aboutsummaryrefslogtreecommitdiffstats
path: root/src/quick/scenegraph/qsgthreadedrenderloop.cpp
Commit message (Collapse)AuthorAgeFilesLines
* Merge remote-tracking branch 'origin/tqtc/lts-5.15.6' into ↵v5.15.6-lts-lgplTarja Sundqvist2022-08-161-7/+6
|\ | | | | | | | | | | tqtc/lts-5.15-opensource Change-Id: Ic3ca76460360cbab866a4e881c4848e6deeaf7d7
| * Sync shadereffect materialType cleanup between renderloopsLaszlo Agocs2021-08-021-7/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The render loops behave differently: the threaded one cleans the "material cache" (ill-named for legacy GL, it is correctly called material _type_ cache on the newer RHI code path) every time the window is unexposed. The basic one keeps it and only cleans the cache (and so destroys the opaque QSGMaterialType objects) only when the scenegraph (i.e. QSGNodes and related geometry and materials) gets released. The former is incorrect, because if the QSGMaterial does not go away, the QSGMaterialType should stay valid too. Strictly speaking this is not a fatal error because the QSGMaterialType pointers are never dereferenced. However, potential reuse of the same address due to a subsequent 'new QSGMaterialType' is an issue. Just switch to the latter everywhere, and also, to be fully pedantic, make the order uniform across all 3 render loops: first destroy the scene tree (where each QSGGeometryNode destroys its QSGMaterial if owned), then the cache (which owns the QSGMaterialTypes for the now-gone shadereffect nodes). Fixes: QTBUG-94844 Change-Id: Iebbf34d36b9eccca85e2557b045e326eef347beb Reviewed-by: Andy Nichols <andy.nichols@qt.io>
* | Revert "Update commercial license headers"v5.15.3-lts-lgplTarja Sundqvist2021-03-241-20/+20
|/ | | | | | | | | | | This reverts commit 7b1331c789df0ac12d7cf5aa5f8967b8651a4cfa. Revert of commercial license headers is required for Qt 5.15.3 opensource release. Task-number: QTBUG-91108 Change-Id: I47eb004523bdb3beb16e9fd51e40da6f738f9f34 Reviewed-by: Jukka Jokiniva <jukka.jokiniva@qt.io>
* Update commercial license headersTarja Sundqvist2021-02-021-20/+20
| | | | | | | | | | | | | | | Updated header.COMM to the files in tqtc-qtdeclarative. Examples, documentation files or tests are not updated. The commercial license header may contain some additional lines so that its line count equals with the earlier license header. Reason for this is that some autotests use hard coded line numbers and a change in the line count causes failures in tests. Task-number: QTQAINFRA-4161 Change-Id: Ic1d011aa01b1fe7242bcbbf72d5ab454d1015cf9 Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
* Move the sharegroup with the context to the render threadLaszlo Agocs2020-09-101-0/+2
| | | | | | | | | | | | ...if the context does not have a share context, i.e. the QOpenGLContextGroup was created together with the QOpenGLContext. This is not applicable to Qt 6 where the context is created on the render thread and so there is no moveToThread() for it. Fixes: QTBUG-86524 Change-Id: Ida8b947548325ea2f7572cdb9373bbc90e4e7e4f Reviewed-by: Andy Nichols <andy.nichols@qt.io>
* rhi: Handle beginFrame() failure correctly in threaded render loopLaszlo Agocs2020-02-061-6/+10
| | | | | | | | Also add some comments to improve maintainability. Task-number: QTBUG-81740 Change-Id: I4be90768f7a8506b1fb493885062b98be9f4f58a Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
* Add trace points for Qt Quick Scene GraphMilian Wolff2019-12-131-3/+22
