summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Complete LOCK_SURFACE_CHANGED ; Introduce NativeVisualID (Daisy chaining ↵Sven Gothel2012-03-0530-301/+577
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | *GraphicsConfiguration) ; ... NativeVisualID: New interface for Capabilities implementations, allowing retrieval of the native 'visual id'. Impl. by WGL, X11 and EGL. JAWTWindow.lockSurface() - Detect surfaceHandle change an return LOCK_SURFACE_CHANGED (see: LOCK_SURFACE_CHANGED) EGLDrawable: - Impl. updateHandle() (see: LOCK_SURFACE_CHANGED) - use NativeVisualID for EGLGraphicsConfiguration selection to respect a native 'visual id' EGLContext.createContextImpl: Use NIO for attributes EGLDisplayUtil: Enhance eglGetDisplay w/ DEBUG code and NativeSurface / EGL_DEFAULT_DISPLAY variation EGL, XGL, WGL GraphicsConfiguration: - Don't set ALPHA_SIZE and STENCIL_SIZE if not requested in attribute list for context creation. - toString() shows proper identification, eg.: egl, x11, win32 .. EGLGraphicsConfigurationFactory: Daisy chain GraphicsConfigurationFactory for native device type (currently only X11). This allows choosing the EGLGraphicsConfiguration and hence native visual id based on EGL when invoked via the factory model (generic). In case EGLGraphicsConfigurationFactory is not suitable or doesn't produce a native visual id, it falls back the the original one. X11AWTGraphicsConfigurationFactory and X11Window: Use generic NativeVisualID which allows EGLGraphicsConfiguration implicit. *GraphicsConfiguration's DEBUG flag is pushed up to DefaultGraphicsConfiguration LOCK_SURFACE_CHANGED: - commit 006e9fe402a0a47b45fd2c4af51296aef895e8b5 - commit a0177c8a1048683e5d43f4712f8f9e37091d4e85 Impact: - Fixes EGL/GLES (wrapper/native) usage on X11, proper Xvisual selection w/ EGL - Fixes EGL/GLES (wrapper/native) usage on Windows, ANGLE works w/ NEWT and forced ES2
* Fix GLArrayDataServer.destroy(GL): vboName was cleared by super class before ↵Sven Gothel2012-03-051-3/+6
| | | | deletion
* NEWT/Windows lockSurfaceImpl(): Detect hdc change and result w/ ↵Sven Gothel2012-03-041-4/+19
| | | | | | | | LOCK_SURFACE_CHANGED This allows an underlying delegation/mapping to update the handles, eg.: EGLDrawable. See commit a0177c8a1048683e5d43f4712f8f9e37091d4e85
* NewtFactory: Since the platform string references are unique, let's just ↵Sven Gothel2012-03-044-20/+18
| | | | compare the references instead of a string compare
* NEWT/GLWindow.display(): No explicit surface locking/unlocking for ↵Sven Gothel2012-03-044-14/+28
| | | | | | | | | | GLDrawableHelper.invokeGL(..) - it's done implicit at makeCurrent()/release() The explicit locking takes away the locking result, eg. SURFACE_CHANGED, which is required to update the delegated drawable handles (e.g.: EGL surface handle). With the followup fix of EGLDrawable.updateHandle()'s recreate EGL surface, an EGL 'wrapper' can work on Windows (eg. ANGLE).
* Add es2readsquare, reassembling JOGL's RedSquareES2 demoSven Gothel2012-03-046-4/+636
|
* Add native es2gears test (mesa-demos, eglut enhanced w/ programmatic lifecycle)Sven Gothel2012-02-2912-0/+2654
|
* TestGearsES2NEWT: Remove redundant GLES2 testSven Gothel2012-02-291-11/+0
|
* CrossTest: Add AWT Unit-Test. Unit-Tests: Refine GLProfile request where we ↵Sven Gothel2012-02-2924-114/+102
| | | | favor a more detailed request than getDefault().
* Software Rasterizer: Add 'Mesa X11' to GL_RENDERER list. EGLDrawable: Remove ↵Sven Gothel2012-02-292-9/+11
| | | | | | | | | 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.
* NativeWindow: Relax Xinerama dependency / Rename Windows impl subfolder to ↵Sven Gothel2012-02-279-92/+190
| | | | | | | common name win32 (same as stub_include) Utilizing dlopen/dlsym in an efficient way relaxes the platform requirement of having Xinerama available. This allows using Nokia N9 MeeGo out of the box.
