On Thu, 6 Apr 2023 03:43:26 GMT, Quan Anh Mai <qa...@openjdk.org> wrote:
>> Thank you for doing this. I am guessing it was always intended to be public
>> but was not spotted in previous reviews.
>
> @PaulSandoz Thanks for your timely review, do I need a second review for this
> patch?
Hello @merykitty,
I'm observing GHA test failure in `jdk/incubator/vector/ShortMaxVectorTests` on
`linux-x86` (jdk master):
test ShortMaxVectorTests.sliceShortMaxVectorTestsMasked(short[-i * 5], short[i
+ 1], mask[true]): failure
java.lang.AssertionError: arrays differ firstly at element [16]; expected value
is <12> but was <1>. (ref: [-135, -140, -145, -150, -155, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27],
res: [-135, -140, -145, -150, -155, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]), at index #0, at origin #27
at org.testng.Assert.fail(Assert.java:99)
at org.testng.Assert.assertEquals(Assert.java:275)
at ShortMaxVectorTests.assertArraysEquals(ShortMaxVectorTests.java:854)
at
ShortMaxVectorTests.sliceShortMaxVectorTestsMasked(ShortMaxVectorTests.java:4812)
Could this be caused by changes in this PR?
See
https://github.com/eirbjo/jdk/actions/runs/4707493730#user-content-jdk_incubator_vector_shortmaxvectortests
-------------
PR Comment: https://git.openjdk.org/jdk/pull/13277#issuecomment-1509840795
