Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

jdk20 MathLoadTest ComparisonFailure such as expected:<1[.8E+3]> but was:<1[800]> #16254

Closed
pshipton opened this issue Nov 2, 2022 · 8 comments

Comments

@pshipton
Copy link
Member

pshipton commented Nov 2, 2022

https://openj9-jenkins.osuosl.org/job/Grinder_testList_1/77 sanity.system running on amac
MathLoadTest_bigdecimal_CS_5m_0

MLT testFailure: testItem_0659(net.adoptopenjdk.test.bigdecimal.TestSuite010): expected:<1[.8E+3]> but was:<1[800]>
MLT org.junit.ComparisonFailure: expected:<1[.8E+3]> but was:<1[800]>
MLT 	at org.junit.Assert.assertEquals(Assert.java:115)
MLT 	at org.junit.Assert.assertEquals(Assert.java:144)
MLT 	at net.adoptopenjdk.test.bigdecimal.TestSuite010.testItem_0659(TestSuite010.java:4571)

https://openj9-jenkins.osuosl.org/job/Grinder_testList_0/175
MathLoadTest_bigdecimal_5m_1

MLT testFailure: testItem_0426(net.adoptopenjdk.test.bigdecimal.TestSuite030): expected:<4[E+2]> but was:<4[00]>
MLT org.junit.ComparisonFailure: expected:<4[E+2]> but was:<4[00]>
MLT 	at org.junit.Assert.assertEquals(Assert.java:115)
MLT 	at org.junit.Assert.assertEquals(Assert.java:144)
MLT 	at net.adoptopenjdk.test.bigdecimal.TestSuite030.testItem_0426(TestSuite030.java:2999)

MLT testFailure: testItem_0691(net.adoptopenjdk.test.bigdecimal.TestSuite030): expected:<-1[E+19]> but was:<-1[0000000000000000000]>
MLT org.junit.ComparisonFailure: expected:<-1[E+19]> but was:<-1[0000000000000000000]>
MLT 	at org.junit.Assert.assertEquals(Assert.java:115)
MLT 	at org.junit.Assert.assertEquals(Assert.java:144)
MLT 	at net.adoptopenjdk.test.bigdecimal.TestSuite030.testItem_0691(TestSuite030.java:4756)
@pshipton
Copy link
Member Author

pshipton commented Nov 8, 2022

Tried MathLoadTest_bigdecimal_5m_1 on xlinux as well, where it also fails.

14:27:15  MLT testFailure: testItem_0659(net.adoptopenjdk.test.bigdecimal.TestSuite010): expected:<1[.8E+3]> but was:<1[800]>
14:27:15  MLT org.junit.ComparisonFailure: expected:<1[.8E+3]> but was:<1[800]>
14:27:15  MLT 	at org.junit.Assert.assertEquals(Assert.java:115)
14:27:15  MLT 	at org.junit.Assert.assertEquals(Assert.java:144)
14:27:15  MLT 	at net.adoptopenjdk.test.bigdecimal.TestSuite010.testItem_0659(TestSuite010.java:4571)

Also tried Hotspot xlinux, which failed

MLT testFailure: testItem_0159(net.adoptopenjdk.test.bigdecimal.TestSuite019): expected:<-4[E+1]> but was:<-4[0]>
MLT org.junit.ComparisonFailure: expected:<-4[E+1]> but was:<-4[0]>
MLT 	at org.junit.Assert.assertEquals(Assert.java:115)
MLT 	at org.junit.Assert.assertEquals(Assert.java:144)
MLT 	at net.adoptopenjdk.test.bigdecimal.TestSuite019.testItem_0159(TestSuite019.java:1064)

Tried Hotspot with -Xint

MLT testFailure: testItem_0740(net.adoptopenjdk.test.bigdecimal.TestSuite060): expected:<-3[.59538626972463141629054847463408713596141135051689993197834953606314521560057077521179117265533756343080917907028764928468642653778928365536935093407075033972099821153102564152490980180778657888151737016910267884609166473806445896331617118664246696549595652408289446337476354361838599762500808052368249716736E+616]> but was:<-3[5953862697246314162905484746340871359614113505168999319783495360631452156005707752117911726553375634308091790702876492846864265377892836553693509340707503397209982115310256415249098018077865788815173701691026788460916647380644589633161711866424669654959565240828944633747635436183859976250080805236824971673600000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000]>
MLT org.junit.ComparisonFailure: expected:<-3[.59538626972463141629054847463408713596141135051689993197834953606314521560057077521179117265533756343080917907028764928468642653778928365536935093407075033972099821153102564152490980180778657888151737016910267884609166473806445896331617118664246696549595652408289446337476354361838599762500808052368249716736E+616]> but was:<-3[5953862697246314162905484746340871359614113505168999319783495360631452156005707752117911726553375634308091790702876492846864265377892836553693509340707503397209982115310256415249098018077865788815173701691026788460916647380644589633161711866424669654959565240828944633747635436183859976250080805236824971673600000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000]>
MLT 	at org.junit.Assert.assertEquals(Assert.java:115)
MLT 	at org.junit.Assert.assertEquals(Assert.java:144)
MLT 	at net.adoptopenjdk.test.bigdecimal.TestSuite060.testItem_0740(TestSuite060.java:5210)

@pshipton
Copy link
Member Author

pshipton commented Nov 8, 2022

I assume the failures are related to the following change in jdk20. I don't know why the failures seem to happen randomly in different places. @Mesbah-Alam the test needs to be fixed.

Restore Behavior of java.math.BigDecimal.movePointLeft() and movePointRight() on a Zero Argument (JDK-8289260)
core-libs/java.math
When these methods are invoked with a zero argument on a target with a negative scale, they return a result which is numerically the same as the target, but with a different unscaled value and scale.
In earlier releases, they returned a result with the same unscaled value and scale, which was against the specification.
The behavior is unchanged when the target has a non-negative scale or when the argument is not zero.

@pshipton
Copy link
Member Author

pshipton commented Nov 8, 2022

Looking at the places which fail, they contains a check for JavaSpecVersionChecker.isJDK13OrNewer().

@pshipton
Copy link
Member Author

pshipton commented Nov 8, 2022

@llxia fyi

@llxia
Copy link
Contributor

llxia commented Nov 10, 2022

@Mesbah-Alam could you look into this? Thanks

@Mesbah-Alam
Copy link
Contributor

These test cases are auto-generated, and it appears some of them (and there are multiple ones based on discussions above) are invalid for JDK20. The validator values will have to either be manually updated or tests regenerated. I will exclude the MathLoadTest on JDK20 temporarily while we fix this.

@Mesbah-Alam
Copy link
Contributor

PR for exclude: adoptium/aqa-tests#4144.

@pshipton
Copy link
Member Author

This seems a dup of adoptium/openj9-systemtest#148 which was resolved via adoptium/aqa-systemtest#482

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants