Christian Thalinger wrote:
> Christian Thalinger wrote:
>> Christian Thalinger wrote:
>>> Still building the newly checked out version...
>> Finally I can reproduce the crashes. Either it's related to debug vs.
>> product build or the additional patches applied. Still investigating...
>
> Oh no! It's so obvious :-/ indy.compiler.patch is not enabled for
> default builds and so the VM crashes when the compiler kicks in.
>
> I just pushed a changeset[1] that enables the patch. Please update your
> mlvm repository and build again.
With a debug build I get this assert:
$ bin/jruby --server -J-Djruby.compile.invokedynamic=true
-J-XX:+EnableInvokeDynamic bench/bench_fib_recursive.rb 100
VM option '+EnableInvokeDynamic'
5.950000 0.000000 5.950000 ( 5.562000)
# Found non oop pointer. Dumping state at failure
------
OopMap{ebp=Oop [80]=Oop [84]=Oop [48]=Oop [52]=Oop off=680}bci: 106
Compiled frame (sp=0xb00fe7e0 unextended sp=0xb00fe810, fp=0xb00fe810,
pc=0x0289da08)
116 nmethod (2)
bench.bench_fib_recursiveInvokermethod__0$RUBY$fib_rubyFixed1::call (15
bytes)
116 nmethod (2)
bench.bench_fib_recursiveInvokermethod__0$RUBY$fib_rubyFixed1::call (15
bytes)
Register map
ebp [0xb00fe7d8] = 0xb00fe810
------
ebp=Oop register r
ebploc = 0xb00fe7d8 *loc = 0xb00fe810
# To suppress the following error report, specify this argument
# after -XX: or in .hotspotrc: SuppressErrorAt=/oopMap.cpp:408
#
# A fatal error has been detected by the Java Runtime Environment:
#
# Internal Error
(/Users/twisti/mlvm/tmp/hotspot/src/share/vm/compiler/oopMap.cpp:408),
pid=6427, tid=2954375168
# Error: assert(Universe::heap()->is_in_or_null(*loc),"found non oop
pointer")
-- Christian
_______________________________________________
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
