Bug 1138175
| Summary: | [abrt] xscreensaver-extras: fireworkx_init(): fireworkx killed by SIGFPE | ||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Daniel Krämer <d.kraemer> | ||||||||||||||||||||||||||
| Component: | mate-screensaver | Assignee: | Dan Mashal <dan.mashal> | ||||||||||||||||||||||||||
| Status: | CLOSED WONTFIX | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||||||||||||||||||||||
| Severity: | unspecified | Docs Contact: | |||||||||||||||||||||||||||
| Priority: | unspecified | ||||||||||||||||||||||||||||
| Version: | 20 | CC: | cheese, dan.mashal, d.kraemer, mtasaka, raveit65.sun, stefano | ||||||||||||||||||||||||||
| Target Milestone: | --- | ||||||||||||||||||||||||||||
| Target Release: | --- | ||||||||||||||||||||||||||||
| Hardware: | x86_64 | ||||||||||||||||||||||||||||
| OS: | Unspecified | ||||||||||||||||||||||||||||
| URL: | https://retrace.fedoraproject.org/faf/reports/bthash/85bdc7ac590e116e912699eabf86ba366dbbc755 | ||||||||||||||||||||||||||||
| Whiteboard: | abrt_hash:f6c0e98e81d27b0d82e2cc7e2fdcfb474c405ba4 | ||||||||||||||||||||||||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||||||||||||||||||||||||
| Doc Text: | Story Points: | --- | |||||||||||||||||||||||||||
| Clone Of: | Environment: | ||||||||||||||||||||||||||||
| Last Closed: | 2014-10-04 15:21:59 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
Daniel Krämer
2014-09-04 07:55:30 UTC
Created attachment 934300 [details]
File: backtrace
Created attachment 934301 [details]
File: cgroup
Created attachment 934302 [details]
File: core_backtrace
Created attachment 934303 [details]
File: dso_list
Created attachment 934304 [details]
File: environ
Created attachment 934305 [details]
File: limits
Created attachment 934306 [details]
File: maps
Created attachment 934307 [details]
File: open_fds
Created attachment 934308 [details]
File: proc_pid_status
Created attachment 934309 [details]
File: var_log_messages
* Would you write in detail how to reproduce this issue? * Also please provide the information about the environment (desktop environment, machine, etc) when this issue happened. Created attachment 934376 [details]
Output of lshw
Created attachment 934377 [details]
Output of lspci
Fedora 20 Mate Spin Reproduce: - Open Control Center - Open Screensaver-GUI - Select Fireworkx - Close Screensaver-GUI - Open Screensaver-GUI again - Bam! Error Thank you. Then please provide the additional information. * So you are using "mate-screensaver-preferences", not "xscreensaver-demo"? * How is this issue reproducible? If reproducible, would you attach the screenshot of "Screensaver-GUI"? I use mate-screensaver-preferences. It is reproducible: - Open mate-screensaver-preferences - Select Fireworkx - Close mate-screensaver-preferences - Open mate-screensaver-preferences again - Bam! Error Screenshot? Its just the GUI with a black demo-screen Switching to mate-screensaver. When xscreensaver-demo launches xscreensaver hack to make it displayed on demo window, xscreensaver-demo passes "--window-id <ID>" as gained by GDK_WINDOW_XWINDOW() to the argument of the hack. For mate-desktop-properties, perhaps tweaking spawn_on_window() is needed (just my guess!!!). Perhaps making "#if GTK_CHECK_VERSION (3, 0, 0)" branch enable (if possible) to make it effective envp as created by spawn_make_environment_for_screen() is sufficient??? Oops... Currently the biggest problem is that as shown in gdb log, XGetWindowAttributes() returns window size of (1,1), which is almost the direct reason for this SIGFPE. The complete Mate stack is compiled with GTK2, so if Fireworkx-ss needs GTK3 code it's impossible to use it in a GUI which use GTK2. A "#if GTK_CHECK_VERSION (3, 0, 0)" check works only if mate-screensaver is compiled for GTK3, otherwise this will be ignored. Maybe we're switching to GTK3 with f22/23, the current GTK3 state of Mate isn't ready for produktion. So, there is nothing what i can do for the momment. Do you agree to close or reassign it back? Well, no, fireworkx is one seperated program and it does not matter whether mate-screensaver is compiled against GTK2 or GTK3. Well, anyway http://stackoverflow.com/questions/10264625/how-do-i-get-the-gdk-window-for-a-gtk-window-in-c says that usage "GTK_WINDOW (window)->screen" is "disallowed" on gtk 2.14 and it seems that GTK2 side code needs fixing. Note that I just looked at mate-screensaver code, not actually having executed so if someone can examine - where mate-screensaver actually forks screensaver hacks - and at that point what does XGetWindowAttributes returns it is much appreciated. Ok, but reporter said it crashs when it runs inside the gui into the preview window. But i'm not C coder and i've currently to much work with maintaining the whole mate stack alone in fedora. For me this is something for mate upstream. Dear supporter, pls open a report at upstream for yourself and include all informations given from Mamoru + backtrace there. https://github.com/mate-desktop/mate-screensaver Due my lack of time i can't do that for external progs. Thank you (Removing needinfo on me per comment 20 .. forgot to remove... I don't really know what piece of software produces this issue and i don't have a clue what you guys are talking at all. I'm not very into those things. I just noticed abrt with the crash and pushed the button to report it. And for me the crashed screensaver-demo is not worth it to put more effort into than i already had. When you guys are not interessted to make your software working (without bugs) for consumers like me, then just close the bug report... If no one wants to forward the issue to upstream than the report is pretty useless. |