Re: pkg/56469 (openjdk11 keytool crashes on pkgsrc build on NetBSD 9.99.88/evbarm (aarch64))

[Alexander Schreiber <als%thangorodrim.ch@localhost>]

On Thu, Nov 04, 2021 at 11:14:39AM +0100, Benny Siegert wrote:
> If memory is the problem, does the issue also occur when you build
> with MAKE_JOBS=1?

Yes, it does. Set MAKE_JOBS=1 and retried on the Raspberry Pi4. Free
RAM never dropped below 2G, with 3G in buffer cache on a 8G machine.

And I just tried something: invoking the keytool binary without any
arguments, where the expected result is a complaint along the lines
of "Hey, dummy, you need to provide arguments foo, bar and baz!".

Nope, not what I got:

/usr/pkgsrc/lang/openjdk11/work.aarch64/bootstrap/bin/keytool
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x0000fc751cc3afb8, pid=16341, tid=277579914009600
#
# JRE version: OpenJDK Runtime Environment (11.0.7) (build 11.0.7-internal+0-adhoc.pkgsrc.openjdk-jdk11u-jdk-11.0.7-10-2)
# Java VM: OpenJDK 64-Bit Server VM (11.0.7-internal+0-adhoc.pkgsrc.openjdk-jdk11u-jdk-11.0.7-10-2, mixed mode, tiered, compressed oops, g1 gc, bsd-aarch64)
# Problematic frame:
# V  [libjvm.so+0x7eafb8]  JVM_RaiseSignal+0x15bc40
#
# Core dump will be written. Default location: /usr/pkgsrc/lang/openjdk11/keytool.core
#
# An error report file with more information is saved as:
# /usr/pkgsrc/lang/openjdk11/hs_err_pid16341.log
#
# Compiler replay data is saved as:
# /usr/pkgsrc/lang/openjdk11/replay_pid16341.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
#
[1]   Abort trap (core dumped) /usr/pkgsrc/lang/openjdk11/work.aarch64/bootstrap/bin/keytool


Running it under /usr/bin/time yields:

/usr/bin/time -l /usr/pkgsrc/lang/openjdk11/work.aarch64/bootstrap/bin/keytool
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x0000fbaf4423afb8, pid=19696, tid=276730170371072
#
# JRE version: OpenJDK Runtime Environment (11.0.7) (build 11.0.7-internal+0-adhoc.pkgsrc.openjdk-jdk11u-jdk-11.0.7-10-2)
# Java VM: OpenJDK 64-Bit Server VM (11.0.7-internal+0-adhoc.pkgsrc.openjdk-jdk11u-jdk-11.0.7-10-2, mixed mode, tiered, compressed oops, g1 gc, bsd-aarch64)
# Problematic frame:
# V  [libjvm.so+0x7eafb8]  JVM_RaiseSignal+0x15bc40
#
# Core dump will be written. Default location: /usr/pkgsrc/lang/openjdk11/keytool.core
#
# An error report file with more information is saved as:
# /usr/pkgsrc/lang/openjdk11/hs_err_pid19696.log
#
# Compiler replay data is saved as:
# /usr/pkgsrc/lang/openjdk11/replay_pid19696.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
#
time: Command terminated abnormally.
        0.21 real         0.16 user         0.14 sys
     22384  maximum resident set size
         2  average shared memory size
         0  average unshared data size
        69  average unshared stack size
      3014  page reclaims
        16  page faults
         0  swaps
         0  block input operations
        17  block output operations
         0  messages sent
         0  messages received
        26  signals received
       247  voluntary context switches
        10  involuntary context switches

So it looks very much not like a "not enough memory" issue.

machine: Raspberry Pi4 8 GB
system info:
NetBSD legolas 9.99.91 NetBSD 9.99.91 (GENERIC64) #0: Sun Oct 17 04:14:49 UTC 2021  mkrepro%mkrepro.NetBSD.org@localhost:/usr/src/sys/arch/evbarm/compile/GENERIC64 evbarm

pkgsrc: current from about 1 week ago or so.

This keytool invocation produced a core file (duh) and two log files:
hs_err_pid17081.log and replay_pid17081.log.

I can drop these somewhere reachable by http if desired (learned the hard
way that gnats doesn't deal with multipart, sorry for the resulting mess).

Happy to provide whatever info/testing I can.

Kind regards,
            Alex.
-- 
"Opportunity is missed by most people because it is dressed in overalls and
 looks like work."                                      -- Thomas A. Edison