aboutsummaryrefslogtreecommitdiffstats
path: root/src/jogl/classes/jogamp/opengl/GLContextImpl.java
Commit message (Collapse)AuthorAgeFilesLines
* Cleanup GLContext special entries: getOffscreenContextPixelDataType(), ↵Sven Gothel2012-12-151-9/+42
| | | | | | | | | | | | | | | | | | | | | | | getOffscreenContextReadBuffer(), .. ; Add GLFBODrawable API entries for multi buffering (no impl. yet); GLJPanel 1st simplification using offscreen drawable - Cleanup GLContext special entries: getOffscreenContextPixelDataType(), getOffscreenContextReadBuffer(), .. ; - add: getDefaultReadBuffer() (-> exposed via GLBase as well) - add: isGLOrientationFlippedVertical() - add: getDefaultPixelDataType() - removed impl: getOffscreenContextPixelDataType() - removed impl: getOffscreenContextReadBuffer() - removed impl: offscreenImageNeedsVerticalFlip() - Add GLFBODrawable API entries for multi buffering (no impl. yet); - TODO: Add implementation code in GLFBODrawableImpl - GLJPanel 1st simplification using FBO - Use above new GL/GLContext entries - Fix: getNativeSurface() and getHandle() - TODO: - Remove distinction of 'pbuffer' and 'software', - Use GLDrawableFactory.createOffscreenDrawable(..) - Use GL for FBO swapping
* GLRendererQuirks.RequiresBoundVAO: Removed, it _is_ in the GL 3.2 core spec ↵Sven Gothel2012-10-291-9/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Setting up default VAO for all GL >= 3.2 core ctx. Refines commit 9b6448b1d54716fd455c0cad0c6133c0edeb3bb8 Due to GL 3.2 core spec: E.2. DEPRECATED AND REMOVED FEATURES (p 331) "There is no more default VAO buffer 0 bound, hence generating and binding one to avoid INVALID_OPERATION at VertexAttribPointer." More clear is GL 4.3 core spec: 10.4 (p 307): "An INVALID_OPERATION error is generated by any commands which modify, draw from, or query vertex array state when no vertex array is bound. This occurs in the initial GL state, and may occur as a result of BindVertexAr- ray or a side effect of DeleteVertexArrays." +++ I just have read (same spec) 2.10 (p 46/47): "An INVALID_OPERATION error is generated if any of the *Pointer commands specifying the location and organization of vertex array data are called while zero is bound to the ARRAY_BUFFER buffer object binding point, and the pointer argu- ment is not NULL." .. which only constraints the *Pointer command use to _VBO_, not forcing a VAO. +++
* GLContextImpl GLRendererQuirks.RequiresBoundVAO: Use isGL3() instead of ↵Sven Gothel2012-10-291-1/+1
| | | | simple major version number check.
* GLRendererQuirks: Add RequiresBoundVAO (w/ impl.), GLSLBuggyDiscard (todo) ; ↵Sven Gothel2012-10-281-4/+25
| | | | | | | | | | | | | | | | GLContextImpl: Bind default VAO if having quirk RequiresBoundVAO. OSX w/ OpenGL >= 3 core context implementation requires a bound VAO for vertex attribute operations, i.e. VertexAttributePointer(..). This has been experienced on OSX 10.7.5, OpenGL 3.2 core w/ Nvidia GPU and in several forum posts. Such 'behavior' violates the GL 3.2 core specification, which does not state this requirement, hence it is a bug. (Please correct me if I am wrong!) GLContextImpl works around this quirk, by generating a default VAO and binds it at 1st makeCurrent (@creation) and deletes it at destroy. This is minimal invasive since no action is required for subsequent makeCurrent or release. We assume if a user uses and binds a VAO herself, she will mind this quirk. Note: We could enhance this workaround by quering for a currently bound VAO at makeCurrent() and bind our default if none. However, we refrain from this operation to minimize the workaround and complexity.
* GLContext: Produce and expose GLSL version as VersionNumber and version ↵Sven Gothel2012-10-241-1/+17
| | | | | | | | | string (for shader programs) Uses GL_SHADING_LANGUAGE_VERSION and parses it via VersionNumber, as well as having a static fallback using the GL context version. The value is valid and can be retrieved after ctx has been made current once.
* GLRendererQuirks.NoDoubleBufferedPBuffer: Set if ( mesa || gallium ) && hwAccelSven Gothel2012-10-021-4/+5
|
* Adding Mesa Quirk 'NoSetSwapIntervalPostRetarget': SIGSEGV on ↵Sven Gothel2012-09-281-14/+24
| | | | | | | | | | | | setSwapInterval() after changing the context's drawable w/ 'Mesa 8.0.4' dri2SetSwapInterval/DRI2 (soft & intel) Analyzing 'TestGLContextDrawableSwitchNEWT' crash at setSwapInterval -> dri2SetSwapInterval after retargeting the context (new drawable association). Turns out Mesa's dri2SetSwapInterval may have a bug. +++ GLContext TRACE_SWITCH: Add drawable handle to debug/trace output.