| | | | | | | | | | | | This adds Q_TRACE tracepoints for all places that are covered by the QML profiler's QSG integration. The big advantage over the existing framework is that these trace points can be visualized next to other system tracepoints to correlate the data better with load and resource consumption induced by other processes on the system. Change-Id: I0c5b70a0870f0b89e4533c351c099e13fd18a55f Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
* Clear RequestRepaint bit set in sync() if we are going to renderLaszlo Agocs2019-12-031-13/+24
| | | | | | | | | | | | ...because if we are going to go on doing a frame then the repaint request is fulfilled. Leaving the bit set would cause the render thread's main loop to start another frame right after the current one is queued, which is kind of bad in certain scheduling and blocking cases because the main thread may end up starving due to the unnecessary extra (and sync-less) frame. Task-number: QTBUG-80498 Change-Id: I6c9a8f60e5f49d771d7d284bb7399eadbde33b8c Reviewed-by: Paul Olav Tvete <paul.tvete@qt.io>
* Make the initialization failure dialog sensible with QRhiLaszlo Agocs2019-11-281-2/+1
| | | | | | | | | | | | | | | So on Windows one now gets a message box with a reasonable message, instead of the OpenGL nonsense. Then the application closes when pressing Abort etc. On other platforms there is a qFatal, printing the same message. Involves simplifying the OpenGL version a bit since passing isES type of flags through multiple layers is not justified here. Task-number: QTBUG-80365 Change-Id: Ie3ea1e9395a283f7e95eda78c1d3894797ff0acf Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
* Prevent infinite rhi init failure in threaded render loopLaszlo Agocs2019-11-281-7/+35
| | | | | | Task-number: QTBUG-80365 Change-Id: I929fb76eb8d023ab048f6d1c91be078de3cfe750 Reviewed-by: Paul Olav Tvete <paul.tvete@qt.io>
* Avoid initializing QFlags with 0 or nullptrAllan Sandfeld Jensen2019-11-221-1/+1
| | | | | | | | | It is being deprecated. Change-Id: I844bd92af85bc53a8fc0371408d05277bd49f511 Reviewed-by: Simon Hausmann <simon.hausmann@qt.io> Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io> Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
* Fix broken scaling with high dpi + rhi + thread loopLaszlo Agocs2019-10-211-3/+1
| | | | | | | | | | | | | | The arguments to renderSceneGraph() were incorrect in the threaded render loop, unlike basic. Fix it up to follow what we do in the basic loop. (The first argument is the QWindow size (logical). The second is only used when rendering via the rhi, and is the output surface/layer size, in pixels. This will get simplified in Qt 6.) This fixes broken rendering with Metal and the threaded render loop for windows with dpr > 1. Change-Id: I427ba5def8cc34900e6fe650e1006ca5f0fa90a4 Reviewed-by: Paul Olav Tvete <paul.tvete@qt.io>
* QQuickWindow: Don't leak the animation controllerUlf Hermann2019-10-021-2/+3
| | | | | | | | | There are ways to close the window without hitting the code paths in the render loops that delete the animation controller. Probably if no frame was ever rendered. Change-Id: If3e9d2051525c4ff50eda19084c967578fe4f4b0 Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* Remove depth-stencil buffer sizing on the rhi pathLaszlo Agocs2019-09-301-5/+3
| | | | | | | | | ...as this will be done automatically by QRhiSwapChain after the corresponding QtGui changes. Task-number: QTBUG-78641 Change-Id: I8edeb728f3aba07bfa6bc6331966fba4bdee71f4 Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
* Recover from device lost on the rhi path in the threaded loopLaszlo Agocs2019-09-231-61/+100
| | | | | Change-Id: I90d43d5daa75bbc52c9c10f4ef920b898bbd39d4 Reviewed-by: Christian Strømme <christian.stromme@qt.io>
* Request correct alpha composition on the rhi code pathLaszlo Agocs2019-09-111-1/+9
