aboutsummaryrefslogtreecommitdiffstats
path: root/src/jogl/classes/jogamp/opengl
Commit message (Collapse)AuthorAgeFilesLines
* Bug 888 - Validate CPU Runtime Performance: Only use one HashMap to collect ↵Sven Gothel2013-11-051-30/+38
| | | | | | | | unified gl- and glx-extension strings. - HashMap is more efficient than HashSet - No need to use sub HashSet's .. we can use global HashMap. 'Loosing' duplicate GLX entries due to GL duplicates is acceptable.
* Bug 888 / Bug 891 - Enhance GLCapabilities-Query: Apply changes of commit ↵Sven Gothel2013-11-057-105/+192
| | | | | | | 613e33ee8ffc1f2b9c5db1e1b5bb5253a159ed6d to EGL and WGL. Note: WGL config query is already performed as a bulk operation. Note: OSX does not perform such queries.
* Bug 888 - Validate CPU Runtime Performance: ↵Sven Gothel2013-11-042-61/+129
| | | | | | | | | | | | | | | | | | | | X11GLXGraphicsConfiguration.GLXFBConfig2GLCapabilities(..) X11GLXGraphicsConfiguration.GLXFBConfig2GLCapabilities(..) ran over all FB configs and for each it grabbed native config values separately. Fetching them in bulk mode saves around 7% of this function's cost. Also reuse XRenderPictFormat instance for 'XRenderDirectFormat XRenderFindVisualFormat(..)' call, saving a few NIO creation cycles w/ StructAccessor. Biggest savior is X11GLXGraphicsConfigurationFactory.chooseGraphicsConfigurationFBConfig()'s fast path w/o chooser and usable 1st FBConfig. Here we only issue 'GLXFBConfig2GLCapabilities(..)' on the first valid entry. Test w/ 50 X11GLXGraphicsConfigurationFactory.chooseGraphicsConfigurationFBConfig() invocations: - pre change: 1.708 ms - post change: 650 ms Time is no spent almost solely on native glXChooseFBConfig (546ms).
* GLContextImpl: Move sharedContextHandle check to makeCurrentWithinLock(..) ↵Sven Gothel2013-11-028-90/+54
| | | | and let it fail there instead of within impl. class, only pass the handle - simplifies and removes redundancy.
* GLRendererQuirks: Add GLSharedContextBuggy ('Mesa Intel 9.2.1' and ↵v2.1.2Sven Gothel2013-11-011-5/+23
| | | | | | | 'Hisilicon Immersion.16') Note: Even though Mesa Intel driver crashes w/ heavy multithreading (Bug 873), it works well w/ our multithreaded GLMediaPlayer.
* Bug 885 - GLMediaPlayer: Allow single threaded mode - Especially where ↵Sven Gothel2013-11-013-55/+133
| | | | | | | | | | | | 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. -
* Clarify Bug 692: Unbinding a VAO does _not_ imply unbinding of set VBOs ↵Sven Gothel2013-11-012-11/+30
| | | | | | | | | | | | | | | | | | | | | (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.
* EGLDrawableFactory.createDummySurfaceImpl: fix caps to pbuffer - since we do ↵Sven Gothel2013-10-312-1/+2
| | | | | | | use EGLDummyUpstreamSurfaceHook .. a pbuffer offscreen fixOffscreenBitOnly(..) may yield FBO queries / chosing .. usually doesn't matter (on devices here), but not accurate.
* GLDrawableFactory: createDummy*(..) Pass GLCapabilitiesImmutable + ↵Sven Gothel2013-10-312-10/+9
| | | | GLCapabilitiesChooser instead of GLProfile, allowing using same or similar caps - important for sharing ctx
* AndroidGLMediaPlayerAPI14: Avoid possible NPEsSven Gothel2013-10-311-44/+50
|
* Fix Bug 875 - Cleanup branch using VersionNumber (fix 'Int' check ↵Sven Gothel2013-10-303-59/+89
| | | | | | | | | | | | 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)
* Bug 875: Safeguard setGLFunctionAvailability(.. strictMatch=false.. ) ↵Sven Gothel2013-10-306-8/+19
| | | | operation, throw InternalError if failing
* Merge pull request #72 from masterzen/tickets/875Sven Gothel2013-10-291-9/+17
|\ | | | | Fix #875 - ES version should be strictly validated
| * Fix #875 - ES version should be strictly validatedBrice Figureau2013-10-291-9/+17
| | | | | | | | | | | | | | | | | | 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 <[email protected]>
* | Bug 776 GLContext Sharing: Fix copy-ctor GLArrayDataClient: Create new ↵Sven Gothel2013-10-305-9/+7
| | | | | | | | | | | | instance of GLArrayHandler of same type; Simplify GLArrayHandler inheritance. Refines commit 9f2a9df0a4b7093925c8854b37fba053469a4b35
* | Bug 877 Concurrency Discussion: Update doc MultiThreading.txt, volatile ↵Sven Gothel2013-10-292-3/+7
| | | | | | | | field usage; GLDrawableImpl: Make read-only fields final.
* | GLDrawableFactory: Add 'GLAutoDrawable createDummyAutoDrawable(..)' for ↵Sven Gothel2013-10-291-0/+10
| | | | | | | | | | | | | | convenience Added: GLAutoDrawable createDummyAutoDrawable(AbstractGraphicsDevice deviceReq, boolean createNewDevice, GLProfile glp)
* | Bug 876: BuildComposablePipeline: getGL*() shall not return ↵Sven Gothel2013-10-291-7/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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;"
* | GLDrawableHelper.reshape(): Catch pre-existing GL errors before ↵Sven Gothel2013-10-291-0/+9
|/ | | | glViewport(..) and dump it (Add stack trace if DEBUG)
* Bug 776 GLContext Sharing: Refine API for relaxed and lazy GLContext sharing ↵Sven Gothel2013-10-276-83/+218
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ; 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 regression of commit 34b35c5a0a379a6b4c0b23b9d347a0b1338f0239 - ↵Sven Gothel2013-10-251-3/+11
| | | | GLContextImpl.createContextARBVersions(..) erroneous upper bounds check.
* Merge remote-tracking branch 'hharrison/master'Sven Gothel2013-10-251-5/+4
|\
| * jogl: replce more unneeded String() constructionHarvey Harrison2013-10-201-5/+4
| | | | | | | | Signed-off-by: Harvey Harrison <[email protected]>
* | Fix GLContext: getGLProfile() add missing GLES3; ↵Sven Gothel2013-10-251-1/+1
| | | | | | | | getAvailableGLProfile(device, ..) shall use GLProfile.get(device, ..)
* | Bug 867 OSX [Common Code]: Trigger GLRendererQuirks.GL4NeedsGL3Request and ↵Sven Gothel2013-10-258-53/+111
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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.
* | Bug 867 - OSX: Allow core >=4 if isMavericksOrLater; Use ↵Sven Gothel2013-10-242-27/+48
| | | | | | | | [kCGLOGLPVersion_GL4_Core, kCGLOGLPVersion_GL3_Core] for major==4 depending on sticky GLRendererQuirks.GL4NeedsGL3Request
* | Fix Bug 866 - Frequent IndexOutOfBoundsException in ↵Brice Figureau2013-10-231-1/+1
|/ | | | jogamp.opengl.egl.EGLGraphicsConfigurationFactory: Add missing 'else' in branch
* Fix Bug 862: Fix GL Version Validation / NVidia GTX550 driver 331.13 - 64bit ↵Sven Gothel2013-10-192-59/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Linux - No compatibility GLProfile (GL2, >= GL3bc) Fix GL Version Validation: We shall not rely on our known good versions when validating a queried GL context version, but allow some 'room' for a higher version post JOGL release while still cutting off 'odd versions'. While GL version detection, we always iterate from the highest known version down to the lowest. Hence 'GLContext.isValidGLVersion(..)' is satisfied by validating the lowest version number but allowing a higher than known one. Now we would return 'invalid' for a version >= 6. It is enough to clip to the maximum known version when iterating, allowing the highest unknown version to be available. GLContext.isValidGLVersion(..): Returns true, if the major.minor is not inferior to the lowest valid version and does not exceed the highest known major number by more than one. The minor version number is ignored by the upper limit validation and the major version number may exceed by one. The upper limit check is relaxed since we don't want to cut-off unforseen new GL version since the release of JOGL. Hence it is important to iterate through GL version from the upper limit and 'decrementGLVersion(..)' until invalid. Add GL Version 4.4 to valid known versions. Remove ES3 desktop detection, which is impossible Regression of commit 3a0d7703da32e9a5ddf08a334f18588a78038d88 (ES3 support)
* jogl: fix two impossible comparisons in glu/tessellatorHarvey Harrison2013-10-171-2/+2