* Add GLRendererQuirks; Fix shared EGL/ES resources;Sven Gothel2012-09-201-10/+45
| | | | | | | | | | | | | | | | | | | | | | | Add GLRendererQuirks: - Contains centralized 'tagged' workarounds for GL renderer bugs (quirks) - Accessible via GLContext and GLDrawableFactory - Initialized in GLContext.setAvailability* - Simplify GLGraphicsConfigurationUtil.fixOffscreenGLCapabilities(..) & use Quirks - use quirks - instead of passing booleans for each config, pass factory & device Fix shared EGL/ES resources: - GLProfile needs to initialize EGLDrawableFactory's shared resources before desktop, so EGLDrawableFactory can use the fallback defaultDisplay & defaultSharedResource for host mapped sharedResources (hack). - If using defaultSharedResources for host mapped ones, do not go through initialization cycles - simply map (sharedResource + context). - EGLDrawableFactory: Use device's unique-key instead of connection only, since the latter causes a collision (EGL-connection == X11-connection).
* Seamless Integration of an FBObject based GLFBODrawable as ↵Sven Gothel2012-09-151-58/+117
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GLOffscreenAutoDrawable.FBO and as an OffscreenLayerSurface's drawable (OSX) - Fix Bugs 569 and 599 Summary: ========= The new FBObject based GLFBODrawable implementation allows the seamless utilization of FBO offscreen rendering in single buffer, double buffer and MSAA mode. The GLFBODrawable uses a parent drawable based on a dummy surface to allow a GLOffscreenAutoDrawable.FBO creation or a mutable surface supporting an existing offscreen layer surface (OSX CALayer). Offscreen GLDrawable's and GLOffscreenAutoDrawable's can be selected via the GLCapabilities. If simply !onscreen is selected in the caps instance w/o enabling FBO, PBuffer or Bitmap, the factory will automatically choose regarding availability: FBO > PBuffer > Bitmap Double buffering is supported in MSAA more (intrinsic) and explicit in non MSAA. It is preferred when delivering resources (texture id's or framebuffer names) to a shared GLContext. This is demonstrated in (emulates our OSX CALayer implementation): TestFBOOffThreadSharedContextMix2DemosES2NEWT, TestFBOOnThreadSharedContext1DemoES2NEWT and with the OSX JAWT OffscreenLayerSurface itself. FBO is the preferred choice. +++ Offscreen drawables can be resized while maintaining a bound GLContext (e.g. w/ GLAutoDrawable). Previously both, drawable and context, needed to be destroyed and recreated at offscreen resize. Common implementation in GLDrawableHelper is used in the implementations (NEWT's GLWindow, AWT GLCanvas, SWT GLCanvas). +++ Tested: ======= Manually run all unit tests on: - Linux x86_64 NVidia/AMD/Mesa3d(ES) - OSX x86_64 NVidia - Windows x86_64 NVidia - Android arm Mali-400/Tegra-2 No regressions. Disclaimer: =========== This feature is committed almost in one patch. Both previous commits were introducing / fixing the capabilities behavior: 90d45928186f2be99999461cfe45f76a783cc961 9036376b7806a5fc61590bf49404eb71830de92f I have to appologize for the huge size and impact (files and platforms) of this commit however, I could not find a better way to inject this feature in one sane piece. NativeWindow Details: ===================== Complete decoupling of platform impl. detail of surfaces implementing ProxySurface. Used to generalize dummy surfaces and EGL surfaces on top of a native platform surface. - ProxySurface.UpstreamSurfaceHook -> UpstreamSurfaceHook - abstract class ProxySurface -> interface ProxySurface + ProxySurfaceImpl - Misc. implementations JOGL Details: ===================== FBOObject: API Change / Simplification & Usability - Removed reference counter to remove complexity, allow user to choose. - Add 'dispose' flag for detachColorbuffer(..), allowing to keep attachment alive - Fix equals operation of Attachment - Check pre-exising GL errors - Interface Colobuffer gets lifecycle methods - Add static factory methods to create Attachments w/o FBObject instance - Reset: - Clip min size to 1 - Keep alive samplingSink, i.e. don't issue resetMSAATexture2DSink(..). It gets called at syncFramebuffer()/use(..) later on before actual usage. This allows the consumer to utilize the GL_FRONT buffer until (e.g.) swap. - misc bugfixes GLOffscreenAutoDrawable: API Change - Reloc and interfacing - class com.jogamp.opengl.OffscreenAutoDrawable -> javax.media.opengl.* interfaces GLOffscreenAutoDrawable extends GLAutoDrawable GLOffscreenAutoDrawable.FBO extends GLOffscreenAutoDrawable, GLFBODrawable - Added general implementation and FBO specialization - Replacing GLPBuffer (deprecated) .. usable for any offscreen GLDrawable via factory GLAutoDrawable: - Add 'GLDrawable getDelegatedDrawable()' - Refine documentation of setContext(..), remove disclaimer and fixme tags GLDrawableFactory: - Refine API doc and it's selection mechanism for offscreen. - Add createOffscreenDrawable(..) - Add createOffscreenAutoDrawable(..) - Add canCreateFBO(..) - Mark createGLPbuffer(..) deprectated Mark GLPBuffer deprecated New: GLFBODrawable extends GLDrawable GLCanvas (AWT and SWT): Add offscreen resize support w/o GLContext recreation GLAutoDrawableBase .. GLWindow: - Add offscreen resize support w/o GLContext recreation - Remove double swapBuffer call - GLBase/GLContext: - Add: - boolean hasBasicFBOSupport() - boolean hasFullFBOSupport() - int getMaxRenderbufferSamples() - boolean isTextureFormatBGRA8888Available() GLContext: Fix version detection and hasGLSL() - Version detection in setGLFunctionAvailability(..) - Query GL_VERSION ASAP and parse it and compare w/ given major/minor - Use parsed version if valid and lower than given _or_ given is invalid. - Use validated version for caching (procaddr, ..), version number, etc. - Fix hasGLSL() Since 'isGL2ES2()' is true if 'isGL2()' and the latter simply alows GL 1.*, we confine the result to a GL >= 2.0 on desktops. FIXME: May consider GL 1.5 w/ extensions. - return isGL2ES2(); + return isGLES2() || + isGL3() || + isGL2() && ctxMajorVersion>1 ; GLDrawableImpl: - Add 'associateContext(GLContext, boolean)' allowing impl. to have a (weak) reference list of bound context. This is was pulled up from the OSX specific drawable impl. - swapBuffersImpl() -> swapBuffersImpl(boolean doubleBuffered) and call it regardless of single buffering. This is required to propagate this event to impl. properly, i.e. FBODrawable requires a swap notification. - Clarify 'contextMadeCurrent(..)' protocol GLDrawableHelper: - Add resize and recreate offscreen drawable util method - Simplify required init/reshape calls for GLEventListener - GLGraphicsConfigurationUtil: - fixWinAttribBitsAndHwAccel: Reflect sharede context hw-accel bits - OSX has no offscreen bitmap, use pbuffer - use proper offscreen auto selection if offscreen and no modes are set EGL Context/Drawable/DrawableFactory: Abstract native platform code out of base classes - Use EGLWrappedSurface w/ UpstreamSurfaceHook to handle upstream (X11, WGL, ..) lifecycle - in case the EGL resource is hooked up on it. Invisible dummy surfaces: All platforms - size is now reduced to 64x64 and decoupled of actual generic mutable size - fix device lifecycle, no more leaks +++ OSX ==== Enable support for GLFBODrawableImpl in offscreen CALayer mode - NSOpenGLImpl: hooks to calayer native code - calayer code: - allows pbuffer and texures (FBO) - decouple size and draw calls avoiding flickering - enable auto resize of calayer tree MacOSXCGLContext: - NSOpenGLImpl: - Fix false pbuffer 'usage', validate the pointer - If !pbuffer, copy other window mode bits of caps - MacOSXCGLGraphicsConfiguration: - Only assume pbuffer if !onscreen - Remove reference of native pixelformat pointer Native code: - use 'respondsToSelector:' query before calling 'new' methods avoiding an error message where unsuported (prev. OSX versions) - if monitor refresh-rate is queried 0, set to default 60hz - add missing NSAutoreleasePool decoration +++ Android / NEWT: =============== Issue setVisible(..) w/o wait, i.e. queue on EDT, @Android surfaceChanged() callback. Otherwise we could deadlock: setVisible(..) -> EDT -> setVisibleImpl(..) -> 'GL-display'. the latter may may cause havoc while Android-EDT is blocked [until it's return].
* Robostness: GLDrawableImpl's contextMadeCurrent()/contextRealized() ; ↵Sven Gothel2012-08-171-2/+19
| | | | | | | | | | | | | | | GLFBODrawableImpl.contextMadeCurrent(false), OffscreenAutoDrawable.setSize(..) GLDrawableImpl's contextMadeCurrent()/contextRealized(): - Catch exception which may appear during callback and cont. w/ GLContextImpl's release()/destroy() while throwing catched exception at end. GLFBODrawableImpl.contextMadeCurrent(false): - Detect null Colorbuffer ASAP and throw exception OffscreenAutoDrawable.setSize(..): - Catch exceptions at 1) GLFBODrawableImpl.setSize(..) and 2) GLContext.release() .. throw it in proper order.
* SWT Update: SWT GLCanvas creates lazy when resource is ready; Create new ↵Sven Gothel2012-07-251-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | NewtCanvasSWT allowing to parent NEWT windows natively. SWT GLCanvas creates lazy when resource is ready - Ensures drawable and context are created when size > zero and native visualID is valid. The latter is platform dependent. - Note that you cannot utilize custom GLCapabilities w/ this one, since the configurations is already realized - use NewtCanvasSWT. Create new NewtCanvasSWT allowing to parent NEWT windows natively: - Similar to NewtCanvasAWT - Allows attaching / detaching NEWT windows NewtCanvasAWT: Public setNEWTChild(..) fixed Added test cases for the above - tested on Linux, OSX and Windows w/ SWT Note: As usual for OSX, add -XstartOnFirstThread Details: - NEWT Display has new method: 'EDTUtil setEDTUtil(EDTUtil)' allowing to set a custom event dispatch utility. We use this to set our SWTEDTUtil for using NEWT w/ SWT complying w/ SWT threading constraints.
