java11 + release progress thread (also: review)

classic Classic list List threaded Threaded
29 messages Options
12
Reply | Threaded
Open this post in threaded view
|

java11 + release progress thread (also: review)

Sven Gothel
Administrator
Dear All,

I just pushed the java11 changes of gluegen
to our jogamp git repo on branch 'java11'
<https://jogamp.org/cgit/gluegen.git/log/?h=java11>.

This has been tested on
- GNU/Linux Debian 10 (Debian OpenJDK Default)
- MacOS 10.14 (AdoptOpenJDK 11.0.3_7)
- Windows 10 (AdoptOpenJDK 11.0.4_11)

using
- OpenJDK 11 as described above
- apache-ant 1.10.5
- and so forth ..

I have validated the gluegen class files
to comply with the class major number 52 (java8).

Jenkins and all my build/test nodes are also
setup and prepared.

Next step is to test my joal, jogl and jocl
changes on the above platforms and push
the changes after validation.

Then I have to complete the Android build setup.

Last but not least, kick off our jenkins build
for a pre-release test.

Many thanks to Wade.
I cherry picked some of his java11 changes
and other patches of him were inspiring.

As usual, my result might look a little bit convoluted
or verbose - however, I try to keep some flexibility
in the SDK layout we may encounter.
Example:
<https://jogamp.org/cgit/gluegen.git/commit/?h=java11&id=f54e4fbaebfec36bd3b523dc4e23f744e8e6a4ef>

Feel free to discuss certain changes here
as we go. Maybe test building as well.
Note the updated 'gluegen/make/doc/HowToBuild.html',
which details the build invocation.

Cheers, Sven
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

gouessej
Administrator
Hello Sven

Thank you for the great job :)

Personally, I find "-release" easier to use and I feel confident in using it as it has proved to work correctly in my projects, I build JogAmp's Ardor3D Continuation with Java11 and I test it with OpenJDK 8, it works like a charm.

We're going to get rid of the nasty warning in com.jogamp.common.os.NativeLibrary, it's a good piece of news :)
Julien Gouesse | Personal blog | Website
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Wade Walker
Administrator
In reply to this post by Sven Gothel
Hurray! I'm glad it's working out :) And of course your commits may be more complex than mine were -- I wrote mine as Java 11-only, whereas you're keeping more backward compatibility :)
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
In reply to this post by Sven Gothel
Little update on the general infrastructure:

<https://jogamp.org/cgit/gluegen.git/log/?h=java11>

Windows
- Mingw64 8.1.0 (compiler update)

Windows/MacOS
- CMake 2.8.10.2 (For JOAL + optionally JOGL)

GlueGen Buffers.Cleaner fix

Others:
- JOAL passed here on all platforms (not pushed)
- JOGL almost through (Windows testing/bugfixing)
- JOCL TODO

.. soon.

~Sven

On 8/18/19 4:54 AM, Sven Gothel [via jogamp] wrote:

> Dear All,
>
> I just pushed the java11 changes of gluegen
> to our jogamp git repo on branch 'java11'
> <https://jogamp.org/cgit/gluegen.git/log/?h=java11>.
>
> This has been tested on
> - GNU/Linux Debian 10 (Debian OpenJDK Default)
> - MacOS 10.14 (AdoptOpenJDK 11.0.3_7)
> - Windows 10 (AdoptOpenJDK 11.0.4_11)
>
> using
> - OpenJDK 11 as described above
> - apache-ant 1.10.5
> - and so forth ..
>
> I have validated the gluegen class files
> to comply with the class major number 52 (java8).
>
> Jenkins and all my build/test nodes are also
> setup and prepared.
>
> Next step is to test my joal, jogl and jocl
> changes on the above platforms and push
> the changes after validation.
>
> Then I have to complete the Android build setup.
>
> Last but not least, kick off our jenkins build
> for a pre-release test.
>
> Many thanks to Wade.
> I cherry picked some of his java11 changes
> and other patches of him were inspiring.
>
> As usual, my result might look a little bit convoluted
> or verbose - however, I try to keep some flexibility
> in the SDK layout we may encounter.
> Example:
> <https://jogamp.org/cgit/gluegen.git/commit/?h=java11&id=f54e4fbaebfec36bd3b523dc4e23f744e8e6a4ef>
>
>
> Feel free to discuss certain changes here
> as we go. Maybe test building as well.
> Note the updated 'gluegen/make/doc/HowToBuild.html',
> which details the build invocation.
>
> Cheers, Sven
>
>
> ------------------------------------------------------------------------------
> If you reply to this email, your message will be added to the discussion below:
> http://forum.jogamp.org/java11-release-progress-thread-also-review-tp4039948.html
> To start a new topic under general, email [hidden email]
> To unsubscribe from general, click here
> <
> NAML
> <
http://forum.jogamp.org/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>

Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

gouessej
Administrator
Hello

I just call the run() method, I don't directly rely on sun.misc.Unsafe:
http://svn.code.sf.net/p/tuer/code/pre_beta/src/main/java/engine/misc/DeallocationHelper.java

jdk.internal.ref.Cleaner implements Runnable in Java >= 1.9, there is no need to call jdk.internal.ref.Cleaner.clean(), jdk.internal.ref.Cleaner.run() just works and doesn't need to be made accessible.
Julien Gouesse | Personal blog | Website
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
On 8/19/19 11:02 AM, gouessej [via jogamp] wrote:
> Hello
>
> I just call the run() method, I don't directly rely on sun.misc.Unsafe:
> http://svn.code.sf.net/p/tuer/code/pre_beta/src/main/java/engine/misc/DeallocationHelper.java

The returned Cleaner is not impl. Runnable here on Java11.
I dumped the class spec, interfaces - no Runnable there.

final Class<?> cleanerType = _mbbCleaner[0].getReturnType();
// Java >= 9: jdk.internal.ref.Cleaner (NOT accessible!)
// Java <= 8: sun.misc.Cleaner OK

This class was returned not impl Runnable: jdk.internal.ref.Cleaner

I saw your post when searching for it ;-)

~Sven
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

gouessej
Administrator
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
On 8/19/19 12:42 PM, gouessej [via jogamp] wrote:
Yup, b/c I was happy with your Runnable finding
and queried the returned class's interfaces.
Oh well, so we stick with Unsafe for now,
as long it is working...
Great job describing the issue on stackexchange.

And thank you!

~Sven
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
In reply to this post by Sven Gothel
OK, I build joal + jogl on Linux, OSX and Windows.
Passed JOAL unit tests.
Sampled some tests for JOGL, complete test run todo.

https://jogamp.org/cgit/gluegen.git/log/?h=java11
https://jogamp.org/cgit/joal.git/log/?h=java11
https://jogamp.org/cgit/jogl.git/log/?h=java11

Wade, I guess all your patches are merged.

I kept the orig behavior alive if possible,
however - I see that the Java<9 JAWTUtil stuff could
be removed as we do have alternatives.

Question is whether the pixelScale value also
works for Java<9? Does it work for Java8?

OK, JOCL and then one Android round and
it is time for Jenkins :)

~Sven

On 8/19/19 7:21 AM, Sven Gothel [via jogamp] wrote:

> Little update on the general infrastructure:
>
> <https://jogamp.org/cgit/gluegen.git/log/?h=java11>
>
> Windows
> - Mingw64 8.1.0 (compiler update)
>
> Windows/MacOS
> - CMake 2.8.10.2 (For JOAL + optionally JOGL)
>
> GlueGen Buffers.Cleaner fix
>
> Others:
> - JOAL passed here on all platforms (not pushed)
> - JOGL almost through (Windows testing/bugfixing)
> - JOCL TODO
>
> .. soon.
>
> ~Sven
>
> On 8/18/19 4:54 AM, Sven Gothel [via jogamp] wrote:
>
>> Dear All,
>>
>> I just pushed the java11 changes of gluegen
>> to our jogamp git repo on branch 'java11'
>> <https://jogamp.org/cgit/gluegen.git/log/?h=java11>.
>>
>> This has been tested on
>> - GNU/Linux Debian 10 (Debian OpenJDK Default)
>> - MacOS 10.14 (AdoptOpenJDK 11.0.3_7)
>> - Windows 10 (AdoptOpenJDK 11.0.4_11)
>>
>> using
>> - OpenJDK 11 as described above
>> - apache-ant 1.10.5
>> - and so forth ..
>>
>> I have validated the gluegen class files
>> to comply with the class major number 52 (java8).
>>
>> Jenkins and all my build/test nodes are also
>> setup and prepared.
>>
>> Next step is to test my joal, jogl and jocl
>> changes on the above platforms and push
>> the changes after validation.
>>
>> Then I have to complete the Android build setup.
>>
>> Last but not least, kick off our jenkins build
>> for a pre-release test.
>>
>> Many thanks to Wade.
>> I cherry picked some of his java11 changes
>> and other patches of him were inspiring.
>>
>> As usual, my result might look a little bit convoluted
>> or verbose - however, I try to keep some flexibility
>> in the SDK layout we may encounter.
>> Example:
>>
> <https://jogamp.org/cgit/gluegen.git/commit/?h=java11&id=f54e4fbaebfec36bd3b523dc4e23f744e8e6a4ef>
>
>>
>>
>> Feel free to discuss certain changes here
>> as we go. Maybe test building as well.
>> Note the updated 'gluegen/make/doc/HowToBuild.html',
>> which details the build invocation.
>>
>> Cheers, Sven
>>
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Wade Walker
Administrator
It seems like getScaleX() works on Java 8 -- see https://docs.oracle.com/javase/8/docs/api/java/awt/geom/AffineTransform.html#getScaleX--, the call at least exists. I'm not sure if it returns the correct value for Java < 9 on Mac, though :)
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
On 8/20/19 12:52 AM, Wade Walker [via jogamp] wrote:
> It seems like getScaleX() works on Java 8 -- see
> https://docs.oracle.com/javase/8/docs/api/java/awt/geom/AffineTransform.html#getScaleX--,
> the call at least exists. I'm not sure if it returns the correct value for
> Java < 9 on Mac, though :)