| | | | | | | | | The comparison to Long.MAX_VALUE will never trigger as it is coparing with an int. The intent of this code appears to be to check against Integer.MAX_VALUE which is used as an error code (unable to allocate sufficiently large array) from the priorityQueue. Signed-off-by: Harvey Harrison <[email protected]>
* jogl: add missing @Override annotationsHarvey Harrison2013-10-1787-1/+245
| | | | Signed-off-by: Harvey Harrison <[email protected]>
* jogl: remove all trailing whitespaceHarvey Harrison2013-10-17181-3591/+3591
| | | | Signed-off-by: Harvey Harrison <[email protected]>
* Refine Int -> Enum conversion (commit ↵Sven Gothel2013-10-171-11/+29
| | | | | | | 40863632d1428de015099b5967e5136425e99f25), throw IllegalArgumentException if ordinal is out-of-range. Add API doc. - FFMPEGNatives - MouseEvent.PointerType
* Int -> Enum using EnumClass.values()[ordinal] instead of for-loop - ↵Sven Gothel2013-10-171-8/+8
| | | | FFMPEGNatives's Enums and new MouseEvent.PointerType.valueOf(int)
* AWTPrintLifecycle.setupPrint(..): Add optional tileWidth and tileHeight, ↵Sven Gothel2013-10-151-8/+13
| | | | allowing user to set custom tile size for performance evaluation/tweak
* Merge pull request #71 from esemplare/masterSven Gothel2013-10-121-14/+63
|\ | | | | | | | | Fix Bug 365: Bug on javax.media.opengl.glu.GLU.gluBuild2DMipmaps I will commit your unit test manually from the bug report (pls add it to your pull request next time) Thank you!
| * Fix Bug 365: Bug on javax.media.opengl.glu.GLU.gluBuild2DMipmapsMichael Esemplare2013-10-071-14/+63
| | | | | | | | Fixed convolution window widths for single column scaling.
* | Bug 860 - AWT Printing (AWTTilePainter): Shall use the enclosing integer ↵Sven Gothel2013-10-121-7/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | rectangle of the scaled double precision clipping rect AWTTilePainter simply rounds the scaled double precision clipping rectangle to receive the integer rectangle. This leads to uncovered drawing areas, since the integer rectangle position could be greater - and the size could be smaller than the double precision source. To get the enclosing rectangle, we need to use iPos = floor(position) iSize = ceil(position+size) - floor(position) .. turns our that the double precision 'Rectangle Rectangle2D.getBounds()' already performs this math.
* | Bug 852: Validate CPU sourced data API is allowed, throw exception if not. ↵v2.1.0Sven Gothel2013-10-101-2/+2
| | | | | | | | GLContext: isCPUSourcedAvail() -> isCPUDataSourcingAvail()
* | Bug 852: Remove CPU sourced data API entry where not allowed (ES3 and GL ↵Sven Gothel2013-10-101-4/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | core >= 3.0) ; GL2 cfg: Ignore GL2ES3 symbols (super) Remove CPU sourced data API entries via new config 'BufferObjectOnly <name>', listed in 'gl-common-gpubufferonly.cfg' and included in ES3 and all GL core >= 3 interfaces. If BufferObjectOnly is defined for a function, only the 'long offset' variant is being emitted. Due to limitations of GlueGen's 'ExtendedInterfaceSymbolsIgnore A.java', which only identifies the function name and not the signature - all CPU sourced variants are manually added to the compatibility and ES2 GL profiles via '*-common-cpubufferJavaCode.java' files. GLContext: Added 'isCPUSourcedAvail()' to determine whether context allows CPU sourced data, i.e. for GL2ES1 and GLES2 ctx. GLContext/GLProfile/GL: isGLES2() now returns false if having a ES3 context due to 'CPU sourced' incompatibility! +++ GL2 cfg: Added ignore GL2ES3 symbols of it's superclass, removing duplicated symbols.
* | NativeWindow/OSX: Fix RunOnThread/RunLater - Properly determine 'forkOnMain' ↵Sven Gothel2013-10-071-1/+1
| | | | | | | | via onMain && ( isOnMain || 0 < delay )
* | FFMPEGMediaPlayer: Use IOUtil.decodeFromURI(..) to unescape %20Sven Gothel2013-10-071-1/+2
| |
* | Bug 789: glDrawArraysIndirect and glDrawElementsIndirect are using ↵Sven Gothel2013-10-061-0/+2
| | | | | | | | GL_DRAW_INDIRECT_BUFFER, add VBO variant and range-check
* | AWTTilePainter: Avoid NPE if Graphic2D's AffineTransform is null, use ↵Sven Gothel2013-10-031-5/+12
| | | | | | | | identity tranform in such case.
* | Bug 818: Enable new quirk GLSLNonCompliant on OSX < 10.7 w/ NV GPU; ↵Sven Gothel2013-10-031-6/+16
| | | | | | | | GLJPanel: Don't use GLSL flip if quirk GLSLNonCompliant is present.
* | Bug 548: GLRendererQuirks.GLFlushBeforeRelease is needed on OSX < 10.7.3 w/ ↵Sven Gothel2013-10-031-2/+2
|/ | | | NV GPU [was comparing against 1.7.3 instead 10.7.3] !
* TileRenderer*: TileRendererNotify -> TileRendererListener w/ clarifying ↵Sven Gothel2013-09-271-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | functionality (reshapeTile(..), ..); Only process GLEventListener impl. TileRendererListener; attachToAutoDrawable -> attachAutoDrawable, etc. -TileRendererNotify -> TileRendererListener - Added methods: - void reshapeTile(TileRendererBase tr,int tileX, int tileY, int tileWidth, int tileHeight, int imageWidth, int imageHeight); - void startTileRendering(TileRendererBase tr); - void endTileRendering(TileRendererBase tr); allowing to clarify user code and API specification, i.e. TR only processes GLEventListener which impl. TileRendererListener. This also allows simplifying the API doc, while having a more descriptive reshape method focusing solely on tile rendering. Further more, the start/end TR methods allow certain GL related actions while the context is current before and after iterating through the tiles. This is even used for RandomTileRenderer (one tile only), to allow to reuse same TileRendererListener for diff TRs. - Fix language, attach and detach usage was vice versa. We do attach an GLAutoDrawable to a TR - attachToAutoDrawable -> attachAutoDrawable - detachFromAutoDrawable -> detachAutoDrawable - Adapted unit tests.
* Fix Bug 839: Clarify whether resetStates(..) shall clearStates() - Pass ↵Sven Gothel2013-09-216-28/+28
| | | | 'isInit' flag.
* AWT Printing: Fix non vertical-flipped printing, i.e. cut-off top-row's tile ↵Sven Gothel2013-09-201-19/+20
| | | | | | | | | | | | | | | upper area, cleanup. TestTiledPrintingGearsSwingAWT2: Provoked !flipped bug where top-row was positioned too low due to using full size tile-height. Cutting of the unused top-row's upper area corrects this issue. vertical-flip mode does not expose this situation, since flipping shifts the payload to the upper tile area. TestTiledPrintingGearsSwingAWT2: Also tests an alternative transparent overlapping mode without layout.
* Bug 839: Clarifying GLStateTracker.clearStates(..) remove 'enable' change - ↵Sven Gothel2013-09-202-9/+12
| | | | | | Part 1 TODO: Only disable state tracker at GLContext.destroy()
* AWTTilePainter.setupGraphics2DAndClipBounds(): Use 'Shape getClip()'; Use ↵Sven Gothel2013-09-191-31/+42
| | | | | | | | | | | | | | | | double precicion clip bounds 'all the way'; Explicitly scale image and clip w/ current scaled transform. - Use 'Shape getClip()' Don't assume Rectangle2D, but use Shape's getBounds2D() - Use double precicion clip bounds 'all the way' Remove rounding error on clip bounds w/ start value, which was _not_ using doubles. - Explicitly scale image and clip w/ current scaled transform. Instead of abusing Graphics2D's clip shape to scale image size and clip-area, explicitly use transform both bounding boxes into transformed space, scale space and transform out (inversion). A possible NoninvertibleTransformException will be thrown while Graphics2D has not been modified.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-01 23:44:37 +0000