| | | | | | Task-number: QTBUG-78089 Change-Id: I22f8bb5ec0af33397df14e064a0306bd4c5a5ef5 Reviewed-by: Christian Strømme <christian.stromme@qt.io>
* Make vulkanunderqml work and update docsLaszlo Agocs2019-09-081-1/+3
| | | | | | | | | | | | | | | What we are doing for now is setting ExternalContentsInPass always. This way vulkanunderqml works as expected. For applications that do not integrate external rendering this means that there is now an additional secondary command buffer per render pass, but we can live with this for now. Later (Qt 6) there should be a way to declare this (that the application will want to issue native rendering stuff) up front in QQuickWindow or somewhere. Change-Id: I736741f9b0eee2f8295b046bacdce862e6a546f5 Reviewed-by: Andy Nichols <andy.nichols@qt.io>
* Remove an unused event type in the threaded render loopLaszlo Agocs2019-08-221-11/+0
| | | | | Change-Id: I505ba09d6a0144f18bf29cda2f549c8b69ada1a5 Reviewed-by: Christian Strømme <christian.stromme@qt.io>
* Change rhi-related debug prints to categorized loggingLaszlo Agocs2019-08-021-3/+3
| | | | | Change-Id: I4e8d3111a2f3b77e984756cc9eef49d42f0b647c Reviewed-by: Andy Nichols <andy.nichols@qt.io>
* Have an autorelease pool for each frame on the render threadLaszlo Agocs2019-07-301-0/+4
| | | | | | | | | | | | | | | | | | Required with Metal. Drawables and various system resources get autoreleased and there is no pool to handle these on the SG render thread by default. This may present a slight perf hit in debug builds due to QMacAutoReleasePool doing smart but - for us - unnecessary stuff every time (so in our case, every frame) but will do for now. This complements the QRhiMetal change for not holding onto the drawable when skipping the present. The pool is essential then to prevent nextDrawable from starving and so blocking. Task-number: QTBUG-76953 Change-Id: Iaf803a0e20504d6b349d3564eda1677868fb29ef Reviewed-by: Andy Nichols <andy.nichols@qt.io>
* Avoid locking up on resize with threaded loop and the rhiLaszlo Agocs2019-07-301-4/+18
| | | | | | | | | | This was visible on X11 only because there beginFrame() happened to fail once or twice with out-of-date swapchain when there were a lot of resizes in a row. Handle this case correctly by waking up the main thread as appropriate. Change-Id: I67dc18522e1c05070267fd355095324f48259276 Reviewed-by: Andy Nichols <andy.nichols@qt.io>
* Enable request setting the (gl) context current on rhi pathLaszlo Agocs2019-07-041-10/+21
| | | | | | | This was disabled due to the unmerged qtbase api change. Change-Id: I38beb8f2aa11dc233765bcfe06e91940b64b5758 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
* Add the graphics api independent scenegraph portLaszlo Agocs2019-07-041-93/+356
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Opt in via environment variables: QSG_RHI=1 -> enable using QRhi instead of GL QSG_RHI_BACKEND -> set to vulkan, metal, d3d11, gl to override the default (the default is d3d11 on Windows, metal on Mac, gl elsewhere) Or force a given rhi backend via the existing QQuickWindow::setSceneGraphBackend(). Otherwise the default behavior is the same as before, the rhi code path is never active by default. -no-opengl builds are supported in the sense that they work and default to the software backend. However, the rhi code path cannot currently be used in such builds, even though QRhi from qtbase is fully functional with Vulkan, D3D, or Metal even when qtbase was configured with -no-opengl. This cannot be utilized by Quick atm due to OpenGL usage being all over the place in the sources corresponding to the default backend, and those host the rhi code path as well. This will be cleaned