bug#29347: 27.0.50; C-g doesn't quit minibuffer

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

C-g does not exit the minibuffer.

Moreover, when I keep typing C-g, eventually Emacs gets hung.  I can
get it unhung by sending it SIGTSTP using another terminal.

I can't find the command to make gdb send the backtrace to a file.



In GNU Emacs 27.0.50 (build 1, x86_64-pc-linux-gnu, GTK+ Version 2.24.23)
 of 2017-11-15 built on freetop
Repository revision: 52d822f31bc7cb57694c1e209b2d02e5efb8f48c
System Description:	Trisquel GNU/Linux 7.0, Belenos

Recent messages:
Saving file /home/rms/outgoing/out-17...
Wrote /home/rms/outgoing/out-17
Mark set
Saving file /home/rms/outgoing/out-16...
Wrote /home/rms/outgoing/out-16
Mark set [3 times]
Saving file /home/rms/outgoing/out-17...
Wrote /home/rms/outgoing/out-17
Buffer has unsaved changes; reinitialize it and discard them? (y or n) y
Disconnect buffer from visited file? (y or n) y

Configured using:
 'configure 'CFLAGS=-O0 -g' --with-gnutls=no'

Configured features:
XPM JPEG TIFF GIF PNG RSVG SOUND GPM DBUS GSETTINGS NOTIFY LIBXML2
FREETYPE M17N_FLT LIBOTF XFT ZLIB TOOLKIT_SCROLL_BARS GTK2 X11

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

Major mode: RMAIL

Minor modes in effect:
  diff-auto-refine-mode: t
  shell-dirtrack-mode: t
  gpm-mouse-mode: t
  tooltip-mode: t
  global-eldoc-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
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  buffer-read-only: t
  line-number-mode: t
  transient-mark-mode: t
  abbrev-mode: t

Load-path shadows:
None found.

Features:
(shadow emacsbug two-column kmacro iso-transl battery quail help-mode
epa-mail etags xref project rmailkwd compare-w diff-mode easy-mmode
rmailout url-util shr svg xml dom browse-url shell pcomplete thingatpt
grep compile comint ansi-color ring rmailsum dabbrev mailalias
misearch multi-isearch qp rmailmm message sendmail rmc puny
format-spec rfc822 mml mml-sec epa epg gnus-util mm-decode mm-bodies
mm-encode mailabbrev gmm-utils mailheader mail-parse rfc2231 rmail
rmail-loaddefs rfc2047 rfc2045 ietf-drums mm-util mail-prsvr
mail-utils dired dired-loaddefs t-mouse term/linux elec-pair view
derived time-date paren cus-start cus-load advice finder-inf package
easymenu epg-config url-handlers url-parse auth-source cl-seq eieio
eieio-core cl-macs eieio-loaddefs password-cache url-vars seq byte-opt
gv bytecomp byte-compile cconv cl-loaddefs cl-lib tooltip eldoc
electric uniquify ediff-hook vc-hooks lisp-float-type mwheel
term/x-win x-win term/common-win x-dnd tool-bar dnd fontset image
regexp-opt fringe tabulated-list replace newcomment text-mode
elisp-mode lisp-mode prog-mode register page menu-bar rfn-eshadow
isearch timer select scroll-bar mouse jit-lock font-lock syntax
facemenu font-core term/tty-colors frame cl-generic cham georgian
utf-8-lang misc-lang vietnamese tibetan thai tai-viet lao korean
japanese eucjp-ms cp51932 hebrew greek romanian slovak czech european
ethiopic indian cyrillic chinese composite charscript charprop
case-table epa-hook jka-cmpr-hook help simple abbrev obarray
minibuffer cl-preloaded nadvice loaddefs button faces cus-face
macroexp files text-properties overlay sha1 md5 base64 format env
code-pages mule custom widget hashtable-print-readable backquote
dbusbind inotify dynamic-setting system-font-setting
font-render-setting move-toolbar gtk x-toolkit x multi-tty
make-network-process emacs)

Memory information:
((conses 16 302080 76236)
 (symbols 48 25226 12)
 (miscs 40 2202 2989)
 (strings 32 48806 9257)
 (string-bytes 1 1231232)
 (vectors 16 29076)
 (vector-slots 8 1398157 226116)
 (floats 8 95 276)
 (intervals 56 60145 1270)
 (buffers 992 67)
 (heap 1024 30747 3090))
[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]


-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Andreas Schwab]

On Nov 17 2017, Richard Stallman <rms@gnu.org> wrote:

> I can't find the command to make gdb send the backtrace to a file.

