bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

I've had a number of crashes today when running exchange-point-and-mark
on editing a particular file shortly after pasting in lots of lines from another
(v large) buffer. Probably replicable but atm hard to give an exact
recipe.

I'll keep the gdb session around in case it is needed.

Looking at gdb it appears to crash in delete_frame??


Program received signal SIGSEGV, Segmentation fault.
delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
2064              struct frame *f1 = XFRAME (frame1);
(gdb) bt full
#0  delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
        f = 0x1317ef0
        sf = 0x1317ef0
        kb = <optimized out>
        frames = XIL(0)
        frame1 = <optimized out>
        minibuffer_selected = <optimized out>
        is_tooltip_frame = <optimized out>
        nochild = <optimized out>
#1  0x00000000004bbdbb in x_connection_closed (dpy=dpy@entry=0x15e9000, error_message=<optimized out>, 
    error_message@entry=0x7fffffffcdd0 "Connection lost to X server ':0'", ioerror=ioerror@entry=true) at xterm.c:9791
        tail = XIL(0x1023733)
#2  0x00000000004bbe76 in x_io_error_quitter (display=0x15e9000) at xterm.c:9915
        buf = "Connection lost to X server ':0'\000\316\377\377\377\177\000\000P\316\377\377\377\177\000\000\000\220^\001\000\000\000\000\276-q\364\377\177\000\000P\242^\001\000\000\000\000\001\000\000\000\000\000\000\000P\367^\001\000\000\000\000\240\330\240\364\377\177\000\000\000\000\000\000\000\000\000\000\001\000\000\000\000\000\000\000p\351\261\002\000\000\000\000\213\000@\005\000\000\000\000\000\220^\001\000\000\000\000\367\237\336\367\377\177\000\000\005", '\000' <repeats 15 times>, "\340i\033\005\000\000\000\000\000}m\364\377\177\000\000\000\220^\001\000\000\000\000\000\220^\001\000\000\000\000^\357y\000\000\000\000\000\025"...
#3  0x00007ffff47154fe in _XIOError () from /usr/lib/x86_64-linux-gnu/libX11.so.6
No symbol table info available.
#4  0x00007ffff4712e5d in _XEventsQueued () from /usr/lib/x86_64-linux-gnu/libX11.so.6
No symbol table info available.
#5  0x00007ffff46f465a in XFlush () from /usr/lib/x86_64-linux-gnu/libX11.so.6
No symbol table info available.
#6  0x00000000004d732c in x_reply_selection_request (dpyinfo=<optimized out>, event=<optimized out>) at xselect.c:694
        reply_base = {
          type = 31, 
          xany = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523
          }, 
          xkey = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            root = 1, 
            subwindow = 324, 
            time = 401, 
            x = 122306683, 
            y = 0, 
---Type <return> to continue, or q <return> to quit---
            x_root = 0, 
            y_root = 500000, 
            state = 0, 
            keycode = 32767, 
            same_screen = 5177591
          }, 
          xbutton = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            root = 1, 
            subwindow = 324, 
            time = 401, 
            x = 122306683, 
            y = 0, 
            x_root = 0, 
            y_root = 500000, 
            state = 0, 
            button = 32767, 
            same_screen = 5177591
          }, 
          xmotion = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            root = 1, 
            subwindow = 324, 
            time = 401, 
            x = 122306683, 
            y = 0, 
            x_root = 0, 
            y_root = 500000, 
            state = 0, 
            is_hint = -1 '\377', 
            same_screen = 5177591
          }, 
          xcrossing = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
---Type <return> to continue, or q <return> to quit---
            root = 1, 
            subwindow = 324, 
            time = 401, 
            x = 122306683, 
            y = 0, 
            x_root = 0, 
            y_root = 500000, 
            mode = 0, 
            detail = 32767, 
            same_screen = 5177591, 
            focus = 0, 
            state = 0
          }, 
          xfocus = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            mode = 1, 
            detail = 0
          }, 
          xexpose = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            x = 1, 
            y = 0, 
            width = 324, 
            height = 0, 
            count = 401
          }, 
          xgraphicsexpose = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            drawable = 88080523, 
            x = 1, 
            y = 0, 
            width = 324, 
            height = 0, 
            count = 401, 
            major_code = 0, 
---Type <return> to continue, or q <return> to quit---
            minor_code = 122306683
          }, 
          xnoexpose = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            drawable = 88080523, 
            major_code = 1, 
            minor_code = 0
          }, 
          xvisibility = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            state = 1
          }, 
          xcreatewindow = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            parent = 88080523, 
            window = 1, 
            x = 324, 
            y = 0, 
            width = 401, 
            height = 0, 
            border_width = 122306683, 
            override_redirect = 0
          }, 
          xdestroywindow = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            event = 88080523, 
            window = 1
          }, 
          xunmap = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
---Type <return> to continue, or q <return> to quit---
            event = 88080523, 
            window = 1, 
            from_configure = 324
          }, 
          xmap = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            event = 88080523, 