up hopefully in Qt 6, with the removal all direct OpenGL usage. Other env.vars.: QSG_RHI_DEBUG_LAYER=1 -> enable D3D debug or Vulkan validation layer (assuming the system is set up for this) QSG_RHI_SHADEREFFECT_DEBUG=1 -> print stuff from ShaderEffect QSG_SAMPLES=1,2,4,... -> MSAA sample count (but QSurfaceFormat works too) QT_D3D_ADAPTER_INDEX=0,1,... -> D3D adapter index QT_VK_PHYSICAL_DEVICE_INDEX=0,1,... -> Vulkan physical device index QSG_RHI_UINT32_INDEX=1 -> always use uint index data (both merged/unmerged, convert when needed - with some rhi backends this is implicit) QSG_RENDER_LOOP -> to override the render loop as usual. The default with RHI is threaded for Metal, threaded for Vulkan on Windows, basic for Vulkan on Linux and Android (to be checked later), while the existing rules apply for OpenGL. Not supported when running with QRhi: - particles - compressed atlases (though this is transparent to the apps) - QSGRenderNode - QQuickRenderControl - QQuickFramebufferObject - certain QQuickWindow functionality that depends directly on OpenGL - anisotropic filtering for textures - native text may lack some gamma correction - QSGEngine applicability unclear - some QML profiler logs may be incorrect or irrelevant Change-Id: I7822e99ad79e342e4166275da6e9e66498d76521 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
* Fix leaking scene graph rendering contextsSimon Hausmann2019-03-151-2/+8
| | | | | | | | | | | | | | | | When using the threaded render loop, the rendering thread assumes ownership of the rendering context. If the rendering thread is never started, that context is leaked. This happens in tests sometimes when we create and destroy a window without waiting for exposure. So this patch maintains the ownership in the render loop for created contexts until the per-window thread is started. Task-number: QTBUG-74348 Change-Id: Ifa397fab0833c82110ac4571c1e3790351c43afd Reviewed-by: BogDan Vatra <bogdan@kdab.com> Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
* Fix more null pointer accesses in QQuickOpenGLShaderEffectMaterialCacheJüri Valdmann2018-11-121-1/+2
| | | | | | | | | This amends 6c08137faf1a53db879701126608833474a2450b. Fixes: QTBUG-71705 Change-Id: I0e9ca137c039802d348af4a29146395155e497c0 Reviewed-by: Liang Qi <liang.qi@qt.io> Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
* use nullptr consistently (clang-tidy)Shawn Rutledge2018-02-261-15/+15
| | | | | | | | | | | | | From now on we prefer nullptr instead of 0 to clarify cases where we are assigning or testing a pointer rather than a numeric zero. Also, replaced cases where 0 was passed as Qt::KeyboardModifiers with Qt::NoModifier (clang-tidy replaced them with nullptr, which waas wrong, so it was just as well to make the tests more readable rather than to revert those lines). Change-Id: I4735d35e4d9f42db5216862ce091429eadc6e65d Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* doc: Fix remaining "Can't tie this..." errorsMartin Smith2018-02-231-8/+8
| | | | | | | | This update mostly removes qdoc comment markers from comments that should not have been qdoc comments. Change-Id: I8ccaa7fd4ae610371e25066e048fcba6cfba8038 Reviewed-by: Martin Smith <martin.smith@qt.io>
* ScenGraph: Prefer printf style loggingKai Koehne2018-02-131-68/+68
| | | | | | | | This generates more compact code, saving 20480 bytes on disk (Linux 64 bit, release build). Change-Id: I73d6d88b7b61b87a5d714e131fcf86ee80c83f38 Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
* Merge remote-tracking branch 'origin/5.9' into 5.10Liang Qi2017-09-051-0/+7
|\ | | | | | | | | | | | | | | Conflicts: src/quick/items/qquickwindow.cpp src/quick/scenegraph/qsgrenderloop.cpp Change-Id: Idd7106995b5545fcac869e9056a365ef9edb36ca