(gdb) help set logging 

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Drew Adams]

> C-g does not exit the minibuffer.
> 
> Moreover, when I keep typing C-g, eventually Emacs gets hung.  I can
> get it unhung by sending it SIGTSTP using another terminal.

FWIW, I believe I've seen the same thing with the 26.1
pretest.  I've had to kill the Emacs process when it's
happened.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> Date: Fri, 17 Nov 2017 22:46:39 -0500
> 
> C-g does not exit the minibuffer.
> 
> Moreover, when I keep typing C-g, eventually Emacs gets hung.  I can
> get it unhung by sending it SIGTSTP using another terminal.

Is this in a GUI frame or a TTY frame?  If the latter, I cannot
reproduce that.  (I have no access to a GUI Emacs built from the
master branch on GNU/Linux.)

Is there any recipe starting from "emacs -Q" that I should try?  E.g.,
what did you type to get into the minibuffer in the first place? does
it matter?  (I tried "M-x" and "C-x C-f".)

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > > Moreover, when I keep typing C-g, eventually Emacs gets hung.  I can
  > > get it unhung by sending it SIGTSTP using another terminal.

  > Is this in a GUI frame or a TTY frame?  If the latter, I cannot
  > reproduce that.  (I have no access to a GUI Emacs built from the
  > master branch on GNU/Linux.)

It is from a tty.

I just tried to make this fail starting from emacs -Q
and couldn't do so.  The initial C-g failure, not quitting out
of the minibuffer, doesn't happen right after startup.

However, once C-g fails to quit, it continues failing reliably.
And it fails regardless of the purpose of the minibuffer.

I learned not to type C-g to get out of a minibuffer.
First I tried giving operands that would be meaningless.
Then I thought of using C-].

As a result of this change in my usage, it does not get hung.
But the bug is still there.

If I can make it fail again, I will send a backtrace.

ISTR that at least once Emacs was in a state where C-g did not turn
off region highlighting.  I suspect that is the same bug, appearing
in a different way.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> CC: 29347@debbugs.gnu.org
> Date: Sat, 18 Nov 2017 22:21:23 -0500
> 
> I just tried to make this fail starting from emacs -Q
> and couldn't do so.  The initial C-g failure, not quitting out
> of the minibuffer, doesn't happen right after startup.
> 
> However, once C-g fails to quit, it continues failing reliably.
> And it fails regardless of the purpose of the minibuffer.
> 
> I learned not to type C-g to get out of a minibuffer.
> First I tried giving operands that would be meaningless.
> Then I thought of using C-].

Did you try "C-x o"?  If that works, then it's not a bug, it means
you've entered recursive exit in the minibuffer in some way.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Andreas Schwab]

On Nov 19 2017, Eli Zaretskii <eliz@gnu.org> wrote:

> Did you try "C-x o"?  If that works, then it's not a bug, it means
> you've entered recursive exit in the minibuffer in some way.

Recusive edit will be indicated in the mode line, and C-g will exit it
in the minibuffer.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: rms@gnu.org,  29347@debbugs.gnu.org
> Date: Sun, 19 Nov 2017 09:33:10 +0100
> 
> On Nov 19 2017, Eli Zaretskii <eliz@gnu.org> wrote:
> 
> > Did you try "C-x o"?  If that works, then it's not a bug, it means
> > you've entered recursive exit in the minibuffer in some way.
> 
> Recusive edit will be indicated in the mode line, and C-g will exit it
> in the minibuffer.

I meant enable-recursive-minibuffers, which AFAIK doesn't have any
indications.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Andreas Schwab]

On Nov 19 2017, Eli Zaretskii <eliz@gnu.org> wrote:

>> From: Andreas Schwab <schwab@linux-m68k.org>
>> Cc: rms@gnu.org,  29347@debbugs.gnu.org
>> Date: Sun, 19 Nov 2017 09:33:10 +0100
>> 
>> On Nov 19 2017, Eli Zaretskii <eliz@gnu.org> wrote:
>> 
>> > Did you try "C-x o"?  If that works, then it's not a bug, it means
>> > you've entered recursive exit in the minibuffer in some way.
>> 
>> Recusive edit will be indicated in the mode line, and C-g will exit it
>> in the minibuffer.
>
> I meant enable-recursive-minibuffers, which AFAIK doesn't have any
> indications.

A recursive minibuffer can be exited like an ordinary minibuffer, and
exiting it will return to the buffer from where you entered it.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: rms@gnu.org,  29347@debbugs.gnu.org
> Date: Sun, 19 Nov 2017 18:21:16 +0100
> 
> > I meant enable-recursive-minibuffers, which AFAIK doesn't have any
> > indications.
> 
> A recursive minibuffer can be exited like an ordinary minibuffer, and
> exiting it will return to the buffer from where you entered it.

And that buffer to which you return is also a minibuffer.  IOW, if you
are N levels deep in a recursive minibuffer, you cannot exit it until
you type C-g N times.  This could be perceived as "C-g doesn't quit
the minibuffer".

I'm not saying this is what happened, I'm just trying to think about
every possible situation which could have been interpreted that way.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Drew Adams]

> > > I meant enable-recursive-minibuffers, which AFAIK doesn't have any
> > > indications.
> >
> > A recursive minibuffer can be exited like an ordinary minibuffer, and
> > exiting it will return to the buffer from where you entered it.
> 
> And that buffer to which you return is also a minibuffer.  IOW, if you
> are N levels deep in a recursive minibuffer, you cannot exit it until
> you type C-g N times.  This could be perceived as "C-g doesn't quit
> the minibuffer".

Clearly C-g _did_ quit the current minibuffer, in that scenario.
If you are in the Nth minibuffer then hitting C-g N times quits
N minibuffers - you are no longer in any minibuffer.  (To quit
all minibuffers you can use C-].)

I don't think that's what the problem reported refers to.
That's the normal behavior, and it always has been.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

I don't enable recursive minibuffers, so my problem wasn't about them.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > > I learned not to type C-g to get out of a minibuffer.
  > > First I tried giving operands that would be meaningless.
  > > Then I thought of using C-].

  > Did you try "C-x o"?

I don't remember whether I tried that at the time this was failing.

It hasn't failed again since the day I noticed it.
Maybe some change has fixed the problem.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

I have not seen C-g fail again.  Strange, because I started Emacs repeatedly
and saw the problem.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Mike Gerwitz]

[-- Attachment #1: Type: text/plain, Size: 8620 bytes --]

On Fri, Nov 17, 2017 at 22:46:39 -0500, Richard Stallman wrote:
> C-g does not exit the minibuffer.
>
> Moreover, when I keep typing C-g, eventually Emacs gets hung.  I can
> get it unhung by sending it SIGTSTP using another terminal.

Yes, I get this as well on 26.0.90.  I've been meaning to report this
for just over a week, but I hadn't gotten around to it.

I can reproduce it simply with `emacs -Q -nsl -nw' (it must be a TTY;
GUI emacs does not suffer from this bug): quickly press `C-f C-f C-g'.
The message "Command attempted to use minibuffer while in a minibuffer"
will display.  If C-g is pressed while it is still visible, emacs
completely hangs.

In the past it seemed like I could issue SIGSHUP and it would stop
whatever it was doing, but usually it kills emacs.  If it doesn't, then
`C-g' stops working until restart.

> I can't find the command to make gdb send the backtrace to a file.

