Changes

jogl: replce more unneeded String() construction

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: remove clone() method that only throws Exceptions

commit:
a4e3f241cfba55e407c68eba91ffcc4beb0758b5 (Analysis of glXMakeCurrent freeze on ATI fglrx 8.78.6; Misc ..)

Removed 'implements Clonable' from NamedDisplay, which means Object.clone() will
always throw CloneNotSupportedException. Kill clone().

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: remove clone() version that can only ever fail

commit:
d75835796900cac602f7e5789601ffba0a27efe2 (Graph: More std. functionality (equals, clone) / Better in-place transformation (cubic -> quadratic))

Added a clone method, but did not implement Cloneable, meaning Object.clone() will
always throw CloneNotSupportedException.  This method never returns anythng but null.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: enable generics annotations on TextRenderer

These were already here, enable them.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: adapt code generator to output @Override annotations in various places

This adds the @Override to methods in the GL implementations (debug, trace, etc) for methods
implementing a GL Profile.  It also adds @Override for the toString() method.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: use .format() to build the method signature when generating pipeline code

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: change code generation for Debug pipelines to avoid building large strings

- split the check for a GL error from the output of the error string
- only build the caller String when there is an error to report
- wrap String building in an if() block rather than using an early return as we
don't know the return type in the postDownstreamCallHook

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: do not generate an extra String for the Debug pipeline wrappers

Build the caller string directly and don't wrap it in an extra constructor.

Update the indentation level for the newly added if-block from the previous commit.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: save on class size in the Debug pipelines

Using a format String and arg list produces smaller class files for the Debug classes
as many of the format strings are identical and shared.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: pass format and arguments directly to the printGLError function

This saves us a bit more code size as the String.format is now in the common
helper rather than in every GL wrapper function.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

Fix Bug 866 - Frequent IndexOutOfBoundsException in jogamp.opengl.egl.EGLGraphicsConfigurationFactory: Add missing 'else' in branch

MacOSXJAWTWindow.CALayer DEBUG: Only Dump AWT location-on-screen if property DEBUG_CALAYER_POS_CRITICAL is explicitly set.

- DEBUG_CALAYER_POS_CRITICAL = nativewindow.debug.JAWT.OSXCALayerPos

Since AWT's location-on-screen query can cause an AWT deadlock,
which is the sole purpose of our custom lock-free impl,
don't enable it's DEBUG output w/ default DEBUG flags.

jogl: add final to Debug wrapper error code and fix indentation

Make the indentation consistent in each function wrapper.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: indent trace statements in the Trace pipelines when we hit glBegin

The indent was being reduced on glEnd/glEndList but was never being incremented.
The intent appeared to be to indent all statements between glBEgin/glEnd pairs to
show the nested context of those calls.  Add the increment after printing the glBegin
trace statement.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

GLContext CTX_IMPL_* bits: Use 10 cached bits (+1), and 6 uncached (-1) ; CTX_IMPL_FP32_COMPAT_API: "FP32 compat-api" -> "FP32 compat"

jogl: add generic annotations to the packed rectangle utility package

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

jogl: add generics annotations to lists in the waveout code

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

Bug 867 - OSX 10.9: Recognize OpenGL Core Profile > 3.0 - Add GLRendererQuirks: Quirk GL4NeedsGL3Request and 'sticky device quirks'

The 'sticky device quirks' are required to share quirks among devices
as collected while mapping the GL versions (audit).

Those are context independent and may only be detected for certain contexts.

They can be pushed/added to the context's quirks, but also queried at context
creation (after mapping) - before the local quirks are being created.

Bug 871 - Add optional xcode.clang support for all modules (Extends Bug 837 w/ xcode's xcrun) - Remove abs. include path.

#include </usr/include/machine/types.h> -> #include <machine/types.h>

Bug 867 - OSX: Allow core >=4 if isMavericksOrLater; Use [kCGLOGLPVersion_GL4_Core, kCGLOGLPVersion_GL3_Core] for major==4 depending on sticky GLRendererQuirks.GL4NeedsGL3Request

Bug 867 OSX [Common Code]: Trigger GLRendererQuirks.GL4NeedsGL3Request and make it sticky; Only alias profiles if HW-Accelerated!

