[Bug 525] New: Shark made not entrant deopt sigsegv regression after bump to b20 and hs17

Tue Jul 27 02:38:22 PDT 2010

http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=525

           Summary: Shark made not entrant deopt sigsegv regression after
                    bump to b20 and hs17
           Product: IcedTea
           Version: 6-hg
          Platform: 32-bit
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Shark
        AssignedTo: unassigned at icedtea.classpath.org
        ReportedBy: xerxes at zafena.se

testcase:
cd openjdk/hotspot/test/compiler/6663621/
javac IVTest.java
java IVTest

This hotspot jtreg test triggers a sigsegv crash when running Shark on hs17
b20.
This are a regression since icedtea6-1.8.

This bug have been tested to be fully reproducable using Shark on ia32 and ARM.

output on ARM:
shark-b20-from-icedtea6-trunk/bin/java -XX:+PrintCompilation IVTest
  1       IVTest::padV15 (23 bytes)
  1%      IVTest::padV15 @ 5 (23 bytes)
  2%      IVTest::main @ 14 (236 bytes)
  2%  made not entrant  IVTest::main @ -2 (236 bytes)
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  Internal Error (os_linux_zero.cpp:236), pid=3897, tid=1097798800
#  Error: caught unhandled signal 11
#
# JRE version: 6.0_20-b20
# Java VM: OpenJDK Shark VM (17.0-b16 mixed mode linux-arm )
# Derivative: IcedTea6 1.9pre+r6bf6ff2ad1ea
# Distribution: Built on Debian GNU/Linux testing (squeeze) (Sat Jul 24
17:30:55 UTC 2010)
# An error report file with more information is saved as:
# /wd/test/hs_err_pid3897.log
#
# If you would like to submit a bug report, please include
# instructions how to reproduce the bug and visit:
#   http://icedtea.classpath.org/bugzilla
#
Aborted

Output on X86:
process 18256 is executing new program:
/home/xerxes/icedtea6-shark-b20/openjdk.build/j2sdk-image/bin/java
[Thread debugging using libthread_db enabled]
[New Thread 0xb7fe5b70 (LWP 18259)]
[New Thread 0x91a2db70 (LWP 18260)]
[New Thread 0x919acb70 (LWP 18261)]
[New Thread 0x9182bb70 (LWP 18262)]
[New Thread 0x9166bb70 (LWP 18263)]
[New Thread 0x914eab70 (LWP 18264)]
[New Thread 0x91269b70 (LWP 18265)]
[New Thread 0x912ffb70 (LWP 18266)]
  1       sun.nio.cs.UTF_8$Decoder::decodeArrayLoop (553 bytes)
  1%      java.lang.String::hashCode @ 28 (60 bytes)
  2       java.lang.String::charAt (33 bytes)
  3       java.lang.String::indexOf (151 bytes)
  2%      sun.nio.cs.UTF_8$Encoder::encodeArrayLoop @ 73 (490 bytes)
  4       IVTest::padV15 (23 bytes)
  3%      IVTest::main @ 14 (236 bytes)
  3%  made not entrant  IVTest::main @ -2 (236 bytes)