* ScreenMode/X11: Use common defaults in case of undefined values. X11: ↵Sven Gothel2012-02-272-9/+17
| | | | Tolerate invalid refresh rate and use default.
* Fix minor test issuesSven Gothel2012-02-273-33/+11
|
* NEWT Screen.getCurrentScreenModeIntern() !ScreenMode case: Attempt to get ↵Sven Gothel2012-02-261-0/+3
| | | | screen size if 0.
* Fix TestGearsGLJPanelAWTBug450 (regression of commit ↵Sven Gothel2012-02-261-1/+1
| | | | 397665ff472d83809a028e4f4a5d332294617623)
* GLContext.isCurrentContextHardwareRasterizer(): Recognize 'softpipe' ↵Sven Gothel2012-02-261-3/+18
| | | | (Gallium) as software rasterizer. Check for NULL and issue warnings (TRIAGE THIS).
* UITestCase: Gracely ignore unsupported tests (ES2) ; Ignore all ↵Sven Gothel2012-02-2518-40/+230
| | | | Shared-Context and more 'known' test cases for ES.
* Include GraphicsConfigurationFactory in lifecycle (initSingleton/shutdown) - ↵Sven Gothel2012-02-253-26/+64
| | | | | | fixes recursion on fallback GraphicsConfigurationFactory .. also validate the X11 GraphicsConfigurationFactory in X11GLXGraphicsConfigurationFactory and fail fast
* X11: Fix unavailable GLX (Server): Use fallback GraphicsConfigurationFactory ↵Sven Gothel2012-02-254-21/+35
| | | | | | | | | | | (X11GLX -> X11) - GraphicsConfigurationFactory.registerFactory(..) returns previous mapped GraphicsConfigurationFactory. This allows daisy chaining of GraphicsConfigurationFactory, in case one is not suitable. - X11GLXGraphicsConfigurationFactory: - Store the previously mapped factory as fallback - choose*Impl(): If GLX is not available, use fallback
* Cleanup DEBUG flags of *GraphicsConfigurationFactory (Use the common ↵Sven Gothel2012-02-254-7/+0
| | | | 'nativewindow.debug.GraphicsConfiguration')
* X11/GLX: Query X Server whether GLX is available (X11/SharedResourceRunner) ↵Sven Gothel2012-02-255-26/+36
| | | | | | | | | | | | ; Minor cleanups. Before gathering GLX details and actually instantiate a X11/GLX shared resource, we shall query whether GLX is actually available. Since GLX availability is being queried on the server side, more changes are required for the X11GLX* impl, eg. not using X11GLXGraphicsConfigurationFactory and fall back to X11GraphicsConfigurationFactory.
* Fix commit 9e66972c193399d6dcdf9e6662f4335bdf15736aSven Gothel2012-02-251-1/+1
|
* enhancement: NWJNILibLoader.loadNativeWindow(..) returns true/false if ↵Sven Gothel2012-02-255-10/+17
| | | | loading was successful. Negative result causes proper exception in caller.
* Fix GLProfile/GLDrawableFactory bug: Recursion on default desktop device, ↵Sven Gothel2012-02-257-31/+71
| | | | | | | | | | | | | | 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).
* JoglVersion: Avoid NPE if no caps are available.Sven Gothel2012-02-251-3/+5
|
* Min. Graph Parameter type change: texSize/width/.. for multipass-renderer: ↵Sven Gothel2012-02-2516-56/+67
| | | | | | | | | | int -> int[] "texWidth desired texture width for multipass-rendering. The actual used texture-width is written back when mp rendering is enabled, otherwise the store is untouched." This allows the 'backend' to correct the texSize, ie in regards to GL_MAX_TEXTURE_SIZE .. etc. Without this write-back, it would re-create the FBO for every frame.
* TestInitConcurrentNEWT: Reduce max threads (16 -> 8) on ARM, also use vsync ↵Sven Gothel2012-02-241-2/+7
| | | | to reduce load
* Surface2File: Use .tga to write pixels, which allows RGB and RGBASven Gothel2012-02-241-1/+1
|
* Fix MultisampleDemo01 -> MultisampleDemoES1 (proper profile, remove ↵Sven Gothel2012-02-243-31/+52
| | | | immediate mode for ES1) ; Fix GLES1Impl.glOrtho()
* Add GLProfile.getGL2GL3() meta profile getter completing getGL2ES[12]()Sven Gothel2012-02-245-8/+58
|
* GLProfile.GL2GL3: Use GL2GL3 profile (required for this test) and check it's ↵Sven Gothel2012-02-231-8/+7
| | | | availability.
* MemoryObject: Use proper generics styleSven Gothel2012-02-231-2/+2
|
* Enhance GLProfile's profile tests - compare the final String references, ↵Sven Gothel2012-02-231-19/+19
| | | | | | | | | instead of String comparison. GLProfile's 'profile' string reference is final and one of the static final GL* string references, with which it is compared. Hence only the references can be used here. Impact: Performance.
* Fix GLProfile.isGL2ES2(): "isGL2() || isGL3()" -> "isGL2GL3()"Sven Gothel2012-02-231-14/+14
| | | | .. was excluding: GL2GL3 common profile.
* NEWT/ScreenMode DEBUG_TEST_SCREENMODE_DISABLED ↵Sven Gothel2012-02-231-1/+6
| | | | (-Dnewt.test.Screen.disableScreenMode) - If set also avoid 'getCurrentScreenModeImpl()'
* NEWT: Add time costs in DEBUG mode. NEWT/XRandR: Reuse ↵Sven Gothel2012-02-235-128/+186
| | | | | | | | | | | XRRScreenConfiguration to reduce perf hit on Linux ARM Omap4. On Linux ARM Omap4, we experience a performance hit when using XRandR: 1st call of XRRSizes: ~ 1668 ms Each call of XRRGetScreenInfo: ~ 1109 ms Even though XRRGetScreenInfo is cached in NEWT's X11Screen initialization, overall init time is ~2s, far too expensive.
* Fine tune unit tests for ES platformsSven Gothel2012-02-2213-49/+134
|
* X11Screen: Only dump RandR version info in DEBUG mode.Sven Gothel2012-02-221-1/+1
|
* EGL Display Lifecycle Robustness Patch (impl. workaround)Sven Gothel2012-02-226-13/+121
| | | | | | | | | | Added EGLDisplayUtil helper class managing the lifecycle of the EGL display handle recursively. This class is required, due to implementation bugs within EGL where EGL.eglTerminate(long) does not mark the resource for deletion when still in use, bug releases them immediatly. This fixes unit test com.jogamp.opengl.test.junit.jogl.acore.TestInitConcurrentNEWT on Linux ARM w/ Omap4 and Tegra2.
* Fix commit 33249b6eca519947b02f3bfbf05b73d73c936094Sven Gothel2012-02-221-4/+0
|
* DEBUG Output: More thread-names to drawable/context lifecycle; Remove ↵Sven Gothel2012-02-2227-119/+125
| | | | massive '!!!' occurence
* TestGPUMemSec01NEWT ES fix: No RGB read pixel available on ES platformSven Gothel2012-02-221-11/+65
|
* Test: UITestCase.get*Name() takes 'String separator' as argument.Sven Gothel2012-02-224-15/+15
|
* GraphicsConfigurationFactory: Add DEBUG dump @ chooserSven Gothel2012-02-223-1/+8
|
* ExtensionAvailabilityCache: Remove 'XXX_NV_vertex_array_range' hackSven Gothel2012-02-221-11/+1
|
* Fix VBORegion2PES2: Don't exceed MAQX_TEXTURE_SIZESven Gothel2012-02-221-0/+8
|
* TextureIO (TGA/NetPbm): Allow GL_BGR[A] and use GL_BGRA if available ; Fix ↵Sven Gothel2012-02-226-61/+284
| | | | | | | | | | | | | NetPbmTextureWriter ; Added unit tests - Allow GL_BGR[A] usage (TGA / NetPbm) - Use GL_BGRA if available (TGA), utilize GLContext.isTextureFormatBGRA8888Available() - Fix NetPbmTextureWriter - Maintain 'auto' magic mode for 'spi' role in TextureIO (was overwritten) - Use FileChannel for nio buffer streaming, instead of array copy
* Test (graph) fixes for ESSven Gothel2012-02-227-52/+108
| | | | | | - MSAATool: Catch gl-errors on glIsEnabled(..) queries - Call MSAATool.dump(..) in base class GPURendererListenerBase01
* Fix GLReadBufferUtil for ES platforms: RGB read format may not be supported. ↵Sven Gothel2012-02-221-13/+44
| | | | Use GL_IMPLEMENTATION_COLOR_READ_FORMAT/TYPE query
* FBObject: Meaningfull error message if TexImage2D fails.Sven Gothel2012-02-221-10/+17
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 04:47:47 +0000