Only alias profiles if HW-Accelerated!
GLContextImpl.mapGLVersions(..) shall not map a higher profile to a lower if it is a software renderer.

+++

GLContextImpl.mapGLVersions(..) attempts to trigger GLRendererQuirks.GL4NeedsGL3Request if OSX 10.9
by creating a GL3 core context first.

+++

GLContextImpl.setGLFunctionAvailability():
  - On OSX 10.9: Detect GLRendererQuirks.GL4NeedsGL3Request and make it sticky (per device)
    while 'withinGLVersionsMapping'

  - Merge sticky quirks w/ local quirks

+++

TestGearsES2NEWT: Add cmdline '-gl2' to force GL2 profile.

Use org.junit.Assert instead of deprecated junit.framework.Assert

Fix GLContext: getGLProfile() add missing GLES3; getAvailableGLProfile(device, ..) shall use GLProfile.get(device, ..)

JoglVersion: "Default Profiles on device" -> "GLProfiles on device" - Since we do use the named device

GLProfile: Fix native profile mapping, i.e. use actual mapped profile-impl as detected and mapped by GLContext ; Enhance glAvailabilityToString(..)

We shall not map profile == profile-impl, i.e. GL3 -> GL3,
but use GLContext.getAvailableGLProfileName(device, reqMajor, reqProfileBits).
The latter reflects the actual mapped context as detected.

glAvailabilityToString(..)
  - Partition result in [Natives, Common and Mappings]

  - Mappings dumps mapped keys to profiles, while excluding default.
    Default is added at last.

  - Add count.

Fix regression of commit 34b35c5a0a379a6b4c0b23b9d347a0b1338f0239 - GLContextImpl.createContextARBVersions(..) erroneous upper bounds check.

Fix Bug 872: ES3 and ES3-GLSL Version not properly Handled

GLContext:
  - Proper API doc for Version* fields

  - getStaticGLSLVersionNumber(): ES3 -> Version300

  - hasGLSL(): Add ES3

ShaderCode:
  - addDefaultShaderPrecision():
    - ES2 default precision: Don't 'tune up' default precision for fragment shader, use 'mediump'
    - Add ES3 default precision (equal to ES2 default precision)

  - requiresDefaultPrecision(): Shall returns 'true' for ES3 as well!

TestGLProfile00NEWT: Dump Desktop's and EGL's default-device GLProfiles

API doc: GLStateKeeper (wording), GLAutoDrawable (enh. init desc.), GLDrawable (Shared Context)

spec-overview.html: Shared OpenGL context must be supported .. (GL spec)

Bug 776 GLContext Sharing: Refine API for relaxed and lazy GLContext sharing ; Fix GLContext memory contract (volatile)

(Unit test remarks see below)

- Add shared GLContext queries
  - Refined GLContextShareSet:
    - Use IdentityHashMap since GLContext's can only be identical w/ same reference (footprint, performance)
    - Add API doc for clarification
    - Add methods:
       - ArrayList<GLContext> getCreatedShares(final GLContext context)
       - ArrayList<GLContext> getDestroyedShares(final GLContext context)
    - Use 'final' where possible

  - Add GLContext methods:
      - boolean isShared()
      - List<GLContext> getCreatedShares()
      - List<GLContext> getDestroyedShares()

- Add GLSharedContextSetter interface defining setting a shared GLContext
  directly (GLContext) or via a GLAutoDrawable:
  - setSharedContext(GLContext)
  - setSharedAutoDrawable(GLAutoDrawable)

  Both cause initialization/creation of GLAutoDrawable's drawable/context to be postponed,
  if the shared GLContext is not yet created natively or
  the shared GLAutoDrawable's GLContext does not yet exist.

  Most of impl. resides in GLDrawableHelper

  Implemented in:
    - GLAutoDrawableBase, GLOffscreenAutoDrawable
    - GLWindow
    - AWT GLCanvas

  TODO:
    - GLJPanel
    - SWT GLCanvas

- GLDrawableFactory:
   - Add 'GLOffscreenAutoDrawable createOffscreenAutoDrawable(..)' variant w/o passing the
     optional shared GLContext _and_ specifying lazy GLContext
     creation. This allows to benefit from GLSharedContextSetter contract.
     Lazy GLContext creation is performed at 2st display() call at the latest.

