-
Bug
-
Resolution: Not an Issue
-
P1
-
None
-
1.1.4
-
generic
-
generic
Running the latest HotJava, I can reliably get the VM to hang here:
"Clock" (TID:0xed9980d0, sys_thread_t:0xed3f1de0, state:CW) prio=12
--> "main" (TID:0xed9980a8, sys_thread_t:0x308c0, state:R, intr) prio=7 *current thread*
Monitor Cache Dump:
(note the lack of any Java stack in thread "main", indicating that this
thread has finished executing any HotJava code). The hang happens _well_
after sunw.hotjava.Main.main() has exited.
It will spin forever in this thread, sending my load to 1.
To reproduce on Solaris: Launch HotJava, and go to Edit->Preferences->Applet
Security. Press reload. I've never had to press reload more than three
times to get the hang (on my ultra).
FWIW, I started noticing the problem right after a change to AppletPanel.
The gist of this change was to make it so that HotJava applets
(applets loaded via the doc: protocol) all share one P5 thread,
instead of each getting its own P3 thread.
bill.foote@Eng 1997-06-17
"Clock" (TID:0xed9980d0, sys_thread_t:0xed3f1de0, state:CW) prio=12
--> "main" (TID:0xed9980a8, sys_thread_t:0x308c0, state:R, intr) prio=7 *current thread*
Monitor Cache Dump:
(note the lack of any Java stack in thread "main", indicating that this
thread has finished executing any HotJava code). The hang happens _well_
after sunw.hotjava.Main.main() has exited.
It will spin forever in this thread, sending my load to 1.
To reproduce on Solaris: Launch HotJava, and go to Edit->Preferences->Applet
Security. Press reload. I've never had to press reload more than three
times to get the hang (on my ultra).
FWIW, I started noticing the problem right after a change to AppletPanel.
The gist of this change was to make it so that HotJava applets
(applets loaded via the doc: protocol) all share one P5 thread,
instead of each getting its own P3 thread.
bill.foote@Eng 1997-06-17