Yeah, keeping it as is for now. Thx.

JOGL status:
    - Builds on MacOS, Windows and GNU/Linux now.

    - All but 2 tests passed on Windows

    - GNU/Linux AMD GPU regression
    -- 1) No compat profiles on X11 device
    -- 2) Remaining core only profiles disclose buggy GLProfile selection
    -- This needs to be fixed pre-release

Then JOCL and Android.

~Sven
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
On 8/20/19 1:55 AM, Sven Gothel [via jogamp] wrote:
>     - GNU/Linux AMD GPU regression
>     -- 1) No compat profiles on X11 device
>     -- 2) Remaining core only profiles disclose buggy GLProfile selection
>     -- This needs to be fixed pre-release
Bug 1382 <https://jogamp.org/bugzilla/show_bug.cgi?id=1382>

Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

gouessej
Administrator
:s You'll have to look at GLContextImpl, I suspect my last attempt of fixing Mesa detection, maybe a renderer quirk has become unnecessary.
Julien Gouesse | Personal blog | Website
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
On 8/20/19 11:13 AM, gouessej [via jogamp] wrote:
> :s You'll have to look at GLContextImpl, I suspect my last attempt of fixing
> Mesa detection, maybe a renderer quirk has become unnecessary.

Well, I was hunting witches it seems:
   - GLRendererQuirks.GLNonCompliant
was triggered, i.e. intentionally set and hence the profile dropped.

Our good old:
     * Non compliant GL context due to a buggy implementation not suitable for
use.
     * <p>
     * Currently, Mesa >= 9.1.3 (may extend back as far as 9.0) OpenGL 3.1
compatibility
     * context is not compliant. Most programs will give completely broken
output (or no
     * output at all. For now, this context is not trusted.
     * </p>
     * The above has been confirmed for the following Mesa 9.* GL_RENDERER
strings:
     * <ul>
     *   <li>Mesa .* Intel(R) Sandybridge Desktop</li>
     *   <li>Gallium 0.4 on AMD RS880</li>
     * </ul>
     * </p>
     * <p>
May need to revalidate whether Mesa has re-enabled support for
OpenGL compat profiles >= 3.1?

The real bug here was that GLProfile didn't chose the right profile,
i.e. the default choice should have been made for the core profile.
So I need to fix that.


Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
This bug became Bug 1383: GLProfile fails to select default core profiles when
no compat profiles are available
<https://jogamp.org/bugzilla/show_bug.cgi?id=1383>

and while working on it

fixing: Bug 1385: Limit Quirk GL3CompatNonCompliant to Mesa < 18.2.0
(yes, new Mesa gives proper GL version/profile replies)

feature: Bug 1384 - Allow GLRendererQuirks to be overridden by user properties
(allowing you to inject or ignore the quirks via 2 properties)

All the above is pushed into java11 branch.

Sadly we got a regression in Mesa:
'Bug 1386 - Mesa 18.3.6 hardware renderer (Intel/AMD) freezes after native
parenting'
Which I 'workaround' (not really) on our test machine
using software rendering only 'LIBGL_ALWAYS_SOFTWARE=true'.