            window = 1, 
            override_redirect = 324
          }, 
          xmaprequest = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            parent = 88080523, 
            window = 1
          }, 
          xreparent = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            event = 88080523, 
            window = 1, 
            parent = 324, 
            x = 401, 
            y = 0, 
            override_redirect = 122306683
          }, 
          xconfigure = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            event = 88080523, 
            window = 1, 
            x = 324, 
            y = 0, 
            width = 401, 
            height = 0, 
            border_width = 122306683, 
            above = 2147483648000000, 
---Type <return> to continue, or q <return> to quit---
            override_redirect = 0
          }, 
          xgravity = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            event = 88080523, 
            window = 1, 
            x = 324, 
            y = 0
          }, 
          xresizerequest = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            width = 1, 
            height = 0
          }, 
          xconfigurerequest = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            parent = 88080523, 
            window = 1, 
            x = 324, 
            y = 0, 
            width = 401, 
            height = 0, 
            border_width = 122306683, 
            above = 2147483648000000, 
            detail = 0, 
            value_mask = 5177591
          }, 
          xcirculate = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            event = 88080523, 
            window = 1, 
            place = 324
          }, 
---Type <return> to continue, or q <return> to quit---
          xcirculaterequest = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            parent = 88080523, 
            window = 1, 
            place = 324
          }, 
          xproperty = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            atom = 1, 
            time = 324, 
            state = 401
          }, 
          xselectionclear = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            selection = 1, 
            time = 324
          }, 
          xselectionrequest = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            owner = 88080523, 
            requestor = 1, 
            selection = 324, 
            target = 401, 
            property = 122306683, 
            time = 2147483648000000
          }, 
          xselection = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            requestor = 88080523, 
---Type <return> to continue, or q <return> to quit---
            selection = 1, 
            target = 324, 
            property = 401, 
            time = 122306683
          }, 
          xcolormap = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            colormap = 1, 
            new = 324, 
            state = 0
          }, 
          xclient = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            message_type = 1, 
            format = 324, 
            data = {
              b = "\221\001\000\000\000\000\000\000{@J\a\000\000\000\000\000\000\000", 
              s = {401, 0, 0, 0, 16507, 1866, 0, 0, 0, 0}, 
              l = {401, 122306683, 2147483648000000, 140733193388032, 5177591}
            }
          }, 
          xmapping = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            request = 1, 
            first_keycode = 0, 
            count = 324
          }, 
          xerror = {
            type = 31, 
            display = 0x57c7ea3, 
            resourceid = 140733210152784, 
            serial = 22974464, 
            error_code = 139 '\213', 
            request_code = 0 '\000', 
---Type <return> to continue, or q <return> to quit---
            minor_code = 64 '@'
          }, 
          xkeymap = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            window = 88080523, 
            key_vector = "\001\000\000\000\000\000\000\000D\001\000\000\000\000\000\000\221\001\000\000\000\000\000\000{@J\a\000\000\000"
          }, 
          xgeneric = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            extension = 88080523, 
            evtype = 0
          }, 
          xcookie = {
            type = 31, 
            serial = 92044963, 
            send_event = 16764752, 
            display = 0x15e9000, 
            extension = 88080523, 
            evtype = 0, 
            cookie = 1, 
            data = 0x144
          }, 
          pad = {31, 92044963, 140733210152784, 22974464, 88080523, 1, 324, 401, 122306683, 2147483648000000, 
            140733193388032, 5177591, 0, 2147483648000000, 140733193388032, 0, 97414500, 5190006, 0, 0, 48139624, 
            1513004943, 92044899, 92044979}
        }
        window = 88080523
#7  x_handle_selection_request (event=<optimized out>) at xselect.c:832
        selection = 262040
        local_selection_time = 0
        local_selection_data = XIL(0)
        dpyinfo = 0x79ef5e
        property = 22974464
        success = false
#8  x_handle_selection_event (event=0x2b1e970) at xselect.c:963
No locals.
#9  0x00000000004f36e2 in process_special_events () at keyboard.c:4246
        copy = {
          kind = SELECTION_REQUEST_EVENT, 
---Type <return> to continue, or q <return> to quit---
          dpyinfo = 0x16632e0, 
          requestor = 88080523, 
          selection = 1, 
          target = 324, 
          property = 401, 
          time = 122306683
        }
        beg = 0xb87e30 <kbd_buffer+14896>
        event = 0xb87e30 <kbd_buffer+14896>
#10 0x00000000004f40bd in swallow_events (do_display=<optimized out>) at keyboard.c:4264
