aboutsummaryrefslogtreecommitdiffstats
path: root/src/newt/native
Commit message (Collapse)AuthorAgeFilesLines
* NEWT/X11 Display: CloseDisplay in same order as creation (ATI); Adding ↵Sven Gothel2011-08-311-22/+45
| | | | | | | | | | | | | | DisplayRelease0; Using 'EDT' suffix for display arguments CloseDisplay in same order as creation (ATI) - This enhanced the erroneous bug 515 (b54497155815852744adb657816cb4057948dae2) situation with closing the display connections. However, some SIGSEGV still slipped through. Adding DisplayRelease0 - Intended for cleaning up resources. Currently a NOP. Using 'EDT' suffix for display arguments - To mark the semantics of the display connection, which may be for window or EDT now.
* Fix NEWT regression due to fix of bug 502, commit ↵Sven Gothel2011-08-301-6/+14
| | | | | | | | | | | | 9ed513e9a9616f6028084df4c650c8caf31ea49d In case of exessive destroy/create (the NEWT reparenting test cases), some dpyEDT events are slipping through the event dispatcher. This fix uses issues more XSync on both Display connection in case of 'requestFocus' and 'closeWindow'. 'requestFocus' also uses the dpyEDT to issue the XSetInputFocus(..), since it's EDT related.
* Workaround (Fix) for Bug 502: Multithreading issue w/ libX11 1.4.2 and ↵Sven Gothel2011-08-301-17/+25
| | | | | | | | | | | | | | | | | | | | | libxcb 1.7 bug 20708 See https://jogamp.org/bugzilla/show_bug.cgi?id=502 Since the libX11/xcb code doesn't seem to be fixed anytime soon a better usable workaround is required than using a system property to enable 'over locking'. It turns out that the race condition is related to the parallel X11 Display connection usage of GLX/OpenGL and event dispatching. This workaround utilizes 2 X11 Display handles, one for windowing/OpenGL and one for event dispatching. This approach allows us to cont. multithreading use w/o locking the display and works on both implementations, the old bug-free libX11 and the 'new' buggy one. Downside is the little resource overhead of the 2nd X11 Display connection .. well. - Removes the property: 'nativewindow.x11.mt-bug'
* Android Cleanup: Remove EGL header & library (compile time)Sven Gothel2011-08-111-1/+0
|
* NEWT/Android: Add missing native codeSven Gothel2011-08-111-0/+47
|
* deployment resturcturing: combine nativewindow/jogl/newt ; newt: 'driver' ↵Sven Gothel2011-08-056-144/+144
| | | | | | | | | | | | | | | | | | | | | separation ; android cleanup remaining all-in-one jnlp's / jars: jogl-all-awt.jnlp -> jogl.all.jar jogl-all-noawt.jnlp -> jogl.all-noawt.jar jogl-all-mobile.jnlp -> jogl.all-mobile.jar native for all above: jogl-all-natives-linux-amd64.jar jogl.all-android.apk jogl.all-android.jar more may follow for each supported platfrom ++++ - newt: proper 'driver' separation - all drivers reside now in jogamp.newt.driver.* - remove intptr.cfg / use gluegen's
* Renamed jogamp.newt.opengl -> jogamp.newt.egl, properly reflecting EGL ↵Sven Gothel2011-08-032-18/+18
| | | | | | dependency, not OpenGL The native impl packages utilize EGL, not opengl, to query the proper pixelformat.
* X11 Nativewindow/NEWT: X11 Error Handler (JNIEnv query for thread, stack ↵Sven Gothel2011-06-261-23/+38
| | | | | | | | | trace), cleanup - X11 Error Handler: if throwing JVM stack trace or fatal JVM error query proper JNIEnv for running thread and attach thread to JVM if necessary. - NEWT/X11: Proper XEvent polling documentation, cleanup window creation event mask
* Minor editing (compile script, X11Window.c - cleanups)Sven Gothel2011-06-121-3/+3
|
* Utilize GlueGen's platform independent header for stdin.h, gluegen_stdint.h, ↵Sven Gothel2011-06-115-50/+8
| | | | | | | .. (remove local copy) - Use them for gluegen code generation - Use them for native compilation (cc)
* NEWT/X11 WindowClosing: End dispatch loop since Display could be destroyed ↵Sven Gothel2011-04-261-3/+15
| | | | after Java callback
* Misc cleanup, strings etcSven Gothel2011-04-241-2/+2
|
* Newt X11: keyPress/Release must use upper case XP_ values to match VK_ ; Set ↵Sven Gothel2011-03-301-50/+65
| | | | modifier properly.