Will post a manual unit test run result snapshot in a bit.
There is more work to do, but not really related to java11,
more related to issues with the updated systems (MacOS, Mesa, ..).

~Sven

On 8/20/19 5:39 PM, Sven Gothel [via jogamp] wrote:

> On 8/20/19 11:13 AM, gouessej [via jogamp] wrote:
>> :s You'll have to look at GLContextImpl, I suspect my last attempt of fixing
>> Mesa detection, maybe a renderer quirk has become unnecessary.
>
> Well, I was hunting witches it seems:
>    - GLRendererQuirks.GLNonCompliant
> was triggered, i.e. intentionally set and hence the profile dropped.
>
> Our good old:
>      * Non compliant GL context due to a buggy implementation not suitable for
> use.
>      * <p>
>      * Currently, Mesa >= 9.1.3 (may extend back as far as 9.0) OpenGL 3.1
> compatibility
>      * context is not compliant. Most programs will give completely broken
> output (or no
>      * output at all. For now, this context is not trusted.
>      * </p>
>      * The above has been confirmed for the following Mesa 9.* GL_RENDERER
> strings:
>      * <ul>
>      *   <li>Mesa .* Intel(R) Sandybridge Desktop</li>
>      *   <li>Gallium 0.4 on AMD RS880</li>
>      * </ul>
>      * </p>
>      * <p>
> May need to revalidate whether Mesa has re-enabled support for
> OpenGL compat profiles >= 3.1?
>
> The real bug here was that GLProfile didn't chose the right profile,
> i.e. the default choice should have been made for the core profile.
> So I need to fix that.
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
In reply to this post by Sven Gothel
Java 11 Branch Status update.
=============================
Build gluegen + joal + jogl on Linux, OSX and Windows.
Passed gluegen and joal unit tests.
Issues with jogl unit tests, see below.

<https://jogamp.org/cgit/gluegen.git/log/?h=java11>
<https://jogamp.org/cgit/joal.git/log/?h=java11>
<https://jogamp.org/cgit/jogl.git/log/?h=java11>

JogAmp-Tests-Java11-20190823
====================================

MacOS 10.14.6 freezes 3 / 273
====================================
(same reason)

MacOS 10.14.6 crashes sigsegv 97 / 273
=======================================
(same reason)

MacOS 10.14.6 errors 98 / 273
====================================

Linux Mesa/AMD
====================================
Bug 1386

Linux Mesa/Soft Errors: Classes 19 / 276
=========================================

Windows failed junit classes - errors 4 / 273
==================================================

++++

I attach my stats file here, dunno how it will
show via forum.

Bottom line TODO (and in that order):
- Fix MacOS Crashes (if not due to openjdk build)
- Fix MacOS Freezes
- Update
- Produce a test build
- Fix Linux Mesa/Soft issues
- Fix Windows Mesa/Soft issues



JogAmp-Tests-Java11-20190823.txt (13K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
In reply to this post by Sven Gothel
On 8/23/19 9:58 AM, Sven Gothel wrote:

> Java 11 Branch Status update.
> =============================
> Build gluegen + joal + jogl on Linux, OSX and Windows.
> Passed gluegen and joal unit tests.
> Issues with jogl unit tests, see below.
>
> <https://jogamp.org/cgit/gluegen.git/log/?h=java11>
> <https://jogamp.org/cgit/joal.git/log/?h=java11>
> <https://jogamp.org/cgit/jogl.git/log/?h=java11>
>

..

> ++++
>
> I attach my stats file here, dunno how it will
> show via forum.
>
> Bottom line TODO (and in that order):
> - Fix MacOS Crashes (if not due to openjdk build)
> - Fix MacOS Freezes
> - Update
> - Produce a test build
> - Fix Linux Mesa/Soft issues
> - Fix Windows Mesa/Soft issues
>

Inline stats (forum email uses download attachment):

