JVM crash on Raspberry Pi 3

Using provided Oracle JDK on my new π

pi@raspberrypi:~/Chronicle-Queue $ uname -a
Linux raspberrypi 4.1.19-v7+ #858 SMP Tue Mar 15 15:56:00 GMT 2016 armv7l GNU/Linux
pi@raspberrypi:~/Chronicle-Queue $ java -version
java version "1.8.0_65"
Java(TM) SE Runtime Environment (build 1.8.0_65-b17)
Java HotSpot(TM) Client VM (build 25.65-b01, mixed mode)
pi@raspberrypi:~/Chronicle-Queue $ mvn -version
Apache Maven 3.3.9 (bb52d8502b132ec0a5a3f4c09453c07478323dc5; 2015-11-10T16:41:47+00:00)
Maven home: /home/pi/apache-maven-3.3.9
Java version: 1.8.0_65, vendor: Oracle Corporation
Java home: /usr/lib/jvm/jdk-8-oracle-arm32-vfp-hflt/jre
Default locale: en_GB, platform encoding: UTF-8
OS name: "linux", version: "4.1.19-v7+", arch: "arm", family: "unix"

Cloned using

git clone https://github.com/OpenHFT/Chronicle-Queue.git

Output :

pi@raspberrypi:~/Chronicle-Queue $ mvn clean install
[INFO] Scanning for projects...
[INFO]
[INFO] ------------------------------------------------------------------------
[INFO] Building OpenHFT/Chronicle-Queue 4.4.0-SNAPSHOT
[INFO] ------------------------------------------------------------------------
[INFO]
[INFO] --- maven-clean-plugin:2.5:clean (default-clean) @ chronicle-queue ---
[INFO] Deleting /home/pi/Chronicle-Queue/target
[INFO]
[INFO] --- maven-enforcer-plugin:1.4.1:enforce (enforce-no-snapshots) @ chronicle-queue ---
[INFO]
[INFO] --- maven-resources-plugin:2.7:resources (default-resources) @ chronicle-queue ---
[INFO] Using 'UTF-8' encoding to copy filtered resources.
[INFO] skip non existing resourceDirectory /home/pi/Chronicle-Queue/src/main/resources
[INFO]
[INFO] --- maven-compiler-plugin:3.3:compile (default-compile) @ chronicle-queue ---
[INFO] Changes detected - recompiling the module!
[INFO] Compiling 34 source files to /home/pi/Chronicle-Queue/target/classes
[INFO] /home/pi/Chronicle-Queue/src/main/java/net/openhft/chronicle/queue/impl/AbstractChronicleQueueBuilder.java: /home/pi/Chronicle-Queue/src/main/java/net/openhft/chronicle/queue/impl/AbstractChronicleQueueBuilder.java uses unchecked or unsafe operations.
[INFO] /home/pi/Chronicle-Queue/src/main/java/net/openhft/chronicle/queue/impl/AbstractChronicleQueueBuilder.java: Recompile with -Xlint:unchecked for details.
[INFO]
[INFO] --- maven-bundle-plugin:3.0.0:manifest (default) @ chronicle-queue ---
[INFO]
[INFO] --- maven-resources-plugin:2.7:testResources (default-testResources) @ chronicle-queue ---
[INFO] Using 'UTF-8' encoding to copy filtered resources.
[INFO] skip non existing resourceDirectory /home/pi/Chronicle-Queue/src/test/resources
[INFO]
[INFO] --- maven-compiler-plugin:3.3:testCompile (default-testCompile) @ chronicle-queue ---
[INFO] Changes detected - recompiling the module!
[INFO] Compiling 36 source files to /home/pi/Chronicle-Queue/target/test-classes
[INFO] /home/pi/Chronicle-Queue/src/test/java/net/openhft/chronicle/queue/impl/single/ToEndTest.java: /home/pi/Chronicle-Queue/src/test/java/net/openhft/chronicle/queue/impl/single/ToEndTest.java uses or overrides a deprecated API.
[INFO] /home/pi/Chronicle-Queue/src/test/java/net/openhft/chronicle/queue/impl/single/ToEndTest.java: Recompile with -Xlint:deprecation for details.
[INFO]
[INFO] --- maven-surefire-plugin:2.19:test (default-test) @ chronicle-queue ---