(gdb) bt
#0  __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1  0x00007f778224c649 in _L_lock_909 () from /lib/x86_64-linux-gnu/libpthread.so.0
#2  0x00007f778224c470 in __GI___pthread_mutex_lock (mutex=0xbd9960 <global_lock>) at ../nptl/pthread_mutex_lock.c:79
#3  0x00000000005c22c5 in sys_mutex_lock (mutex=mutex@entry=0xbd9960 <global_lock>) at systhread.c:106
#4  0x00000000005c1a95 in acquire_global_lock (self=0xbd9988 <main_thread>) at thread.c:100
#5  really_call_select (arg=0x7ffcfc8f6910) at thread.c:576
#6  0x00000000005c2057 in thread_select (func=<optimized out>, max_fds=max_fds@entry=7, rfds=rfds@entry=0x7ffcfc8f69d0, wfds=wfds@entry=0x7ffcfc8f6a50, efds=efds@entry=0x0,
    timeout=timeout@entry=0x7ffcfc8f6fe0, sigmask=sigmask@entry=0x0) at thread.c:595
#7  0x00000000005de450 in xg_select (fds_lim=7, rfds=rfds@entry=0x7ffcfc8f70d0, wfds=wfds@entry=0x7ffcfc8f7150, efds=efds@entry=0x0, timeout=timeout@entry=0x7ffcfc8f6fe0,
    sigmask=sigmask@entry=0x0) at xgselect.c:117
#8  0x00000000005a4fc5 in wait_reading_process_output (time_limit=time_limit@entry=2, nsecs=nsecs@entry=0, read_kbd=read_kbd@entry=1, do_display=do_display@entry=true,
    wait_for_cell=wait_for_cell@entry=0, wait_proc=wait_proc@entry=0x0, just_wait_proc=just_wait_proc@entry=0) at process.c:5375
#9  0x0000000000422c72 in sit_for (timeout=<optimized out>, reading=reading@entry=false, display_option=display_option@entry=2) at dispnew.c:5793
#10 0x00000000004fe73e in command_loop_1 () at keyboard.c:1340
#11 0x0000000000562d14 in internal_condition_case (bfun=bfun@entry=0x4fdb60 <command_loop_1>, handlers=handlers@entry=21072, hfun=hfun@entry=0x4f4590 <cmd_error>)
    at eval.c:1332
#12 0x00000000004ef92c in command_loop_2 (ignore=ignore@entry=0) at keyboard.c:1110
#13 0x0000000000562cc4 in internal_catch (tag=tag@entry=21744, func=func@entry=0x4ef910 <command_loop_2>, arg=arg@entry=0) at eval.c:1097
#14 0x00000000004ef89a in command_loop () at keyboard.c:1081
#15 0x00000000004f41c3 in recursive_edit_1 () at keyboard.c:695
#16 0x000000000051cbba in read_minibuf (map=map@entry=21543315, initial=initial@entry=22883060, prompt=prompt@entry=9012828, expflag=<optimized out>, histvar=<optimized out>,
    histpos=<optimized out>, defalt=defalt@entry=22879220, allow_props=false, inherit_input_method=false) at minibuf.c:685
#17 0x000000000051d3bf in Fread_from_minibuffer (prompt=9012828, initial_contents=22883060, keymap=21543315, read=0, hist=<optimized out>, default_value=22879220,
    inherit_input_method=0) at minibuf.c:992
#18 0x0000000000565654 in funcall_subr (subr=0x85a6e8 <Sread_from_minibuffer>, numargs=numargs@entry=7, args=args@entry=0x7ffcfc8f7800) at eval.c:2861
#19 0x00000000005646c6 in Ffuncall (nargs=nargs@entry=8, args=args@entry=0x7ffcfc8f77f8) at eval.c:2766
#20 0x000000000059a6f1 in exec_byte_code (bytestr=<optimized out>, vector=9553853, maxdepth=<optimized out>, args_template=<optimized out>,
    nargs=nargs@entry=72057594037927944, args=<optimized out>, args@entry=0x91c7c0 <pure+771808>) at bytecode.c:629
#21 0x00000000005643cc in funcall_lambda (fun=140724545747065, nargs=72057594037927944, nargs@entry=8, arg_vector=0x91c7c0 <pure+771808>, arg_vector@entry=0x7ffcfc8f79b8)
    at eval.c:2967
#22 0x0000000000564643 in Ffuncall (nargs=nargs@entry=9, args=args@entry=0x7ffcfc8f79b0) at eval.c:2780
#23 0x000000000051b9d0 in Fcompleting_read (prompt=9012828, collection=4729664, predicate=23040, require_match=4745440, initial_input=22883060, hist=23472, def=22879220,
    inherit_input_method=0) at minibuf.c:1696
#24 0x0000000000565525 in funcall_subr (subr=0x85a568 <Scompleting_read>, numargs=numargs@entry=7, args=args@entry=0x7ffcfc8f7b70) at eval.c:2866