* Move implementation private files from com.jogamp.<module>.impl. to ↵Sven Gothel2011-02-096-144/+144
| | | | | | | | | | | | | | | jogamp.<module> (2/2) - edit files - com.jogamp.opengl.impl -> jogamp.opengl - com.jogamp.opengl.util.glsl.fixedfunc.impl -> jogamp.opengl.util.glsl.fixedfunc - com.jogamp.nativewindow.impl -> jogamp.nativewindow - com.jogamp.newt.impl -> jogamp.newt This sorts implementation details from the top level, ie skipping the public 'com', allowing a better seperation of public classes and implementation details and also reduces strings. This approach of public/private seperation is also used in the OpenJDK.
* NEWT: Add WindowListener.windowDestroyed() ; Remove WindowImpl.windowDestroyed()Sven Gothel2010-12-235-27/+9
| | | | | | | | | | | | | Add WindowListener.windowDestroyed() To expose a proper window lifecycle, ie destroy-notify and destroyed, this notification is added. This will be used at least in unit tests, where we verify destruction. Remove WindowImpl.windowDestroyed(): This native hook (planned to be called by native destroy notification) is unreliable or not supported for all platforms. NEWT relies on the pre destroy native hooks and handles the final destroy notification itself.
* Windows RegisterClass: Use new RegisteredClassFactory (window class), Misc.Sven Gothel2010-12-121-83/+22
| | | | | | | | | | | | | | | | | This solves the issue when an applet is started/stop and started again, or another applet runs in the same JVM. Also soves the issue for multiple JVMs. RegisteredClassFactory can be instanced to manage one shared window class, currently in use for GDI's dummy window and NEWT. A class base name and a window proc handle must be passed in the factory cstr. Before registering, the class is tested if already exists, eg another applet in the same JVM. If registration fails, the class name will iterate until successful or MAX_INT reached, eg if multiple JVMs are running. Added NativeWindow Common Native Code.
* NEWT/Windows: Use GDI GetDC/ReleaseDCSven Gothel2010-12-091-22/+0
|
* X11 DummyWindow: Pass size and set size in proxySven Gothel2010-12-041-1/+0
|
* NEWT/X11: Don't erase background with pixmap or pixelSven Gothel2010-11-281-2/+2
|
* Moving NEWT X11Window GetRelativeLocation() native implementation to ↵Sven Gothel2010-11-171-58/+0
| | | | NativeWindow X11.
* Move GDI GlueGen wrapping from JOGL -> NativeWindow (following X11). Moving ↵Sven Gothel2010-11-171-44/+0
| | | | NEWT WindowsWindow GetRelativeLocation() native implementation to GDI as well.
* Complete removal of NV KD and OMX started with ↵Sven Gothel2010-11-061-1/+6
| | | | | | | | 46f17013c7cd59d551371edb2c1a4a57f8cbd84f (code dependencies) This currently removes KD fullscreen for NV devices and the whole OMX NV stream/file type detection, which renders the OMX hack useless. However, updated EGL sync (NV proprietary) to EGL_KHR_reusable_sync and EGL_KHR_fence_sync.