        old_timers_run = <optimized out>
#11 0x00000000005a3e20 in wait_reading_process_output (time_limit=1, nsecs=0, read_kbd=0, do_display=false, 
    wait_for_cell=XIL(0x1317e00), wait_proc=0x2a8f4b0, just_wait_proc=0) at process.c:5510
        old_timers_run = 617
        old_buffer = 0x2a8f4b0
        old_window = XIL(0x5cf3845)
        leave = true
        process_skipped = true
        nfds = 1
        Available = {
          fds_bits = {64, 0 <repeats 15 times>}
        }
        Writeok = {
          fds_bits = {0 <repeats 16 times>}
        }
        check_write = false
        check_delay = 1
        no_avail = false
        xerrno = 11
        timeout = {
          tv_sec = 0, 
          tv_nsec = 0
        }
        end_time = {
          tv_sec = 0, 
          tv_nsec = 0
        }
#12 0x0000000000421112 in sit_for (timeout=make_number(50), reading=253, display_option=0) at dispnew.c:5793
        sec = 202
        nsec = 0
#13 0x00000000004f8c15 in read_char (commandflag=commandflag@entry=1, map=XIL(0xfffffffffffffffd), 
    map@entry=XIL(0x10249e3), prev_event=XIL(0), used_mouse_menu=0x0, used_mouse_menu@entry=0x7fffffffdb6b, 
    end_time=0x1317e00, end_time@entry=0x0) at keyboard.c:2717
        tem0 = XIL(0)
        buffer_size = 0
        local_getcjmp = {{
---Type <return> to continue, or q <return> to quit---
            __jmpbuf = {23476993, -5963633980627533669, 3, 23477216, 140737488346176, 0, 5963636584858206363, 
              -5963636162151628645}, 
            __mask_was_saved = 0, 
            __saved_mask = {
              __val = {44627125, 56832, 101300596, 19429125, 16741059, 19429125, 18249683, 19622325, 101300596, 23072208, 
                4217902893331710400, 9448903869477172611, 3855962269045293109, 15062352596681216, 3855962132650836675, 
                14394386273282803212}
            }
          }}
        save_jump = {{
            __jmpbuf = {0, 0, 0, 0, 0, 0, 0, 0}, 
            __mask_was_saved = 0, 
            __saved_mask = {
              __val = {0 <repeats 16 times>}
            }
          }}
        save = XIL(0x1663be0)
        previous_echo_area_message = XIL(0)
        also_record = XIL(0)
        reread = false
        recorded = false
        polling_stopped_here = false
#14 0x00000000004f99ca in read_key_sequence (keybuf=keybuf@entry=0x7fffffffdc40, prompt=prompt@entry=XIL(0), 
    dont_downcase_last=dont_downcase_last@entry=false, can_return_switch_frame=can_return_switch_frame@entry=true, 
    fix_current_buffer=fix_current_buffer@entry=true, prevent_redisplay=prevent_redisplay@entry=false, bufsize=30)
    at keyboard.c:9150
        interrupted_kboard = 0x1663be0
        interrupted_frame = 0x1317ef0
        key = <optimized out>
        used_mouse_menu = false
        echo_local_start = 0
        last_real_key_start = <optimized out>
        keys_local_start = <optimized out>
        new_binding = <optimized out>
        t = <optimized out>
        echo_start = 0
        keys_start = 0
        current_binding = XIL(0x10249e3)
        first_event = XIL(0)
        first_unbound = 31
        mock_input = 0
        fkey = {
          parent = XIL(0x1034243), 
          map = XIL(0x1034243), 
          start = 0, 
          end = 0
---Type <return> to continue, or q <return> to quit---
        }
        keytran = {
          parent = XIL(0xc31a93), 
          map = XIL(0xc31a93), 
          start = 0, 
          end = 0
        }
        indec = {
          parent = XIL(0x1034253), 
          map = XIL(0x1034253), 
          start = 0, 
          end = 0
        }
        shift_translated = false
        delayed_switch_frame = XIL(0)
        original_uppercase = XIL(0x3ee8)
        original_uppercase_position = -1
        dummyflag = false
        starting_buffer = 0x2a8f4b0
        fake_prefixed_keys = XIL(0)
#15 0x00000000004fb4a6 in command_loop_1 () at keyboard.c:1368
        cmd = <optimized out>
        keybuf = {make_number(24), make_number(24), make_number(98), XIL(0x4ef405), XIL(0), XIL(0x3), XIL(0x3), 
          make_number(1410489), XIL(0x5ce320), XIL(0x3464ea3), XIL(0x7fffffffdd00), XIL(0x3464ea3), XIL(0xffffffff), 
          XIL(0x563614), XIL(0x69320), XIL(0x3464ea3), XIL(0x85ed64), XIL(0x5ce320), XIL(0x3464ea3), make_number(1296728), 
          XIL(0x7fffffffdd00), XIL(0), XIL(0), XIL(0x4f2697), XIL(0xfd3100), make_number(1000), XIL(0xffffffff), 
          make_number(1409433), XIL(0x5), XIL(0x7a10)}
        i = <optimized out>
        prev_modiff = 3
        prev_buffer = 0x2a8f4b0
#16 0x000000000055fe67 in internal_condition_case (bfun=bfun@entry=0x4fb2a0 <command_loop_1>, 
    handlers=handlers@entry=XIL(0x5250), hfun=hfun@entry=0x4f2590 <cmd_error>) at eval.c:1332
        val = XIL(0)
        c = 0x15cf9d0
#17 0x00000000004ed3ac in command_loop_2 (ignore=ignore@entry=XIL(0)) at keyboard.c:1110
        val = XIL(0)
#18 0x000000000055fdf4 in internal_catch (tag=tag@entry=XIL(0xc720), func=func@entry=0x4ed390 <command_loop_2>, 
    arg=arg@entry=XIL(0)) at eval.c:1097
        val = XIL(0)
        c = 0x15cf8b0
#19 0x00000000004ed369 in command_loop () at keyboard.c:1089
No locals.
#20 0x00000000004f2196 in recursive_edit_1 () at keyboard.c:695
        val = <optimized out>
#21 0x00000000004f24b3 in Frecursive_edit () at keyboard.c:766
        buffer = <optimized out>
---Type <return> to continue, or q <return> to quit---
#22 0x00000000004177de in main (argc=1, argv=0x7fffffffdfb8) at emacs.c:1711
        stack_bottom_variable = 0x7fffffffde60
        skip_args = 0
        junk = 0x0
        dname_arg = 0x0
        ch_to_dir = 0x0
        rlim = {
          rlim_cur = 10022912, 
          rlim_max = 18446744073709551615
        }
(gdb) xbacktrace
(gdb)