     All JOGL code and unit tests use this new method now.

  - Mark 'createOffscreenAutoDrawable(..)' w/ shared GLContext argument
    and immediate GLContext creation deprecated - shall be removed in 2.2.0

- Make reference to GLContext and it's native handle volatile
  Since we rely on the query 'GLContext.isCreated()' to properly allow GLAutoDrawable's to query whether
  a shared GLContext is natively created (already), the handle must be volatile
  since such query and the actual creation may operate on different threads.

+++++

- Add/Refine shared GLContext unit tests demonstrating diff. sharing methods.

  All variants of using shared GLContext:
      com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextVBO*

  Most convenient way to share via setSharedAutoDrawable(GLAutoDrawable):
      com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextVBOES2[NEWT|AWT]3

  AWT use w/ JTabbedPane using setSharedAutoDrawable(GLAutoDrawable):
      com.jogamp.opengl.test.junit.jogl.acore.TestSharedContextWithJTabbedPaneAWT

Fix GLCanvas ctor shared GLContext: Use helper.setSharedContext(..) remove local shared ctx fields.

Regression of 7f7a23dd0ddf106e6f0c69fc2a05ff92ac56200e

Bug 776 GLContext Sharing: Add GLSharedContextSetter to GLJPanel

Bug 776 GLContext Sharing: Add note about driver stability (destruction order) ; Fix unit tests (Shared Gears, wait for created context and destruction order)

- Add note about driver stability (destruction order)
  - See GLSharedContextSetter: Don't destroy master context before slaves!

- Fix spec-overview.html#SHARED links, add link to GLSharedContextSetter in SHARED subsection.

- Fix unit tests (Shared Gears, wait for created context and destruction order)
  - The GearsObject sharing was completly bogus!
    It simply used the _same_ GLArrayDataServer instance for sharing,
    but it should use a _copy_ of the shared GLArrayDataServer while
    only preserving the VBO object!
    Fixed, while adding required methods to GLArrayDataServer.

  - Waiting for the created GLContext of a GLAutoDrawable required us
    to pass the latter _and_ check whether it's GLContext exists and is natively created.

  - Accomodated the destruction order - see above!

Bug 776 GLContext Sharing: Add GLSharedContextSetter to SWT GLCanvas

GLDrawableHelper.reshape(): Catch pre-existing GL errors before glViewport(..) and dump it (Add stack trace if DEBUG)

Fix TestSharedContextNewtAWTBug523: Don't use VAO's w/ GL2 context - No VAO on GL2 ctx on OSX ; GLWindow #2 shall also use sharedDrawable; Cleanup  ..

Bug 876: BuildComposablePipeline: getGL*() shall not return downstream.getGL*() [TraceGL* / DebugGL*] ; Simplify GLContextImpl's set Debug/Trace Pipeline

Regression of commit 0002fccdcd6383874b2813dc6bbe3e33f5f00924:
  "Trace/Debug shall utilize downstream identification for isGL*() and getGL*() methods."

  Using the downstream identification commit is right for the isGL*() case,
  however, getGL*() returned the downstream object which makes the caller loosing the pipeline!

  Instead, we shall produce !GEN_GL_IDENTITY_BY_ASSIGNABLE_CLASS:

     "if( isGL<type>() ) { return this; }
     throw new GLException("Not a <type> implementation");"

  or for GEN_GL_IDENTITY_BY_ASSIGNABLE_CLASS:

     "return this;"

GLDrawableFactory: Add 'GLAutoDrawable createDummyAutoDrawable(..)' for convenience

Added:
  GLAutoDrawable createDummyAutoDrawable(AbstractGraphicsDevice deviceReq, boolean createNewDevice, GLProfile glp)

Bug 776 GLContext Sharing: Add copy-ctor to GLArrayData* w/ sliced Buffer; Refine GearsObject* GLArrayDataServer copying; GearsES*: Init VBO eagerly

Add copy-ctor to GLArrayData* w/ sliced Buffer to allow general sharing of VBO via these high-level types.