JogAmp-Tests-Java11-20190823
====================================

MacOS 10.14.6 freezes 3 / 273
====================================
com.jogamp.opengl.test.junit.jogl.acore.ect.TestExclusiveContext01VSyncAnimNEWT
com.jogamp.opengl.test.junit.jogl.acore.ect.TestExclusiveContext12FPSAnimNEWT
com.jogamp.opengl.test.junit.jogl.acore.ect.TestExclusiveContext02FPSAnimNEWT
"Thread-2-UserApp-JUnitTestRunner-FPSAnimator#06-Timer9" #34 prio=5 os_prio=31
cpu=37573.30ms elapsed=414.47s allocated=7316K defined_classes=2
tid=0x00007fb363937000 nid=0xe843 runnable  [0x0000700005bba000]
   java.lang.Thread.State: RUNNABLE
        at jogamp.opengl.macosx.cgl.CGL.flushBuffer(Native Method)
        at
jogamp.opengl.macosx.cgl.MacOSXCGLContext$NSOpenGLImpl.swapBuffers(MacOSXCGLContext.java:1150)
        at
jogamp.opengl.macosx.cgl.MacOSXCGLContext.swapBuffers(MacOSXCGLContext.java:426)
        at
jogamp.opengl.macosx.cgl.MacOSXCGLDrawable.swapBuffersImpl(MacOSXCGLDrawable.java:140)
        - locked <0x0000000787593cf8> (a java.util.ArrayList)
        at jogamp.opengl.GLDrawableImpl.swapBuffers(GLDrawableImpl.java:88)
        at jogamp.opengl.GLDrawableHelper.invokeGLImpl(GLDrawableHelper.java:1295)
        at jogamp.opengl.GLDrawableHelper.invokeGL(GLDrawableHelper.java:1147)
        at com.jogamp.newt.opengl.GLWindow.display(GLWindow.java:759)
        at
com.jogamp.opengl.util.DefaultAnimatorImpl.display(DefaultAnimatorImpl.java:58)
        at com.jogamp.opengl.util.AnimatorBase.display(AnimatorBase.java:453)
        at com.jogamp.opengl.util.FPSAnimator$MainTask.run(FPSAnimator.java:178)
        at java.util.TimerThread.mainLoop(java.base@11.0.3/Timer.java:556)
        at java.util.TimerThread.run(java.base@11.0.3/Timer.java:506)

   Locked ownable synchronizers:
        - <0x0000000787593e18> (a
jogamp.common.util.locks.RecursiveLockImpl01Unfairish$SingleThreadSync)
        - <0x000000078783a458> (a
jogamp.common.util.locks.RecursiveLockImpl01Unfairish$SingleThreadSync)

MacOS 10.14.6 crashes sigsegv 97 / 273
=======================================
# JRE version: OpenJDK Runtime Environment (11.0.3+7) (build 11.0.3+7)
# Java VM: OpenJDK 64-Bit Server VM (11.0.3+7, mixed mode, tiered, compressed
oops, g1 gc, bsd-amd64)
# Problematic frame:
# C  [libosxapp.dylib+0x2b08]  -[NSApplicationAWT sendEvent:]+0x17c

Current thread (0x00007fa0140b3800):  JavaThread "AppKit Thread" daemon
[_thread_in_native, id=775, stack(0x00007ffeea52a000,0x00007ffeead2a000)]

Stack: [0x00007ffeea52a000,0x00007ffeead2a000],  sp=0x00007ffeead26c80,  free
space=8179k
Native frames: (J=compiled Java code, A=aot compiled Java code, j=interpreted,
Vv=VM code, C=native code)
C  [libosxapp.dylib+0x2b08]  -[NSApplicationAWT sendEvent:]+0x17c
C  [AppKit+0x135e0]  -[NSApplication run]+0x2f3
C  [libosxapp.dylib+0x2773]  +[NSApplicationAWT runAWTLoopWithApp:]+0x9d
C  [libawt_lwawt.dylib+0x39af7]  +[AWTStarter starter:headless:]+0x342
C  [JavaNativeFoundation+0x6fce]  +[JNFRunLoop _performCopiedBlock:]+0x11
C  [Foundation+0xb0742]  __NSThreadPerformPerform+0x148
C  [CoreFoundation+0x57683]
__CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION__+0x11
C  [CoreFoundation+0x57629]  __CFRunLoopDoSource0+0x6c
C  [CoreFoundation+0x3afeb]  __CFRunLoopDoSources0+0xc3
C  [CoreFoundation+0x3a5b5]  __CFRunLoopRun+0x4a5
C  [CoreFoundation+0x39ebe]  CFRunLoopRunSpecific+0x1c7
C  [java+0x681c]  CreateExecutionEnvironment+0x191
C  [java+0x28ea]  JLI_Launch+0x5ea
C  [java+0x16f8]  main+0x198
C  [libdyld.dylib+0x163d5]  start+0x1