In GNU Emacs 27.0.50 (build 1, x86_64-pc-linux-gnu, GTK+ Version 2.24.25)
 of 2017-11-14 built on ct-lt-579
Repository revision: 13248f7444630508cfc3b78a07e8d96613af11c8
Windowing system distributor 'The X.Org Foundation', version 11.0.11604000
System Description:	Debian GNU/Linux 8.9 (jessie)


Configured features:
XPM JPEG TIFF GIF PNG RSVG SOUND DBUS GSETTINGS NOTIFY GNUTLS LIBXML2
FREETYPE XFT ZLIB TOOLKIT_SCROLL_BARS GTK2 X11

Important settings:
  value of $LANG: en_GB.UTF-8
  locale-coding-system: utf-8-unix

Major mode: Help

Minor modes in effect:
  shell-dirtrack-mode: t
  tooltip-mode: t
  global-eldoc-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  buffer-read-only: t
  line-number-mode: t
  transient-mark-mode: t

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

 > Program received signal SIGSEGV, Segmentation fault.
 > delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
 > 2064              struct frame *f1 = XFRAME (frame1);

So it seems we always have to check Vframe_list for emptyness before
running FOR_EACH_FRAME.  Funny this was never seen before ...

martin

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

martin rudalics <rudalics@gmx.at> writes:

>> Program received signal SIGSEGV, Segmentation fault.
>> delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
>> 2064              struct frame *f1 = XFRAME (frame1);
>
> So it seems we always have to check Vframe_list for emptyness before
> running FOR_EACH_FRAME.  Funny this was never seen before ...
>
I can also replicate this crash  on version 24.4.1 - but maybe you're
saying it's been a long term problem.

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

 > I can also replicate this crash  on version 24.4.1 - but maybe you're
 > saying it's been a long term problem.

Great (you earlier only said that it's "probably replicable").  Could
you check that Vframe_list is empty at the time FOR_EACH_FRAME is run
somewhere around line 2064 (or 2067 on current master, as I believe).

Thanks, martin

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

martin rudalics <rudalics@gmx.at> writes:

>> I can also replicate this crash  on version 24.4.1 - but maybe you're
>> saying it's been a long term problem.
>
> Great (you earlier only said that it's "probably replicable").  Could
> you check that Vframe_list is empty at the time FOR_EACH_FRAME is run
> somewhere around line 2064 (or 2067 on current master, as I believe).
>
> Thanks, martin

Breakpoint 3, delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2062
2062          FOR_EACH_FRAME (frames, frame1)
(gdb) p Vframe_list
$1 = XIL(0)
(gdb) c
Continuing.

Program received signal SIGSEGV, Segmentation fault.
delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
2064              struct frame *f1 = XFRAME (frame1);

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

[-- Attachment #1: Type: text/plain, Size: 565 bytes --]

 > Breakpoint 3, delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2062
 > 2062          FOR_EACH_FRAME (frames, frame1)
 > (gdb) p Vframe_list
 > $1 = XIL(0)
 > (gdb) c
 > Continuing.
 >
 > Program received signal SIGSEGV, Segmentation fault.
 > delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
 > 2064              struct frame *f1 = XFRAME (frame1);

Thanks.  The attached patch should fix that.  Please check.

But I have no idea how to proceed with the x_reply_selection_request
failure.

martin

[-- Attachment #2: frame.c.diff --]
[-- Type: text/plain, Size: 1394 bytes --]

diff --git a/src/frame.c b/src/frame.c
index 5bafbed..66d1b5c 100644
--- a/src/frame.c
+++ b/src/frame.c
@@ -2058,7 +2058,7 @@ of them (the selected terminal frame) is actually displayed.

   /* If we've deleted the last_nonminibuf_frame, then try to find
      another one.  */
-  if (f == last_nonminibuf_frame)
+  if (f == last_nonminibuf_frame && !NILP (Vframe_list))
     {
       last_nonminibuf_frame = 0;

@@ -2076,7 +2076,7 @@ of them (the selected terminal frame) is actually displayed.

   /* If there's no other frame on the same kboard, get out of
      single-kboard state if we're in it for this kboard.  */
-  if (kb != NULL)
+  if (kb != NULL && !NILP (Vframe_list))
     {
       /* Some frame we found on the same kboard, or nil if there are none.  */
       Lisp_Object frame_on_same_kboard = Qnil;
@@ -2093,7 +2093,9 @@ of them (the selected terminal frame) is actually displayed.
   /* If we've deleted this keyboard's default_minibuffer_frame, try to
      find another one.  Prefer minibuffer-only frames, but also notice
      frames with other windows.  */
-  if (kb != NULL && EQ (frame, KVAR (kb, Vdefault_minibuffer_frame)))
+  if (kb != NULL
+      && EQ (frame, KVAR (kb, Vdefault_minibuffer_frame))
+      && !NILP (Vframe_list))
     {
       /* The last frame we saw with a minibuffer, minibuffer-only or not.  */
       Lisp_Object frame_with_minibuf = Qnil;

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

martin rudalics <rudalics@gmx.at> writes:

