NativeLibLoader - installLibraryFromResource - RuntimeImage
Kevin Rushforth
kevin.rushforth at oracle.com
Fri Jul 27 14:49:03 UTC 2018
Those are all excellent points. This suggests we might want to adopt a
general policy of "the current OpenJFX dev works with the latest
released OpenJDK plus the current OpenJDK under development" is the
better way to go. Otherwise it will be several years before OpenJFX can
adopt, say, a new JDK 12 or JDK 13 feature.
I note that this matches the model and spirit of the OpenJDK
development. If you want to get the latest and greatest features, use
the latest feature release. If you want stability, use a supported LTS
release. Makes sense to me.
-- Kevin
On 7/27/2018 7:35 AM, Johan Vos wrote:
> I think we have to distinguish between the technical work done by
> OpenJFX and the distributions/support offered by companies; similar to
> the technical work in OpenJDK and the distributions/support offered by
> Oracle and others.
>
> The technical work done here should imo be able to use all
> functionality offered by the latest released JDK (11/12/...)
>
> I don't think there is a good reason on why the head-development of
> OpenJFX can not use the latest JDK release. If developers/companies
> want to switch to the latest and greatest OpenJFX release, they can
> also upgrade to the latest released OpenJDK.
>
> And if there are company rules preventing this, well, that is a reason
> to consider commercial support.
> I don't like to use my commercial hat in this list, but this is why we
> have JavaFX Enterprise support with Gluon:
> https://gluonhq.com/commercial-support-for-javafx/
> Also, not that it is relevant on this list, but the revenues from
> commercial enterprise support allow us to spend resources on working
> on OpenJFX.
>
> - Johan
>
> On Fri, Jul 27, 2018 at 4:12 PM Kevin Rushforth
> <kevin.rushforth at oracle.com <mailto:kevin.rushforth at oracle.com>> wrote:
>
> Hi Steve,
>
>
>> I don't know if this discussion already comes up but I didn't
>> found any related answers.
>> Is there a self-defined goal about the compability of versions
>> OpenJFX and the OracleJDK LTS versions?
>> E.g. OpenJFX will be compatible to OracleJDK 11 until the next
>> LTS version of Oracle will be released
>
> That's an excellent question. While I might expect OpenJFX 12 to
> stop running on JDK 10 at some point (JDK 10 is EOLed by JDK 11,
> and dropping support for it would allow us to get rid of the old
> swing interop implementation and build logic), having it run on
> JDK 11 is a requirement: At a minimum we will need to run on the
> latest released version at any point, since we are no longer
> bundled. Beyond that, it's up for discussion among the OpenJFX
> community: I can see value in allowing, say, OpenJFX 13 or 14 to
> continue to run with JDK 11 LTS, but there is also a cost to doing
> so, both in terms of testing and also in either avoiding new JDK
> features or using reflection or other techniques to optionally use
> them.
>
> What do others think?
>
>
> -- Kevin
>
>
>
> On 7/27/2018 6:43 AM, Steve Hruda wrote:
>> Hi Kevin,
>> at the moment it's not really a use-case. I didn't used the jmods
>> because I test if the maven artifacts could easier our life in
>> case of maintaining different product versions.
>> I love the easy distribution of the maven artifacts across
>> different machines (developers & build servers).
>>
>> But so far I'm not sure if I the maven artifacts bring me a real
>> benefit compared to the jmod's.
>>
>> Off-Topic:
>> I don't know if this discussion already comes up but I didn't
>> found any related answers.
>> Is there a self-defined goal about the compability of versions
>> OpenJFX and the OracleJDK LTS versions?
>> E.g. OpenJFX will be compatible to OracleJDK 11 until the next
>> LTS version of Oracle will be released
>>
>> Best Regards,
>> Steve
>> Kevin Rushforth <kevin.rushforth at oracle.com
>> <mailto:kevin.rushforth at oracle.com>> schrieb am Do., 26. Juli
>> 2018, 14:42:
>>
>> Hi Steve,
>>
>> Seems it will be an easy bug for Johan to fix, but I'm
>> curious about how
>> you created your custom image. The jmods are more suitable
>> for that
>> approach, since their purpose is to be used with jlink to
>> create a
>> custom image, in which case the native libraries will be
>> linked into the
>> image. Maybe I'm missing some use case, though.
>>
>> -- Kevin
>>
>>
>> On 7/26/2018 2:46 AM, Johan Vos wrote:
>> > Hi Steve,
>> >
>> > That looks like a bug. The libPrefix and libSuffix are
>> indeed not set in
>> > cases where usingModules is true, which is only the case
>> when the jrt
>> > protocol is used.
>> > It seems to me the prefix/suffix should always be computed.
>> It doesn't look
>> > right that they are computed inside the
>> loadLibraryFullPath. But it looks
>> > even worse that the "reallib" is calculated based on
>> statements that might
>> > not be reached, where it could have used
>> System.mapLibraryName()
>> >
>> > I'll create an issue and a PR to fix this.