* NEWT Window Lifecycle / ScreenMode:Sven Gothel2010-11-031-27/+57
| | | | | | | | | | Lifecycle.reparentActionPre()/reparentActionPost() -> pauseRenderingAction()/resumeRenderingAction() for a more generic use, ie reparenting and screen mode change. ScreenMode change: No more visibility/fullscreen changes, no more locking, just pause/resume animation. X11 ScreenMode set: move from thread/wait to simple polling over time (timeout)
* Minor cleanup: strings, unused vars, ..Sven Gothel2010-11-021-5/+4
| | | | | | | | | | DEBUG strings w/ thread name nativewindow.TraceLock -> nativewindow.debug.ToolkitLock.TraceLock Sync Xmisc (DummyWindow) with NEWT's creation test scripts: awt and non-awt usage
* Fix macosx NEWT warningsSven Gothel2010-11-021-1/+2
|
* Fix macosx compilationSven Gothel2010-11-021-10/+10
|
* Windows: define DISPLAY_DEVICE_ACTIVE, if not alreadySven Gothel2010-11-021-4/+6
|
* ScreenMode: Ignore modes < 15bppSven Gothel2010-11-021-1/+0
|
* Newt/Windows: Turn off native VERBOSESven Gothel2010-11-011-1/+1
|
* Fix: Newt Rotation (Windows) - added description (CCW)Sven Gothel2010-11-011-100/+80
|
* Fix ScreenMode ; Add FatalError to NewtCommon.c ; Fix Windows BuildSven Gothel2010-10-294-81/+85
| | | | | | | | Fix ScreenMode - Avoid NPE/Out-of-memory: Return zero sized NewIntArrays instead of NULL. Fix Windows Build - ScreenMode still has a regression
* remove positive RANDR messageSven Gothel2010-10-291-2/+0
|
* NEWT: ScreenMode changesSven Gothel2010-10-265-417/+520
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - New type definition: ScreenMode { MonitorMode { SurfaceSize { Resolution, bpp }, ScreenSizeMM, refreshRate }, rotation }, where Resolution and ScreenSizeMM are of type DimensionReadOnly - ScreenMute instance is - immutable - hashable - cloneable The above allows fast query and storage w/o redundancies. More than 300 modes via permutation could be expected. ScreenMode impl. changes: ScreenImpl: To be implemented methods by native specialization: - protected int[] getScreenModeFirstImpl() - protected int[] getScreenModeNextImpl() - protected ScreenMode getCurrentScreenModeImpl() - protected boolean setCurrentScreenModeImpl(ScreenMode screenMode) The data unification etc is implemented generic using ScreenModeUtil and the 'int[]' streaming. ScreenModeStatus holds all ScreenMode related data and provides a locking strategy. ScreenModeListener provides a callback facility for ScreenMode change events. - Screens listen to ScreenModeStatus, so all FQN referenced Screen's receive the change. - Windows listen to Screen, to take appropriate action for the event (fullscreen, reshape). Misc: - Screen/Display: promoting 'addReference'/'removeReference' to public interface, so a user may trigger construction/destruction (-> junit tests, plus other clients than WindowImpl). - Gears: 'setSwapInterval' at 'reshape' instead of 'init', so it's reset when ScreenMode is changing. -
* Merge branch 'master' of http://github.com/sgothel/joglRami Santina2010-10-214-396/+559
|\ | | | | | | | | | | | | | | | | | | | | 18bf27fa86da1f26fd085565f501736816d2f2e9 Conflicts resolved: src/newt/classes/com/jogamp/newt/impl/WindowImpl.java src/newt/classes/com/jogamp/newt/impl/windows/WindowsWindow.java src/newt/classes/com/jogamp/newt/impl/x11/X11Window.java src/newt/native/WindowsWindow.c src/newt/native/X11Window.c
| * NEWT: Fix / Stabilize Fullscreen/Decoration/Reparenting Mode ChangesSven Gothel2010-10-214-381/+539
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - setSizeImpl/setPositionImpl/reparent -> reconfigureWindowImpl - setVisible(boolean) is state checked (500ms) for better reliability on resource creation. Guarantees valid surface. - reparentWindow: start pos of child -> top is current position on screen - reparentWindow: Recheck success (setVisible), if failed fall back to recreate, which gets rid of a lost child windows (1/20) .. - reparentWindow: if size failed, reconfigure for size again - add toggle decoration - unify nfs_ size/pos state - WindowsWindow.c/X11Window.c: Unify size/pos settings - X11Window.c: - NewtWindows_setFullscreen: use 'root of screen' instead of 'default root of display' - Adding SubstructureNotifyMask incl event semantics - Parse ReparentNotify (debugging of reparenting) Misc: - Add native getLocationOnScreen() impl to avoid possible AWT deadlock - setSize/setPosition/setFullScreen -> EDT - More documentation on expected native implementation semantics
