[
https://issues.apache.org/jira/browse/FLINK-13450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16899908#comment-16899908
]
wangxiyuan commented on FLINK-13450:
------------------------------------
There is an openjdk bug for StrictMath performance:
[https://bugs.openjdk.java.net/browse/JDK-8210416]
According to its test, after this bug fix, some calculation using StricMath is
still slow than Math. For example:
Function | java.lang.Math | java.lang.StrictMath
---------------------------------------------------------------------------------
sin | 1.713649346570452 0m5.800s | 1.7136493465700542 0m18.731s
cos | 0.17098435541810225 0m5.765s | 0.1709843554185943 0m18.796s
tan | -5.5500322522995315E7 0m6.031s |-5.5500322522995315E7 0m21.093s
log | 1.7420680845245087E9 0m2.321s | 1.7420680845245087E9 0m4.439s
log10 | 7.565705562087342E8 0m2.263s | 7.565705562087342E8 0m5.543s
> Adjust tests to tolerate arithmetic differences between x86 and ARM
> -------------------------------------------------------------------
>
> Key: FLINK-13450
> URL: https://issues.apache.org/jira/browse/FLINK-13450
> Project: Flink
> Issue Type: Sub-task
> Components: Tests
> Affects Versions: 1.9.0
> Reporter: Stephan Ewen
> Priority: Major
>
> Certain arithmetic operations have different precision/rounding on ARM versus
> x86.
> Tests using floating point numbers should be changed to tolerate a certain
> minimal deviation.
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