MacOS 10.14.6 errors 98 / 273
====================================
com.jogamp.opengl.test.junit.jogl.acore.TestAWTCloseX11DisplayBug565
com.jogamp.opengl.test.junit.jogl.acore.TestAddRemove01GLCanvasSwingAWT
com.jogamp.opengl.test.junit.jogl.acore.TestAddRemove02GLWindowNewtCanvasAWT
com.jogamp.opengl.test.junit.jogl.acore.TestBug1146GLContextDialogToolTipAWT
com.jogamp.opengl.test.junit.jogl.acore.TestFBOAutoDrawableDeadlockAWT
com.jogamp.opengl.test.junit.jogl.acore.TestGLAutoDrawableGLCanvasOnOffscrnCapsAWT
com.jogamp.opengl.test.junit.jogl.acore.TestGLAutoDrawableNewtCanvasAWTOnOffscrnCapsAWT
com.jogamp.opengl.test.junit.jogl.acore.TestGLReadBuffer01GLCanvasAWT
com.jogamp.opengl.test.junit.jogl.acore.TestMainVersionGLCanvasAWT
com.jogamp.opengl.test.junit.jogl.acore.TestOffscreenLayer01GLCanvasAWT
com.jogamp.opengl.test.junit.jogl.acore.TestOffscreenLayer02NewtCanvasAWT
com.jogamp.opengl.test.junit.jogl.acore.TestPBufferDeadlockAWT
com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextListAWT
com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextNewtAWTBug523
com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextVBOES2AWT3
com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextVBOES2NEWT5
com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextWithJTabbedPaneAWT
com.jogamp.opengl.test.junit.jogl.acore.TestShutdownCompleteAWT
com.jogamp.opengl.test.junit.jogl.acore.anim.TestAWTCardLayoutAnimatorStartStopBug532
com.jogamp.opengl.test.junit.jogl.acore.ect.TestExclusiveContext01VSyncAnimAWT
com.jogamp.opengl.test.junit.jogl.acore.ect.TestExclusiveContext02FPSAnimAWT
com.jogamp.opengl.test.junit.jogl.acore.glels.TestGLContextDrawableSwitch02AWT
com.jogamp.opengl.test.junit.jogl.acore.glels.TestGLContextDrawableSwitch11NewtAWT
com.jogamp.opengl.test.junit.jogl.acore.glels.TestGLContextDrawableSwitch12AWT
com.jogamp.opengl.test.junit.jogl.acore.glels.TestGLContextDrawableSwitch13Newt2AWT
com.jogamp.opengl.test.junit.jogl.awt.TestAWT01GLn
com.jogamp.opengl.test.junit.jogl.awt.TestAWT03GLCanvasRecreate01
com.jogamp.opengl.test.junit.jogl.awt.TestBug1225EventQueueInterruptedAWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug1245JTabbedPanelCrashAWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug551AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug572AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug642JSplitPaneMixHwLw01AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug664GLCanvasSetVisibleSwingAWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816GLCanvasFrameHoppingB849B889AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816JTabbedPanelVisibilityB849B878AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos01AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos02AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos03aB729AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos03bB849AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos03cB849AWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos04aAWT
com.jogamp.opengl.test.junit.jogl.awt.TestBug816OSXCALayerPos04bAWT
com.jogamp.opengl.test.junit.jogl.awt.TestGLCanvasAWTActionDeadlock00AWT
com.jogamp.opengl.test.junit.jogl.awt.TestGLCanvasAWTActionDeadlock01AWT
com.jogamp.opengl.test.junit.jogl.awt.TestGLCanvasAWTActionDeadlock02AWT
com.jogamp.opengl.test.junit.jogl.awt.TestIsRealizedConcurrency01AWT
com.jogamp.opengl.test.junit.jogl.awt.TestJScrollPaneMixHwLw01AWT
com.jogamp.opengl.test.junit.jogl.awt.TestSwingAWT01GLn
com.jogamp.opengl.test.junit.jogl.awt.text.TestAWTTextRendererUseVertexArrayBug464
com.jogamp.opengl.test.junit.jogl.caps.TestMultisampleES1AWT
com.jogamp.opengl.test.junit.jogl.caps.TestTranslucencyAWT
com.jogamp.opengl.test.junit.jogl.demos.es2.awt.TestGearsES2AWT
com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestGearsES2NewtCanvasAWT
com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestLandscapeES2NewtCanvasAWT
com.jogamp.opengl.test.junit.jogl.demos.gl2.awt.TestGearsAWT
com.jogamp.opengl.test.junit.jogl.demos.gl2.awt.TestGearsAWTAnalyzeBug455
com.jogamp.opengl.test.junit.jogl.demos.gl2.newt.TestGearsNewtAWTWrapper
com.jogamp.opengl.test.junit.jogl.glsl.TestShaderCompilationBug459AWT
com.jogamp.opengl.test.junit.jogl.glu.TestBug463ScaleImageMemoryAWT
com.jogamp.opengl.test.junit.jogl.glu.TestBug694ScaleImageUnpackBufferSizeAWT
com.jogamp.opengl.test.junit.jogl.newt.TestSwingAWTRobotUsageBeforeJOGLInitBug411
com.jogamp.opengl.test.junit.jogl.perf.TestPerf001GLJPanelInit01AWT
com.jogamp.opengl.test.junit.jogl.perf.TestPerf001GLJPanelInit02AWT
com.jogamp.opengl.test.junit.jogl.tile.TestRandomTiledRendering3GL2AWT
com.jogamp.opengl.test.junit.jogl.tile.TestTiledPrintingGearsAWT
com.jogamp.opengl.test.junit.jogl.tile.TestTiledPrintingGearsNewtAWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestGLReadBufferUtilTextureIOWrite01AWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestGLReadBufferUtilTextureIOWrite02AWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestJPEGJoglAWTCompareNewtAWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestPNGTextureFromFileAWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestTexture01AWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestTexture02AWT
com.jogamp.opengl.test.junit.jogl.util.texture.TestTextureSequence01AWT
com.jogamp.opengl.test.junit.newt.TestCloseNewtAWT
com.jogamp.opengl.test.junit.newt.TestEventSourceNotAWTBug
com.jogamp.opengl.test.junit.newt.TestListenerCom01AWT
com.jogamp.opengl.test.junit.newt.TestMultipleNewtCanvasAWT
com.jogamp.opengl.test.junit.newt.TestWindowClosingProtocol01AWT
com.jogamp.opengl.test.junit.newt.TestWindowClosingProtocol03NewtAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersAWTCanvas
com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNEWTWindowAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNewtCanvasAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyCodeModifiersAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyCodesAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyEventAutoRepeatAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyEventOrderAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyPressReleaseUnmaskRepeatAWT
com.jogamp.opengl.test.junit.newt.event.TestParentingFocus01SwingAWTRobot
com.jogamp.opengl.test.junit.newt.event.TestParentingFocus03KeyTraversalAWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting01aAWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting01bAWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting01cAWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting01cSwingAWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting01dAWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting02AWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting03AWT
com.jogamp.opengl.test.junit.newt.parenting.TestParenting04AWT
com.jogamp.opengl.test.junit.newt.parenting.TestTranslucentParentingAWT