>> >
>> > Thanks for reporting,
>> >
>> > - Johan
>> >
>> >
>> > On Thu, Jul 26, 2018 at 11:03 AM Steve Hruda
>> <steve.hruda at gmail.com <mailto:steve.hruda at gmail.com>> wrote:
>> >
>> >> Hi,
>> >> I created a custom runtime image (windows x64) for my
>> JavaFX application
>> >> and used the OpenJFX 11-ea+19 maven artifacts.
>> >>
>> >> I get the following exception if I try to execute my
>> application:
>> >>
>> >> Graphics Device initialization failed for : d3d, sw
>> >> Error initializing QuantumRenderer: no suitable pipeline found
>> >> java.lang.RuntimeException: java.lang.RuntimeException:
>> Error initializing
>> >> QuantumRenderer: no suitable pipeline found
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumRenderer.getInstance(QuantumRenderer.java:280)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumToolkit.init(QuantumToolkit.java:222)
>> >> at
>> >>
>> javafx.graphics/com.sun.javafx.tk.Toolkit.getToolkit(Toolkit.java:260)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:263)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:157)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.startToolkit(LauncherImpl.java:658)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplicationWithArgs(LauncherImpl.java:409)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplication(LauncherImpl.java:363)
>> >> at
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
>> >> Method)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>> >> at
>> java.base/java.lang.reflect.Method.invoke(Method.java:564)
>> >> at
>> >>
>> >>
>> java.base/sun.launcher.LauncherHelper$FXHelper.main(LauncherHelper.java:941)
>> >> Caused by: java.lang.RuntimeException: Error initializing
>> QuantumRenderer:
>> >> no suitable pipeline found
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumRenderer$PipelineRunnable.init(QuantumRenderer.java:94)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumRenderer$PipelineRunnable.run(QuantumRenderer.java:124)
>> >> at java.base/java.lang.Thread.run(Thread.java:844)
>> >> Exception in thread "main"
>> java.lang.reflect.InvocationTargetException
>> >> at
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
>> >> Method)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>> >> at
>> java.base/java.lang.reflect.Method.invoke(Method.java:564)
>> >> at
>> >>
>> >>
>> java.base/sun.launcher.LauncherHelper$FXHelper.main(LauncherHelper.java:941)
>> >> Caused by: java.lang.RuntimeException: No toolkit found
>> >> at
>> >>
>> javafx.graphics/com.sun.javafx.tk.Toolkit.getToolkit(Toolkit.java:272)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:263)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:157)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.startToolkit(LauncherImpl.java:658)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplicationWithArgs(LauncherImpl.java:409)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplication(LauncherImpl.java:363)
>> >> ... 5 more
>> >>
>> >>
>> >> -Djavafx.verbose=true shows that JavaFX isn't able to load
>> the native libs
>> >> from the javafx-graphics jar.
>> >>
>> >>
>> >> I debugged the NativeLibLoader and loadLibraryFromResource and
>> >> installLibraryFromResource will be executed which is ok.
>> But the
>> >> *reallib *value for
>> >> all dll's is wrong because of the empty *libSuffix*.
>> >>
>> >> e.g. the reallib value of *api-ms-win-core-console-l1-1-0* is
>> >> */api-ms-win-core-console-l1-1-0* and not
>> >> */api-ms-win-core-console-l1-1-0.dll*
>> >>
>> >> Is it intentional that *libPrefix *& *libSuffix *will not
>> be set in case of
>> >> the jrt protocol ?
>> >>
>> >>
>> >>
>> https://github.com/javafxports/openjdk-jfx/blob/c168ab56accd7e74d53737bc0832495dbc318e52/modules/javafx.graphics/src/main/java/com/sun/glass/utils/NativeLibLoader.java#L308
>> >>
>> >>
>> >>
>> >>
>> >>
>> https://github.com/javafxports/openjdkjfx/blob/c168ab56accd7e74d53737bc0832495dbc318e52/modules/javafx.graphics/src/main/java/com/sun/glass/utils/NativeLibLoader.java#L338
>> >>
>> >>
>> >>
>> >>
>> >> Best Regards,
>> >> Steve
>> >>
>>
>>
>> Kevin Rushforth <kevin.rushforth at oracle.com
>> <mailto:kevin.rushforth at oracle.com>> schrieb am Do., 26. Juli
>> 2018, 14:42:
>>
>> Hi Steve,
>>
>> Seems it will be an easy bug for Johan to fix, but I'm
>> curious about how
>> you created your custom image. The jmods are more suitable
>> for that
>> approach, since their purpose is to be used with jlink to
>> create a
>> custom image, in which case the native libraries will be
>> linked into the
>> image. Maybe I'm missing some use case, though.
>>
>> -- Kevin
>>
>>
>> On 7/26/2018 2:46 AM, Johan Vos wrote:
>> > Hi Steve,
>> >
>> > That looks like a bug. The libPrefix and libSuffix are
>> indeed not set in
>> > cases where usingModules is true, which is only the case
>> when the jrt
>> > protocol is used.
