platform/core/uifw/dali-core.git
4 days agoMerge "Cleaned up render pass creation comments" into devel/master
David Steele [Mon, 15 Jul 2024 13:20:00 +0000 (13:20 +0000)]
Merge "Cleaned up render pass creation comments" into devel/master

4 days ago(TestApplication) Fix SVACE issue 61/314561/1
Adeel Kazmi [Mon, 15 Jul 2024 09:47:33 +0000 (10:47 +0100)]
(TestApplication) Fix SVACE issue

Change-Id: I654a48b18ef90fb7c0a4737ac9dc273f4c3e661a

6 days agoMerge changes I03cb1f37,I66ce8530,Iad1ace55,Ib9eef3cf into devel/master
David Steele [Fri, 12 Jul 2024 15:44:43 +0000 (15:44 +0000)]
Merge changes I03cb1f37,I66ce8530,Iad1ace55,Ib9eef3cf into devel/master

* changes:
  Making test-render-surface.h/cpp common across repos
  Updating framebuffer to use renderpass
  Minor test harness updates
  Updating graphics-api for Vulkan

6 days agoCleaned up render pass creation comments 14/314514/1
David Steele [Tue, 9 Jul 2024 17:58:31 +0000 (18:58 +0100)]
Cleaned up render pass creation comments

Change-Id: I92a7a4c4beb236fd2caa9207c7726079374bc480

7 days agoDALi Version 2.3.32 95/314495/1 dali_2.3.32
David Steele [Fri, 12 Jul 2024 10:12:04 +0000 (11:12 +0100)]
DALi Version 2.3.32

Change-Id: If9a370530f2668add53306b58726d6cfb7c5f60a

8 days agoCalculate Has with std::string_view 20/314420/2
Eunki, Hong [Thu, 11 Jul 2024 10:26:56 +0000 (19:26 +0900)]
Calculate Has with std::string_view

Since convert from std::string_view to std::string
might requred memory copy.

To avoid this overhead, let we make a API that calculate
hash as string_view, instead of full string

Change-Id: Ie5455a5001e1c1f423023708a4988d53badff74c
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
11 days agoMaking test-render-surface.h/cpp common across repos 52/313552/4
David Steele [Thu, 27 Jun 2024 17:12:20 +0000 (18:12 +0100)]
Making test-render-surface.h/cpp common across repos

Change-Id: I03cb1f37ad9f51bc96b08bb31b6ee9d7d20cd083
Signed-off-by: David Steele <david.steele@samsung.com>
11 days agoUpdating framebuffer to use renderpass 80/313480/5
David Steele [Thu, 20 Jun 2024 17:21:04 +0000 (18:21 +0100)]
Updating framebuffer to use renderpass

Vulkan framebuffer requires a renderpass to be defined before
the framebuffer can be created. Currently in gles backend, this
order is irrelevant, and is not defined in the graphics api.

Added Graphics::RenderPass vector to the
Graphics::FramebufferCreateInfo - this vector may have more than one
compatible render pass in it. The first render pass will be used to
create the framebuffer, but BeginRenderPass may use any of the defined
render passes with that framebuffer (as long as they are compatible in
vulkan, i.e. only differ by layout or load/store ops).

Change-Id: I66ce8530a34bcb1da57461db5644771efe0af48c

11 days agoMinor test harness updates 70/313170/6
David Steele [Wed, 19 Jun 2024 09:07:58 +0000 (10:07 +0100)]
Minor test harness updates

Adding test-render-surface to dali-core

Change-Id: Iad1ace557fd4bab95a5e5cae5cf8281a9f4d2c64

11 days agoUpdating graphics-api for Vulkan 69/313169/5
David Steele [Fri, 14 Jun 2024 08:56:47 +0000 (09:56 +0100)]
Updating graphics-api for Vulkan

Change-Id: Ib9eef3cf2901b4fae094891d9ac82939c5be9d49

2 weeks agoDALi Version 2.3.31 19/314119/1 dali_2.3.31
Adam Bialogonski [Fri, 5 Jul 2024 09:08:01 +0000 (10:08 +0100)]
DALi Version 2.3.31

Change-Id: Ifa20f9c8ef8faea482e55bcfd24cf71f1460af05

2 weeks agoMerge "Print log if worker thread destruction occured for UI items." into devel/master
Adeel Kazmi [Thu, 4 Jul 2024 15:56:28 +0000 (15:56 +0000)]
Merge "Print log if worker thread destruction occured for UI items." into devel/master

2 weeks agoPrint log if worker thread destruction occured for UI items. 79/313979/7
Eunki, Hong [Wed, 3 Jul 2024 11:24:22 +0000 (20:24 +0900)]
Print log if worker thread destruction occured for UI items.

