RFR+RFA [11u]: 8214189: test/hotspot/jtreg/compiler/intrinsics/mathexact/MulExactLConstantTest.java fails on Windows x64 when run with -XX:-TieredCompilation

[Aleksey Shipilev]

On 2/22/19 12:23 PM, Langer, Christoph wrote:
> I think the regression was introduced when https://bugs.openjdk.java.net/browse/JDK-8213419 and
> https://bugs.openjdk.java.net/browse/JDK-8214206 were brought to jdk11u at about that time. The
> issue is fixed with 8214189. I consider this severe enough to warrant a backport of the fix to 11.0.3.

I agree, this should go to 11.0.3. No way we would release 11.0.3 without passing JCK.

Andrew, please approve?

> Bug: https://bugs.openjdk.java.net/browse/JDK-8214189
> Original Change: http://hg.openjdk.java.net/jdk/jdk/rev/0037ea3c7322
> Backport Webrev: http://cr.openjdk.java.net/~clanger/webrevs/8214189.jdk11u/

Looks good.

> The patch applies cleanly, apart from the change to the exclusion list
> test/hotspot/jtreg/ProblemList.txt which is not valid in 11u. It did also not show further
> regressions in the nightly tests.

Since patch applies cleanly and fixes the tests, I don't think it needs more reviews.

> I guess it should also be brought to 8u as well, since the regressing patches have been checked in
> there as well.

Marked for 8u backport.

-Aleksey