+++

Linux Mesa/AMD
====================================
Bug 1386

Linux Mesa/Soft Errors: Classes 19 / 276
=========================================

com.jogamp.opengl.test.junit.jogl.acore.TestGLAutoDrawableDelegateOnOffscrnCapsNEWT

com.jogamp.opengl.test.junit.jogl.acore.TestGLAutoDrawableFactoryGL2OffscrnCapsNEWT

com.jogamp.opengl.test.junit.jogl.acore.TestGLAutoDrawableFactoryGLnBitmapCapsNEWT

com.jogamp.opengl.test.junit.jogl.acore.TestGLAutoDrawableGLWindowOnOffscrnCapsNEWT
    com.jogamp.opengl.test.junit.jogl.acore.TestGLProfile03NEWTOffscreen
    com.jogamp.opengl.test.junit.jogl.acore.TestGLProfile04NEWTOffscreenNoARBCtx
    com.jogamp.opengl.test.junit.jogl.acore.TestPBufferDeadlockAWT
    com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextVBOES2NEWT5
    com.jogamp.opengl.test.junit.jogl.awt.TestBug461PBufferSupersamplingSwingAWT
    com.jogamp.opengl.test.junit.jogl.caps.TestBug605FlippedImageAWT
    com.jogamp.opengl.test.junit.jogl.caps.TestBug605FlippedImageNEWT
    com.jogamp.opengl.test.junit.jogl.caps.TestMultisampleES1NEWT
    com.jogamp.opengl.test.junit.jogl.caps.TestMultisampleES2NEWT
    com.jogamp.opengl.test.junit.jogl.math.TestPMVMatrix01NEWT

com.jogamp.opengl.test.junit.jogl.newt.TestSwingAWTRobotUsageBeforeJOGLInitBug411
    com.jogamp.opengl.test.junit.jogl.offscreen.TestOffscreen01GLPBufferNEWT

com.jogamp.opengl.test.junit.jogl.util.texture.TestBug817GLReadBufferUtilGLCTXDefFormatTypeES2NEWT

com.jogamp.opengl.test.junit.jogl.util.texture.TestGLReadBufferUtilTextureIOWrite01NEWT

com.jogamp.opengl.test.junit.jogl.util.texture.TestGLReadBufferUtilTextureIOWrite02NEWT