| * Make releaseResources() free the renderer's shader cacheLaszlo Agocs2017-09-041-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Normally all shader programs from materials are cached in a simple table that is alive as long as the renderer exists, in practice this means being tied to the QQuickWindow which is often the entire lifetime of an application. Use releaseResources(), which is mostly a no-op on the scenegraph level at the moment, to free this cache. Task-number: QTBUG-62392 Change-Id: I279e595874fee1ab4fe0dede0843a3686dea5806 Reviewed-by: Gunnar Sletta <gunnar@crimson.no>
* | Merge remote-tracking branch 'origin/5.9' into devSimon Hausmann2017-08-231-0/+1
|\| | | | | | | | | | | | | Conflicts: tests/auto/qml/debugger/qqmlenginedebuginspectorintegrationtest/tst_qqmlenginedebuginspectorintegration.cpp Change-Id: I31375151eb239f348bec988d2d0506c2b4d9604c
| * Fix memory leaks with threaded rendererJoni Poikelin2017-08-221-0/+1
| | | | | | | | | | | | | | | | | | | | Reverts 39061af50cc3092289cdd71d17802139590ecb59 and calls endSync() instead after syncSceneGraph(). Task-number: QTBUG-61754 Change-Id: I7b51ffdc93557bacd93927dd642ec3715980b21c Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io> Reviewed-by: Roger Zanoni <rogerzanoni@gmail.com>
* | Merge remote-tracking branch 'origin/5.9' into devLiang Qi2017-07-041-0/+1
|\| | | | | | | | | | | | | | | Conflicts: .qmake.conf src/quick/doc/src/qmltypereference.qdoc Change-Id: Ia58f1c5a98309c32ef8d8e5c893faf261215f19f
| * Delay the deletion of QSGTextures until all windows are synchronizedDominik Holland2017-06-161-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the 'basic' and the 'windows' render loop the scene graph context is shared. Because of this we cannot start deleting textures after the first window is synchronized as it may contain textures needed by the another window, which is not yet synchronized. QWindowPrivate::syncSceneGraph() is not calling endSync() anymore as it doesn't know whether it is the last window or not. Instead the renderloop is now responsible for calling endSync() once this is safe to do. Change-Id: Icb50ebfb447c928e38b41df7e26f3bfafdb4a811 Reviewed-by: Robert Griebl <robert.griebl@pelagicore.com> Reviewed-by: Gunnar Sletta <gunnar@crimson.no>
* | Merge remote-tracking branch 'origin/5.9' into devLars Knoll2017-05-021-0/+1
|\| | | | | | | Change-Id: I71275a2076c3d32ee2896571be882067320a2e9e
| * Run includemocs in qtdeclarativeThiago Macieira2017-04-261-0/+1
| | | | | | | | | | Change-Id: I84e363d735b443cb9beefffd14b8c023a37aa489 Reviewed-by: Lars Knoll <lars.knoll@qt.io>
* | Merge remote-tracking branch 'origin/5.9' into devSimon Hausmann2017-04-111-0/+1
|\| | | | | | | Change-Id: I95914046889022e0ef9bdc17338dc5895c5223fe
| * Merge remote-tracking branch 'origin/5.8' into 5.9Liang Qi2017-04-111-0/+1
| |\ | | | | | | | | | Change-Id: Ib3d81ad33a6ba28d891da91271a64d5fcc4874e6
| | * Fix hidpi support for opengl window grabbingFilipe Azevedo2017-04-091-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now set the QImage devicePixelRatio so the content is correct on all screens. Task-number: QTBUG-53795 Change-Id: If94edf901da1285afe9bb847b8973d568a2b7082 Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io> Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
* | | sources: Add missing override and remove redundant virtualAlexander Volkov2017-02-231-2/+2
|/ / | | | | | | | | | | | | Change-Id: I48e26b1491024d41ebf75d40f0d1a7e9a1b20f22 Reviewed-by: Anton Kudryavtsev <antkudr@mail.ru> Reviewed-by: Marc Mutz <marc.mutz@kdab.com> Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
