| Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
|---|---|---|---|---|---|---|
| JDK-2051074 | 1.4.1 | Btplusnull User | P4 | Closed | Fixed | hopper |
The testcase is /net/jano.sfbay/export/disk20/serviceability/ws/regressionTestsInWaiting/RedefineBorland.sh
The test just runs to a bkpt and then does two redefineClasses naming
the original .class file, ie, no changes. The 2nd redefineClasses
causes the assertion (must be run with java_g).
This happens on solaris, and it will happen on windows too after
4559100 has been fixed.
The test just runs to a bkpt and then does two redefineClasses naming
the original .class file, ie, no changes. The 2nd redefineClasses
causes the assertion (must be run with java_g).
This happens on solaris, and it will happen on windows too after
4559100 has been fixed.
- backported by
-
JDK-2051074 RedefineClasses gets assertion: "Should be a method entry in cpcache!"
-
- Closed
-
- relates to
-
JDK-4663134 RedefineClasses hits assertion "Bad cpcache index!" on x86 platforms
-
- Closed
-