()
> [...]
> (add-hook 'window-size-change-functions #'zoom--handler)
>
> (defun zoom--handler (&optional window-or-frame norecord)
> [...]
> (with-selected-window
> [...]
> (zoom--update))))
>
> (defun zoom--update ()
> "Update the window layout in the current frame."
> (let (;; temporarily disables this mode during resize to avoid
> infinite
> ;; recursion (probably not needed thanks to the following)
> (zoom-mode nil)
> ;; temporarily disable all (even external) hooks about window
> ;; configuration changes to try to avoid potential flickering since
> ;; `balance-windows' calls them
> (window-configuration-change-hook nil)
> ;; make sure that other windows are resized nicely after resizing the
> ;; selected one
> (window-combination-resize t)
> ;; make sure that the exact same amount of pixels is assigned to all
> the
> ;; siblings
> (window-resize-pixelwise t))
> ;; start from a balanced layout anyway
> (balance-windows)
> ;; check if the selected window is not ignored
> (unless (zoom--window-ignored-p)
> (zoom--resize)
> (zoom--fix-scroll))))
>
> ;; doom/core/core-popups.el:
> (advice-add #'balance-windows :around #'doom*popups-save)
>
> (defun doom*popups-save (orig-fn &rest args)
> "Sets aside all popups before executing the original function, usually
> to
> prevent the popup(s) from messing up the UI (or vice versa)."
> (save-popups! (apply orig-fn args)))
>
> (defmacro save-popups! (&rest body)
> "Sets aside all popups before executing the original function, usually
> to
> prevent the popup(s) from messing up the UI (or vice versa)."
> `(let ((in-popup-p (doom-popup-p))
> (popups (doom-popup-windows))
> (doom-popup-remember-history t)
> (doom-popup-inhibit-autokill t))
> (when popups
> (mapc #'doom/popup-close popups))
> (unwind-protect
> (progn ,@body)
> (when popups
> (let ((origin (selected-window)))
> (doom/popup-restore)
> (unless in-popup-p
> (select-window origin)))))))
So we need the same defense after prepare_menu_bars as we
have after do_pending_changes, I think.
next prev parent reply other threads:[~2018-05-02 12:45 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-29 10:56 bug#31312: Segmentation fault with doom-emacs, NeoTree and Zoom Andrea Cardaci
2018-05-01 13:25 ` Noam Postavsky
2018-05-01 14:33 ` martin rudalics
2018-05-01 23:10 ` Noam Postavsky
2018-05-02 6:16 ` martin rudalics
2018-05-02 11:52 ` Eli Zaretskii
2018-05-02 12:21 ` Noam Postavsky
2018-05-02 12:45 ` Eli Zaretskii [this message]
2018-05-02 13:27 ` Andrea Cardaci
2018-05-02 13:47 ` martin rudalics
2018-05-02 15:06 ` Andrea Cardaci
2018-05-02 15:06 ` Eli Zaretskii
2018-05-02 15:14 ` Andrea Cardaci
2018-05-02 15:30 ` Eli Zaretskii
2018-05-02 13:42 ` martin rudalics
2018-05-02 15:03 ` Eli Zaretskii
2018-05-02 18:43 ` martin rudalics
2018-05-02 19:02 ` Eli Zaretskii
2018-05-03 7:11 ` martin rudalics
2018-05-03 0:04 ` Noam Postavsky
2018-05-03 2:32 ` Eli Zaretskii
2018-05-03 12:21 ` Noam Postavsky
2018-05-03 17:54 ` Eli Zaretskii
2018-05-04 1:18 ` Noam Postavsky
2018-05-02 13:42 ` martin rudalics
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=0DFE6C67-38C5-45EE-B658-8BEF80B595FF@gnu.org \
--to=eliz@gnu.org \
--cc=31312@debbugs.gnu.org \
--cc=cyrus.and@gmail.com \
--cc=npostavs@gmail.com \
--cc=rudalics@gmx.at \
/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).