#25 0x00000000005646c6 in Ffuncall (nargs=nargs@entry=8, args=args@entry=0x7ffcfc8f7b68) at eval.c:2766
#26 0x000000000059a6f1 in exec_byte_code (bytestr=<optimized out>, vector=9547005, maxdepth=<optimized out>, args_template=<optimized out>,
    nargs=nargs@entry=72057594037927942, args=<optimized out>, args@entry=0x91ad00 <pure+764960>) at bytecode.c:629
#27 0x00000000005643cc in funcall_lambda (fun=140724545748086, nargs=72057594037927942, nargs@entry=6, arg_vector=0x91ad00 <pure+764960>, arg_vector@entry=0x7ffcfc8f7ed0)
    at eval.c:2967
#28 0x0000000000564643 in Ffuncall (nargs=nargs@entry=7, args=args@entry=0x7ffcfc8f7ec8) at eval.c:2780
#29 0x000000000059a6f1 in exec_byte_code (bytestr=<optimized out>, vector=9546893, maxdepth=<optimized out>, args_template=<optimized out>,
    nargs=nargs@entry=72057594037927940, args=<optimized out>, args@entry=0x91ac90 <pure+764848>) at bytecode.c:629
#30 0x00000000005643cc in funcall_lambda (fun=140724545748754, nargs=72057594037927940, nargs@entry=4, arg_vector=0x91ac90 <pure+764848>, arg_vector@entry=0x7ffcfc8f8050)
    at eval.c:2967
#31 0x0000000000564643 in Ffuncall (nargs=nargs@entry=5, args=args@entry=0x7ffcfc8f8048) at eval.c:2780
#32 0x000000000059a6f1 in exec_byte_code (bytestr=<optimized out>, vector=9131805, maxdepth=<optimized out>, args_template=<optimized out>,
    nargs=nargs@entry=72057594037927938, args=<optimized out>, args@entry=0x8b5720 <pure+349760>) at bytecode.c:629
#33 0x00000000005643cc in funcall_lambda (fun=140724545749110, nargs=72057594037927938, nargs@entry=2, arg_vector=0x8b5720 <pure+349760>, arg_vector@entry=0x7ffcfc8f81a0)
    at eval.c:2967
#34 0x0000000000564643 in Ffuncall (nargs=nargs@entry=3, args=args@entry=0x7ffcfc8f8198) at eval.c:2780
#35 0x000000000059a6f1 in exec_byte_code (bytestr=<optimized out>, vector=9132077, maxdepth=<optimized out>, args_template=<optimized out>, nargs=<optimized out>,
    args=<optimized out>) at bytecode.c:629
#36 0x0000000000563c9b in eval_sub (form=form@entry=9131995) at eval.c:2237
#37 0x0000000000567680 in Feval (form=form@entry=9131995, lexical=<optimized out>) at eval.c:2051
#38 0x0000000000560888 in Fcall_interactively (function=4338912, record_flag=0, keys=12823093) at callint.c:357
#39 0x00000000005646c6 in Ffuncall (nargs=nargs@entry=4, args=args@entry=0x7ffcfc8f8518) at eval.c:2766
#40 0x000000000059a6f1 in exec_byte_code (bytestr=<optimized out>, vector=9587005, maxdepth=<optimized out>, args_template=<optimized out>,
    nargs=nargs@entry=72057594037927937, args=<optimized out>, args@entry=0x924940 <pure+804960>) at bytecode.c:629
#41 0x00000000005643cc in funcall_lambda (fun=140724545750475, nargs=72057594037927937, nargs@entry=1, arg_vector=0x924940 <pure+804960>, arg_vector@entry=0x7ffcfc8f8718)
    at eval.c:2967
#42 0x0000000000564643 in Ffuncall (nargs=nargs@entry=2, args=args@entry=0x7ffcfc8f8710) at eval.c:2780
#43 0x000000000056476a in call1 (fn=fn@entry=16128, arg1=<optimized out>) at eval.c:2617
#44 0x00000000004fdf4d in command_loop_1 () at keyboard.c:1482
#45 0x0000000000562d14 in internal_condition_case (bfun=bfun@entry=0x4fdb60 <command_loop_1>, handlers=handlers@entry=21072, hfun=hfun@entry=0x4f4590 <cmd_error>)
    at eval.c:1332
#46 0x00000000004ef92c in command_loop_2 (ignore=ignore@entry=0) at keyboard.c:1110
#47 0x0000000000562cc4 in internal_catch (tag=tag@entry=50928, func=func@entry=0x4ef910 <command_loop_2>, arg=arg@entry=0) at eval.c:1097
#48 0x00000000004ef8e7 in command_loop () at keyboard.c:1089
#49 0x00000000004f41c3 in recursive_edit_1 () at keyboard.c:695
#50 0x00000000004f44d5 in Frecursive_edit () at keyboard.c:766
#51 0x0000000000419093 in main (argc=3, argv=0x7ffcfc8f8af8) at emacs.c:1713