| * Fix: X11 Fullscreen mode behaviourRami Santina2010-10-111-26/+23
| | | | | | | | | | | | | | | | | | Event of fullscreen is sent to the display root and properties of fullscreen are set on our window/screen defined source attachment of data sent using XSendEvent to be normal application (required) cleaned up the code. replaced tabs with spaces
| * Added Fullscreen WM Hit to the X11 windowRami Santina2010-10-111-4/+56
| | | | | | | | | | when true the request is send after configuring the window otherwize before decoration. needs tweeking to ensure that the fullscreen is on-off
* | Added Windows Impl for ScreenMode and rotation change; Windows Fullscreen HintRami Santina2010-10-201-3/+260
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added windows implemenation for screen mode change lifecycle and screen rotation screen modes. +++++++++++ Added getHighestAvailableRate() to Screen mode as a helper function which loops over the available rates of this screen mode and returns the highest available. Added toString impl to ScreenMode. +++++++++++ Unit Tests: TestScreenMode01NEWT,TestScreenMode02NEWT are modified adding the gears as GL event listeners. +++++++++++ Notes: 1- Screen modes with bits per pixel values not same as current are currently filtered out. 2- Since windows stores the Display settings differently you will see some duplicate values which the documentation says is usual. These modes are not filtered out in getScreenModes(). Keeping values same as provided from windows enum. 3- BitsPerPixel is currently natively read on windows only.
* | Added Screen Rotation manipulation API (with X11 impl)Rami Santina2010-10-181-4/+102
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added screen rotation change capability with impl for X11 (using Xrandr) com.jogamp.newt.Screen: added 2 methods which cover screen roation lifecycle setScreenRotation(int rot): Change the Screen Rotation to one of the rotations defined in ScreenMode, namely: ROTATE_0, ROTATE_90, ROTATE_180, ROTATE_270 int getCurrentScreenRotation(): Get the Current screen rotation returns -1 if not implemented natively. +++++++++++++++++++++++++ Notes: 1- At init the original rotation is called natively and cached, when screen is destroyed the rotation is reverted. 2- On X11 with Nvidia: you need to edit /etc/X11/xorg.conf: Add the following line: Option "RandRRotation" "on" in Section "Device" after BoardName. +++++++++++++++ Added TestScreenMode02NEWT which includes 4 tests 1- Rotate 90 2- Rotate 180 3- Rotate 270 4- Rotate with screen mode change should fail if screen rotation not implemented natively. (4) withh fail if screen mode not impl natively as well.
* | Screen Mode manipulation API (with X11 impl) ; X11 Fullscreen implRami Santina2010-10-171-5/+210
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added screen mode change capability with impl for X11 (using Xrandr) com.jogamp.newt.Screen: added 4 methods which cover screen mode lifecycle ScreenMode[] getScreenModes(): Get list of available screen modes returns null if not natively implemented. On first call will fetch the list thru a set of native calls and stores them in the ScreenModeState. Consequent calls will retreive the data from the ScreenModeStatus object. Native calls are done on creation stage, so subsequent API calls will return cached data. int getDesktopScreenModeIndex(): Get the Current Desktop Screen mode index returns -1 if functionality not natively implemented. The index returned refers to the screen mode location in the ScreenMode array. Used to get initial mode for reverting back. Initial call will retreive the mode thru a native call (done on initialization), subsequent API calls will retreive it from the ScreenModeStatus object. Note: If Window is closed without reverting back and screen is set to unreachable. The api will revert the screen. A call to destoy(unreachable) is needed for revert to take place on close windowDestroyNotify(WindowEvent e) { ... window.destroy(true); } short getCurrentScreenRate(): Get the current screen rate returns -1 if not natively implemented. Used along with getDesktopScreenModeIndex You can identify the current screen mode (w x h) and the rate. Initial call will retreive the mode thru a native call (done on initialization), subsequent API calls will retreive