JDK 9 RFR of 8150206: Demote LFMultiThreadCachingTest.java to tier 2 until resolution of JDK-8150014

Martin Buchholz martinrb at google.com
Fri Feb 19 21:37:33 UTC 2016


I will commit by Monday.

The only other change I'm thinking of would be to move the Unsafe
initialization to the top of file (i.e. beginning of <clinit>) instead
of bottom of file where we've traditionally placed it.  Here a static
method call that used U found it to be null - we must still be inside
<clinit>, eh?

On Fri, Feb 19, 2016 at 10:44 AM, joe darcy <joe.darcy at oracle.com> wrote:
> Hello,
>
> I'm happy if Martin's delambdafication is the right solution.
>
> However, if that is not pushed by Monday Feb. 22, I'll want to go ahead and
> problem list the test.
>
> Thanks,
>
> -Joe
>
>
> On 2/19/2016 1:40 AM, Paul Sandoz wrote:
>>
>> Hi Martin,
>>
>> Your patch to TLR and SR looks good:
>>
>>
>> http://cr.openjdk.java.net/~martin/webrevs/openjdk9/jsr166-jdk9-integration/unlambdafyBootstrap/jsr166-jdk9-integration-
>>
>> Sorry, i should of caught this in review, did notice the change to use
>> lambdas but i forgot that this might cause issues at VM startup.
>>
>> ConcurrentHashMap is used by infrastructure of j.l.invoke. CHM cannot
>> itself leverage or depend on stuff that leverages lambdas otherwise it may
>> result in a circular dependency at VM startup.
>>
>> In this case the test tickles the code paths that results in the
>> initialization of ThreadLocalRandom by CHM during startup which explains why
>> it was yet to be observed in other cases.
>>
>> Paul.
>>
>>> On 19 Feb 2016, at 07:18, Martin Buchholz <martinrb at google.com> wrote:
>>>
>>> It's easy enough to get the test to pass by delambdafication.
>>>
>>> But we may not want to fix the test because it's not clear whether
>>> it's a jdk lambda bug or not, and we don't sweep bugs under the rug.
>>>
>>> Maybe instead of a "tier2" bucket we really want a "test found a bug
>>> that's not yet fixed" bucket?
>>>
>>> On Thu, Feb 18, 2016 at 4:48 PM, joe darcy <joe.darcy at oracle.com> wrote:
>>>>
>>>> Hello,
>>>>
>>>> Right now, we are seeing some cross-platform intermittent failures of
>>>>
>>>>     java/lang/invoke/LFCaching/LFMultiThreadCachingTest.java
>>>>
>>>> as noted in bug JDK-8150014. Until that test instability is resolved,
>>>> the
>>>> test should be demoted from tier 1 to tier 2.
>>>>
>>>> Please see the patch below.
>>>>
>>>> (Circumstantial evidence points to the changes in JDK-8145485 being
>>>> involved.)
>>>>
>>>> -Joe
>>>>
>>>> diff -r 3973fe856db2 test/TEST.groups
>>>> --- a/test/TEST.groups    Wed Feb 17 12:47:35 2016 -0800
>>>> +++ b/test/TEST.groups    Thu Feb 18 16:44:40 2016 -0800
>>>> @@ -29,6 +29,7 @@
>>>>      :jdk_lang \
>>>>      -java/lang/ProcessHandle/TreeTest.java \
>>>>      -java/util/zip/TestLocalTime.java \
>>>> +    -java/lang/invoke/LFCaching/LFMultiThreadCachingTest.java \
>>>>      :jdk_util \
>>>>      -java/util/WeakHashMap/GCDuringIteration.java \
>>>>      -java/util/concurrent/ThreadPoolExecutor/ConfigChanges.java \
>>>> @@ -39,6 +40,7 @@
>>>>      :jdk_math
>>>>
>>>> tier2 = \
>>>> +    java/lang/invoke/LFCaching/LFMultiThreadCachingTest.java \
>>>>      java/lang/ProcessHandle/TreeTest.java \
>>>>      java/util/zip/TestLocalTime.java \
>>>>      java/util/WeakHashMap/GCDuringIteration.java \
>>>>
>



More information about the core-libs-dev mailing list