GTK warnings printed by Transmission

[gnu]

I'm running 1.50b3 (7776) on Ubuntu Hardy, using the GTK+ GUI, on a T1 line, with 24 ipv4 torrents seeding, and one or two IPv6 torrents downloading or seeding.

I have a Properties window open, set to the Peers display.

On the terminal window where I started transmission, I just got two messages like this:

(transmission:30319): Gtk-WARNING : A floating object was finalized. This means that someone called g_object_unref() on an object that had only a floating reference; the initial floating reference is not owned by anyone and must be removed with g_object_ref_sink().

A third copy just came out as I exposed the main window; I don't know if that triggered it. A fourth copy came out a few minutes later, with no obvious trigger.

There's another warning message that came out dozens of times, when picking files in the Add dialog earlier:

(transmission:30319): atk-bridge-WARNING : failure: no device event controller found.

[charles]

I'm not seeing any of these errors. could you run Transmission in gdb (and tell it to ignore SIGPIPE), set a breakpoint for g_log, and get a backtrace for them?

[gnu]

OK, here's what happened so far:

(gdb) handle SIGPIPE nostop noprint pass

Signal Stop Print Pass to program Description

SIGPIPE No No Yes Broken pipe

(gdb) b g_log

Breakpoint 2 at 0xb7998e66

(gdb) i b

Num Type Disp Enb Address What

1 breakpoint keep y 0x0806833f <main+17>

breakpoint already hit 1 time

2 breakpoint keep y 0xb7998e66 <g_log+6>

(gdb) c

Continuing.

[New Thread 0xb5d2fb90 (LWP 4597)]

Breakpoint 2, 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

(gdb) bt

#0 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

#1 0xb6d990d5 in spi_atk_bridge_key_listener (event=0x8c36e38, data=0x0)

at bridge.c:473

#2 0xb6d58d18 in notify_hf (key=0x0, value=0xb6d98f60, data=0x8d18948)

at gailutil.c:270

#3 0xb7982f55 in ?? () from /usr/lib/libglib-2.0.so.0

#4 0xb6d59257 in gail_key_snooper (the_widget=0x8caa020, event=0x81cd868,

func_data=0x0) at gailutil.c:292

#5 0xb7c43f3c in IAgtk_main_do_event (event=0x81cd868)

at /build/buildd/gtk+2.0-2.12.9/gtk/gtkmain.c:1852

#6 0xb7abca9a in gdk_event_dispatch (source=0x810d5b8, callback=0,

user_data=0x0) at /build/buildd/gtk+2.0-2.12.9/gdk/x11/gdkevents-x11.c:2351

#7 0xb798fbf8 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0

#8 0xb7992e5e in ?? () from /usr/lib/libglib-2.0.so.0

#9 0xb79931e7 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0

#10 0xb7c44264 in IAgtk_main ()

at /build/buildd/gtk+2.0-2.12.9/gtk/gtkmain.c:1163

#11 0x0806889f in main ()

(gdb) finish

Run till exit from #0 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

(transmission:4594): atk-bridge-WARNING : failure: no device event controller found.

spi_atk_bridge_key_listener (event=0x8c36e38, data=0x0) at bridge.c:474

474 bridge.c: No such file or directory.

in bridge.c

This happened while I had a Properties window up and was trying to limit the download speed of the torrent.

[gnu]

Oh, sweet! I got the other one, too. After about six of the atk-bridge warning, it printed the other warning about floating pointers. Luckily I got it twice, since I'd continued past it the first time:

Breakpoint 2, 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

(gdb) c

Continuing.

(transmission:4594): atk-bridge-WARNING : failure: no device event controller found.

Breakpoint 2, 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

(gdb) c

Continuing.

(transmission:4594): Gtk-WARNING : A floating object was finalized. This means that someone called g_object_unref() on an object that had only a floating reference; the initial floating reference is not owned by anyone and must be removed with g_object_ref_sink().

Breakpoint 2, 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

(gdb) bt

#0 0xb7998e66 in g_log () from /usr/lib/libglib-2.0.so.0

#1 0xb7c6a5cb in gtk_object_finalize (gobject=0x8f239f0)

at /build/buildd/gtk+2.0-2.12.9/gtk/gtkobject.c:439

#2 0xb7b81180 in gtk_cell_renderer_text_finalize (object=0x8f239f0)

at /build/buildd/gtk+2.0-2.12.9/gtk/gtkcellrenderertext.c:606

#3 0xb7a158db in g_object_unref () from /usr/lib/libgobject-2.0.so.0

#4 0xb6d4810a in gail_renderer_cell_finalize (object=0x818d170)

at gailrenderercell.c:81

#5 0xb6d4c392 in gail_text_cell_finalize (object=0x818d170)