Change-Id: I15db9dd5d6763b403b4b5015c95e3491c18443dc
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 weeks agoMake Log::LogMessage print to stderr / stdout if log function not installed 48/314048/5
Eunki, Hong [Thu, 4 Jul 2024 10:16:18 +0000 (19:16 +0900)]
Make Log::LogMessage print to stderr / stdout if log function not installed

Until now, we just ignore the messages when we use Dali::Integration::Log::LogMessage(Something like, DALI_LOG_ERROR macro) when we don't install log function.

The log function exist per each threads.
But it is difficult to debug something at worker thread jobs
what DALi cannot controled. (Like GC thread for dotnet Application)

For more general and each debugging, let we print 'something' even if
log function is not installed, like stderr.

For now, Tizen platform re-open the stderr / stdout as dlog.
and Linux-friendly + Windows support to print logs into stderr, stdout file.

Change-Id: Iaf6a49d8aa9f53108b4c83bbdb18df406e3b57fa
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 weeks agoFix coverity issue : Uninitialized value at HitTestResult + PanGestureDetector SG... 87/313787/3
Eunki, Hong [Tue, 2 Jul 2024 03:35:07 +0000 (12:35 +0900)]
Fix coverity issue : Uninitialized value at HitTestResult + PanGestureDetector SG delete

Since PanGestureDetector use SG::PanGesture, which the memory ownership is on UpdateManager,
we need to take carefully to control the memory usage.

Change-Id: I7c2929c97b57e6e14072e0598bcd1788aa43af98
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 weeks agoMerge "Added element stride to the UniformInfo" into devel/master
David Steele [Fri, 28 Jun 2024 14:36:30 +0000 (14:36 +0000)]
Merge "Added element stride to the UniformInfo" into devel/master

3 weeks agoDALi Version 2.3.30 24/313624/1 dali_2.3.30
Richard Huang [Fri, 28 Jun 2024 10:10:21 +0000 (11:10 +0100)]
DALi Version 2.3.30

Change-Id: I4ac6554c928cb7a90cea3574308c40634476147d

3 weeks agoAdded element stride to the UniformInfo 03/313403/5
Adam Bialogonski [Tue, 25 Jun 2024 11:27:33 +0000 (12:27 +0100)]
Added element stride to the UniformInfo

Added element stride to the UniformInfo when uniform is an array.

Change-Id: Ib006d0ce50589b2ef5cf1a5bd1e6b3ab7fe9ec45

3 weeks agoMerge "Add TouchPropagationType enum. PARENT events are propagated to parents in...
joogab yun [Tue, 25 Jun 2024 07:15:28 +0000 (07:15 +0000)]
Merge "Add TouchPropagationType enum. PARENT events are propagated to parents in the traditional way. GEOMETRY is a new way in which events are propagated based on geometry." into devel/master

3 weeks agoAdd TouchPropagationType enum. 65/313365/3
joogab.yun [Tue, 25 Jun 2024 01:24:26 +0000 (10:24 +0900)]
Add TouchPropagationType enum.
PARENT events are propagated to parents in the traditional way.
GEOMETRY is a new way in which events are propagated based on geometry.

Change-Id: I3a717d5433e73f22e835ed3e6cd384a337160688

3 weeks agoMerge changes I576e6929,Ib8bae68c into devel/master
Eunki Hong [Mon, 24 Jun 2024 02:09:50 +0000 (02:09 +0000)]
Merge changes I576e6929,Ib8bae68c into devel/master

* changes:
  Mark dirty flag into SG::Renderer directly if visual property changed
  Remove static_cast when we try to get decorated visual properties

4 weeks agoDALi Version 2.3.29 84/313284/1 dali_2.3.29
David Steele [Fri, 21 Jun 2024 11:28:01 +0000 (12:28 +0100)]
DALi Version 2.3.29

Change-Id: I130c77353c8488d8d62b02bbbff076bd016acdb5

4 weeks agoMark dirty flag into SG::Renderer directly if visual property changed 05/313205/4
Eunki, Hong [Thu, 20 Jun 2024 05:29:47 +0000 (14:29 +0900)]
Mark dirty flag into SG::Renderer directly if visual property changed

Since age-down the visual property every frames make performance down,
let we update VisualRenderer's property send updated flag into SG::Renderer directly.