* Bug 599 - FBObject / Offscreen Support - Part 1Sven Gothel2012-07-191-57/+133
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - New FBObject implementation handling FBO and it's attachments *** API CHANGE: Util -> Core *** while it's size and sample-count can be reconfigured on the fly. - com.jogamp.opengl.util.FBObject -> com.jogamp.opengl.FBObject - agnostic to texture unit - separate attachments using OO hierarchy reflecting FBO - handling MSAA and blitting - no FBO destruction for reconfig (attach/detach) - New GLFBODrawableImpl impl. an FBObject based GLDrawable - Instantiated by a dummy native surface (onscreen and invisible) hooked up to a dummy GLDrawable, which is the delegation for context creation. - Utilizies ProxySurface.UpstreamSurfaceHook for dummy surface avoiding specialization for native platforms. - TODO: Allow to utilize common surface interface as a dummy-surface to supporting API seperation of windowing/GL. The latter allows impl. of createGLDrawable(NativeSurface) with FBO. - New OffscreenAutoDrawable (extends GLAutoDrawableDelegate) for all offscreen drawables. Shall replace GLPbuffer. - New GLCapabilities*.isFBO() / setFBO(boolean) to request FBO offscreen, similar to isPBuffer(). Rule: if both are requested, FBO shall be favored. - GLContext adds raw FBO availability query (min. FBO avail), FBObject contains fine grained queries (TODO: Move parts to GLContext for efficiency). - Add framebuffer tracking, allowing fast querying: - GLBase/GLContext: public int getBoundFramebuffer(int target); public int getDefaultDrawFramebuffer(); public int getDefaultReadFramebuffer(); - GLContextImpl public final void setBoundFramebuffer(int target, int framebufferName) .. called by GL impl bind framebuffer - GL: getDefaultDrawFramebuffer(), getDefaultReadFramebuffer() Adding default framebuffer queries being issued by GL.glBindFramebuffer(target, 0) w/ a default framebuffer, o.e. zero. This allows a transparent use of a custom FBO even in case the applications attempts to reset FBO to zero. Value flow: GL <- GLContext <- GLDrawable, - GLCapabilities handle fbo/pbuffer seperate, don't disable the other - GLContext/GL track read/write framebuffer to be queried by FBObject to determine whether to bind/unbind a framebuffer - Test cases for multiple FBO w/ and w/o MSAA Other Features: - New interface ProxySurface.UpstreamSurfaceHook, allowing to hook an upstream surface of unknown type providing lifecycle and information (size, ..) callbacks. Used for all new dummy NativeSurface impl and SWT GLCanvas. - GLContext -> GLDrawable propagation context/drawable lifecycle via ProxySurface.UpstreamSurfaceHook allowing dynamic resources to react (create, init, ..) - contextRealized() - contextMadeCurrent() - SurfaceChangeable -> MutableSurface currently only contains setting the surface handle. TODO: May need to move ProxySurface.UpstreamSurfaceHook -> MutableSurface.UpstreamSurfaceHook, allowing other impl. classes (NEWT OffscreenWindow) to utilize the new upstream hookup mechanism - will allow FBO/Dummy window to work. - SWT GLCanvas using ProxySurface.UpstreamSurfaceHook for proper size propagation. - New GLAutoDrawable::getUpstreamWidget(), allowing GLEventListener to fetch the owning Java side UI element (NEWT, SWT, AWT, ..). - GLDrawableFactory: Removed createOffscreenSurface() - unused and not GL related - EGLDrawableFactory handles device/profile avail. mapping while actually creating context/drawable. This allows us to learn whether the ES context is software/hardware as well as FBO avail. - EGLDrawable: Removed secret buckets of EGL configs :) Employ native surface (X11, WGL, ..) to EGL 'mapping' in EGLDrawableFactory utilizing new EGLUpstreamSurfaceHook (implements ProxySurface.UpstreamSurfaceHook). Other Bugs: - Add CTX_OPTION_DEBUG to ctx/extension cache key since only a debug ctx may expose the ARB debug capability. This bug caused lack of ARB/AMD debug functionality. - Fix GLProfile deadlock (debug mode, w/ EGL/ES, no X11), dump availability information _after_ lock. - ImmModeSink draw(): Use GL's glDrawElements(..), don't cast for GL2ES1. Fixes use for GL2ES2. - Fix KeyEvent.getKeyChar() comment (-> only stable for keyTyped(..)) Misc: - Refined alot of API doc - New GLExtensions holds commonly used GL extension strings, allows better referencing and usage lookup. - Move GL (interface) decl. to GLBase - GLBuffers: Cleanup API doc (format, types) - TextureIO: Add PAM and PPM static suffix identifier - GLCapabilities getNumSamples() returns 0 if sampleBuffers is disabled, this seems to be more natural. - finalized a lot
* Enhance Bootsrapping of JOGL around 37% - 40% (1st start in new JVM) - ↵Sven Gothel2012-07-061-17/+83
| | | | | | | | | | | | | | | | | | | | | | | | GLProfile and GLContext* GLProfile: Enhance bootsrapping performance of loading GL*Impl class - Offthread classloading of all GL*Impl via reflection at startup reduces startup time here around 12% (800ms down to 700ms). GLContext*: Enhance bootsrapping performance of querying available GL profiles - Add PROFILE_ALIASING mode, defaults to true - can be disabled w/ property 'jogl.debug.GLContext.NoProfileAliasing' - PROFILE_ALIASING: If true (default), bootstrapping the available GL profiles will use the highest compatible GL context for each profile, hence skipping querying lower profiles if a compatible higher one is found. Linux x86_64 - Nvidia: 28%, 700ms down to 500ms Linux x86_64 - AMD : 40%, 1500ms down to 900ms - GL*Impl: - make fields final: glProfile, _context, buffer*Tracker and glStateTracker - allow null _context/glProfile in initialization (bootstrapping) - JoglVersion.getDefaultOpenGLInfo(..) - add arg: 'boolean withCapabilitiesInfo', allowing to suppres the list of caps
* GLContext adds FBO availability to profile mapping; Fix ↵Sven Gothel2012-07-051-29/+68
| | | | | | | | | | | | | | | | GL[Auto]Drawable/GLContext re-association (switch) incl. unit test. - GLContext adds FBO availability to profile mapping - GLContext added 'GLDrawable setGLDrawable(GLDrawable readWrite, boolean setWriteOnly)' allowing to set the write GLDrawable. This method enables switching context/drawable. Fix GL[Auto]Drawable/GLContext re-association (switch) incl. unit test Commit eed8508ae1132e5f45f788e9cb3f3d5a1050ac70 impl. of GLAutoDrawable's setContext(..) enabled proper setting of the GLAutoDrawable context incl. updating the context's drawables. Test covers: - remove/set (GLContext, GLEventListener) of GL[Auto]Drawable - switch (GLContext, GLEventListener) of 2 GLAutoDrawables
* GLContextImpl/GLDrawableImpl: More fail-safe, cleanup, mark some methods finalSven Gothel2012-06-291-8/+11
| | | | | | | | | | | GLContextImpl: Cleanup - release(): simplify the conditions - destroy(): allow locked twice before (destroy case) GLDrawableImpl: - fail safe: swapBuffers - final methods: getFactoryImpl, toHexString, getGLProfile, getRequestedGLCapabilities, getFactory, isRealized, lockSurface, unlockSurface
* GLContext* TRACE_SWITCH: Add GLContext instance hash value to allow tracking.Sven Gothel2012-06-281-11/+13
|
* Misc cleanup: Add @Override, remove trailing whitespaceSven Gothel2012-06-271-119/+146
|
* NativeSurface Cleanup (API Change) - Adapt to GlueGen Lock cleanup commit: ↵Sven Gothel2012-06-271-3/+3
| | | | | | | | | | | | 834b9e530e652b7ff7c5e222720bce3ad2b11c5f - adapt to GlueGen Lock cleanup - remove isSurfaceLocked(), use 'null != getSurfaceLockOwner()' instead Misc: - remove unused priv./impl. methods - add @Override
* GLContext.destroy() error case: Lock hold > 1 - More explicit error message.Sven Gothel2012-05-141-2/+2
|
* GLContext*: Remove '[set/is]Synchronized(..)' - Defaults to wait for locks: ↵Sven Gothel2012-05-131-70/+49
| | | | | | | | | | | | | | | | | | | | | | | | 1. Drawable, 2. GLContext Remove deadlock situation where thread-1 (Animator Thread) holds the GLContext-Lock and acquires the Surface-Lock, while thread-2 (UI/Main/EDT) holds the Surface-Lock and attempts to create the GLContext and hence acquires the GLContext-Lock. A GLContext-Lock and hence makeing the GLContext current requires to hold the Surface-Lock. The prev. code acquired the locks in reverse order and allowed the deadlock as described above. This fix acquires the locks in the proper natural order 1 - Surface-Lock 2 - GLContext-Lock This fix also renders the use of the non-synchronized behavior invalid, since it is bogus not to wait for the GLContext lock where it waits for the Surface lock. It also seems nonsense not to wait for any of both locks and our code always waited for both (synchronized := true). The GLContext [set/is]Synchronized(..) methods are removed and waiting for the lock per default is the correct behavior.
* GLContext: Don't enable TRACE_SWITCH if DEBUG is enabled.Sven Gothel2012-04-031-2/+2
|
* Enhance and generalize AWT Threading* implementation; Minor changes ..Sven Gothel2012-03-251-19/+9
| | | | | | | | | | | | | | | | | | | | | | | | | Threading*: - add invoke(..) generalizing the Therading decision GLCanvas: - remove 'manual' Threading decision, simply call Threading.invoke(..) - use anonymous Runnable instances - remove drawable lock, drawable is volatile instead GLJPanel: - remove 'manual' Threading decision, simply call Threading.invoke(..) - use anonymous Runnable instances - DEBUG: Use getThreadName() prefix GLContextImpl: - Remove GLWorkerThread idle command on makeCurrent(), no holding of context in worker thread while idle. - DEBUG: Use getThreadName() prefix X11GLXContext: - DEBUG: Use getThreadName() prefix TODO: Validate whether it's OK for GLCanvas and GLJPanel to set Threading.Mode.MT as the default mode!
* Fix bug 564 - part 2 (X11 Mesa 8.0.1 GL 3.0 w/ failing GLX_ARB_create_context)Sven Gothel2012-03-201-26/+34
| | | | | | | | | | | | | | | | | | Tolerate a buggy ARB createContext impl (always failing to create a context). Now we only reset the GLContext state and set the context/device availability flag if ARB create context actually was successful. Otherwise the fallback 'old' context is being used w/o having it's context state cleared and context/device availability is set later on. This makes the general use of ARB_create_context more stable. +++ Minor changes: - Non compatibility fallback: Try forward context instead. - Limit ARB context 'seeking' to >= OpenGL 2.0
* GLContextImpl: Detect "llvmpipe" as a software renderer (Mesa 8.0.1)Sven Gothel2012-03-191-1/+2
|
* Fix bug 564 (X11 Mesa 8.0.1 GL 3.0 w/o GLX_ARB_create_context)Sven Gothel2012-03-181-11/+14
| | | | | | | | | | | | X11/Mesa 8.0.1 offers a GL 3.0 context w/o having the GLX_ARB_create_context extension available (even though the func-ptr glXCreateContextAttribsARB is not null). We assumed that if no GLContext device availability is set, it can be only GL 2.0 or ES1/ES2. Fix: Relaxed these (false) constrains and map the created context reflecting using it's actual attributes.
* GL/GLContext: Properly define swapInterval incl. default value for EGL (1) ↵Sven Gothel2012-03-101-0/+2
| | | | | | and desktop (undefined). *GLContext.setSwapIntervalImpl: Simple return success, set state in GLContext.
* GLContextImpl: Expose GL_RENDERER for impl. / EGLContext workaround Android ↵Sven Gothel2012-03-071-22/+36
| | | | | | | | | | 4.0.3, Pandaboard ES, PowerVR SGX 540 Bug GLContextImpl: Expose GL_RENDERER for impl. - This allows reusing the 'early' string. EGLContext workaround Android 4.0.3, Pandaboard ES, PowerVR SGX 540 Bug - On mentioned platform, the app crashes when eglSwapInterval() has been called.
* Software Rasterizer: Add 'Mesa X11' to GL_RENDERER list. EGLDrawable: Remove ↵Sven Gothel2012-02-291-2/+4
| | | | | | | | | getWidth()/getHeight() EGL surface queries. EGLDrawable: Remove getWidth()/getHeight() EGL surface queries. This code is not only redunant, since the surface dimension is known in the NativeWindow surface returned by GLDrawableImpl., but also causes an exception since the EGL surface might not be realized at the time it's dimension is being queried.
* GLContext.isCurrentContextHardwareRasterizer(): Recognize 'softpipe' ↵Sven Gothel2012-02-261-3/+18
| | | | (Gallium) as software rasterizer. Check for NULL and issue warnings (TRIAGE THIS).
* Fix GLProfile/GLDrawableFactory bug: Recursion on default desktop device, ↵Sven Gothel2012-02-251-1/+1
| | | | | | | | | | | | | | since no profile was mapped. GLDrawableFactory*: - Initialize defaultDevice, even if impl. is not available (no GL libraries for impl.), hence - getDefaultDevice() always returns a valid device - getIsDeviceCompatible() only returns 'true' if device is supported _and_ drawable factory is functional GLProfile: - default-desktop-device becomes default-device even if the desktop-factory itself is not functional. This is due to the fact that the subsequent EGL-factory always handles desktop-devices (X11->EGL, GDI->EGL, etc).
* DEBUG Output: More thread-names to drawable/context lifecycle; Remove ↵Sven Gothel2012-02-221-17/+17
| | | | massive '!!!' occurence
* API Change [GLProfile/GLContext]: Add notion of hardware acceleration in ↵Sven Gothel2012-02-201-12/+61
| | | | | | | | | | | | | | | | | | | | | | | | | | GLProfile.get<Profile>() methods. We need to distinguish between software and hardware accelerated OpenGL profiles to allow choosing the proper profiles [default, GL2ES1, GL2ES2, ..] on platforms where both, software and hardware implementations exist (GL, GLES2, ..). Where no preference is being requested, hardware acceleration is favored: GLProfile.getDefault() GLProfile.getGL2ES1() GLProfile.getGL2ES2() Some method signatures needed to change GLProfile: getMaxProgrammable(AbstractGraphicsDevice device) -> getMaxProgrammable(AbstractGraphicsDevice device, boolean favorHardwareRasterizer) GLProfile adds: isHardwareRasterizer() Determination whether a hardware acceleration is being used or not is extended in GLContextImpl by querying the current context's GL_RENDERER string. If the latter contains 'software' (case insensitive) it is not hardware accelerated. At least this works w/ newer Mesa3D impl, where GLX_SLOW_CONFIG is not set!
* Fix ExtensionAvailabilityCache ; Enhance caching.Sven Gothel2012-02-141-15/+24
| | | | | | | | | | | | | | | | | | | ExtensionAvailabilityCache regression / enhancement: - Set context version (w/o string) before caching. This is required since we query the ctx version. Regression from 4011e70eed8c88aee0fcd051a50ab3f15bb94f68 - Remove GLContextImpl state. Only use the passed value at initialization. - Defined initialization, due to the 'new' cache/instantiation logic Remove redundant GLContext profile bits: - CTX_OPTION_ANY: implicit if !CTX_OPTION_FORWARD - CTX_IMPL_ACCEL_HARD: implicit if !CTX_IMPL_ACCEL_SOFT Cache key (ProcAddressTable, Extensions): - Mask out GLContext.CTX_OPTION_DEBUG | GLContext.CTX_IMPL_ES2_COMPAT, since they don't influence the cached values.
* Enhance ExtensionAvailabilityCache ; Expose extension count in GLContext ↵Sven Gothel2012-02-131-25/+12
| | | | | | | | | | | | | | (and clean up) - GLContext - Expose isFunctionAvailable(), isExtensionAvailable(), getPlatformExtensionCount(), getGLExtensionCount() - sort methods a bit ExtensionAvailabilityCache: - Favor StringBuilder instead of StringBuffer (faster) - Resuse set's - Hold dedicated counts of extensions, platform and GL
* Fix JOGL GLContextImpl ProcAddressTable and Extension CachingSven Gothel2012-02-131-15/+11
| | | | | | | | | Bug caused using wrong extension cache and probably the wrong procAddress table. - do not reuse local field value if not cached - issue resetState() after each createContextARBMapVersionsAvailable(..) query - remove cache entry for extension (copy/paste bug) - resetState() shall clean platform extProcAddressTable in specializations
* OpenGL ES/EGL OverhaulSven Gothel2012-02-131-25/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - GLProfile properly detects native EGL/ES1/ES2 on the 'desktop' device factory. This allows usage of Mesa's EGL/ES or Imageon's PVR emulation, etc. - GLProfile drops getDefaultDesktopDevice() and getDefaultEGLDevice() since both are aligned by getDefaultDevice(). - Fix GL_ARB_ES2_compatibility detection and utilize resulting isGLES2Compatible() where possible. This allows ES2 compatible desktop profiles to use core ES2 functionality (glShaderBinary() .. etc) even with a GL2ES2 desktop implementation. - EGLDrawable: If createSurface(..) fails (BAD_NATIVE_WINDOW) w/ surfaceHandle it uses windowHandle if available and differs. This allows the ANGLE impl. to work. - Properly order of EGL/ES library lookup: ES2: libGLESv2.so.2, libGLESv2.so, GLES20, GLESv2_CM EGL: libEGL.so.1, libEGL.so, EGL - *DynamicLookupHelper reference will be null if it's library is not complete (all tool libs, all glue libs and a ProcAddressFunc lookup function - if named). - Enhance GL version string (incl. ES2 compatible, hw/sw, ..) - GLBase: Fix docs and remove redundancies - Prepared (disabled) DesktopES2DynamicLibraryBundleInfo to be used for a real EGL/ES2 implementation within the desktop GL lib (AMD). Sadly it currenly crashed within eglGetDisplay(EGL_DEFAULT_DISPLAY), hence it's disabled.
* OSX Fix: Catch releaseImpl's glFinish() exception (DebugGL); Make ↵Sven Gothel2012-01-131-6/+10
| | | | | | | | | | | | | | GLContext.release's setCurrent(null) exception prone. Catch releaseImpl's glFinish() exception (DebugGL) glGetError() after glFinish() (eg. w/ debug pipeline) produced unknown error 0x0506 on OS X (10.7.2 NV). Make GLContext.release's setCurrent(null) exception prone Call setCurrent(null) in finalizer block to ensure it's 'released' out of the TLS even when an exception is being thrown. Make MacOSX Shared Resources 'destroy' more error prone (catch exceptions)
* GLContext/NativeSurface Impl's toString(): Add lock.toString()Sven Gothel2012-01-081-5/+0
|
* GLContext*/GLDrawableHelper: Fix consistency of recursive ↵Sven Gothel2012-01-081-128/+160
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | makeCurrent()/release()/destroy() calls ; Enable context switch tracing ; GLCanvas: proper AbstractGraphicsDevice destruction GLContext*/GLDrawableHelper: Fix consistency of recursive makeCurrent()/release()/destroy() calls Utilizing volatile and lock.tryLock(0) for lockConsiderFailFast(), reducing redundant synchronization and using RecursiveLock implicit sync. GLContext 'early-out' is the case where the thread already holds the context, ie. context is already current and the native makeCurrent is skipped. makeCurrent()'s 'early-out' w/o incr. the recursive lock of GLContext and it's NativeSurface could lead to asymetry in lock/unlock count with release()/destroy() calls. The 1st release actually released the native ctx already. Properly utilize recursive lock/unlock in all cases and impl. 'early-out' after locking. Following the above in GLDrawableHelper.invokeGL()'s 'early-out' case as well, ie calling makeCurrent()/release() symmetrical. Introduce GLDrawableHelper.disposeGL(), which issues dispose on all GLEventListeners within a current context and issued context destruction directly. This simplifies GLAutodrawable's destroy/dispose calls and ensures that the above sequence of events happens atomically (lock is being hold until destruction). Enable context switch tracing If property 'jogl.debug.GLContext.TraceSwitch' is defined, trace context switch. GLCanvas: proper AbstractGraphicsDevice destruction
* setGLFunctionAvailability(..): Reduce the calls to resetProcAddressTable() ↵Sven Gothel2011-12-131-34/+60
| | | | | | | | | | | 12 -> 7 in initialization. Reducing the calls to resetProcAddressTable() 12 -> 7 in initialization, saves: Linux/AMD: 600ms -> 300ms Linux/NV: 161ms -> 112ms OSX 10.7/NV: 522ms -> 397ms Still some freezes on OSX 10.6.8/NV .. further analysis is going on.
* Fix concurrency bug of GLProfile initialization ; Fix SharedResourceRunner ↵Sven Gothel2011-12-111-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 'dead' thread (Applets) GLDrawableFactory: - clarify: public getWasSharedContextCreated(..) -> protected createSharedResource(..) - add: getSharesResourceThread() GLProfile: - proper locking of initSingletion(..) path: - Use RecursiveThreadGroupLock and add/remove GLDrawableFactory's sharesResourceThread while creating it's the sharedResource. This simplifies and fixes GLProfile's locking code. - Fix and simplify initSingleton(boolean) API doc - mark it deprecated. - Add initSingleton() for controlled initialization only, pairing w/ shutdown(..) Remove initSingleton(boolean) calls in code and test! +++ Fix SharedResourceRunner 'dead' thread (Applets) In Applets, stopping an Applet makes the browser Java plugin interrupting and killing all related threads, including our SharedResourceRunner thread. - Validate whether the shared resource thread is alive - Catch interruption in shared resource thread and assume it's a kill signal - releaseSharedResource: clear devicesTried set
* GLContext fix shutdown(): Clear proc address tables and extension cache ↵Sven Gothel2011-12-011-2/+6
| | | | (GLContextImpl)
* GLContextImpl*: createImpl() / makeCurrentImpl() refinement / robostnessSven Gothel2011-11-291-11/+27
| | | | | | | | | | | | | | | createImpl(): If successful must leave context current. makeCurrentImpl(): Is only called if context is not just created, hence the boolean parameter 'boolean newCreatedContext' is removed. This clearifies and actually cleans up the native makeContextCurrent/releaseContext call pairs. MacOSXCGLContext: CGL and NS impl. of native makeContextCurrent/releaseContext uses CGL locking to provide a thread safety. This is recommended in OS X OpenGL documentation on [shared context] multithreaded use cases. Post creation code, as seen in some pbuffer cases is moved to overriden createImpl() methods.
* Fix Bug 527: Creating a context w/ shared context, while the latter is in ↵Sven Gothel2011-11-281-2/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | use (threading) Bug 527 describes the situation where wglShareLists() fails (throws exception) while the shared context is in use by another thread (some Animator). It was reported by Jerome Jouvie. The exception happens not everytime, but at least around 20% on manual tests I have performed on the Windows platform. The context in question are all JOGL's GL2, which natively where bound to an OpenGL 3.0 profile. The WGL_ARB_create_context spec says http://www.opengl.org/registry/specs/ARB/wgl_create_context.txt: +++ Future versions of OpenGL may only support being added to a share group at context creation time. Specifying such a version of a context as either the <hglrc1> or <hglrc2> arguments to wglShareLists will fail. wglShareLists will return FALSE, and GetLastError will return ERROR_INCOMPATIBLE_DEVICE_CONTEXTS_ARB. +++ Hence the 1st patch was to remove 'wglShareLists()' usage in case we use the new WGL_ARB_create_context context creation method. Even though this is a desired change, and works in general, it didn't fix the issue. It seems that the shared context, which is passed @ new context creation, cannot be used while it is in use itself in another thread. This conclusion leads to the actual fix, ie. locking the shared context while creating the new context which shares it. Manual tests using this patch could not reproduce this issue (40 attempts). Test: TestSharedContextListNEWT2
* NEWT/GLContextImpl: Refine API doc, NEWT: Display name -> connection nameSven Gothel2011-11-261-1/+4
|
* NativeSurface's getGraphicsConfiguration() returns the native (delegated) ↵Sven Gothel2011-11-231-4/+4
| | | | | | | | | | | AbstractGraphicsConfiguration, if delegation is used. This change restricts the usage of AbstractGraphicsConfiguration's getNativeGraphicsConfiguration() to NativeSurface implementations and hence reduces complexity. NativeSurface implementations are adapted and access to it's AbstractGraphicsConfiguration is controlled via get/set method avoiding flawed usage (read/write), since read access shall return the delegated AbstractGraphicsConfiguration, if used.
* GLProfile: Add GLProfile getImpl(); Use getImpl() for getGL2ES[12](..), ↵Sven Gothel2011-10-261-1/+1
| | | | dropping GL2ES[12] lists
* MacOSX: Always release ctx on main thread (offscreen was frozen as well); ↵Sven Gothel2011-10-131-1/+1
| | | | | | Cleanup Warnings and OSX ContextUpdater -
* MacOSX: Fix shared ctx release [onMainThread]; Make GLContextShareSet ↵Sven Gothel2011-10-131-19/+9
| | | | | | | | | | | | | | | | | lifecycle deterministic; Remove warnings Fix shared ctx release [onMainThread] - Releasing the shared contexts caused a freeze of about 10s from one of the shared release operations. [NSOpenGLContext release] - Thorough triage concluded the workaround to release the shared ctx on the main thread. - Using enhanced GLContextShareSet, see below Make GLContextShareSet lifecycle deterministic - Programmatically control the lifecycle of tracked shared ctx allows us using 'hard' references. - Features queries for isShared() and ofc unregister a share set if all are destroyed. Remove warnings - MacOSXWindowSystemInterface.m used 'long', where 'GLint' was requested.
* JOGL/OSX: Properly utilize NSOpenGLContext update() via ContextUpdater, ↵Sven Gothel2011-10-131-3/+13
| | | | | | | | | | | | | | | which only holds the 'update' state now. Avoid calling updater() for every makeCurrent(), but if view's frame has changed only. This solves the pixel flickering experienced on OSX. - GLContextImpl:update() -> drawableUpdatedNotify() w/ comments - ContextUpdater holds context, view, old view frame and the update state. It doesn't issue NSOpenGLContext update() by itself, but allows querying and clearing the update flag. - MacOSXOnscreenCGLContext impl drawableUpdatedNotify() - register via ContextUpdater, and use it if available.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-02 10:02:59 +0000