Refine GearsObject* GLArrayDataServer copying (commit bcfaa149b9803ce33c5a356cbcb45f7dfd3e4361):
  Utilize new GLArrayData* copy-ctor.

GearsES*: Init VBO eagerly, allowing VBO usage after init(..).

Fix #875 - ES version should be strictly validated

When initializing the context in GLContextImpl.setGLFuncAvailability
ES devices must be validated by strictly matching the major version,
otherwise on ES3 devices we were mixing ES1 implementation with ES3
contexts, ultimately crashing in a safeguard.

Signed-off-by: Brice Figureau <brice@daysofwonder.com>

Bug 877 Concurrency Discussion: Update doc MultiThreading.txt, volatile field usage; GLDrawableImpl: Make read-only fields final.

Bug 776 GLContext Sharing: Fix copy-ctor GLArrayDataClient: Create new instance of GLArrayHandler of same type; Simplify GLArrayHandler inheritance.

Refines commit 9f2a9df0a4b7093925c8854b37fba053469a4b35

Bug 776 GLContext Sharing: GLSharedContextSetter API Doc: No 'Driver stability constraints' ; Fixing Test cases: Enable all, GearsObject*: Check VBO

- GLSharedContextSetter API Doc: No 'Driver stability constraints'
  - No driver issues ..
  - Use 'Lifecycle Considerations' to describe usage issues ..

- Fixing Test cases: Enable all, GearsObject*: Check VBO
  - GearsObject* needs to check whether VBO is 'still alive'
    if sharing is enabled.

  - Enable all unit tests.

Bug 776 GLContext Sharing: GLSharedContextSetter API Doc: Add 'glFinish()' to lifecycle considerations ; GearsES2: Add glFinish() after init().

Bug 875: Safeguard setGLFunctionAvailability(.. strictMatch=false.. ) operation, throw InternalError if failing

Fix Bug 875 - Cleanup branch using VersionNumber (fix 'Int' check strictMatch minor) and reuse isES; EGLContext: Use strictMatch for setGLFunctionAvailability() and handle failure; EGLDrawableFactory: Either detect ES3 or ES2.

- Cleanup branch using VersionNumber (fix 'Int' check strictMatch minor) and reuse isES

- EGLContext: Use strictMatch for setGLFunctionAvailability() and handle failure
  On ES, we require strictMatch, cleanup if failing.

- EGLDrawableFactory: Either detect ES3 or ES2.
  Both only available with proper EGL context creation for ES profiles (TODO)

Android NEWT.ScreenDriver MonitorSize: Use xdpi for for width (fix); Add DEBUG output.

Add GLES1Impl 'finalizeInit()' to avoid a 'catched exception roundtrip'

Fix Unit Test Regression (commit 9f2a9df0a4b7093925c8854b37fba053469a4b35): GearsObject used getGL2ES2(), which is not allowed.

TestSharedContextVBOES2NEWT1: Fix copy/paste (mistakenly used sharedGears), add println; GearsES: Add init/shared state to 'toString()'

Add functions for converting to and from binary16 half-precision floating point values. Derived from http://mvn.io7m.com/ieee754b16, of which I am the original author.

Rename Binary*Test -> TestBinary*NOUI to get picked up by build-test.xml's junit.run.noui

Test Rename: Add Bug 729 / Bug 849 to Bug 816 Unit Test Names - Allowing a better unit test lookup

Fix Bug 878 - JAWTWindow's HierarchyListener doesn't set component visible (again) on 'addNotify(..)' - GLCanvas in JtabbedPane disappear

Regression of commit e33e6374e0be0454f7e9732b5f897f84dbc3c4dc (Fix for Bug 729 and Bug 849) !

+++

JAWTWindow's HierarchyListener doesn't set component visible (again) on 'addNotify(..)'

It only renders the component invisible after removeNotify() which is performed implicit anyways ..

Case java.awt.event.HierarchyEvent.DISPLAYABILITY_CHANGED
shall perform similar as our java.awt.event.HierarchyEvent.SHOWING_CHANGED impl.

