RFR: 8265956: JVM crashes when matching LShiftVB Node [v5]

[Wang Huang]

On Mon, 10 May 2021 07:13:48 GMT, Jie Fu <jiefu at openjdk.org> wrote:

>> Wang Huang has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   change to static final
>
> test/hotspot/jtreg/compiler/vectorapi/TestVectorShuffleIotaByte.java line 42:
> 
>> 40: 
>> 41: @Test
>> 42: public class TestVectorShuffleIotaByte {
> 
> Can this jtreg test reproduce the bug on x86?
> 
> I tested it on our x86 platforms, but failed to reproduce it.
> Thanks.

Thank you for your wonderful review. I will change that.

> test/hotspot/jtreg/compiler/vectorapi/TestVectorShuffleIotaByte.java line 74:
> 
>> 72:             bv1.intoArray(ab_64, 0);
>> 73:         }
>> 74:         Assert.assertEquals(ab_64, expected_64);
> 
> How about putting the expected value as the first parameter, which is suggested by [1] ?
>  
> [1] https://github.com/openjdk/jdk/blob/master/doc/hotspot-unit-tests.md#first-parameter-is-expected-value

Thank you for your review.
* I think this document might be the advice of GoogleTest in jdk(`test/hotspot/gtest`) instead of jtreg (`test/hotspot/jtreg`)
* The common way which can pass expect result to method is using `@DataProvider`. However, in this small test it will inflate the whole test case codes. :-)

-------------

PR: https://git.openjdk.java.net/jdk/pull/3747