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb7fe5b70 (LWP 18259)]
0x892cec83 in ?? ()
(gdb) bt
#0  0x892cec83 in ?? ()
#1  0x003bf26f in ZeroEntry::invoke (recurse=0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/entry_zero.hpp:56
#2  Interpreter::invoke_method (recurse=0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/interpreter_zero.hpp:28
#3  CppInterpreter::main_loop (recurse=0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:95
#4  0x003bf42a in ZeroEntry::maybe_deoptimize (recurse=0, 
    __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/entry_zero.hpp:67
#5  ZeroEntry::invoke_osr (recurse=0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/entry_zero.hpp:61
#6  Interpreter::invoke_osr (recurse=0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/interpreter_zero.hpp:34
#7  CppInterpreter::main_loop (recurse=0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:152
---Type <return> to continue, or q <return> to quit---
#8  0x003bffbe in CppInterpreter::normal_entry (method=0xb1fa5348, 
    UNUSED=-1243197120, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:48
#9  0x005cb069 in ZeroEntry::invoke (call_wrapper=0xb7fe50c8, 
    result=0xb7fe52f0, result_type=T_INT, method=0xb1fa5348, 
    entry_point=0xb5e65140 "`\377;", parameters=0xb7fe51c4, parameter_words=1, 
    __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/entry_zero.hpp:56
#10 Interpreter::invoke_method (call_wrapper=0xb7fe50c8, result=0xb7fe52f0, 
    result_type=T_INT, method=0xb1fa5348, entry_point=0xb5e65140 "`\377;", 
    parameters=0xb7fe51c4, parameter_words=1, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/interpreter_zero.hpp:28
#11 StubGenerator::call_stub (call_wrapper=0xb7fe50c8, result=0xb7fe52f0, 
    result_type=T_INT, method=0xb1fa5348, entry_point=0xb5e65140 "`\377;", 
    parameters=0xb7fe51c4, parameter_words=1, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/cpu/zero/vm/stubGenerator_zero.cpp:68
#12 0x004515d7 in JavaCalls::call_helper (result=0xb7fe52ec, m=0xb7fe5174, 
    args=0xb7fe51c0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/share/vm/runtime/java---Type
<return> to continue, or q <return> to quit---
Calls.cpp:387
#13 0x0053aeb4 in os::os_exception_wrapper (
    f=0x451340 <JavaCalls::call_helper(JavaValue*, methodHandle*,
JavaCallArguments*, Thread*)>, value=0xb7fe52ec, method=0xb7fe5174,
args=0xb7fe51c0, 
    thread=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/os/linux/vm/os_linux.cpp:4052
#14 0x0045036c in JavaCalls::call (result=0xb7fe52ec, method=..., 
    args=0xb7fe51c0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/share/vm/runtime/javaCalls.cpp:293
#15 0x0045d651 in jni_invoke_static (env=<value optimized out>, 
    result=<value optimized out>, receiver=0x0, call_type=JNI_STATIC, 
    method_id=0x80c03e0, args=0xb7fe52c0, __the_thread__=0x8070768)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/share/vm/prims/jni.cpp:1017
#16 0x0046baf0 in jni_CallStaticVoidMethod (env=0x8070880, cls=0x8072b6c, 
    methodID=0x80c03e0)
    at
/home/xerxes/icedtea6-shark-b20/openjdk/hotspot/src/share/vm/prims/jni.cpp:1589
#17 0x0804b12d in JavaMain (_args=0xbfffc220)
    at ../../../../src/share/bin/java.c:614

Dump of assembler code from 0x892cec73 to 0x892cec93:
   0x892cec73:  Cannot access memory at address 0x892cec73
(gdb) disassemble  0x892cec83,0x892cec93
Dump of assembler code from 0x892cec83 to 0x892cec93:
=> 0x892cec83:  Cannot access memory at address 0x892cec83
(gdb)

(11.06.35) xranby: gbenson: what does the -2   at  2%  made not entrant 
IVTest::main @ -2 (236 bytes)  mean?

(11.09.39) gbenson: / Handy constants for deciding which compiler mode to use.
(11.09.39) gbenson: enum MethodCompilation {
(11.09.39) gbenson:   InvocationEntryBci = -1,     // i.e., not a on-stack
replacement compilation
(11.09.39) gbenson:   InvalidOSREntryBci = -2
(11.09.39) gbenson: };
(11.09.47) gbenson: there's your -2
(11.09.52) gbenson: in src/share/vm/utilities/globalDefinitions.hpp

(11.12.44) gbenson: 2% IVTest::main @ 14 (236 bytes)
(11.12.44) gbenson: 2% made not entrant IVTest::main @ -2 (236 bytes)
(11.12.56) gbenson: so it is compiled with one bci, but deoptimized with
another
(11.13.35) gbenson: ah, that's nmethod::invalidate_osr_method()

(11.32.44) gbenson: xranby: it would be nice to see what
-XX:+TraceDeoptimization says

-- 
Configure bugmail: http://icedtea.classpath.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.