Now it will reduce some useless call of PrepareProperties of visual renderer.

Change-Id: I576e6929fca6945379b72988b4f3633e2ad70e22
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
4 weeks agoRemove static_cast when we try to get decorated visual properties 94/313194/3
Eunki, Hong [Thu, 20 Jun 2024 04:35:07 +0000 (13:35 +0900)]
Remove static_cast when we try to get decorated visual properties

Since we don't have any plan to add extra data for VisualRenderer now,
let we specify the class of Extension of VisualRenderer, instead of void*

Change-Id: Ib8bae68c58082c93fb9a7dbe673c589f3b523766
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
4 weeks agoMerge "If Hittable is false, actor should not receive events except INTERRUPTED"...
joogab yun [Thu, 20 Jun 2024 05:02:42 +0000 (05:02 +0000)]
Merge "If Hittable is false, actor should not receive events except INTERRUPTED" into devel/master

4 weeks agoMerge "For Frame Event" into devel/master
joogab yun [Thu, 20 Jun 2024 01:52:41 +0000 (01:52 +0000)]
Merge "For Frame Event" into devel/master

4 weeks agoIf Hittable is false, actor should not receive events except INTERRUPTED 75/313175/2
joogab.yun [Thu, 20 Jun 2024 01:02:48 +0000 (10:02 +0900)]
If Hittable is false, actor should not receive events except INTERRUPTED

Change-Id: I6f782f0e3021ea1d3bcc8a9ce630956bd2128fda

4 weeks agoCall OnBake(), OnSet only if the value is really changed 49/313149/1
Eunki, Hong [Wed, 19 Jun 2024 09:57:05 +0000 (18:57 +0900)]
Call OnBake(), OnSet only if the value is really changed

Let we check the current value and inputed value, so if they are same, ignore.

It will be useful when we don't want to mark dirty for animatable property doesn't changed.

For performance issue, let we just check only for bool / int / float.

TODO : Chould we check them also at Vector2,3,4?

Change-Id: I70b36eab97585a09c2c021da4f37a1d983a771af
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
4 weeks agoMerge "If USER_INTERACTION_ENABLED is false, actor should not receive events." into...
joogab yun [Tue, 18 Jun 2024 07:47:36 +0000 (07:47 +0000)]
Merge "If USER_INTERACTION_ENABLED is false, actor should not receive events." into devel/master

4 weeks agoMerge "A new gesture recognition method. 1. Add HandleEvent(actor, touch) api at...
joogab yun [Tue, 18 Jun 2024 07:33:01 +0000 (07:33 +0000)]
Merge "A new gesture recognition method. 1. Add HandleEvent(actor, touch) api at GestureDetector    This is an API that recognizes gestures by passing the actor for which the gesture should be recognized and the touch event to GestureDetector.    You can use it in the touch event callback. This should be used with SetGeometryHittestEnabled(true). 2. To support multi-gesture, we hit-test each multi-touch" into devel/master

4 weeks agoMerge "Reverts the previously implemented FeedTouch. The new Logic FeedTouch is comin...
joogab yun [Tue, 18 Jun 2024 07:32:41 +0000 (07:32 +0000)]
Merge "Reverts the previously implemented FeedTouch. The new Logic FeedTouch is coming soon." into devel/master

4 weeks agoFor Frame Event 04/311804/9
joogab.yun [Tue, 28 May 2024 07:12:24 +0000 (16:12 +0900)]
For Frame Event

When processing multiple multi-touches, they are grouped and processed. Add handledMultiTouch condition for grouping.

