Experimental options in official builds ?
Kirk Pepperdine
kirk at kodewerk.com
Wed Oct 9 01:42:05 PDT 2013
this is normal
-- Kirk
On 2013-10-02, at 6:27 PM, Pierre Laporte <pierre at pingtimeout.fr> wrote:
> Hi guys
>
> I just discovered that you can unlock experimental VM options in a regular
> build of the JDK.
>
> This page<http://www.programmingforliving.com/2013/08/how-to-speed-up-eclipse-sts-startup.html>tells
> that unlocking some experimental features can speed up Spring
> ToolSuite. I made some tests on jdk 1.7.0_25 and 1.7.0_40 on linux machines
> and I can indeed modify experimental flags. I thought that experimental
> flags were only for debug build.
>
> Is this normal ?
>
>
> $ java -XX:+PrintFlagsFinal -version | grep experimental
> java version "1.7.0_40"
> OpenJDK Runtime Environment (IcedTea 2.4.2) (ArchLinux build
> 7.u40_2.4.2-1-x86_64)
> OpenJDK 64-Bit Server VM (build 24.0-b56, mixed mode)
>
>
> $ java *-XX:+UnlockExperimentalVMOptions* -XX:+PrintFlagsFinal -version |
> grep experimental
> uintx ArrayAllocatorMallocLimit =
> 18446744073709551615{experimental}
> intx G1ExpandByPercentOfAvailable = 20
> {experimental}
> ccstr G1LogLevel =
> {experimental}
> uintx G1MaxNewSizePercent = 60
> {experimental}
> uintx G1MixedGCLiveThresholdPercent = 65
> {experimental}
> uintx G1NewSizePercent = 5
> {experimental}
> uintx G1OldCSetRegionThresholdPercent = 10
> {experimental}
> bool G1UseConcMarkReferenceProcessing = true
> {experimental}
> bool G1UseParallelRSetScanning = true
> {experimental}
> bool G1UseParallelRSetUpdating = true
> {experimental}
> double ObjectCountCutOffPercent = 0.500000
> {experimental}
> intx PredictedLoadedClassCount = 0
> {experimental}
> bool TrustFinalNonStaticFields = false
> {experimental}
> * bool UnlockExperimentalVMOptions := true
> {experimental} *
> bool UseCriticalCMSThreadPriority = false
> {experimental}
> bool UseCriticalCompilerThreadPriority = false
> {experimental}
> bool UseCriticalJavaThreadPriority = false
> {experimental}
> bool UseMemSetInBOT = true
> {experimental}
> uintx WorkStealingHardSpins = 4096
> {experimental}
> intx WorkStealingSleepMillis = 1
> {experimental}
> uintx WorkStealingSpinToYieldRatio = 10
> {experimental}
> uintx WorkStealingYieldsBeforeSleep = 5000
> {experimental}
> java version "1.7.0_40"
> OpenJDK Runtime Environment (IcedTea 2.4.2) (ArchLinux build
> 7.u40_2.4.2-1-x86_64)
> OpenJDK 64-Bit Server VM (build 24.0-b56, mixed mode)
>
>
> $ java *-XX:+UnlockExperimentalVMOptions* -XX:+PrintFlagsFinal *
> -XX:+TrustFinalNonStaticFields* -version | grep experimental
> uintx ArrayAllocatorMallocLimit =
> 18446744073709551615{experimental}
> intx G1ExpandByPercentOfAvailable = 20
> {experimental}
> ccstr G1LogLevel =
> {experimental}
> uintx G1MaxNewSizePercent = 60
> {experimental}
> uintx G1MixedGCLiveThresholdPercent = 65
> {experimental}
> uintx G1NewSizePercent = 5
> {experimental}
> uintx G1OldCSetRegionThresholdPercent = 10
> {experimental}
> bool G1UseConcMarkReferenceProcessing = true
> {experimental}
> bool G1UseParallelRSetScanning = true
> {experimental}
> bool G1UseParallelRSetUpdating = true
> {experimental}
> double ObjectCountCutOffPercent = 0.500000
> {experimental}
> intx PredictedLoadedClassCount = 0
> {experimental}
> * bool TrustFinalNonStaticFields := true
> {experimental}
> bool UnlockExperimentalVMOptions := true
> {experimental} *
> bool UseCriticalCMSThreadPriority = false
> {experimental}
> bool UseCriticalCompilerThreadPriority = false
> {experimental}
> bool UseCriticalJavaThreadPriority = false
> {experimental}
> bool UseMemSetInBOT = true
> {experimental}
> uintx WorkStealingHardSpins = 4096
> {experimental}
> intx WorkStealingSleepMillis = 1
> {experimental}
> uintx WorkStealingSpinToYieldRatio = 10
> {experimental}
> uintx WorkStealingYieldsBeforeSleep = 5000
> {experimental}
> java version "1.7.0_40"
> OpenJDK Runtime Environment (IcedTea 2.4.2) (ArchLinux build
> 7.u40_2.4.2-1-x86_64)
> OpenJDK 64-Bit Server VM (build 24.0-b56, mixed mode)
>
> --
>
> Pierre Laporte
> @pingtimeout <https://twitter.com/pingtimeout>
> pierre at pingtimeout.fr
> http://www.pingtimeout.fr/
More information about the hotspot-dev
mailing list