[rvm-research] problems using advice with MarkCompact and Jikes-3.1.3
"Jennifer B. Sartor" <[email protected]> Wed, 6 Sep 2017 11:33:03 +0200
| Newsgroups | gmane.comp.java.jikes.rvm.devel |
|---|---|
| Message-ID | <[email protected]> |
Hi All, I’m a long time Jikes developer. I’m trying to use Jikes’ many collectors in a course I teach, and have students compare the various collectors’ performance. However, I’m having a problem getting MarkCompact to work with replay. Is anyone familiar with this problem? I’m using (unmodified) Jikes-3.1.3. I generated fresh advice when running with the MarkCompact collector, then tried to use it, and it always fails with a message such as “recursive use of hardware exception registers” after it prints a message about Replay probe ending. I’m using Steve Blackburn’s probe code. A sample command line, when using the generated advice, is here: mkdir -p /tmp/runbms-jbsartor 2>/dev/null; cd /tmp/runbms-jbsartor; /home/jbsartor/newRunning/bin/timedrun -t 900 /home/jbsartor/newRunning/JikesRVM-3.1.3/dist/FastAdaptiveMarkCompact_x86_64-linux/rvm -Xms110M -X:gc:variableSizeHeap=false -Dprobes=Replay,MMTk -X:aos:initial_compiler=base -X:aos:enable_bulk_compile=true -X:aos:enable_recompilation=false -X:aos:cafi=/home/jbsartor/newRunning/results/advice-3.1.3-MarkCompact/xalan.2000.80.FastAdaptiveMarkCompact.4.ca -X:aos:dcfi=/home/jbsartor/newRunning/results/advice-3.1.3-MarkCompact/xalan.2000.80.FastAdaptiveMarkCompact.4.dc -X:vm:edgeCounterFile=/home/jbsartor/newRunning/results/advice-3.1.3-MarkCompact/xalan.2000.80.FastAdaptiveMarkCompact.4.ec -X:availableProcessors=4 -X:gc:threads=2 -cp /home/jbsartor/newRunning/bin/jbsCallback/SteveProbeEssential/ProbeEssential.jar:/usr/share/benchmarks/dacapo/dacapo-9.12-bach.jar Harness -c DacapoBachCallback -n 2 xalan I’ll attach the error message and stack trace produced by xalan to this email. It seems to fail during collection, during compaction, when trying to get the end address of an object. I’ve also tried this with other benchmarks, and they all fail when trying to use replay advice. I can definitely run DaCapo benchmarks with MarkCompact without replay successfully. I just would prefer to use replay, so any advice is appreciated. Thank you! Best, Jennifer Jennifer B. Sartor ------------------------------------ Assistant Professor at Vrije Universiteit Brussel Department of Computer Science Brussels, Belgium http://soft.vub.ac.be/~jsartor/index.html <http://soft.vub.ac.be/~jsartor/index.html> [email protected] <mailto:[email protected]> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Jikesrvm-researchers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers