JDK-8152168 : SumRedSqrt_Double.java fails with "Invalid sum of elements"
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 9,16
  • Priority: P2
  • Status: Closed
  • Resolution: Duplicate
  • Submitted: 2016-03-18
  • Updated: 2020-11-18
  • Resolved: 2020-11-18
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 16
16Resolved
Related Reports
Duplicate :  
JDK-8249936 - [AArch64] compiler/loopopts/superword/SumRed_Double.java fails "Invalid sum of elements variable in total: 2.1130768538923856E19"
Relates :  
JDK-8148175 - C1: G1 barriers don't preserve FP registers
Description
----------messages:(3/398)----------
command: main -XX:+IgnoreUnrecognizedVMOptions -XX:-SuperWordReductions -XX:LoopUnrollLimit=250 -XX:LoopMaxUnroll=2 -XX:CompileThresholdScaling=0.1 SumRedSqrt_Double
reason: User specified action: run main/othervm -XX:+IgnoreUnrecognizedVMOptions -XX:-SuperWordReductions -XX:LoopUnrollLimit=250 -XX:LoopMaxUnroll=2 -XX:CompileThresholdScaling=0.1 SumRedSqrt_Double 
elapsed time (seconds): 26.322
----------System.out:(2/97)----------
Invalid sum of elements variable in total: 2.06036466451497E14
Expected value = 2.06157643776E14
----------System.err:(13/670)----------
java.lang.Exception: Failed
	at SumRedSqrt_Double.main(SumRedSqrt_Double.java:62)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.lang.reflect.Method.invoke(Method.java:520)
	at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:110)
	at java.lang.Thread.run(Thread.java:804)

JavaTest Message: Test threw exception: java.lang.Exception: Failed
JavaTest Message: shutting down test
Comments
Failures on AArch64 are tracked by JDK-8249936.
13-11-2020
Here's a snippet from the log file for the jdk-16+19-938-tier3 failure sighting: #section:main ----------messages:(4/489)---------- command: main -XX:+IgnoreUnrecognizedVMOptions -XX:LoopUnrollLimit=250 -XX:CompileThresholdScaling=0.1 -XX:-SuperWordReductions -XX:LoopMaxUnroll=8 compiler.loopopts.superword.SumRedSqrt_Double reason: User specified action: run main/othervm -XX:+IgnoreUnrecognizedVMOptions -XX:LoopUnrollLimit=250 -XX:CompileThresholdScaling=0.1 -XX:-SuperWordReductions -XX:LoopMaxUnroll=8 compiler.loopopts.superword.SumRedSqrt_Double Mode: othervm [/othervm specified] elapsed time (seconds): 19.573 ----------configuration:(0/0)---------- ----------System.out:(2/97)---------- Invalid sum of elements variable in total: 1.98814471229952E14 Expected value = 2.06157643776E14 ----------System.err:(16/947)---------- java version "16-ea" 2021-03-16 Java(TM) SE Runtime Environment (fastdebug build 16-ea+19-938) Java HotSpot(TM) 64-Bit Server VM (fastdebug build 16-ea+19-938, mixed mode) java.lang.Exception: Failed at compiler.loopopts.superword.SumRedSqrt_Double.main(SumRedSqrt_Double.java:95) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:64) at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.base/java.lang.reflect.Method.invoke(Method.java:564) at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127) at java.base/java.lang.Thread.run(Thread.java:832) JavaTest Message: Test threw exception: java.lang.Exception: Failed JavaTest Message: shutting down test STATUS:Failed.`main' threw exception: java.lang.Exception: Failed ----------rerun:(40/6339)*----------
02-10-2020
I tried to reproduce the failure, but without any success. Closing as CNR.
05-04-2016
It doesn't look related to JDK-8148175. The test works with double[] arrays, so GC barriers shouldn't be a problem.
05-04-2016
[~vlivanov] mentioned that this could be related to "JDK-8148175, which is caused by G1 GC barriers in C1 don���t preserve FP regs".
18-03-2016
ILW = invalid result, rare, no workaround = HLH = P2
18-03-2016
Hi Vladimir, You've already looked into JDK-8148175. As this problem could be related, I'm assigning it to you. Please feel free assign in any way that you think is more appropriate. Thank you and best regards, Zoltan
18-03-2016