Re: [oe] [meta-java] Sporadic segfaults when compiling jamvm-native and classpath-native
On Mon, Nov 9, 2015 at 6:10 AM, Kaaria, Erkka wrote: > While the workaround would leave the underlying cause unfixed, it only > requires changing one line. Since the cacao-initial-native is only used > briefly during the build process, I think small workaround is better than > large scale changes to the old codebase. I also didn't notice any difference > in classpath-native or jamvm-native build times, so the performance impact is > negligible. > > On a related note, I ran the clean\recompile script over the weekend and saw > no segfaults, so I think the workaround works. So please cook a patch adding the workaround and add a proper comment in the recipe with FIXME so we are aware of it. If possible, consider the possibility of packaging the staging repository so we could use a newer release, after that, and likely drop patches :) -- Otavio Salvador O.S. Systems http://www.ossystems.com.brhttp://code.ossystems.com.br Mobile: +55 (53) 9981-7854Mobile: +1 (347) 903-9750 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] Sporadic segfaults when compiling jamvm-native and classpath-native
> From: openembedded-devel-boun...@lists.openembedded.org > [mailto:openembedded-devel-boun...@lists.openembedded.org] On > Behalf Of Otavio Salvador > Sent: Friday, November 6, 2015 4:40 PM > To: OpenEmbedded Devel List de...@lists.openembedded.org> > Subject: Re: [oe] [meta-java] Sporadic segfaults when compiling jamvm- > native and classpath-native > > On Fri, Nov 6, 2015 at 9:44 AM, Kaaria, Erkka > wrote: > > There seems to be something wrong with cacao-initial-native JIT compiler > that causes sporadic segfaults when compiling jamvm-native or classpath- > native. This is fairly rare, I estimate around 1% of the builds fail due to > segfault. I use 64 bit Ubuntu 15.04 with GCC 4.9.2 as build machine. > > Did you check if cacao-staging tree does not include a fix for it? > Hi, I briefly glanced at the cacao-staging commit history, but I didn't find any patches that would fix this segfault specifically. There are several patches that fix issues and race conditions in the jit compiler, for example https://bitbucket.org/cacaovm/cacao-staging/commits/7e6eef2b4c94 and https://bitbucket.org/cacaovm/cacao-staging/commits/56697326cf27. One of these patches may very well fix the segfault, but the problem is that the cacao version that is used in the bootstrapping process is positively ancient (0.98, released in 2007). The linked patches for example are patching files that are either substantially different from the ones in 0.98 or don't even exist in 0.98. Backporting any patches to would not be easy. While the workaround would leave the underlying cause unfixed, it only requires changing one line. Since the cacao-initial-native is only used briefly during the build process, I think small workaround is better than large scale changes to the old codebase. I also didn't notice any difference in classpath-native or jamvm-native build times, so the performance impact is negligible. On a related note, I ran the clean\recompile script over the weekend and saw no segfaults, so I think the workaround works. Best Regards, Erkka Kääriä - Intel Finland Oy Registered Address: PL 281, 00181 Helsinki Business Identity Code: 0357606 - 4 Domiciled in Helsinki This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java] Sporadic segfaults when compiling jamvm-native and classpath-native
On Fri, Nov 6, 2015 at 9:44 AM, Kaaria, Erkka wrote: > There seems to be something wrong with cacao-initial-native JIT compiler that > causes sporadic segfaults when compiling jamvm-native or classpath-native. > This is fairly rare, I estimate around 1% of the builds fail due to segfault. > I use 64 bit Ubuntu 15.04 with GCC 4.9.2 as build machine. Did you check if cacao-staging tree does not include a fix for it? -- Otavio Salvador O.S. Systems http://www.ossystems.com.brhttp://code.ossystems.com.br Mobile: +55 (53) 9981-7854Mobile: +1 (347) 903-9750 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-java] Sporadic segfaults when compiling jamvm-native and classpath-native
Hi everyone,
There seems to be something wrong with cacao-initial-native JIT compiler that
causes sporadic segfaults when compiling jamvm-native or classpath-native. This
is fairly rare, I estimate around 1% of the builds fail due to segfault. I use
64 bit Ubuntu 15.04 with GCC 4.9.2 as build machine.
Example do_compile log:
make[6]: Entering directory
'/build/tmp/work/x86_64-linux/jamvm-native/1.5.5+1.6.0-devel+gitAUTOINC+ebd11bde0a-r0/build/src/classlib/gnuclasspath/lib'
mkdir classes
LOG: [0x7f32c46a7700] We received a SIGSEGV and tried to handle it, but we
were
LOG: [0x7f32c46a7700] unable to find a Java method at:
LOG: [0x7f32c46a7700]
LOG: [0x7f32c46a7700] PC=0x00440c09
LOG: [0x7f32c46a7700]
LOG: [0x7f32c46a7700] Dumping the current stacktrace:
<>
LOG: [0x7f32c46a7700] Exiting...
Aborted (core dumped)
Makefile:662: recipe for target 'classes.zip' failed
Examination of the core dump in this particular case indicates that the
segfault is caused by null pointer dereference in builtin_new at builtin.c:764:
#5
#6 builtin_new (c=0x0) at builtin.c:764
#7 0x7f23badb5a5f in ?? ()
#8 0x in ?? ()
Gdb is unable to show the calling method, and judging by the fact that the old
instruction pointer does not match any shared library addresses, I think it
originates from JIT compiled code.
This http://c1.complang.tuwien.ac.at/pipermail/cacao/2011-March/001350.html
thread seems to indicate that there could be a race condition in cacao JIT
compiler when multiple Java threads are involved. I forced the ecj to only use
one thread and segfaulting seems to have stopped.
Given the non-deterministic nature of the bug, I am not sure if I actually
managed to fix this or if I have just been lucky. As such, I request help to
a) Verify the presence of the bug
b) Verify that forcing ecj to only use single thread fixes the issue.
Multihreading can be disabled by adding -Djdt.compiler.useSingleThread=true to
the third line in "ecj-initial" script in build/tmp/sysroots//usr/bin/
example line:
${RUNTIME} -Xmx512m -Djdt.compiler.useSingleThread=true -cp ${ECJ_JAR}
org.eclipse.jdt.internal.compiler.batch.Main ${1+"$@"}
Script I use for the recompilation. I usually get a failure within 90 minutes,
but this naturally varies between computers. Run from the build folder
#!/bin/bash
while [ true ]; do
echo $(date)
echo "Cleaning..."
bitbake -c cleansstate jamvm-native > /dev/null
echo "Compiling..."
bitbake -c compile jamvm-native > /dev/null || { echo "Stopping"; exit 1; }
done
Best regards,
Erkka Kääriä
-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki
Business Identity Code: 0357606 - 4
Domiciled in Helsinki
This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
--
___
Openembedded-devel mailing list
Openembedded-devel@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-devel
