-
Bug
-
Resolution: Fixed
-
P4
-
1.3.0, 1.4.0
-
03
-
x86, sparc
-
solaris_7, windows_nt, windows_2000
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-2117231 | 1.4.1 | Chris Hegarty | P4 | Closed | Fixed | hopper |
JDK-2117230 | 1.4.0_01 | Chris Hegarty | P4 | Resolved | Fixed | 01 |
Name: tb29552 Date: 02/23/2001
For:
java version "1.3.1-beta"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.3.1-beta-b16)
Java HotSpot(TM) Client VM (build 1.3.1beta-b16, mixed mode)
With -Xcomp mode, JVMPI_EVENT_MONITOR_DUMP sometimes shows a monitor
with a non-zero "waiting to enter" count, but the thread ID waiting
is 0 (NULL).
Compile the attached JVMPI client, jvmpi_tool.cpp to make libjvmpi_tool.dll
and compile the Java example Deadlock.java. Run java as:
java -Xcomp -Xrunjvmpi_tool Deadlock
The two threads will deadlock, and after several seconds you will see
a monitor dump:
Start 0x800fd4: "Monitor dumper"
Start 0x800fd4: "Monitor dumper"
Start 0x7fcd14: "Finalizer"
Start 0x8990994: "Reference Handler"
Start 0x23493c: "main"
Start 0x802374: "Signal Dispatcher"
Start 0x802e34: "CompileThread0"
Start 0x8a0867c: "One"
Start 0x23493c: "Thread-0"
Start 0x8a10064: "Two"
Thread One: lockA
Thread Two: lockB
Monitor Dump:
Monitor 0x69ee1c0: Owner: 0x8a10064 Entry count: 1
1 waiting to enter
0
0 awaiting notification
Monitor 0x69ee190: Owner: 0x8a0867c Entry count: 1
1 waiting to enter
0
0 awaiting notification
Monitor 0x2930328: Owner: 0 Entry count: 1
0 waiting to enter
1 awaiting notification
0x7fcd14
Monitor 0x2930238: Owner: 0 Entry count: 1
0 waiting to enter
1 awaiting notification
0x8990994
Note the IDs for the threads waiting to enter monitors 0x69ee190 and
0x69ee1c0 are 0, but should be 0x8a10064 and 0x8a0867c respectively.
Tim B. attached an archive with the test program files along with
a Makefile so I'm deleting the copies from the description.
(Review ID: 117569)
======================================================================
Name: elR10090 Date: 05/07/2001
This bug affects the following testbase_nsk test:
nsk/regression/b4311308
======================================================================
- backported by
-
JDK-2117230 NULL thread ID in MONITOR_DUMP
- Resolved
-
JDK-2117231 NULL thread ID in MONITOR_DUMP
- Closed
- duplicates
-
JDK-4318767 MONITOR_DUMP event info is incorrect for threads in the deadlock state
- Closed
-
JDK-4329135 RequestEvent for JVMPI_EVENT_MONITOR_DUMP crashes HotSpot VM in compiled mode
- Closed
- relates to
-
JDK-4311308 JVMPI_EVENT_MONITOR_DUMP is wrong or has missing info
- Closed