* | Remove remnants of blackberry platform supportRobin Burchell2017-01-161-1/+1
| | | | | | | | | | | | | | | | The last remnants were removed from qtbase in 5.7 making this all dead code, so match here too. Change-Id: I10f3f1c614562f2a97ade7cdf5002065d6f79e07 Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* | QSGThreadedRenderLoop: Correctly stop the animation timer ...Robin Burchell2017-01-121-1/+7
|/ | | | | | | | | | | | | | | | | | | ... when destroying the last window while it is still visible. For the case of windowDestroyed(), the QQuickWindow can still be isVisible true and isExposed true. Given these circumstances, we would not stop the timer (assuming that we needed to keep it running), but this is actually wrong when we are evaluating a timer change depending on the window being destroyed. By calling startOrStopAnimationTimer again, after the window has been removed from m_windows, we will successfully stop the timer -- the only cost being an unnecessary polish event being sent from the first call (which will be ignored). Task-number: QTBUG-58091 Change-Id: I043ce5d88eafee15019e1c8d41c06d531d1da059 Reviewed-by: Gunnar Sletta <gunnar@crimson.no>
* Merge remote-tracking branch 'origin/5.7' into 5.8Liang Qi2016-11-281-8/+16
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The renderers added in 5.8 had to be adapted to the changed profiling macros from 5.6. Conflicts: src/plugins/scenegraph/d3d12/qsgd3d12renderloop.cpp src/plugins/scenegraph/d3d12/qsgd3d12threadedrenderloop.cpp src/quick/scenegraph/adaptations/software/qsgsoftwarerenderloop.cpp src/quick/scenegraph/adaptations/software/qsgsoftwarethreadedrenderloop.cpp src/quick/util/qquickprofiler_p.h tests/auto/qml/qjsengine/tst_qjsengine.cpp tests/auto/qml/qqmlvaluetypes/tst_qqmlvaluetypes.cpp Change-Id: Icb370b7c95aab12589ad73881ac6d178759a5c6b
| * Merge remote-tracking branch 'origin/5.6' into 5.75.7Liang Qi2016-11-251-8/+16
| |\ | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: src/qml/compiler/qv4ssa.cpp src/qml/qml/v8/qqmlbuiltinfunctions.cpp src/quick/util/qquickprofiler_p.h Change-Id: I11a89c2a166115d6697adfba09928805643e709e
| | * QmlProfiler: Explicitly specify the offsets for scene graph eventsUlf Hermann2016-11-241-8/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The profiler can be switched on in the middle of a frame. In that case the last offset into the timing data would be some random number, which may lead to a crash when recording the sample. However, as we know all the data points we are going to record, we can as well specify where they are supposed to go. The timings themselves may still be random for frames of which we only recorded parts, but the clients can deal with this. Task-number: QTBUG-57304 Change-Id: I1d507f2591516e43d5b3cd25f7939716f2b64ed9 Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* | | Make shader effects configurableLars Knoll2016-10-101-0/+4
| | | | | | | | | | | | | | | | | | Change-Id: I4e7fd5e9781dec7ee6ed8807ca1a51c937f6f9f3 Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com> Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
* | | Merge remote-tracking branch 'origin/5.7' into devEdward Welbourne2016-08-021-1/+1
