Bugzilla – Full Text Bug Listing
| Summary: | Eclipse crash | ||
|---|---|---|---|
| Product: | [Eclipse Project] Platform | Reporter: | Roman Galaktionov <3draven> |
| Component: | SWT | Assignee: | Platform-SWT-Inbox <platform-swt-inbox> |
| Status: | RESOLVED WORKSFORME | QA Contact: | |
| Severity: | normal | ||
| Priority: | P3 | CC: | Alexander.Mitin, clayberg, ericwill, tom.schindl |
| Version: | 4.4 | Keywords: | triaged |
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Attachments: | |||
|
Description
Roman Galaktionov
I see a lot of trouble in the log file but I don't see how JDT could cause any of this. Moving to e(fx)clipse for comment. Could also be an Xtext issue, or ... X.org server segfault or close only when i work in eclipse+javafx+swing (see example app). Sad but true. Wow! I download eclipse Version: 4.4.0 Build id: N20131005-1500 no plugins installed! Open project (example in first attach) Run project. x.org server segfault! 2215.595] (EE) [ 2215.595] (EE) Backtrace: [ 2215.595] (EE) 0: /usr/bin/X (xorg_backtrace+0x36) [0x7fc3109c2516] [ 2215.595] (EE) 1: /usr/bin/X (0x7fc310812000+0x1b4359) [0x7fc3109c6359] [ 2215.595] (EE) 2: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7fc30f915000+0xfbd0) [0x7fc30f924bd0] [ 2215.595] (EE) 3: /usr/lib/x86_64-linux-gnu/libpixman-1.so.0 (pixman_region_intersect+0x24) [0x7fc30f4c7034] [ 2215.595] (EE) 4: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7fc30d0a6000+0x5667f) [0x7fc30d0fc67f] [ 2215.596] (EE) 5: /usr/bin/X (0x7fc310812000+0x132d35) [0x7fc310944d35] [ 2215.596] (EE) 6: /usr/bin/X (0x7fc310812000+0x58af1) [0x7fc31086aaf1] [ 2215.596] (EE) 7: /usr/bin/X (0x7fc310812000+0x475aa) [0x7fc3108595aa] [ 2215.596] (EE) 8: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xf5) [0x7fc30e561ea5] [ 2215.596] (EE) 9: /usr/bin/X (0x7fc310812000+0x478f1) [0x7fc3108598f1] [ 2215.596] (EE) [ 2215.596] (EE) Segmentation fault at address 0x6e00000063 [ 2215.596] Fatal server error: [ 2215.596] Caught signal 11 (Segmentation fault). Server aborting [ 2215.596] [ 2215.596] (EE) Please change bug to eclipse bug. Created attachment 236159 [details]
xorg server segfault log
My notebook with "Optimus" (hybrid APU+GPU video system Nvidia GEForce 640m+Intel HD 4000). If i run ./eclipse then crash x server with segfault If i run optirun ./eclipse (its bumblebee project optirun utility) with 325 nvidia driver, then i recieve: # # A fatal error has been detected by the Java Runtime Environment: # # SIGSEGV (0xb) at pc=0x0000000000000000, pid=4011, tid=140479057688320 # # JRE version: Java(TM) SE Runtime Environment (7.0_40-b43) (build 1.7.0_40-b43) # Java VM: Java HotSpot(TM) 64-Bit Server VM (24.0-b56 mixed mode linux-amd64 compressed oops) # Problematic frame: # C 0x0000000000000000 # # Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again # # An error report file with more information is saved as: # /mnt/data/Personal/Projects/Sleep/SleepGraph/hs_err_pid4011.log # # If you would like to submit a bug report, please visit: # http://bugreport.sun.com/bugreport/crash.jsp # The crash happened outside the Java Virtual Machine in native code. # See problematic frame for where to report the bug. # In attachment full log files (several). Created attachment 236182 [details]
full jvm crash logs