at gailtextcell.c:226

#6 0xb7a158db in g_object_unref () from /usr/lib/libgobject-2.0.so.0

#7 0xb6ceaace in spi_base_object_dispose (gobject=0x8f33200) at base.c:46

#8 0xb7a1583c in g_object_unref () from /usr/lib/libgobject-2.0.so.0

#9 0xb6c80a77 in bonobo_object_unref (obj=0x8f33200) at bonobo-object.c:201

#10 0xb6d98c07 in spi_atk_bridge_signal_listener (signal_hint=0xbfbf1e9c,

n_param_values=3, param_values=0xbfbf1f94, data=0x81368b0) at bridge.c:1218

#11 0xb7a27b20 in ?? () from /usr/lib/libgobject-2.0.so.0

#12 0xb7a29916 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0

#13 0xb7a29bd2 in g_signal_emit_by_name () from /usr/lib/libgobject-2.0.so.0

#14 0xb6d544f1 in model_row_inserted (tree_model=0x8ed78c0, path=0x8ba4980,

iter=0x8d70e90, user_data=0x8240bb0) at gailtreeview.c:2977

#15 0xb7c47d61 in _gtk_marshal_VOIDBOXED_BOXED (closure=0x8ba7978,

return_value=0x0, n_param_values=3, param_values=0xbfbf2544, invocation_hint=0xbfbf244c, marshal_data=0xb6d543a0) at /build/buildd/gtk+2.0-2.12.9/gtk/gtkmarshalers.c:1432

#16 0xb7a13759 in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0

#17 0xb7a280a3 in ?? () from /usr/lib/libgobject-2.0.so.0

#18 0xb7a29916 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0

#19 0xb7a29c59 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0

#20 0xb7d28b09 in IAgtk_tree_model_row_inserted (tree_model=0x8ed78c0,

path=0x8ba4980, iter=0xbfbf280c) at /build/buildd/gtk+2.0-2.12.9/gtk/gtktreemodel.c:1496

#21 0xb7d31d34 in gtk_tree_model_sort_row_inserted (s_model=0x8b9cb50,

s_path=0x8c29e68, s_iter=0x8c85110, data=0x8ed78c0) at /build/buildd/gtk+2.0-2.12.9/gtk/gtktreemodelsort.c:686

#22 0xb7c47d61 in _gtk_marshal_VOIDBOXED_BOXED (closure=0x8d26b70,

return_value=0x0, n_param_values=3, param_values=0xbfbf2b44, invocation_hint=0xbfbf2a4c, marshal_data=0xb7d31b60) at /build/buildd/gtk+2.0-2.12.9/gtk/gtkmarshalers.c:1432

#23 0xb7a13759 in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0

#24 0xb7a27d1d in ?? () from /usr/lib/libgobject-2.0.so.0

#25 0xb7a29916 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0

#26 0xb7a29c59 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0

#27 0xb7d28b09 in IAgtk_tree_model_row_inserted (tree_model=0x8b9cb50,

path=0x8c29e68, iter=0xbfbf2e24) at /build/buildd/gtk+2.0-2.12.9/gtk/gtktreemodel.c:1496

#28 0xb7c40c6d in IAgtk_list_store_insert (list_store=0x8b9cb50,

iter=0xbfbf2e24, position=3) at /build/buildd/gtk+2.0-2.12.9/gtk/gtkliststore.c:1032

#29 0xb7c40d27 in IAgtk_list_store_append (list_store=0x8b9cb50,

iter=0xbfbf2e24) at /build/buildd/gtk+2.0-2.12.9/gtk/gtkliststore.c:1137

#30 0x08061654 in ?? ()

#31 0x08061d9e in ?? ()

#32 0x08064c3a in ?? ()

---Type <return> to continue, or q <return> to quit---

#33 0xb7990336 in ?? () from /usr/lib/libglib-2.0.so.0

#34 0xb798fbf8 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0

#35 0xb7992e5e in ?? () from /usr/lib/libglib-2.0.so.0

#36 0xb79931e7 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0

#37 0xb7c44264 in IAgtk_main ()

at /build/buildd/gtk+2.0-2.12.9/gtk/gtkmain.c:1163

#38 0x0806889f in main ()

(gdb)

I'll leave it hanging here for some hours; if you have further suggestions, let me know and I'll be glad to keep poking at this.

[gnu]

After some IRC discussion I switched to using the nightly deb, and also ran it under valgrind. I didn't see the messages that I got using the 1.50b3 prerelease. I'm appending the valgrind output, which reported a few uninitialized variables and such during startup, reported nothing while running overnight, and then reported lots of lost memory and such when I quit the GUI.

The "no device event controller" report also occurs in xchat, and I see references to it all over the Internet; it's presumably not Transmission's problem.

Uh, this stupid TRAC system won't let me upload the valgrind output, even after maximum compression, it only allows 256K attachments, but the output is 610K (28 MB uncompressed). Clue-less!

[charles]

This ticket is turning into quite a catch-all. :)

the uninitialized variables found by valgrind are from openssl's randomize functions and are intentional/unavoidable.

Likewise the lost memory at the end is from freetype and cairo, which is out of Transmission's hands.

If the "no device controller" error is occurring elsewhere too, as you say it's probably not a Transmission issue.

I'm glad that the nightly deb is running better for you than the previous builds. :)