>> Breakpoint 3, delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2062
>> 2062          FOR_EACH_FRAME (frames, frame1)
>> (gdb) p Vframe_list
>> $1 = XIL(0)
>> (gdb) c
>> Continuing.
>>
>> Program received signal SIGSEGV, Segmentation fault.
>> delete_frame (frame=<optimized out>, force=force@entry=XIL(0x98a0)) at frame.c:2064
>> 2064              struct frame *f1 = XFRAME (frame1);
>
> Thanks.  The attached patch should fix that.  Please check.
>
> But I have no idea how to proceed with the x_reply_selection_request
> failure.
>
> martin
>
> diff --git a/src/frame.c b/src/frame.c
> index 5bafbed..66d1b5c 100644
> --- a/src/frame.c
> +++ b/src/frame.c
> @@ -2058,7 +2058,7 @@ of them (the selected terminal frame) is actually displayed.
>
>    /* If we've deleted the last_nonminibuf_frame, then try to find
>       another one.  */
> -  if (f == last_nonminibuf_frame)
> +  if (f == last_nonminibuf_frame && !NILP (Vframe_list))
>      {
>        last_nonminibuf_frame = 0;
>
> @@ -2076,7 +2076,7 @@ of them (the selected terminal frame) is actually displayed.
>
>    /* If there's no other frame on the same kboard, get out of
>       single-kboard state if we're in it for this kboard.  */
> -  if (kb != NULL)
> +  if (kb != NULL && !NILP (Vframe_list))
>      {
>        /* Some frame we found on the same kboard, or nil if there are none.  */
>        Lisp_Object frame_on_same_kboard = Qnil;
> @@ -2093,7 +2093,9 @@ of them (the selected terminal frame) is actually displayed.
>    /* If we've deleted this keyboard's default_minibuffer_frame, try to
>       find another one.  Prefer minibuffer-only frames, but also notice
>       frames with other windows.  */
> -  if (kb != NULL && EQ (frame, KVAR (kb, Vdefault_minibuffer_frame)))
> +  if (kb != NULL
> +      && EQ (frame, KVAR (kb, Vdefault_minibuffer_frame))
> +      && !NILP (Vframe_list))
>      {
>        /* The last frame we saw with a minibuffer, minibuffer-only or not.  */
>        Lisp_Object frame_with_minibuf = Qnil;

Thanks, it now no longer crashes there, but crashes elsewhere but looks
to about to terminate in a more controlled manner?

#1  check_glyph_memory () at dispnew.c:2264
        tail = XIL(0)
#2  0x00000000004ecead in shut_down_emacs (sig=sig@entry=0, stuff=stuff@entry=XIL(0)) at emacs.c:2107
No locals.
#3  0x00000000004bbdeb in x_connection_closed (dpy=dpy@entry=0x15e5800, error_message=<optimized out>, 
    error_message@entry=0x7fffffffcda0 "Connection lost to X server ':0'", ioerror=ioerror@entry=true) at xterm.c:9806
        tail = <optimized out>
#4  0x00000000004bbe96 in x_io_error_quitter (display=0x15e5800) at xterm.c:9915
        buf = "Connection lost to X server ':0'\000\315\377\377\377\177\000\000 \316\377\377\377\177\000\000\000X^\001\000\000\000\000\276-q\364\377\177\000\000Pj^\001\000\000\000\000\001\000\000\000\000\000\000\000\360\274^\001\000\000\000\000\240\330\240\364\377\177\000\000\000\000\000\000\000\000\000\000\001\000\000\000\000\000\000\000\300{\355\000\000\000\000\000\363\000@\005\000\000\000\000\000X^\001\000\000\000\000\367\237\336\367\377\177\000\000\005", '\000' <repeats 15 times>, "\340\177\373\000\000\000\000\000\000}m\364\377\177\000\000\000X^\001\000\000\000\000\000X^\001\000\000\000\000t\031@\002\000\000\000\000\025"...
#5  0x00007ffff47154fe in _XIOError () from /usr/lib/x86_64-linux-gnu/libX11.so.6
No symbol table info available.
#6  0x00007ffff4712e5d in _XEventsQueued () from /usr/lib/x86_64-linux-gnu/libX11.so.6
No symbol table info available.
#7  0x00007ffff46f465a in XFlush () from /usr/lib/x86_64-linux-gnu/libX11.so.6
No symbol table info available.
#8  0x00000000004d734c in x_reply_selection_request (dpyinfo=<optimized out>, event=<optimized out>) at xselect.c:694
        reply_base = {
          type = 31, 
          xany = {
            type = 31, 
            serial = 16882611, 
            send_event = 2, 
            display = 0x15e5800, 
            window = 88080627
          }, 
          xkey = {
            type = 31, 
            serial = 16882611, 
            send_event = 2, 
            display = 0x15e5800, 
            window = 88080627, 
            root = 1, 
            subwindow = 324, 
            time = 401, 
            x = 162228637, 
            y = 0, 
            x_root = 0, 
            y_root = 500000, 
            state = 0, 
etc 

I can reliably replicate this error by opening this large file,
searching for a string that only appears on the last line and
exchange-point-and-mark'ing

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

Robert Marshall <robert.marshall@codethink.co.uk> writes:

>
> I can reliably replicate this error by opening this large file,
> searching for a string that only appears on the last line and
> exchange-point-and-mark'ing
>

So this python

#!/usr/bin/python
num = 1000000
while num:
    print "123456789012345678901234567890"
    num = num -1
print "eureka"

redirecting output to a file, open that file with emacs -Q
incremental search for eur then C-x C-x  move the cursor down one or two
lines and emacs consistently crashes for me

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

[-- Attachment #1: Type: text/plain, Size: 707 bytes --]

 > Thanks, it now no longer crashes there, but crashes elsewhere but looks
 > to about to terminate in a more controlled manner?

IIUC what Emacs is trying to do here is to avoid the infinite loop
mentioned in etc/PROBLEMS:

   There is a long-standing bug in GTK that prevents it from recovering
   from disconnects: http://bugzilla.gnome.org/show_bug.cgi?id=85715.

It also says:

   If you need Emacs to be able to recover from closing displays, compile
   it with the Lucid toolkit instead of GTK.

