Bug 752936

Summary: emacs -nw does not accept any keyboard input
Product: [Fedora] Fedora Reporter: Michal Jaegermann <michal>
Component: emacsAssignee: Karel Klíč <kklic>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: rawhideCC: jonathan.underwood, kklic, rjones, rvokal
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-11-25 13:54:52 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
results of strace for ' emacs --no-init-file --no-site-file -nw'
none
trace lines added when the process above got killed none

Description Michal Jaegermann 2011-11-10 19:41:22 UTC
Description of problem:

Trying to start emacs in a terminal windown by using '-nw' brings up an emacs display and that is it.  There is no input which is recognized. One is stuck with options except of loging from somwhere else and killing a wayward process.  Once this is accomplishe the following shows up at a terminal prompt:

1;1400;0c

Attempts to simplify startup, in order to eliminate possible troubles there, like 'emacs -nw --no-site-file', has no effect save slightly different "decorations" on a resulting display.

Starting the same emacs in a graphics mode does show any issues.

Version-Release number of selected component (if applicable):
emacs-23.3-11.fc17.x86_64

How reproducible:
always

Additional info:
All of the above from a remote login/remote display as a local desktop is currenly completely broken and nothing can be started there.

Comment 1 Karel Klíč 2011-11-14 16:02:29 UTC
I have tested the emacs-23.3-11 build prepared locally for Fedora 16 and `emacs -nw` works well here.

Some X server issue and glibc issue occurred recently in Rawhide, so that might be the cause of the emacs input problem. Please try the following actions:

1. Update to the latest Rawhide packages.
   A rebuilt Emacs packages should be installed: emacs-23.3-12.fc17
   http://koji.fedoraproject.org/koji/packageinfo?packageID=560
   Does the rebuilt package solved the issue?

2. Switch into textual multi user mode and try using Emacs from here.
   Command to switch into multi user mode: systemctl isolate multi-user.target
   Does emacs respond to input?

Thanks.

Comment 2 Michal Jaegermann 2011-11-14 19:49:42 UTC
(In reply to comment #1)
> I have tested the emacs-23.3-11 build prepared locally for Fedora 16 and `emacs
> -nw` works well here.

> Please try the following actions:
> 
> 1. Update to the latest Rawhide packages.

Not available from mirrors yet but I pulled out from koji and installed
emacs-23.3-12.fc17.x86_64
emacs-common-23.3-12.fc17.x86_64
emacs-filesystem-23.3-12.fc17.x86_64

>    Does the rebuilt package solved the issue?

Not at all.  A behaviour is exactly the same. 'emacs' does work, locally or
with a remote display, 'emacs -nw' gets stuck.

> 2. Switch into textual multi user mode and try using Emacs from here.

Yes, I tried that.  It does not make a difference.  After 'emacs -nw' a splash
screen shows up.  A keyboard is not locked in that sense that I can switch
to another console (and kill emacs from there) but no input is accepted by
emacs.  I also tried 'TERM=vt100 emacs -nw'.  Resulting "decorations" are different, as expected, but otherwise the same thing.

>    Does emacs respond to input?

I backed off assorted xorg-* packages so I can have at least some display.
Running 'emacs -nw' locally in a terminal window has only this effect that
after killing it the following shows up '1;3001;0c' instead of '1;1400;0c'
but not much else.

BTW - my current glibc is 2.14.90-15.2.

I tried to strace 'emacs --no-init-file --no-site-file -nw'.
I attach resulting to the moment when emacs gets stuck.  Another trace
includes lines which are added after emacs process is killed.

Comment 3 Michal Jaegermann 2011-11-14 19:51:27 UTC
Created attachment 533605 [details]
results of strace for ' emacs --no-init-file --no-site-file  -nw'

Comment 4 Michal Jaegermann 2011-11-14 19:52:48 UTC
Created attachment 533606 [details]
trace lines added when the process above got killed

Comment 5 Michal Jaegermann 2011-11-15 00:57:42 UTC
I should add to clarify.  While booted into a local text console if I will type
'emacs some_file' then a content of 'some_file' shows up in an emacs window and no further input is accepted (emacs keystrokes, "normal" typing, arrow keys, whatever ...).  Neither /usr/bin/vim nor /bin/vi are afflicted by such disease.

Comment 6 Michal Jaegermann 2011-11-18 18:24:30 UTC
An update to emacs-23.3-13.fc17.x86_64 does not bring any joy, I am afraid.

Comment 7 Karel Klíč 2011-11-22 16:15:20 UTC
An update to emacs-23.3-14.fc17 should definitely help, then.

Comment 8 Karel Klíč 2011-11-22 16:23:50 UTC
*** Bug 754837 has been marked as a duplicate of this bug. ***

Comment 9 Richard W.M. Jones 2011-11-22 16:59:49 UTC
-14 doesn't fix it for me.

Comment 10 Michal Jaegermann 2011-11-23 21:39:37 UTC
(In reply to comment #9)
> -14 doesn't fix it for me.
Does not work for me either. It is not 100% CPU but "only" 90%+ but that is not of a great help.  One can attach strace to a running process but this does not show anything; just sits there.

Comment 11 Karel Klíč 2011-11-25 13:54:52 UTC
This is a duplicate of the bug describing the same problem with `emacs --daemon`, which also fails because it doesn't have an X window.

*** This bug has been marked as a duplicate of bug 751154 ***