-- 
Mike Gerwitz
Free Software Hacker+Activist | GNU Maintainer & Volunteer
GPG: D6E9 B930 028A 6C38 F43B  2388 FEF6 3574 5E6F 6D05
https://mikegerwitz.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 818 bytes --]

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Mike Gerwitz <mtg@gnu.org>
> Date: Mon, 20 Nov 2017 13:09:06 -0500
> Cc: 29347@debbugs.gnu.org
> 
> I can reproduce it simply with `emacs -Q -nsl -nw' (it must be a TTY;
> GUI emacs does not suffer from this bug): quickly press `C-f C-f C-g'.

Do you mean "C-x C-f C-g"?

> The message "Command attempted to use minibuffer while in a minibuffer"
> will display.  If C-g is pressed while it is still visible, emacs
> completely hangs.

I cannot reproduce this with the current emacs-26 branch.  Can you try
that branch?

Thanks.

P.S. I'm not at all sure this is the same problem that Richard
reported.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Mike Gerwitz]

[-- Attachment #1: Type: text/plain, Size: 808 bytes --]

On Mon, Nov 20, 2017 at 20:36:38 +0200, Eli Zaretskii wrote:
>> I can reproduce it simply with `emacs -Q -nsl -nw' (it must be a TTY;
>> GUI emacs does not suffer from this bug): quickly press `C-f C-f C-g'.
>
> Do you mean "C-x C-f C-g"?

I'm sorry, I meant `C-x C-f C-x C-f C-g` (attempt a recursive
minibuffer).  I was in a rush typing the original message.

> I cannot reproduce this with the current emacs-26 branch.  Can you try
> that branch?

I'm at work but I'll give that a try tonight.

> P.S. I'm not at all sure this is the same problem that Richard
> reported.

If it isn't I'll be happy to open a new bug.  Lmk.

-- 
Mike Gerwitz
Free Software Hacker+Activist | GNU Maintainer & Volunteer
GPG: D6E9 B930 028A 6C38 F43B  2388 FEF6 3574 5E6F 6D05
https://mikegerwitz.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 818 bytes --]

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Mike Gerwitz <mtg@gnu.org>
> Cc: rms@gnu.org,  29347@debbugs.gnu.org
> Date: Mon, 20 Nov 2017 15:26:36 -0500
> 
> On Mon, Nov 20, 2017 at 20:36:38 +0200, Eli Zaretskii wrote:
> >> I can reproduce it simply with `emacs -Q -nsl -nw' (it must be a TTY;
> >> GUI emacs does not suffer from this bug): quickly press `C-f C-f C-g'.
> >
> > Do you mean "C-x C-f C-g"?
> 
> I'm sorry, I meant `C-x C-f C-x C-f C-g` (attempt a recursive
> minibuffer).  I was in a rush typing the original message.

I think I fixed the problem which caused Emacs to hang in this
particular scenario, but I'm not sure Richard is seeing the same
problem.  The change I installed on the emacs-26 branch is below;
Richard, please try applying it, and see if your problems go away.

diff --git a/src/thread.c b/src/thread.c
index c03cdda..1ded8f5 100644
--- a/src/thread.c
+++ b/src/thread.c
@@ -97,7 +97,12 @@ post_acquire_global_lock (struct thread_state *self)
 static void
 acquire_global_lock (struct thread_state *self)
 {
-  sys_mutex_lock (&global_lock);
+  /* If some Lisp was interrupted by C-g while inside pselect, the
+     signal handler could have called maybe_reacquire_global_lock, in
+     which case we are already holding the lock and shouldn't try
+     taking it again, or else we will hang forever.  */
+  if (!(self && self->not_holding_lock))
+    sys_mutex_lock (&global_lock);
   post_acquire_global_lock (self);
 }
 

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> Date: Fri, 24 Nov 2017 18:17:58 +0200
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: 29370@debbugs.gnu.org, 29347@debbugs.gnu.org, rms@gnu.org
> 
> I think I fixed the problem which caused Emacs to hang in this
> particular scenario, but I'm not sure Richard is seeing the same
> problem.  The change I installed on the emacs-26 branch is below;
> Richard, please try applying it, and see if your problems go away.

Sorry, wrong patch.  Please use the one below instead.

diff --git a/src/thread.c b/src/thread.c
index c03cdda..1ded8f5 100644
--- a/src/thread.c
+++ b/src/thread.c
@@ -97,7 +97,12 @@ post_acquire_global_lock (struct thread_state *self)
 static void
 acquire_global_lock (struct thread_state *self)
 {
-  sys_mutex_lock (&global_lock);
+  /* If some Lisp was interrupted by C-g while inside pselect, the
+     signal handler could have called maybe_reacquire_global_lock, in
+     which case we are already holding the lock and shouldn't try
+     taking it again, or else we will hang forever.  */
+  if (!(self && !self->not_holding_lock))
+    sys_mutex_lock (&global_lock);
   post_acquire_global_lock (self);
 }
 

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> Date: Fri, 24 Nov 2017 23:14:16 +0200
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: 29370@debbugs.gnu.org, 29347@debbugs.gnu.org, rms@gnu.org
> 
> > Date: Fri, 24 Nov 2017 18:17:58 +0200
> > From: Eli Zaretskii <eliz@gnu.org>
> > Cc: 29370@debbugs.gnu.org, 29347@debbugs.gnu.org, rms@gnu.org
> > 
> > I think I fixed the problem which caused Emacs to hang in this
> > particular scenario, but I'm not sure Richard is seeing the same
> > problem.  The change I installed on the emacs-26 branch is below;
> > Richard, please try applying it, and see if your problems go away.
> 
> Sorry, wrong patch.  Please use the one below instead.