|\| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: src/quick/items/qquickshadereffect.cpp 5.7 had a bug-fix in code dev has replaced wholesale. src/quick/items/qquickwindow.cpp src/quick/items/qquickwindow_p.h One side changed a method's signature; the other side renamed a method declared adjacent to it and changed some code using it, moving some from the public class to its private partner. tests/auto/qml/debugger/qqmlprofilerservice/tst_qqmlprofilerservice.cpp One side added a blank line before a comment the other re-wrote. Kept the re-write, killed the stray blank. .qmake.conf Ignore 5.7's change to MODULE_VERSION. src/qml/compiler/qqmltypecompiler.cpp src/qml/compiler/qqmlpropertyvalidator.cpp 5.7 changed code in the former that dev moved to the latter. Reflect 5.7's changes there, adapted to dev's form. src/qml/qml/qqmlobjectcreator.cpp One side added new QVariant types; the other changed how it handled each type of QVariant (without git seeing any conflict); adapted the new stanzas to work the same as the transformed ones. tests/manual/v4/test262 dev had a broken sha1 for it; so used 5.7's 9741ac4655808ac46c127e3d1d8ba3d27ada618e Change-Id: I1fbe2255b97d6ef405cdd1d0cea7fab8dc351d6f
| * | Merge remote-tracking branch 'origin/5.6' into 5.7Liang Qi2016-08-011-1/+1
| |\| | | | | | | | | | | | | | | | | | | | | | Conflicts: tests/auto/qml/qqmllanguage/tst_qqmllanguage.cpp tests/auto/quick/qquickitem/tst_qquickitem.cpp Change-Id: If261f8eea84dfa5944bb55de999d1f70aba528fd
| | * Remove vsync sleep before rendering first frameMilian Wolff2016-07-181-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The QSGRenderThread slept for up to one vsync interval before the first frame was rendered. This happened when the QSGRenderContext was not yet initialized and thus a sync could not result in changes. With LTTNG and custom trace points this latency was easily visible before the first frame swap. With perf it can also be checked, when one does: QSG_RENDER_LOOP=threaded perf trace record -m 10M \ -e syscalls:sys_enter_nanosleep,syscalls:sys_enter_ioctl \ --call-graph dwarf qml main.qml ... [ perf record: Captured and wrote 116.731 MB perf.data (14309 samples) ] Then afterwards have a look at the output of perf script --comms QSGRenderThread And you will notice the sleep directly at the start, followed by the bulk of ioctl required to setup the OpenGL contexts: QSGRenderThread 10875 [001] 13940.801449: syscalls:sys_enter_nanosleep: rqtp: 0x7f1ab954cd60, rmtp: 0x7f1ab954cd60 7f1ada3997fd __nanosleep+0x2d (/usr/lib/libpthread-2.23.so) 7f1ada9683ed qt_nanosleep+0x2d (/home/milian/projects/compiled/qt5-5.7-opt/lib/libQt5Core.so.5.7.0) 7f1ada8c73b8 QThread::msleep+0x38 (/home/milian/projects/compiled/qt5-5.7-opt/lib/libQt5Core.so.5.7.0) 7f1abb594320 QSGRenderThread::syncAndRender+0x320 (/home/milian/projects/compiled/qt5-5.7-opt/lib/libQt5Quick.so.5.7.0) 7f1abb598b0c QSGRenderThread::run+0x19c (/home/milian/projects/compiled/qt5-5.7-opt/lib/libQt5Quick.so.5.7.0) 7f1ada8c7f89 QThreadPrivate::start+0x149 (/home/milian/projects/compiled/qt5-5.7-opt/lib/libQt5Core.so.5.7.0) 7f1ada390484 start_thread+0xc4 (/usr/lib/libpthread-2.23.so) 7f1ad982e6dd __clone+0x6d (/usr/lib/libc-2.23.so) ... ioctl follow This change here checks the validity of the render context before sleeping, thereby removing the vsync latency before rendering the first frame. Note that simply skipping the calls to `syncAndRender` from `run` is not an option, as those also trigger the expose event that is required for the initialization to happen eventually. [ChangeLog][QtQuick] The threaded scene graph renderer does not sleep up to one vsync interval before the first frame anymore. Change-Id: If7474d5420e0d4a1d05ccb664c7c6932fa989127 Reviewed-by: Gunnar Sletta <gunnar@sletta.org> Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
* | | Merge remote-tracking branch 'origin/dev' into scenegraphngAndy Nichols2016-05-311-1/+1
|\ \ \ | | | | | | | | | | | | Change-Id: I35715e07b6f837f21cd8e8898f19d97af92c6b69
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-09 00:07:02 +0000