FULL PRODUCT VERSION :
java version "1.8.0_25"
Java(TM) SE Runtime Environment (build 1.8.0_25-b18)
Java HotSpot(TM) 64-Bit Server VM (build 25.25-b02, mixed mode)

FULL OS VERSION :
Microsoft Windows [Version 6.1.7601]

EXTRA RELEVANT SYSTEM CONFIGURATION :
Intel(R) Core(TM) i7-3540M CPU @ 3.0.0GHz 3.00 GHz
8.00 GB (7.87 GB usable)
64-bit Operating System

A DESCRIPTION OF THE PROBLEM :
CPU goes up to 100% after running my java application for about 16 minutes. I can reproduce it inside eclipse or outside. It does not happen to 1.8u11, 1.8 u40(EA), Linux nor with 32-bit JVM but is 100% reproducible for 1.8u20 and 1.8u25 on Windows 7 64-bit machine. My application code was compiled with jdk1.7.

I tried running it inside eclipse and when it happened, I suspended all JAVA Threads. The cpu didn't come down.

However, if I'm inside Progress explorer, I can see a native thread which cause the cpu spike, if I suspend that thread, the cpu comes back down. With Process Explorer, I suspended that thread and get the following stack:

ntoskrnl.exe!KeWaitForMultipleObjects+0xc0a
ntoskrnl.exe!KeAcquireSpinLockAtDpcLevel+0x732
jvm.dll!JVM_FindSignal+0x24efd7
jvm.dll!JVM_FindSignal+0x2344c0
jvm.dll!JVM_FindSignal+0x2393d5
jvm.dll!JVM_FindSignal+0x23b7b1
jvm.dll!JVM_FindSignal+0x1d3aab
jvm.dll!JVM_FindSignal+0x1d4764
jvm.dll!JVM_FindSignal+0x1b8272
jvm.dll+0x98a02
jvm.dll+0x991f3
jvm.dll!JVM_ResolveClass+0x590b4
jvm.dll!JVM_FindSignal+0x61fa
msvcr100.dll!endthreadex+0x43
msvcr100.dll!endthreadex+0xdf
kernel32.dll!BaseThreadInitThunk+0xd
ntdll.dll!RtlUserThreadStart+0x21



It didn't happen if I run with -Xint

THE PROBLEM WAS REPRODUCIBLE WITH -Xint FLAG: No

THE PROBLEM WAS REPRODUCIBLE WITH -server FLAG: Yes

REGRESSION. Last worked in version 8u11

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
Unfortunately, I don't have a reproducible java file. My application was compiled by using jdk1.7. We're upgrading to jre1.8u25 in the upcoming release and noticed this issue. I can junction different version of jre (1.7, 1.8u11, 1.8u20, 1.8u25 and 1.8u40) to show the issue is actually happening on 1.8u20 and 1.8u25.

EXPECTED VERSUS ACTUAL BEHAVIOR :
Shouldn't spike the cpu.
REPRODUCIBILITY :
This bug can be reproduced always.

CUSTOMER SUBMITTED WORKAROUND :
Run with -Xint