Sorry again, that was not the best idea.  Here is a better patch:

diff --git a/src/thread.c b/src/thread.c
index 9e799ce..dd46681 100644
--- a/src/thread.c
+++ b/src/thread.c
@@ -578,8 +573,15 @@ really_call_select (void *arg)
 			   sa->timeout, sa->sigmask);
 
   block_interrupt_signal (&oldset);
-  acquire_global_lock (self);
-  self->not_holding_lock = 0;
+  /* If we were interrupted by C-g while inside sa->func above, the
+     signal handler could have called maybe_reacquire_global_lock, in
+     which case we are already holding the lock and shouldn't try
+     taking it again, or else we will hang forever.  */
+  if (self->not_holding_lock)
+    {
+      acquire_global_lock (self);
+      self->not_holding_lock = 0;
+    }
   restore_signal_mask (&oldset);
 }
 

bug#29370: bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  >    block_interrupt_signal (&oldset);
  > -  acquire_global_lock (self);
  > -  self->not_holding_lock = 0;
  > +  /* If we were interrupted by C-g while inside sa->func above, the
  > +     signal handler could have called maybe_reacquire_global_lock, in
  > +     which case we are already holding the lock and shouldn't try
  > +     taking it again, or else we will hang forever.  */
  > +  if (self->not_holding_lock)
  > +    {
  > +      acquire_global_lock (self);
  > +      self->not_holding_lock = 0;
  > +    }
  >    restore_signal_mask (&oldset);

I am running with this patch, and I will report the results.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Mike Gerwitz]

[-- Attachment #1: Type: text/plain, Size: 601 bytes --]

On Fri, Nov 24, 2017 at 18:17:58 +0200, Eli Zaretskii wrote:
> I think I fixed the problem which caused Emacs to hang in this
> particular scenario, but I'm not sure Richard is seeing the same
> problem.  The change I installed on the emacs-26 branch is below;
> Richard, please try applying it, and see if your problems go away.

The issue has been fixed on the current emacs-26 branch
(6ec5d49).  Thanks for the quick fix, Eli!

-- 
Mike Gerwitz
Free Software Hacker+Activist | GNU Maintainer & Volunteer
GPG: D6E9 B930 028A 6C38 F43B  2388 FEF6 3574 5E6F 6D05
https://mikegerwitz.com

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 818 bytes --]

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Mike Gerwitz <mtg@gnu.org>
> Cc: rms@gnu.org,  29347@debbugs.gnu.org,  29370@debbugs.gnu.org
> Date: Sun, 26 Nov 2017 23:48:51 -0500
> 
> On Fri, Nov 24, 2017 at 18:17:58 +0200, Eli Zaretskii wrote:
> > I think I fixed the problem which caused Emacs to hang in this
> > particular scenario, but I'm not sure Richard is seeing the same
> > problem.  The change I installed on the emacs-26 branch is below;
> > Richard, please try applying it, and see if your problems go away.
> 
> The issue has been fixed on the current emacs-26 branch
> (6ec5d49).  Thanks for the quick fix, Eli!

Thanks for testing.  This scenario was simple enough to understand
what was wrong.  I just hope this also solves Richard's problems.

bug#29370: bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

I have not noticed that problem since installing the change.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

bug#29370: bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> CC: eliz@gnu.org, 29347@debbugs.gnu.org, 29370@debbugs.gnu.org
> Date: Mon, 27 Nov 2017 18:28:57 -0500
> 
> I have not noticed that problem since installing the change.

Good to know, thanks.  I will close the bug in a few days, if not new
issues like that come up.

bug#29347: bug#29370: bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Eli Zaretskii]

> Date: Tue, 28 Nov 2017 05:33:46 +0200
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: 29370@debbugs.gnu.org, mtg@gnu.org, 29347@debbugs.gnu.org
> 
> > From: Richard Stallman <rms@gnu.org>
> > CC: eliz@gnu.org, 29347@debbugs.gnu.org, 29370@debbugs.gnu.org
> > Date: Mon, 27 Nov 2017 18:28:57 -0500
> > 
> > I have not noticed that problem since installing the change.
> 
> Good to know, thanks.  I will close the bug in a few days, if not new
> issues like that come up.

Nothing new, so I'm closing these bugs.

bug#29347: 27.0.50; C-g doesn't quit minibuffer

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

Thanks for investigating this.  Having a reproducible case
means it will get fixed.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.