+++

Windows failed junit classes - failures 4 / 273
==================================================
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyCodeModifiersAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyCodesAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyEventAutoRepeatAWT
com.jogamp.opengl.test.junit.newt.event.TestNewtKeyEventOrderAWT

Windows failed junit classes - errors 4 / 273
==================================================
com.jogamp.opengl.test.junit.jogl.acore.glels.TestBug722GLContextDrawableSwitchNewt2AWT
com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextVBOES2NEWT5
com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersAWTCanvas
com.jogamp.opengl.test.junit.newt.event.TestNewtEventModifiersNEWTWindowAWT



Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
In reply to this post by Sven Gothel
A little overview might help, here the updated wiki pages for

2.4.0
<https://jogamp.org/wiki/index.php?title=SW_Tracking_Report_Objectives_for_the_release_2.4.0>

Features
<https://jogamp.org/wiki/index.php?title=SW_Tracking_Report_Feature_Objectives_Overview>

~Sven

On 8/23/19 9:58 AM, Sven Gothel [via jogamp] wrote:

> Java 11 Branch Status update.
> =============================
> Build gluegen + joal + jogl on Linux, OSX and Windows.
> Passed gluegen and joal unit tests.
> Issues with jogl unit tests, see below.
>
> <https://jogamp.org/cgit/gluegen.git/log/?h=java11>
> <https://jogamp.org/cgit/joal.git/log/?h=java11>
> <https://jogamp.org/cgit/jogl.git/log/?h=java11>
>
> JogAmp-Tests-Java11-20190823
> ====================================
>
> MacOS 10.14.6 freezes 3 / 273
> ====================================
> (same reason)
>
> MacOS 10.14.6 crashes sigsegv 97 / 273
> =======================================
> (same reason)
>
> MacOS 10.14.6 errors 98 / 273
> ====================================
>
> Linux Mesa/AMD
> ====================================
> Bug 1386
>
> Linux Mesa/Soft Errors: Classes 19 / 276
> =========================================
>
> Windows failed junit classes - errors 4 / 273
> ==================================================
>
> ++++
>
> I attach my stats file here, dunno how it will
> show via forum.
>
> Bottom line TODO (and in that order):
> - Fix MacOS Crashes (if not due to openjdk build)
> - Fix MacOS Freezes
> - Update
> - Produce a test build
> - Fix Linux Mesa/Soft issues
> - Fix Windows Mesa/Soft issues
>
>
>
> *JogAmp-Tests-Java11-20190823.txt* (13K) Download Attachment
> <http://forum.jogamp.org/attachment/4039992/0/JogAmp-Tests-Java11-20190823.txt>
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

gouessej
Administrator
In reply to this post by Sven Gothel
Are some crashes under OS X happening when releasing some resources? Do some problems with AWT concern middle and right mouse clicks?
Julien Gouesse | Personal blog | Website
Reply | Threaded
Open this post in threaded view
|

Re: java11 + release progress thread (also: review)

Sven Gothel
Administrator
In reply to this post by Sven Gothel
I had to change access to internal non-exported
functionality of module java.desktop etc.

This is a workaround (hack) to access certain
modularized encapsulated (hidden) functionality
where there is no replacement.

We simply cannot avoid it foremost for JOGL/AWT.

Therefor I added UnsafeUtil to GlueGen to have
access well centralized via its
'T doWithoutIllegalAccessLogger(Action<T> action)' method.
Here, we simply disable the IllegalAccessLogger temporarily.

<https://jogamp.org/cgit/gluegen.git/commit/?h=java11&id=07c1885e9a3d1f3a3853414648c06fb3864bc69f>

You can see its usage also in JAWTUtil now:

<https://jogamp.org/cgit/jogl.git/commit/?h=java11&id=c5431f46b7bf64f109315ec78461859dd88f202a>

~Sven
12