JDK-6773776 : CTW ZKM.jar: assert(Compile::current()->unique() < (uint)MaxNodeLimit,"Node limit exceeded")
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: hs14,9,10
  • Priority: P4
  • Status: Closed
  • Resolution: Won't Fix
  • OS: solaris_10
  • CPU: sparc
  • Submitted: 2008-11-19
  • Updated: 2022-10-03
  • Resolved: 2018-03-08
Related Reports
Relates :  
JDK-7081842 - assert(Compile::current()->unique() < (uint)MaxNodeLimit) failed: Node limit exceeded
Description
% /export/kvn/jdk7_b39/fastdebug/bin/java -d64  -Xss4m -Xverify:all -XX:+CompileTheWorld -Xbootclasspath/p:/net/vmsqe.sfbay/export/backup/testbase/CompileTheWorld/jarfiles/ZKM.jar
...
CompileTheWorld (491) : a/a/d/e/h
# To suppress the following error report, specify this argument
# after -XX: or in .hotspotrc:  SuppressErrorAt=/node.cpp:60
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  Internal Error (/net/dhcp-usca22-158-108/export/home/kvn/work2/hg/6766316/src/share/vm/opto/node.cpp:60), pid=5664, tid=20
#  Error: assert(Compile::current()->unique() < (uint)MaxNodeLimit,"Node limit exceeded")
#
# Java VM: OpenJDK 64-Bit Server VM (14.0-b06-internal-fastdebug mixed mode solaris-amd64 )
nsk/stress/jck60/jck60021
Crashed 1/21/09 in rt_baseline with this assert, but I don't know if it's related.  Should it be a different bug??  thanks.

solaris-amd64 with
-d64 -server -Xmixed -Dsun.jvm.hotspot.runtime.VM.disableVersionCheck=1 -XX:+UseCompressedOops -XX:DefaultMaxRAMFraction=8 -XX:-PrintVMOptions

log file:
http://sqeweb.sfbay.sun.com/nfs/results/vm/gtee/JDK7/NIGHTLY/VM/2009-01-21/RT_Baseline/vm/solaris-amd64/server/mixed/solaris-amd64_server_mixed_nsk.stress.testlist/ResultDir/jck60021/jck60021.log

hs_err file:
http://sqeweb.sfbay.sun.com/nfs/results/vm/gtee/JDK7/NIGHTLY/VM/2009-01-21/RT_Baseline/vm/solaris-amd64/server/mixed/solaris-amd64_server_mixed_nsk.stress.testlist/ResultDir/jck60021//hs_err_pid2208.log
Comments
Saw only with CTW testing. We are moving from C2 to Graal which doesn't have such limitations.
08-03-2018
ILW = M(compilation bailout in product build ~ performance penalty)L(very large methods)H(none) = P4
11-04-2016
Issue is still there: CompileTheWorld (491) : a/a/d/e/h # To suppress the following error report, specify this argument # after -XX: or in .hotspotrc: SuppressErrorAt=/node.cpp:77 # # A fatal error has been detected by the Java Runtime Environment: # # Internal Error (/HUDSON/workspace/9-2-build-linux-amd64-phase2/jdk9/4571/hotspot/src/share/vm/opto/node.cpp:77), pid=55116, tid=55261 # assert(Compile::current()->live_nodes() < Compile::current()->max_node_limit()) failed: Live Node limit exceeded limit # # JRE version: Java(TM) SE Runtime Environment (9.0+108) (fastdebug build 9-ea+108-2016-03-02-173138.javare.4571) # Java VM: Java HotSpot(TM) 64-Bit Server VM (fastdebug 9-ea+108-2016-03-02-173138.javare.4571, mixed mode, tiered, compressed oops, g1 gc, linux-amd64)
07-03-2016
ILW=Assert (Benign since we will exit gracefully when hitting node limit), this method in this test, none=LLH=P5 Try to reproduce, we have changed how we count live nodes
05-06-2014