So it might be interesting if Lucid (or non-toolkit) builds work around
this.

 > #1  check_glyph_memory () at dispnew.c:2264
 >          tail = XIL(0)

Yet another one.  I attached a new patch.

martin


[-- Attachment #2: FOR_EACH_FRAME.c.diff --]
[-- Type: text/plain, Size: 1888 bytes --]

diff --git a/src/dispnew.c b/src/dispnew.c
index b0fc5c3..d078647 100644
--- a/src/dispnew.c
+++ b/src/dispnew.c
@@ -2260,8 +2260,9 @@ struct glyph_row *
   Lisp_Object tail, frame;

   /* Free glyph memory for all frames.  */
-  FOR_EACH_FRAME (tail, frame)
-    free_glyphs (XFRAME (frame));
+  if (!NILP (Vframe_list))
+    FOR_EACH_FRAME (tail, frame)
+      free_glyphs (XFRAME (frame));

 #if defined GLYPH_DEBUG && defined ENABLE_CHECKING
   /* Check that nothing is left allocated.  */
diff --git a/src/frame.c b/src/frame.c
index 5bafbed..66d1b5c 100644
--- a/src/frame.c
+++ b/src/frame.c
@@ -2058,7 +2058,7 @@ of them (the selected terminal frame) is actually displayed.

   /* If we've deleted the last_nonminibuf_frame, then try to find
      another one.  */
-  if (f == last_nonminibuf_frame)
+  if (f == last_nonminibuf_frame && !NILP (Vframe_list))
     {
       last_nonminibuf_frame = 0;

@@ -2076,7 +2076,7 @@ of them (the selected terminal frame) is actually displayed.

   /* If there's no other frame on the same kboard, get out of
      single-kboard state if we're in it for this kboard.  */
-  if (kb != NULL)
+  if (kb != NULL && !NILP (Vframe_list))
     {
       /* Some frame we found on the same kboard, or nil if there are none.  */
       Lisp_Object frame_on_same_kboard = Qnil;
@@ -2093,7 +2093,9 @@ of them (the selected terminal frame) is actually displayed.
   /* If we've deleted this keyboard's default_minibuffer_frame, try to
      find another one.  Prefer minibuffer-only frames, but also notice
      frames with other windows.  */
-  if (kb != NULL && EQ (frame, KVAR (kb, Vdefault_minibuffer_frame)))
+  if (kb != NULL
+      && EQ (frame, KVAR (kb, Vdefault_minibuffer_frame))
+      && !NILP (Vframe_list))
     {
       /* The last frame we saw with a minibuffer, minibuffer-only or not.  */
       Lisp_Object frame_with_minibuf = Qnil;

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

 > So this python
 >
 > #!/usr/bin/python
 > num = 1000000
 > while num:
 >      print "123456789012345678901234567890"
 >      num = num -1
 > print "eureka"
 >
 > redirecting output to a file, open that file with emacs -Q
 > incremental search for eur then C-x C-x  move the cursor down one or two
 > lines and emacs consistently crashes for me

30 MB without newlines?  This makes my Emacs pretty unresponsive here.
However, on Windows I apparently don't get such crashes.

martin

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

martin rudalics <rudalics@gmx.at> writes:

>> So this python
>>
>> #!/usr/bin/python
>> num = 1000000
>> while num:
>>      print "123456789012345678901234567890"
>>      num = num -1
>> print "eureka"
>>
>> redirecting output to a file, open that file with emacs -Q
>> incremental search for eur then C-x C-x  move the cursor down one or two
>> lines and emacs consistently crashes for me
>
> 30 MB without newlines?  This makes my Emacs pretty unresponsive here.
> However, on Windows I apparently don't get such crashes.
>

A python print automatically adds a new line
I assume the windows failure to crash is because there isn't an X failure.

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Eli Zaretskii]

> From: Robert Marshall <robert.marshall@codethink.co.uk>
> Date: Thu, 14 Dec 2017 08:35:21 +0000
> Cc: 29661@debbugs.gnu.org
> 
> I assume the windows failure to crash is because there isn't an X failure.

Do you perhaps have some utility installed that attempts to manage the
X clipboard and selections?  If so, can you disable it and see if the
problem goes away?

Basically, the question is why does X fail or close the connection
with Emacs.

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

 > I now get - but maybe this is anticipated:
 >
 > Connection lost to X server ':0'
 > When compiled with GTK, Emacs cannot recover from X disconnects.
 > This is a GTK bug: https://bugzilla.gnome.org/show_bug.cgi?id=85715
 > For details, see etc/PROBLEMS.

This is the expected behavior.  I now installed the guards to get there
without segfaulting before.

martin

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[martin rudalics]

 > Basically, the question is why does X fail or close the connection
 > with Emacs.

Incidentally, this could be related to Bug#29690 (and the ancient
Bug#13486).

martin

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Robert Marshall <robert.marshall@codethink.co.uk>
>> Date: Thu, 14 Dec 2017 08:35:21 +0000
>> Cc: 29661@debbugs.gnu.org
>> 
>> I assume the windows failure to crash is because there isn't an X failure.
>
> Do you perhaps have some utility installed that attempts to manage the
> X clipboard and selections?  If so, can you disable it and see if the
> problem goes away?
>
> Basically, the question is why does X fail or close the connection
> with Emacs.