-------------------------------------------------------
 T E S T S
-------------------------------------------------------
Running net.openhft.chronicle.queue.LastIndexAppendedTest
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0x7) at pc=0x740ab5dc, pid=6450, tid=1982784608
#
# JRE version: Java(TM) SE Runtime Environment (8.0_65-b17) (build 1.8.0_65-b17)
# Java VM: Java HotSpot(TM) Client VM (25.65-b01 mixed mode linux-arm )
# Problematic frame:
# v  ~StubRoutines::atomic_store_long
#
# 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:
# /home/pi/Chronicle-Queue/hs_err_pid6450.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
#
Aborted

Results :

Tests run: 0, Failures: 0, Errors: 0, Skipped: 0

[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 20.233 s
[INFO] Finished at: 2016-05-12T09:48:21+00:00
[INFO] Final Memory: 19M/51M
[INFO] ------------------------------------------------------------------------
[ERROR] Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.19:test (default-test) on project chronicle-queue: Execution default-test of goal org.apache.maven.plugins:maven-surefire-plugin:2.19:test failed: The forked VM terminated without properly saying goodbye. VM crash or System.exit called?
[ERROR] Command was /bin/sh -c cd /home/pi/Chronicle-Queue && /usr/lib/jvm/jdk-8-oracle-arm32-vfp-hflt/jre/bin/java -jar /home/pi/Chronicle-Queue/target/surefire/surefirebooter8772016599844544191.jar /home/pi/Chronicle-Queue/target/surefire/surefire8678532781266343868tmp /home/pi/Chronicle-Queue/target/surefire/surefire_02798566282531128178tmp
[ERROR] -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/PluginExecutionException

The report : hs_err_pid6450.zip

May be an issue with the JVM…

[peter-lawrey]

I have changed the alignment for atomic long operations for queue in the latest snapshot. Changing it for int values would be much harder, but at least you shouldn’t get this error for atomic longs in future.

Peter.

On 12 May 2016 at 13:57, franz1981 [e-mail] wrote:

@peter-lawrey https://github.com/peter-lawrey In theory every memory operation with address on primitive type is eligible for this kind of issue, thus adding a proper check on Memory with a configuration flag could rise an exception during the tests (putXXX,getXXX,cas and wait-free getAndAdd etc etc)…

@glandais https://github.com/glandais ARMs are a NightARMare 😛 MMM…generally i do not encounter that issue because i perform on the rasp only aligned accesses…but “maybe”: http://lxr.free-electrons.com/source/arch/arm/mm/alignment.c http://url

/* 104 * ARMv6 and later CPUs can perform unaligned accesses for 105 * most single load and store instructions up to word size. 106 * LDM, STM, LDRD and STRD still need to be handled. 107 * 108 * Ignoring the alignment fault is not an option on these 109 * CPUs since we spin re-faulting the instruction without 110 * making any progress. 111 */

there are a lot of possibilities:

• “LDM, STM, LDRD and STRD still need to be handled”; i’ve to check what kind of java call issue them…
• “…spin re-faulting the instruction…”, not cool ( - WITH A LOT OF IMAGINATION: http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.faqs/ka15414.html http://url at Device Memory section)

Anyway, here: http://psy-lob-saw.blogspot.it/2013/01/direct-memory-alignment-in-java.html http://url you could find tests and info to understand the issue…

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/OpenHFT/Chronicle-Queue/issues/253#issuecomment-218748904

[glandais]

Thank for you responses

I’m getting the same problem on bytes (not in core but I guess atomic_store_long is never called).

Thanks franz1981 for your answer, I’m a bit disappointed, I don’t really like to recompile a kernel 😃