>> > It seems to me the prefix/suffix should always be computed.
>> It doesn't look
>> > right that they are computed inside the
>> loadLibraryFullPath. But it looks
>> > even worse that the "reallib" is calculated based on
>> statements that might
>> > not be reached, where it could have used
>> System.mapLibraryName()
>> >
>> > I'll create an issue and a PR to fix this.
>> >
>> > Thanks for reporting,
>> >
>> > - Johan
>> >
>> >
>> > On Thu, Jul 26, 2018 at 11:03 AM Steve Hruda
>> <steve.hruda at gmail.com <mailto:steve.hruda at gmail.com>> wrote:
>> >
>> >> Hi,
>> >> I created a custom runtime image (windows x64) for my
>> JavaFX application
>> >> and used the OpenJFX 11-ea+19 maven artifacts.
>> >>
>> >> I get the following exception if I try to execute my
>> application:
>> >>
>> >> Graphics Device initialization failed for : d3d, sw
>> >> Error initializing QuantumRenderer: no suitable pipeline found
>> >> java.lang.RuntimeException: java.lang.RuntimeException:
>> Error initializing
>> >> QuantumRenderer: no suitable pipeline found
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumRenderer.getInstance(QuantumRenderer.java:280)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumToolkit.init(QuantumToolkit.java:222)
>> >> at
>> >>
>> javafx.graphics/com.sun.javafx.tk.Toolkit.getToolkit(Toolkit.java:260)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:263)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:157)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.startToolkit(LauncherImpl.java:658)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplicationWithArgs(LauncherImpl.java:409)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplication(LauncherImpl.java:363)
>> >> at
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
>> >> Method)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>> >> at
>> java.base/java.lang.reflect.Method.invoke(Method.java:564)
>> >> at
>> >>
>> >>
>> java.base/sun.launcher.LauncherHelper$FXHelper.main(LauncherHelper.java:941)
>> >> Caused by: java.lang.RuntimeException: Error initializing
>> QuantumRenderer:
>> >> no suitable pipeline found
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumRenderer$PipelineRunnable.init(QuantumRenderer.java:94)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.tk.quantum.QuantumRenderer$PipelineRunnable.run(QuantumRenderer.java:124)
>> >> at java.base/java.lang.Thread.run(Thread.java:844)
>> >> Exception in thread "main"
>> java.lang.reflect.InvocationTargetException
>> >> at
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
>> >> Method)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
>> >> at
>> >>
>> >>
>> java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>> >> at
>> java.base/java.lang.reflect.Method.invoke(Method.java:564)
>> >> at
>> >>
>> >>
>> java.base/sun.launcher.LauncherHelper$FXHelper.main(LauncherHelper.java:941)
>> >> Caused by: java.lang.RuntimeException: No toolkit found
>> >> at
>> >>
>> javafx.graphics/com.sun.javafx.tk.Toolkit.getToolkit(Toolkit.java:272)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:263)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.PlatformImpl.startup(PlatformImpl.java:157)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.startToolkit(LauncherImpl.java:658)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplicationWithArgs(LauncherImpl.java:409)
>> >> at
>> >>
>> >>
>> javafx.graphics/com.sun.javafx.application.LauncherImpl.launchApplication(LauncherImpl.java:363)
>> >> ... 5 more
>> >>
>> >>
>> >> -Djavafx.verbose=true shows that JavaFX isn't able to load
>> the native libs
>> >> from the javafx-graphics jar.
>> >>
>> >>
>> >> I debugged the NativeLibLoader and loadLibraryFromResource and
>> >> installLibraryFromResource will be executed which is ok.
>> But the
>> >> *reallib *value for
>> >> all dll's is wrong because of the empty *libSuffix*.
>> >>
>> >> e.g. the reallib value of *api-ms-win-core-console-l1-1-0* is
>> >> */api-ms-win-core-console-l1-1-0* and not
>> >> */api-ms-win-core-console-l1-1-0.dll*
>> >>
>> >> Is it intentional that *libPrefix *& *libSuffix *will not
>> be set in case of
>> >> the jrt protocol ?
>> >>
>> >>
>> >>
>> https://github.com/javafxports/openjdk-jfx/blob/c168ab56accd7e74d53737bc0832495dbc318e52/modules/javafx.graphics/src/main/java/com/sun/glass/utils/NativeLibLoader.java#L308
>> >>
>> >>
>> >>
>> >>
>> >>
>> https://github.com/javafxports/openjdkjfx/blob/c168ab56accd7e74d53737bc0832495dbc318e52/modules/javafx.graphics/src/main/java/com/sun/glass/utils/NativeLibLoader.java#L338
>> >>
>> >>
>> >>
>> >>
>> >> Best Regards,
>> >> Steve
>> >>
>>
>
More information about the openjfx-dev
mailing list