I'm running the standard kde desktop which comes with klipper. If I
stop klipper - at the moment! - I can no longer replicate the error

When running klipper from the command line and replicating the problem I
see

QClipboardWatcher::empty: Internal error: Application owns the selection

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Eli Zaretskii]

> From: Robert Marshall <robert.marshall@codethink.co.uk>
> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
> Date: Fri, 15 Dec 2017 09:24:21 +0000
> 
> > Do you perhaps have some utility installed that attempts to manage the
> > X clipboard and selections?  If so, can you disable it and see if the
> > problem goes away?
> >
> > Basically, the question is why does X fail or close the connection
> > with Emacs.
> 
> I'm running the standard kde desktop which comes with klipper. If I
> stop klipper - at the moment! - I can no longer replicate the error
> 
> When running klipper from the command line and replicating the problem I
> see
> 
> QClipboardWatcher::empty: Internal error: Application owns the selection

Ah, in that case we have an entry in PROBLEMS about klipper.  Does
upgrading to a newer version as indicated there solve the problem?  If
not, perhaps we should update that entry with some new information.

Thanks.

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Robert Marshall <robert.marshall@codethink.co.uk>
>> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
>> Date: Fri, 15 Dec 2017 09:24:21 +0000
>> 
>> > Do you perhaps have some utility installed that attempts to manage the
>> > X clipboard and selections?  If so, can you disable it and see if the
>> > problem goes away?
>> >
>> > Basically, the question is why does X fail or close the connection
>> > with Emacs.
>> 
>> I'm running the standard kde desktop which comes with klipper. If I
>> stop klipper - at the moment! - I can no longer replicate the error
>> 
>> When running klipper from the command line and replicating the problem I
>> see
>> 
>> QClipboardWatcher::empty: Internal error: Application owns the selection
>
> Ah, in that case we have an entry in PROBLEMS about klipper.  Does
> upgrading to a newer version as indicated there solve the problem?  If
> not, perhaps we should update that entry with some new information.
>
I have 4.14.2 of kde, Klipper: 0.9.7

Which would appear to be newer than the entry in PROBLEMS - update
to say emacs may crash? And mention select-active-regions?

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Eli Zaretskii]

> From: Robert Marshall <robert.marshall@codethink.co.uk>
> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
> Date: Fri, 15 Dec 2017 10:58:47 +0000
> 
> >> When running klipper from the command line and replicating the problem I
> >> see
> >> 
> >> QClipboardWatcher::empty: Internal error: Application owns the selection
> >
> > Ah, in that case we have an entry in PROBLEMS about klipper.  Does
> > upgrading to a newer version as indicated there solve the problem?  If
> > not, perhaps we should update that entry with some new information.
> >
> I have 4.14.2 of kde, Klipper: 0.9.7
> 
> Which would appear to be newer than the entry in PROBLEMS - update
> to say emacs may crash? And mention select-active-regions?

Please propose the text to add there, and thanks.

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Robert Marshall <robert.marshall@codethink.co.uk>
>> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
>> Date: Fri, 15 Dec 2017 10:58:47 +0000
>> 
>> >> When running klipper from the command line and replicating the problem I
>> >> see
>> >> 
>> >> QClipboardWatcher::empty: Internal error: Application owns the selection
>> >
>> > Ah, in that case we have an entry in PROBLEMS about klipper.  Does
>> > upgrading to a newer version as indicated there solve the problem?  If
>> > not, perhaps we should update that entry with some new information.
>> >
>> I have 4.14.2 of kde, Klipper: 0.9.7
>> 
>> Which would appear to be newer than the entry in PROBLEMS - update
>> to say emacs may crash? And mention select-active-regions?
>
> Please propose the text to add there, and thanks.

I'll take a look at my home machine tonight - which probably has a later
version to see if the problem is still present and then suggest an
amendment.

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Eli Zaretskii]

> From: Robert Marshall <robert.marshall@codethink.co.uk>
> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
> Date: Fri, 15 Dec 2017 13:00:58 +0000
> 
> >> I have 4.14.2 of kde, Klipper: 0.9.7
> >> 
> >> Which would appear to be newer than the entry in PROBLEMS - update
> >> to say emacs may crash? And mention select-active-regions?
> >
> > Please propose the text to add there, and thanks.
> 
> I'll take a look at my home machine tonight - which probably has a later
> version to see if the problem is still present and then suggest an
> amendment.

Any news on this?

TIA

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Robert Marshall <robert.marshall@codethink.co.uk>
>> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
>> Date: Fri, 15 Dec 2017 13:00:58 +0000
>> 
>> >> I have 4.14.2 of kde, Klipper: 0.9.7
>> >> 
>> >> Which would appear to be newer than the entry in PROBLEMS - update
>> >> to say emacs may crash? And mention select-active-regions?
>> >
>> > Please propose the text to add there, and thanks.
>> 
>> I'll take a look at my home machine tonight - which probably has a later
>> version to see if the problem is still present and then suggest an
>> amendment.
>
> Any news on this?
>

Apologies - last weekend was busier than expected - hopefully in the
break next week!

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

On Fri, Dec 15 2017, Robert Marshall <robert.marshall@codethink.co.uk> wrote:

> Eli Zaretskii <eliz@gnu.org> writes:
>
>>> From: Robert Marshall <robert.marshall@codethink.co.uk>
>>> Cc: rudalics@gmx.at,  29661@debbugs.gnu.org
>>> Date: Fri, 15 Dec 2017 10:58:47 +0000
>>> 
>>> >> When running klipper from the command line and replicating the problem I
>>> >> see
>>> >> 
>>> >> QClipboardWatcher::empty: Internal error: Application owns the selection
>>> >
>>> > Ah, in that case we have an entry in PROBLEMS about klipper.  Does
>>> > upgrading to a newer version as indicated there solve the problem?  If
>>> > not, perhaps we should update that entry with some new information.
>>> >
>>> I have 4.14.2 of kde, Klipper: 0.9.7

With 4.14.34 of kde Klipper 5.10.5 (that's a huge version leap!) I also
see the crash. I also have the problem with clipit (v 1.4.2) which I use in
another desktop environment for clipboard management

>>> 
>>> Which would appear to be newer than the entry in PROBLEMS - update
>>> to say emacs may crash? And mention select-active-regions?