(In reply to Roman Galaktionov from comment #6) > My notebook with "Optimus" (hybrid APU+GPU video system Nvidia GEForce > 640m+Intel HD 4000). > If i run ./eclipse then crash x server with segfault > If i run optirun ./eclipse (its bumblebee project optirun utility) with 325 > nvidia driver, then i recieve: > > # > # A fatal error has been detected by the Java Runtime Environment: > # > # SIGSEGV (0xb) at pc=0x0000000000000000, pid=4011, tid=140479057688320 > # > # JRE version: Java(TM) SE Runtime Environment (7.0_40-b43) (build > 1.7.0_40-b43) > # Java VM: Java HotSpot(TM) 64-Bit Server VM (24.0-b56 mixed mode > linux-amd64 compressed oops) > # Problematic frame: > # C 0x0000000000000000 > # > # Failed to write core dump. Core dumps have been disabled. To enable core > dumping, try "ulimit -c unlimited" before starting Java again > # > # An error report file with more information is saved as: > # /mnt/data/Personal/Projects/Sleep/SleepGraph/hs_err_pid4011.log > # > # If you would like to submit a bug report, please visit: > # http://bugreport.sun.com/bugreport/crash.jsp > # The crash happened outside the Java Virtual Machine in native code. > # See problematic frame for where to report the bug. > # > > > In attachment full log files (several). Sorry. If i run ./eclipse and then run javafx project in eclipse -> many xorg server segfault and If i run optirun ./eclipse and then run javafx project in eclipse -> sigserv segfault jvm It's not a bug, but a few bugs. 1. WindowBuilder not parse javafx swing app with JFXPanel 2. JavaFX project crash xorg server 3. JavaFX project run on optimus (bumblebee optirun) crash jvm (In reply to Roman Galaktionov from comment #7) > Created attachment 236182 [details] > full jvm crash logs The stacktrace in the logs show that the crash happens in JavaFx, not in eclipse. I think the bug has to be logged on JavaFx. Yes, I have come to the same conclusion. Has already sent a bug in Oracle. But. WindowBuilder still does not work. I check java 8 on same bugs. (In reply to Roman Galaktionov from comment #10) > Yes, I have come to the same conclusion. Has already sent a bug in Oracle. > But. WindowBuilder still does not work. Moving bug to WindowBuilder It seems Java 8 does not cause segfolts! 1. i run project.jar. It work fine. 2. i run optirun project.jar It work fine. 3. WindowBuilder crash eclipse when parse swing app with JFXPanel Eclipse crash log in new attachment. Created attachment 236212 [details]
next start log when windowbuilder crash eclipse (silent crash, no logs)
I don't see what this has to do with WindowBuilder, and I don't see anything WB related in any of the logs. If JavaFX has issues running in Eclipse, it certainly won't work with WB either. When WB opens a file for edit, all of the widgets are instantiated in order to get the WYSIWYG visual presentation. If those widgets have problems and throw exceptions, WB will trap those exceptions and show an appropriate error to the user. If those widgets actually crash the system, there is nothing we can do other than suggest not using those components in the future until the underlying problem is fixed. I don't see anything here for WB to "fix" or even investigate further. I don't think we can do anything to make things work, at least not e(fx)clipse which is not involved at all. When 1. I run eclipse 2. Open swing project with JFXPanel 3. Open "design" tab in windowbuilder 4. Eclipse show progress windowbuilder "parsing"->"refreshing"->crash In apport (ubuntu crash report system): ERROR: apport (pid 23135) Tue Oct 8 21:52:12 2013: called for pid 23072, signal 6, core limit 0 ERROR: apport (pid 23135) Tue Oct 8 21:52:12 2013: executable: /usr/lib/jvm/java-8-oracle/jre/bin/java (command line "/usr/bin/java -Dosgi.requiredJavaVersion=1.7 -XX:MaxPermSize=1024m -Xms256m -Xmx2048m -jar /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar -os linux -ws gtk -arch x86_64 -showsplash /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.platform_4.3.1.v20130911-1000/splash.bmp -launcher /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64/eclipse -name Eclipse --launcher.library /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.200.v20130807-1835/eclipse_1506.so -startup /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar --launcher.appendVmargs -exitdata 3d2001f -product org.eclipse.epp.package.java.product -vm /usr/bin/java -vmargs -Dosgi.requiredJavaVersion=1.7 -XX:MaxPermSize=1024m -Xms256m -Xmx2048m -jar /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar") ERROR: apport (pid 23135) Tue Oct 8 21:52:12 2013: debug: session gdbus call: (true,) ERROR: apport (pid 23135) Tue Oct 8 21:57:16 2013: wrote report /var/crash/_usr_lib_jvm_java-8-oracle_jre_bin_java.1000.crash I add new attachment: _usr_lib_jvm_java-8-oracle_jre_bin_java.1000.crash without crash dump Created attachment 236221 [details]