it from the ScreenModeStatus object. void setScreenMode(int modeIndex, short rate): change the screen mode to the mode index which refers to one of the screen modes retreived by getScreenModes() and a rate which can be one of screenModes.getRates(). If the rate provided doesnt belong to the list of available rates the first rate is chosen. build-newt.xml: Added header files generation for Screen and Display Added Xrandr syslink to x11 based builds. +++++++++++++ Fullscreen for X11Window: Added Fullscreen WM Hint for X11Window. Event of Fullscreen is sent to the display root. When changing to fullscreen the request is send after configuring the window otherwize before decoration. Not Stable yet, needs tweeking on other Linux dist. Window looses focus after switching to fullscreen mode. +++++++++++++++ Added TestScreenMode01NEWT which includes 4 tests 1- Fullscreen change mode 2- Screen mode change 3- Screen change with fullscreen (revert screenmode then revert fullscreen) 4- Screen change with fullscreen (revert fullscreen then revert screenmode) should fail if not screenmode not implemented on platform. +++++++++++++++ Cleaned up some unused imports
* NativeWindow/NativeSurface Refactoring ; Added mouseClick NEWT/AWT unit testSven Gothel2010-10-062-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | | NativeWindow/NativeSurface Refactoring - Using NativeSurface interface - NativeWindow extends NativeSurface, adds getLocationOnScreen(Point) - NativeWindow add: getParent() - NativeWindow/Surface: Removed 'invalidate()', use 'destroy()' if you must. - NullWindow -> ProxySurface impl NativeSurface - JOGL: Uses NativeSurface only. - GLDrawable.getNativeWindow() -> GLDrawable.getNativeSurface() Added mouseClick NEWT/AWT unit test JOGL: - GLAnimatorControl add: resetCounter() - NEWT: - GLWindow counters: return GLWindow counters always - WindowImpl - requestFocus() wait until done - reparent: readded requestFocusImpl(true), native impl skips java focusAction if reparented - X11Window: Add XRaiseWindow() in requestFocus()
* NEWT: Fix Display/Window/Screen OO Identity, Reparenting and requestFocusSven Gothel2010-09-232-44/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | NativeWindow: Interface NativeWindow changes: - Remove 'throws' qualifier in lockSurface(), since it is not - Adding convenient 'one call' isSurfaceLockedByOtherThread() - Adding getSurfaceLockOwner() NEWT Window/GLWindow: - Unclutter Window/GLWindow relationship - save Window's indentity GLWindow's role is a GLAutoDrawable implementation aggregating (maybe even compositioning) a Window. The previous implementation just derived from the Window implementation, overwriting methods and fields - impossible to ensure sanity / completness. It was also not ensured that the added functionality of GLWindow (setVisible, destroy, ..) has been issued in case of handling the aggregated Window alone (window callbacks, ..). To solve this issue in a 1st attempt without changing the GLWindow API, Window is just an interface, being implemented by their specializations, hence sanity is intrinsic. GLWindow's added functionality is ensured by a Window.LifecycleHook interfaced implementation, registered at the aggregated Window. - Screen and Window are interfaces now (new files) - Display is an abstract class. - Their (abstract) implementations resides in impl/<BaseName>Impl - GLWindow implements Window as well - Remove Screen reference handled by setScreen(Screen) method. - Lock native parentWindow if used (createNative/reparenting) - Move lockSurface/unlockSurface from unchecked override pattern to an callback style using abstract methods lockSurfaceImpl/... - Sorting all methods to semantic sections, abstract, superinterface, .. - Reparenting: Handling different reparenting situations: - Unchanged - No change - Native Reparenting - Compatible Display/Screen, try native reparenting - Native (Re)Creation - Use destroy/create pattern - Native Creation Pending - Create later - setUndecorated() calls reconfigure Window now, ie tries to change the window actually - Don't issue 'requestFocus()' directly from the native implementation anymore, call it from the Java code. - Window/GLWindow/NewtFactory: Constructor simplification Avoid explosion of constructor overloading, ie removing the 'undecorated' variant, since this is redundant due to the 'setUndecorated(boolean)' method. - Fixed/added API documentation
* NEWT: Focus Fix + CleanupSven Gothel2010-09-037-28/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Issueing 'requestFocus' via the native EDT dispatch loop may cause a deadlock, due to a possible implicite AWT requestFocus call (NewtCanvasAWT). Approach: RequestFocus issued directly, by Window.requestFocus() and the native EDT dispatch loop, is queued for later execution by EDT. This shall decouple a possible native windowing TK resource collision. - X11Windows.c: Add missing 'reparented' param for requestFocus to force requestFocus after reparenting. - AWTWindow.java: Add requestFocusImpl() +++ NEWT: Cleanup - Remove Event Type Bits in: - EventListener.h - NEWTEventListener.java - Remove InputEvent 'consume' status -
* Fix OSX 2: Newt EDT (MainThread) and I/O methods ; Build jogl.build.numberSven Gothel2010-07-262-34/+35
|
* Newt: X11 ICC window properties ; Still bugs with composite manager and Newt ↵Sven Gothel2010-07-171-26/+27
| | | | | | | | | reparenting Set _NET_WM_WINDOW_TYPE to specify decoration etc. On KDE4 having compositing enabled the reparent window is sometimes made invisible (UnmapWindow happened). FIXME.
* Fix file modesSven Gothel2010-07-075-0/+0
|
* GLAutoDrawable: setAnimator/getAnimator/invoke/display changes; NEWT: Adding ↵Sven Gothel2010-07-072-60/+98
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | native repaint; Fix reparent/fullscreen New: NEWT Native Repaint ========================= Support for native repaint, which shall call display() in case no animator is running. GLAutoDrawable invoke(GLRunnable) impl. handles case if invoked on animator thread, or no animator thread is running (issueing a display() call). The impl resides in GLDrawableHelper. The Animator un-/registers itself at the GLAutoDrawable via setAnimator. New: NEWT AWT/NEWT Parenting Focus Handling ============================================ Introducing Window.FocusRunnable, to be registered at the NEWT Window, which will be executed before the native focus claim. Window.FocusRunnable's run method returns a boolean, which determines whether the native implementation shall proceed claiming the native focus. This API focus hook is necessary to allow an optional underlying windowing toolkit, ie AWT (see usage NewtCanvasAWT), to make the focus traversal transparent. Fix: GLEventListener / GLDrawableHelper ======================================== GLEventListener's init() and glViewport()/reshape() method must be called before the 1st display() and after a dispose() call. It could miss the 1st display() call if added after the setVisible(true) call - due to the native repainting. The impl resides in GLDrawableHelper. Fix: Misc NEWT ============== Window reparent issues a resize() and display() call, if it is visible. native Window uses direct send.*Event for input events (again), instead of enqueueing it for performance. Window impl all status change native event Java callbacks, instead of having duplicated code in all implementations. Fullscreen, reposition at zero. Reparent/Fullscreen repaint if visible. Native reparent/fullscreen, fix glitches on Windows (visibility while reparenting)
* GLAutoDrawable: setAnimator/getAnimator/invoke/display changes; NEWT: Adding ↵Sven Gothel2010-06-262-76/+118
| | | | | | | | | | | | | | | | | | | | | | | | | | native repaint; NewtCanvasAWT focus fix Support for native repaint, which shall call display() in case no animator is running. GLAutoDrawable invoke(GLRunnable) impl. handles case if invoked on animator thread, or no animator thread is running (issueing a display() call). The impl resides in GLDrawableHelper. GLEventListener's init() and glViewport()/reshape() method must be called before the 1st display() and after a dispose() call. It could miss the 1st display() call if added after the setVisible(true) call - due to the native repainting. The impl resides in GLDrawableHelper. The Animator un-/registers itself at the GLAutoDrawable via setAnimator. NEWT Window reparent always issues a resize() and display() call. NEWT native Window uses direct send.*Event for input events (again), instead of enqueueing it for performance. NEWT Window implements all status change and Java native event callbacks, instead of having duplicated code in all implementations. NewtCanvasAWT if the Newt window is focused, the AWT/Swing component[s] will loose the focus.
