After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 99220 - Eats all the CPU time when creating a new tab, it's fixed when swithing to existing tab and then back
Eats all the CPU time when creating a new tab, it's fixed when swithing to ex...
Status: RESOLVED FIXED
Product: gnome-terminal
Classification: Core
Component: general
2.1.x
Other Linux
: High normal
: ---
Assigned To: Havoc Pennington
GNOME Terminal Maintainers
: 99963 100438 100484 100543 100690 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2002-11-21 19:33 UTC by Tuomas Jormola
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: 2.1/2.2


Description Tuomas Jormola 2002-11-21 19:43:20 UTC 
Package: gnome-terminal
Severity: normal
Version: GNOME2.1.2 2.1.1.20021121
Synopsis: Eats all the CPU time when creating a new tab, it's fixed when swithing to existing tab and then back
Bugzilla-Product: gnome-terminal
Bugzilla-Component: general
Description:
Description of Problem:
I'm not quite sure if this is a gnome-terminal problem or vte problem.
When a new tab is created, no shell window is displayed in the window.
Instead gnome-terminal eats all the CPU it can. Program is still
responsive, though. Switch back to 1st tab and then back to the newly
created 2nd tab. CPU usage drops and the shell exists there as exepcted.
This problem only occurrs when creating the 1st new tab during a
gnome-terminal session, all further tabs are created OK.

Steps to reproduce the problem:
1. Launch a new gnome-terminal session
2. Create a new tab
3. New shell fails to appear. Switch back to 1st tab and then back to
this new 2nd tab.

Actual Results:
After creation, new tab is frozen with no shell.

Expected Results:
New tab with shell should be created.

How often does this happen?
Each time.

Additional Information:
Debian GNU/Linux, Athlon PC, Gnome 2.2 snapshot compiled from Gnome CVS
on the 11/20/2002 and installed in clean prefix.




------- Bug moved to this database by unknown@bugzilla.gnome.org 2002-11-21 14:43 -------

Unknown version 1.1.x in product gnome-terminal. Setting version to the default, "unspecified".
The original reporter (tj@shakti.tky.hut.fi) of this bug does not have an account here.
Reassigning to the exporter, unknown@bugzilla.gnome.org.
Reassigning to the default owner of the component, hp@redhat.com.
Comment 1 Heath Harrelson 2002-11-21 23:26:08 UTC 
I've seen this, and others on IRC mentioned it, including Luis. 
Confirming.
Comment 2 Heath Harrelson 2002-11-30 12:15:59 UTC 
*** Bug 99963 has been marked as a duplicate of this bug. ***
Comment 3 Heath Harrelson 2002-12-05 17:16:09 UTC 
*** Bug 100438 has been marked as a duplicate of this bug. ***
Comment 4 Heath Harrelson 2002-12-06 01:00:02 UTC 
*** Bug 100484 has been marked as a duplicate of this bug. ***
Comment 5 Heath Harrelson 2002-12-06 20:57:18 UTC 
*** Bug 100543 has been marked as a duplicate of this bug. ***
Comment 6 Heath Harrelson 2002-12-08 21:21:07 UTC 
*** Bug 100690 has been marked as a duplicate of this bug. ***
Comment 7 Heath Harrelson 2002-12-08 22:04:47 UTC 
I finally decided to get some traces of what gnome-terminal is doing.  

First, here's what happens when I do Shft+Control+T when attached to
gnome-terminal with gdb:

Program received signal SIGPIPE, Broken pipe.
[Switching to Thread 1024 (LWP 1662)]
0x409d5754 in write () from /lib/libc.so.6
(gdb) thread apply all bt

+ Trace 31465

Thread 1 (Thread 1024 (LWP 1662))

  • #0 write
    from /lib/libc.so.6
  • #1 __DTOR_END__
    from /lib/libpthread.so.0
  • #2 _vte_pty_open_with_helper
    at pty.c line 852
  • #3 _vte_pty_open
    at pty.c line 956
  • #4 vte_terminal_fork_command
    at vte.c line 6397
  • #5 terminal_widget_fork_command
    at terminal-widget-vte.c line 494
  • #6 terminal_screen_launch_child
    at terminal-screen.c line 982
  • #7 new_tab_callback
    at terminal-window.c line 2043
  • #8 g_cclosure_marshal_VOID__VOID
    at gmarshal.c line 77
  • #9 g_closure_invoke
    at gclosure.c line 437
  • #10 signal_emit_unlocked_R
    at gsignal.c line 2814
  • #11 g_signal_emit_valist
    at gsignal.c line 2546
  • #12 g_signal_emit
    at gsignal.c line 2604
  • #13 gtk_widget_activate
    at gtkwidget.c line 3153
  • #14 gtk_menu_shell_activate_item
    at gtkmenushell.c line 941
  • #15 gtk_menu_shell_button_release
    at gtkmenushell.c line 557
  • #16 gtk_menu_button_release
    at gtkmenu.c line 1893
  • #17 _gtk_marshal_BOOLEAN__BOXED
    at gtkmarshalers.c line 82
  • #18 g_type_class_meta_marshal
    at gclosure.c line 514
  • #19 g_closure_invoke
    at gclosure.c line 437
  • #20 signal_emit_unlocked_R
    at gsignal.c line 2852
  • #21 g_signal_emit_valist
    at gsignal.c line 2556
  • #22 g_signal_emit
    at gsignal.c line 2604
  • #23 gtk_widget_event_internal
    at gtkwidget.c line 3121
  • #24 gtk_propagate_event
    at gtkmain.c line 2248
  • #25 gtk_main_do_event
    at gtkmain.c line 1497
  • #26 gdk_event_dispatch
    at gdkevents-x11.c line 1971
  • #27 g_main_dispatch
    at gmain.c line 1653
  • #28 g_main_context_dispatch
    at gmain.c line 2197
  • #29 g_main_context_iterate
    at gmain.c line 2278
  • #30 g_main_loop_run
    at gmain.c line 2498
  • #31 gtk_main
    at gtkmain.c line 1091
  • #32 main
    at terminal.c line 1322
  • #33 __libc_start_main
    from /lib/libc.so.6 
Hopefully this helps.

I'm bumping this bug up to high priority since it's getting so many
duplicates.
Comment 8 Tuomas Jormola 2002-12-29 09:19:35 UTC 
As the original reporter I'm happy to say that this bug seems to be
vanished by some change recently. I compiled a fresh GNOME 2.2
snapshot on the 27th of December and gnome-terminal is now working as
expected. Probably improved vte helped here. Bug #100660 is also
fixed. I guess you maintainers can close both bugs now.
Comment 9 Nalin Dahyabhai 2003-01-03 22:50:43 UTC 
Closing.
Comment 10 Heath Harrelson 2003-01-07 02:57:53 UTC 
*** Bug 102713 has been marked as a duplicate of this bug. ***