Bug 348963 - kdeinit5 crash on "Special Window Settings"
Summary: kdeinit5 crash on "Special Window Settings"
Status: RESOLVED DUPLICATE of bug 348472
Alias: None
Product: kwin
Classification: Plasma
Component: general (show other bugs)
Version: 5.3.1
Platform: Kubuntu Linux
: NOR normal
Target Milestone: ---
Assignee: KWin default assignee
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-06-10 10:35 UTC by Greg Lepore
Modified: 2015-06-18 12:03 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments
No reporting address for kdeinit5 (34.11 KB, image/jpeg)
2015-06-10 12:59 UTC, Greg Lepore
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Greg Lepore 2015-06-10 10:35:53 UTC
I get a kdeinit5 crash when attempting to set "Special Window Settings". Crash is 100% on all apps that I've tried (both KDE and other.) Appears similar to fixed but 346486 and current bug 348849, but with a different stack trace.

Reproducible: Always

Steps to Reproduce:
1. Right click on icon, select More Actions
2. Select Special Window Settings
3. Crash

Actual Results:  
Crash

Expected Results:  
Allowed to set window settings

Application: kdeinit5 (kdeinit5), signal: Segmentation fault
Using host libthread_db library "/lib/i386-linux-gnu/libthread_db.so.1".
[Current thread is 1 (Thread 0xb1ac89c0 (LWP 12552))]

Thread 2 (Thread 0xaf8d7b40 (LWP 12553)):
#0  0xb775abe0 in __kernel_vsyscall ()
#1  0xb69903db in poll () at ../sysdeps/unix/syscall-template.S:81
#2  0xb5e1a3a8 in poll (__timeout=-1, __nfds=1, __fds=0xaf8d721c) at /usr/include/i386-linux-gnu/bits/poll2.h:46
#3  _xcb_conn_wait (c=c@entry=0x95d8430, cond=cond@entry=0x95d8454, vector=vector@entry=0x0, count=count@entry=0x0) at ../../src/xcb_conn.c:447
#4  0xb5e1c394 in xcb_wait_for_event (c=0x95d8430) at ../../src/xcb_in.c:622
#5  0xafa4c7d3 in QXcbEventReader::run (this=0x95e4858) at qxcbconnection.cpp:1105
#6  0xb6af67eb in QThreadPrivate::start (arg=0x95e4858) at thread/qthread_unix.cpp:337
#7  0xb5d721aa in start_thread (arg=0xaf8d7b40) at pthread_create.c:333
#8  0xb699afde in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:122

Thread 1 (Thread 0xb1ac89c0 (LWP 12552)):
[KCrash Handler]
#7  xcb_send_request (c=c@entry=0x0, flags=flags@entry=1, vector=vector@entry=0xbf8dab20, req=req@entry=0xb5e2ae90 <xcb_req>) at ../../src/xcb_out.c:160
#8  0xb5e20f7a in xcb_get_property (c=0x0, _delete=0 '\000', window=56623105, property=36, type=31, long_offset=0, long_length=100000) at xproto.c:3901
#9  0xb7173f60 in NETWinInfo::update (this=0xbf8db0a0, dirtyProperties=..., dirtyProperties2=...) at ../../src/netwm.cpp:4161
#10 0xb717655b in NETWinInfo::NETWinInfo (this=0xbf8db0a0, connection=0x0, window=56623105, rootWindow=0, properties=..., properties2=..., role=NET::Client) at ../../src/netwm.cpp:2814
#11 0xafbc03fc in KWin::ClientMachine::resolve (this=0xbf8db18c, window=56623105, clientLeader=0) at ../../../kcmkwin/kwinrules/../../client_machine.cpp:179
#12 0xafb9a185 in findRule (whole_app=<optimized out>, wid=<optimized out>, rules=...) at ../../../kcmkwin/kwinrules/main.cpp:77
#13 edit (whole_app=<optimized out>, wid=<optimized out>) at ../../../kcmkwin/kwinrules/main.cpp:212
#14 kdemain (argc=3, argv=0x95bb340) at ../../../kcmkwin/kwinrules/main.cpp:266
#15 0x0804f679 in launch (argc=2, argc@entry=3, _name=0xbf8dac20 "\320%c\tp\254\215\277X\254\215\277\n", _name@entry=0x956e0dc "kwin_rules_dialog", args=0xafb99690 <kdemain(int, char**)> "UWVS\350\267\376\377\377\201\303g9\005", cwd=0x0, envc=0, envs=0xbf8db254 "H_\372\266\240\016[\t\b\301[\t\001\304Y\t", reset_env=false, tty=0x0, avoid_loops=false, startup_id_str=0x80524aa "0") at ../../../src/kdeinit/kinit.cpp:741
#16 0x08050f53 in handle_launcher_request (sock=8, who=<optimized out>) at ../../../src/kdeinit/kinit.cpp:1180
#17 0x080514b1 in handle_requests (waitForPid=2, waitForPid@entry=0) at ../../../src/kdeinit/kinit.cpp:1369
#18 0x0804bf12 in main (argc=5, argv=0xbf8db974) at ../../../src/kdeinit/kinit.cpp:1860
Comment 1 Greg Lepore 2015-06-10 10:37:03 UTC
Also, should it be a separate bug or wish that kdeinit5 should provide a bug reporting address? It doesn't, at least according to the crash dialog.
Comment 2 Thomas Lübking 2015-06-10 12:54:37 UTC
What do you mean by "should provide a bug reporting address"? Like "https://bugs.kde.org"?
(I assume kdeinit5 cannot be mapped to the proper component, because the process that crashes is a generic library loader)

*** This bug has been marked as a duplicate of bug 348472 ***
Comment 3 Greg Lepore 2015-06-10 12:59:14 UTC
Created attachment 93105 [details]
No reporting address for kdeinit5
Comment 4 Greg Lepore 2015-06-10 12:59:45 UTC
I missed Bug 348472. As for the bug reporting address, see the attachment for the crash dialog.
Comment 5 Thomas Lübking 2015-06-10 13:20:52 UTC
Yupp, the problem here is that kdeinit doesn't know that what it launched is somehow related to kwin - nor does then DrKonqi have a chance to figure that.

The solution would be to extend KToolInvocation::kdeinitExec with a parameter "QString bugComponent" which it would then wire down to kdeinit5 so that it can "somehow" (no idea how it works exactly) feed that into DrKonqi (or we allow to report crashes against kdeinit and hire some dispatcher monkey ;-)


Btw, you can still create/edit rules running "kcmshell5 kwinrules".
Comment 6 Greg Lepore 2015-06-10 13:28:07 UTC
Unfortunately my problems with window rules are pretty extensive. For example, I currently have no rules set, but Kate insists on opening non-maximized and to the left of the screen. Setting a rule via kcmshell5 kwinrules doesn't have any effect. Nor does maximizing, then closing and re-opening. Ditto for Gwenview. I'll search for bugs on it... It may have something to do with the window state when a plasma crash occurred.
Comment 7 Thomas Lübking 2015-06-10 13:41:01 UTC
As the bug is dead-as-dupe anyway, feel free to export and attach the failing rules.
Atm., I've no problem creating rules for at least kate.
Comment 8 Greg Lepore 2015-06-10 13:51:58 UTC
Of course it's working now! I'll keep an eye on it and report a bug if the misbehavior reappears.