[11u] RFR: 8208655: use JTreg skipped status in hotspot tests

[Severin Gehwolf]

Hi Christoph,

On Wed, 2019-08-07 at 21:42 +0000, Langer, Christoph wrote:
> Hi Severin
> 
> thanks for looking into this.
> 
> > > Firstly, the change did not apply exactly cleanly. I had to make modifications to
> > > 
> > test/hotspot/jtreg/runtime/CompressedOops/CompressedClassPointers.java.
> > 
> > What changes did you have to make to that file? I see that jdk/jdk
> > patch has:
> > 
> > + * @requires vm.bits == 64
> > 
> > which seems to be missing from your patch. Intentional?
> 
> Yes, because it can already be found in the current version of the file:
> line 28: 28  * @requires vm.bits == 64 & vm.opt.final.UseCompressedOops == true

Ah, right. Missed that, sorry.

> > > Then, I also modified the following 3 tests which already contain
> > > backports to 11u during which SkippedException was removed. So I
> > > added back SkippedException there – making the tests resemble what
> > > they currently look like in jdk/jdk:
> > > test/hotspot/jtreg/compiler/codegen/aes/TestAESMain.java
> > > 
> > test/hotspot/jtreg/compiler/codegen/aes/TestCipherBlockChainingEncrypt.j
> > ava
> > > test/hotspot/jtreg/compiler/intrinsics/base64/TestBase64.java
> > 
> > How did you determine these? I wonder whether it would be more
> > appropriate to get these updates in a follow-up.
> 
> Actually, I scanned the communication on jdk-updates-dev where
> SkippedException was mentioned - this was the cause for some manual
> backport adaptions. I believe this backport is the right place to
> "rectify" these modified backports.

Fair enough. I guess it's not worth doing a separate changeset with
these.

> > I see that
> > test/hotspot/jtreg/runtime/containers/docker/TestCPUSets.java
> > should receive a similar treatment too.
> 
> Well, the SkippedException there was introduced with JDK-8220672
> which you seem about to backport. So I guess you'll do the backport
> after I pushed this one and you should have one less reject to
> resolve ��

Right. Too many different backports ;-)

> > FWIW, the diff stat of 7b1ddbafa134 from HEAD doesn't list any of them.
> > 
> > > As the original change of 8208655 wasn’t perfect and would cause a
> > > few test failures if applied alone, I’ll also have to bring in JDK-
> > > 8208701 [2] and JDK-8208706 [3]. So, once pushing this, I’ll push all
> > > of these 3 items.
> > 
> > Sounds good.
> 
> Great. So, good to push after my clarifications?

Yes, but please with JDK-8208701 and JDK-8208706 as this patch would
otherwise introduce tier1 failures:

TEST: runtime/XCheckJniJsig/XCheckJSig.java
TEST JDK: /disk/openjdk/upstream-sources/openjdk-11-dev/build/linux-x86_64-normal-server-release/images/jdk

TEST RESULT: Error. Parse Exception: Syntax error in @requires expression: invalid name: vm.family
--------------------------------------------------


TEST: runtime/memory/LargePages/TestLargePagesFlags.java
TEST JDK: /disk/openjdk/upstream-sources/openjdk-11-dev/build/linux-x86_64-normal-server-release/images/jdk

TEST RESULT: Error. Parse Exception: Syntax error in @requires expression: invalid name: vm.family
--------------------------------------------------


TEST: runtime/CompressedOops/CompressedKlassPointerAndOops.java
TEST JDK: /disk/openjdk/upstream-sources/openjdk-11-dev/build/linux-x86_64-normal-server-release/images/jdk

TEST RESULT: Error. Parse Exception: Invalid tag: requiers
--------------------------------------------------


TEST: runtime/6819213/TestBootNativeLibraryPath.java
TEST JDK: /disk/openjdk/upstream-sources/openjdk-11-dev/build/linux-x86_64-normal-server-release/images/jdk

TEST RESULT: Error. Parse Exception: Invalid tag: requiers
--------------------------------------------------

Thanks,
Severin