[11] RTM tests fail

[Christian Thalinger]

> On Jun 25, 2018, at 8:35 AM, Lindenmaier, Goetz <goetz.lindenmaier at sap.com> wrote:
> 
> Hi Christian,
> 
> We also saw lot's of failures here, but Gustavo fixed a row of issues: 
> 8204134: jtreg: Fix RTM abort provoker for various tests after "8149159: Clean up Unsafe"
> 8204135: jtreg: Fix failing RTM test TestUseRTMXendForLockBusy
> 8204136: jtreg: Fix failing RTM test RTMSpinLoopCount
> 
> Did you check with or without these fixes?

Without, unfortunately.  Are all of the failures fixed now (in jdk-11+19)?

> 
> Best regards,
>  Goetz.
> 
>> -----Original Message-----
>> From: hotspot-compiler-dev [mailto:hotspot-compiler-dev-
>> bounces at openjdk.java.net] On Behalf Of Christian Thalinger
>> Sent: Montag, 25. Juni 2018 14:17
>> To: hotspot compiler <hotspot-compiler-dev at openjdk.java.net>
>> Subject: [11] RTM tests fail
>> 
>> Hi!  Is anyone running the compiler/rtm/ tests on a machine that actually
>> supports RTM?
>> 
>> Every time one of our nightly jobs ends up on a machine that supports RTM a
>> bunch of locking tests and the print test fail:
>> 
>> 	• compiler/rtm/locking/TestRTMAbortRatio.java.TestRTMAbortRatio
>> 	•
>> compiler/rtm/locking/TestRTMAbortThreshold.java.TestRTMAbortThreshold
>> 	•
>> compiler/rtm/locking/TestRTMAfterNonRTMDeopt.java.TestRTMAfterNonR
>> TMDeopt
>> 	•
>> compiler/rtm/locking/TestRTMDeoptOnHighAbortRatio.java.TestRTMDeopt
>> OnHighAbortRatio
>> 	•
>> compiler/rtm/locking/TestRTMDeoptOnLowAbortRatio.java.TestRTMDeopt
>> OnLowAbortRatio
>> 	•
>> compiler/rtm/locking/TestRTMLockingThreshold.java.TestRTMLockingThresh
>> old
>> 	•
>> compiler/rtm/locking/TestUseRTMXendForLockBusy.java.TestUseRTMXend
>> ForLockBusy
>> 	•
>> compiler/rtm/print/TestPrintPreciseRTMLockingStatistics.java.TestPrintPreci
>> seRTMLockingStatistics
>> 
>> Exceptions look like this (for all the failures above in order):
>> 
>> java.lang.RuntimeException: Actual abort ratio (10000) should lower or equal
>> to specified (100).: expected that 10000 <= 100
>> 
>> java.lang.RuntimeException: Expected that method with rtm lock elision was
>> deoptimized after 1 lock attempts: expected 3320 to equal 1
>> 
>> java.lang.RuntimeException: Two uncommon traps with reason
>> rtm_state_change should be fired.: expected 0 to equal 2
>> 
>> java.lang.RuntimeException: Expected to get only one deoptimization due to
>> rtm state change: expected 0 to equal 1
>> 
>> java.lang.RuntimeException: Expected to get only one deoptimization due to
>> rtm state change: expected 0 to equal 1
>> 
>> java.lang.RuntimeException: There should be exactly one statistics entry
>> corresponding to ProfileRTM state.: expected null to not equal null
>> 
>> java.lang.RuntimeException: VM output should contain exactly one rtm
>> locking statistics entry for method compiler.testlibrary.rtm.BusyLock::test:
>> expected 0 to equal 1
>> 
>> java.lang.RuntimeException: RTM locking statistics should contain non zero
>> aborts count for abort reason 0: expected 0 > 0
>> 
>> Since these tests are not excluded I have to assume that either (a) they work
>> for you (really?) or (b) that Oracle (or someone else) doesn’t have machines
>> that support RTM.