* Fix: Locking/Threading; Common IntIntHashMap and Buffers; Fix: ↵Sven Gothel2010-06-104-228/+215
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | glMap*Buffer*; GLX/WGL/CgGL: All runtime dynamic; Misc .. TODO: Compile and test on MacOSX .. Fix: ===== Multithreading/Locking: See jogl/doc/Implementation/MultiThreading.txt - Locking layer is not platform agnostic, ie GLContextImpl, GLDrawableImpl, .. and NEWT: Window/Display - No more use of JAWT global lock necessary, removed. - No need for X11 Display lock, on the contrary, this made the NV driver hang. - Use common window/surface lock - All NativeWindow surfaceLock's are recursive now glMapBuffer: If size is 0, don't do cont with the native call. glMapBufferRange: Fix capacity. glNamedBufferDataEXT: Track the size. glMapNamedBufferEXT: Manual impl. - use the tracked size glXGetVisualFromFBConfig, glXChooseFBConfig, glXChooseVisual: Instead of ignoring and implement a renamed version (*Copied), we just use ManualImplementation for the proper copy-result code. DesktopGLDynamicLookupHelper: Initialize _hasGLBinding* attributes in the determing loadGLJNILibrary() method, which is called by super(). Otherwise static init will overwrite them after the super() call. X11GLXDrawableFactory: Don't release anything at shutdown (removed sharedContext.destroy()), since this caused a freeze/SEGV sometimes. Fixed NEWT's reparentWindow() functionality incl NewtCanvasAWT usage. - Native: if not visible, don't focus, etc - NewtCanvasAWT: Use the container size to start with - Run the command on the EDT Using GlueGen's new DynamicLibraryBundle utility: - X11, Windows and MacOSX OpenGL adapted to DynamicLibraryBundleInfo. - X11GLXDynamicLookupHelper -> X11GLXDynamicLibraryBundleInfo - Remove all path from lib names. - GL order: libGL.so.1, libGL.so, GL - shallLinkGlobal: true -> to server some 'old' DRI systems -> http://dri.sourceforge.net/doc/DRIuserguide.html - shallLookupGlobal: false - Try both : glXGetProcAddressARB and glXGetProcAddress - Using bootstrap: GLX.glXGetProcAddress(long glxGetProcAddressHandle, String glFuncName) Found the issue with LIBGL_DRIVERS_PATH, ie if not set no valid GL instance can be found (ie ATI fglrx/DRI). This may happen if using a differen user than the desktop user for whom the env var is set within some /etc/X11/Xsession.d/ script. Enhancements: ============= GLBufferSizeTracker: Use IntIntHashMap and add DirectState size tracking. GLBufferStateTracker: Use IntIntHashMap. GLStateTracker: Use IntIntHashMap. GLDynamicLookupHelper: More generic (global loading/lookup and GetProcAddress function name list), remove redundant code. FIXME: MacOSXCGLDynamicLookupHelper: - Not tested - Not using NSImage lookup anymore as recommended by OSX API Doc, so dlsym is used always (to be tested) WindowsWGLDynamicLookupHelper: - Not tested GLX/WGL/CgGL is all runtime-dynamic as now, ie loaded and looked-up at runtime, no compile time dependencies to GL anymore, nor a need to specify CgGL. Split up WGL in GDI and WGL, to allow proper dynamic runtime linkage of OpenGL32 while using static binding to GDI32 NEWT events generated by native code are enqueued and not send directly. This should ease locking mechanisms .. if any are necessary. NEWT: More platform specific code moved to *Impl method, simplifying the generic code of the superclass and impl protocol. Cleanup: ========= Replace all InternalBufferUtil's with com.jogamp.common.nio.Buffers Removed all InternalBufferUtil's from repository Removed GLContextImpl notion of 'optimized' surface locking, where the surface gets unlocked during makeCurrent/release. This just makes no sense and would impact multithreading in a horrible way.
* Allow build on OSX 10.5 (thanks to sed)Sven Gothel2010-05-311-1/+1
|