eclipse crash log
In java 7 same crash eclipse Created attachment 236222 [details]
eclipse crash with java 7
I agree with Eric: the problem is not related to WB and we can do nothing if JavaFX crashes Eclipse. The only thing I can suggest you is to try to uninstall bumblebee (since it is not very stable yet and I've heard about many issues with it and even cancelled to use Linux on my dual-GPU system). Created attachment 236223 [details]
Fully work project swing+JFXPanel ->crash windowbuilder+eclipse
In java 8 javafx NOT crash jvm or eclipse. Full and fine work! In every eclipse+windowbuilder->swing+JFXPanel->crash eclipse. Now I do not use bumblebee. Eclipse+windowBuilder+Swing+JFXPanel->Crash. In ubuntu linux: 1. In java 8 javafx work fine. 2. WindowBuilder crash when parse+refreshing swing+JFXPanel code. In windows XP: 1. I download Eclipse for windows. 2. setup java 7 3. Open same project (in last attacment). 4. Open WindowBuilder Design Tab. 5. WindowBuilder work fine I loaded up WB source code and attempted to reproduce the issue using provided sample code. I wasn't able to get it crashes, but I can see a bunch of "invalid thread access" messages: "Exception in thread "JavaFX Application Thread" org.eclipse.swt.SWTException: Invalid thread access" (see attachment for JavaFX thread dump). I think this is incompatibility between SWT & JavaFX. Created attachment 236235 [details]
Invalid thread access message
May be reopen bug? It might be worth having the SWT team comment. JavaFX and SWT share the same event-thread => Swing and JavaFX/SWT don't - I can only repeat this has nothing to do with e(fx)clipse, not sure where this bug should get move too, it clearly does not belong into our bucket. I just do not know where to move it. If you have any suggestions, move. Possibly Platform > SWT...? Ok I do not know, the messages in the SWT team will come right after the redirect bug, or need more messages in the comments? What is no reaction from them no. https://bugs.freedesktop.org/show_bug.cgi?id=70342 May be its intel driver bug. I install new intel driver (from freedesctop bug tracker). 1. Start eclipse 2. Open example attached project 3. Open windowbuilder design tab. 4. Eclipse crash. JVM terminated. Exit code=1 /usr/bin/java -Dosgi.requiredJavaVersion=1.7 -XX:MaxPermSize=1024m -Xms256m -Xmx2048m -jar /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar -os linux -ws gtk -arch x86_64 -showsplash /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.platform_4.3.1.v20130911-1000/splash.bmp -launcher /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64/eclipse -name Eclipse --launcher.library /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.200.v20130807-1835/eclipse_1506.so -startup /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar --launcher.appendVmargs -exitdata 6a000c -product org.eclipse.epp.package.java.product -vm /usr/bin/java -vmargs -Dosgi.requiredJavaVersion=1.7 -XX:MaxPermSize=1024m -Xms256m -Xmx2048m -jar /mnt/data/IDE/eclipse-java-kepler-R-linux-gtk-x86_64//plugins/org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar No other logs. Is this still reproducible? Not sure this is even an SWT bug, and no response from the reporter. Please file a new bug if the issue reproduces on GTK3.22 with a 4.8 I-build. |