Reading bug#29889 I set select-active-regions to 'only and the problem no
longer happens.

>> Please propose the text to add there, and thanks.
>
> I'll take a look at my home machine tonight - which probably has a later
> version to see if the problem is still present and then suggest an
> amendment.
>

Here's a suggested change:
----------------------------------------------------------------
diff --git a/etc/PROBLEMS b/etc/PROBLEMS
index c9b11fbc56..36df5493f4 100644
--- a/etc/PROBLEMS
+++ b/etc/PROBLEMS
@@ -1034,18 +1034,17 @@ It is also reported that a bug in the gtk-engines-qt engine can cause this if
 Emacs is compiled with Gtk+.
 The bug is fixed in version 0.7 or newer of gtk-engines-qt.
 
-*** KDE: Emacs hangs on KDE when a large portion of text is killed.
+*** Clipboard: Emacs hangs or crashes when a large portion of text is selected or killed.
 
-This is caused by a bug in the KDE applet 'klipper' which periodically
-requests the X clipboard contents from applications.  Early versions
-of klipper don't implement the ICCCM protocol for large selections,
-which leads to Emacs being flooded with selection requests.  After a
-while, Emacs may print a message:
+This is caused by a bug in the clipboard management applets (it has
+been observer in 'klipper' and 'clipit') which periodically request
+the X clipboard contents from applications.  After a while, Emacs may
+print a message:
 
   Timed out waiting for property-notify event
 
-A workaround is to not use 'klipper'.  Upgrading 'klipper' to the one
-coming with KDE 3.3 or later also solves the problem.
+A workaround is to not use 'klipper'or 'clipit'. Alternatively setting
+select-active-regions to 'only also solves the problem.
 
 *** KDE / Plasma 5: Emacs exhausts memory and needs to be killed
----------------------------------------------------------------

Robert

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Eli Zaretskii]

> From: Robert Marshall <robert@capuchin.co.uk>
> Cc: 29661@debbugs.gnu.org, Robert Marshall <robert.marshall@codethink.co.uk>
> Date: Mon, 01 Jan 2018 10:01:44 +0000
> 
> >>> I have 4.14.2 of kde, Klipper: 0.9.7
> 
> With 4.14.34 of kde Klipper 5.10.5 (that's a huge version leap!) I also
> see the crash. I also have the problem with clipit (v 1.4.2) which I use in
> another desktop environment for clipboard management
> 
> >>> 
> >>> Which would appear to be newer than the entry in PROBLEMS - update
> >>> to say emacs may crash? And mention select-active-regions?
> 
> Reading bug#29889 I set select-active-regions to 'only and the problem no
> longer happens.
> 
> >> Please propose the text to add there, and thanks.
> >
> > I'll take a look at my home machine tonight - which probably has a later
> > version to see if the problem is still present and then suggest an
> > amendment.
> >
> 
> Here's a suggested change:

Thanks, I committed it with a few changes.

I guess we can now close this bug?

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Robert Marshall]

Eli Zaretskii writes:
 > > From: Robert Marshall <robert@capuchin.co.uk>
 > > Cc: 29661@debbugs.gnu.org, Robert Marshall <robert.marshall@codethink.co.uk>
 > > Date: Mon, 01 Jan 2018 10:01:44 +0000
 > > 
 > > >>> I have 4.14.2 of kde, Klipper: 0.9.7
 > > 
 > > With 4.14.34 of kde Klipper 5.10.5 (that's a huge version leap!) I also
 > > see the crash. I also have the problem with clipit (v 1.4.2) which I use in
 > > another desktop environment for clipboard management
 > > 
 > > >>> 
 > > >>> Which would appear to be newer than the entry in PROBLEMS - update
 > > >>> to say emacs may crash? And mention select-active-regions?
 > > 
 > > Reading bug#29889 I set select-active-regions to 'only and the problem no
 > > longer happens.
 > > 
 > > >> Please propose the text to add there, and thanks.
 > > >
 > > > I'll take a look at my home machine tonight - which probably has a later
 > > > version to see if the problem is still present and then suggest an
 > > > amendment.
 > > >
 > > 
 > > Here's a suggested change:
 > 
 > Thanks, I committed it with a few changes.
 > 
 > I guess we can now close this bug?
 > 

I'm happy for that to happen - though I wondered if it should be moved
(or a reference placed ) in the Crash bugs section of that file?

Robert
-- 
Robert Marshall               twitter: @rajm

bug#29661: 27.0.50; emacs crashing on exchange-point-and-mark

[Eli Zaretskii]

> Date: Mon, 1 Jan 2018 19:11:51 +0000
> From: Robert Marshall <robert@capuchin.co.uk>
> Cc: 29661@debbugs.gnu.org
> 
>  > I guess we can now close this bug?
>  > 
> 
> I'm happy for that to happen - though I wondered if it should be moved
> (or a reference placed ) in the Crash bugs section of that file?

I'm not sure, because Emacs doesn't always crash due to this issue.