+++

Tested on Gnu/Linux X11 and OSX incl. re-test Bug 729 and Bug 849 unit tests.

Bug 754 - Remove Ubuntu fonts from jogl-all.jar, provide it separately to reduce footprint for the masses.

Remove the ubuntu fonts from atomic/jogl-util-graph.jar and hence all derivated 'all' JAR files.
The Android jar files still contain the fonts as assets!

atomic/jogl-util-graph-fonts-p0.jar contains the fonts and is either referenced by:
  - UbuntuFontLoader: Using class based Jar URI derivation using TempJarCache to [down]load
    and extract the jar file (similar to native lib-loading).

  - Explicitly via traditional classpath, see jnlp-files/jogl-applet-runner-newt-GraphTextDemo01b-napplet.html

The pack200 jogl-all.jar file is now below 1MB

AndroidGLMediaPlayerAPI14: Avoid possible NPEs

TestBinary16NOUI: Remove hardship from test node - Disable 'verbose' avoids 'out of memory' and saves most of the time; Also run test exclusively.

GLDrawableFactory: createDummy*(..) Pass GLCapabilitiesImmutable + GLCapabilitiesChooser instead of GLProfile, allowing using same or similar caps - important for sharing ctx

EGLDrawableFactory.createDummySurfaceImpl: fix caps to pbuffer - since we do use EGLDummyUpstreamSurfaceHook .. a pbuffer offscreen

fixOffscreenBitOnly(..) may yield FBO queries / chosing .. usually doesn't matter (on devices here),
but not accurate.

Bug 882 - Crash on OSX when closing NEWT window - Check JavaVM and JNIEnv handles before usage in NewtMacWindow (Not the culprit .. but more safe)

Bug 882 - Crash on OSX when closing NEWT window - Fix: Release NewtMacWindow  manually in close0()

Release NewtMacWindow  manually in close0()
  - Mark [NewtMacWindow setReleasedWhenClosed: NO] in init0(..)
  - Release NewtMacWindow manually in close0(..)

Check pointer args in close0(..)

Bug 881 - Add 'Application-Name' in Jar's manifest to avoid Java6 NPEs ..

Clarify Bug 692: Unbinding a VAO does _not_ imply unbinding of set VBOs (spec doesn't mention it, and it does not show results w/ CPU sourced rendering) ; Clean up GLBuffer*Tracker

+ * Note that VAO initialization does unbind the VBO .. since otherwise they are still bound
+ * and the CPU_SRC test will fail!<br/>
+ * The OpenGL spec does not mention that unbinding a VAO will also unbind the bound VBOs
+ * during their setup.<br/>
+ * Local tests here on NV and AMD proprietary driver resulted in <i>no ourput image</i>
+ * when not unbinding said VBOs before the CPU_SRC tests.<br/>
+ * Hence Bug 692 Comment 5 is invalid, i.e. <https://jogamp.org/bugzilla/show_bug.cgi?id=692#c5>,
+ * and we should throw an exception to give users a hint!

Leaving uncommented code in GLBufferStateTracker ..

+++

- Clean up GLBuffer*Tracker
  - Use final
  - Use static final keyNotFound value.

Bug 852: Add unit test TestCPUSourcingAPINEWT validating CPU sourcing, i.e. expecting exception w/ core profile!

Bug 885 - GLMediaPlayer: Allow single threaded mode - Especially where multiple media textures (Android) or shared GL context are not usable.

- GLMediaPlayer:
  - TEXTURE_COUNT_MIN is the new minimum: '1' - i.e. no multithreading, single threaded player
  - TEXTURE_COUNT_DEFAULT is '4' - multithreaded

- GLMediaPlayerImpl:
  - Add Single threaded mode, but perform initStreamImpl(..) off-thread.
  -

GLRendererQuirks: Add GLSharedContextBuggy ('Mesa Intel 9.2.1' and 'Hisilicon Immersion.16')

Note: Even though Mesa Intel driver crashes w/ heavy multithreading (Bug 873),
it works well w/ our multithreaded GLMediaPlayer.