Change-Id: I9870485f03c0ef16f7ba7231f47d5bd3e8bc1f84

4 weeks agoMerge "Make Transform Component dirty as 2 frames, not for true/false" into devel...
Eunki Hong [Mon, 17 Jun 2024 05:52:25 +0000 (05:52 +0000)]
Merge "Make Transform Component dirty as 2 frames, not for true/false" into devel/master

5 weeks agoDALi Version 2.3.28 01/312801/1 dali_2.3.28
Adam Bialogonski [Fri, 14 Jun 2024 08:32:14 +0000 (09:32 +0100)]
DALi Version 2.3.28

Change-Id: I6208c856efbdcce2f56dc53d8745062c3c012f94

5 weeks agoIf USER_INTERACTION_ENABLED is false, actor should not receive events. 18/312718/1
joogab.yun [Thu, 13 Jun 2024 09:07:11 +0000 (18:07 +0900)]
If USER_INTERACTION_ENABLED is false, actor should not receive events.

Change-Id: I21ca9b8587dfecf7412d6dff736f0bbcbc9b8bed

5 weeks agoRemove old pan gesture history when guesture started. 89/312689/1 devel/async_text
Eunki, Hong [Thu, 13 Jun 2024 05:02:29 +0000 (14:02 +0900)]
Remove old pan gesture history when guesture started.

Change-Id: I2cec5ea23b801c5dc6e7ac2e0fd0da0c780c42c2
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
5 weeks agoA new gesture recognition method. 69/309669/84
joogab.yun [Mon, 15 Apr 2024 07:02:41 +0000 (16:02 +0900)]
A new gesture recognition method.
1. Add HandleEvent(actor, touch) api at GestureDetector
   This is an API that recognizes gestures by passing the actor for which the gesture should be recognized and the touch event to GestureDetector.
   You can use it in the touch event callback. This should be used with SetGeometryHittestEnabled(true).
2. To support multi-gesture, we hit-test each multi-touch

Change-Id: I6ea2d42fe54d1ebd7ce8a1844f91434765f9b478

5 weeks agoFix partial update issue after VisualRenderer patch applied 47/312647/1
Eunki, Hong [Wed, 12 Jun 2024 13:18:42 +0000 (22:18 +0900)]
Fix partial update issue after VisualRenderer patch applied

Change-Id: I0df716d6a4607ada01119c67f77f127a70566ab2
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
5 weeks agoMake Transform Component dirty as 2 frames, not for true/false 15/312615/4
Eunki, Hong [Wed, 12 Jun 2024 09:11:58 +0000 (18:11 +0900)]
Make Transform Component dirty as 2 frames, not for true/false

Since ResetToBaseValue for TransfromProperty doesn't use basic
AnimatableProperty logic, the dirty flag is not match with common sence.

To match it as frame-by-frame, let we make component dirty as 2 bit,
and age-down every frame.

And after we allow to use dirty flag feature for transform,
we can determine whether we need to re-calculate matrix or not.

If that flag be used, we can 'skip' heavy matrix multiply operation.

Change-Id: Icf51e0be37962bee5147c08bb0cd3302bf203f89
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
5 weeks agoReverts the previously implemented FeedTouch. The new Logic FeedTouch is coming soon. 63/309663/16
joogab.yun [Mon, 15 Apr 2024 05:52:37 +0000 (14:52 +0900)]
Reverts the previously implemented FeedTouch. The new Logic FeedTouch is coming soon.

Revert "1. Fixes a bug where events do not occur when longPressGestures overlap. This is when using gestures with FeedTouch."

This reverts commit b0516efeefaff68139bee29a7317eccd3729adbb.

Revert "actorCoordinates were missing from TapGesture when FeedTouch was used"

This reverts commit 9e32a10a253c53d66ec41fe915d6d378bf2d8aa4.

Revert "Touch and Hover event propagrated by geometry way.(5)"

This reverts commit c1ead670ee0d10718449be889b6bdacf98d93a60.

Revert "Add FeedTouch api to GestureDetector."

This reverts commit 54e52fd9d4556e19b2e97082b057ad886952e55e.

Change-Id: Iffe838047df670006ad23f7c5d46eb7f938cd16c

5 weeks agoMerge "Fix for GCC14" into devel/master
Seungho BAEK [Wed, 12 Jun 2024 04:56:04 +0000 (04:56 +0000)]
Merge "Fix for GCC14" into devel/master

5 weeks agoCall ResetToBaseValue only 1 times per PropertyBase 60/312360/4
Eunki Hong [Fri, 7 Jun 2024 08:26:06 +0000 (17:26 +0900)]
Call ResetToBaseValue only 1 times per PropertyBase

If multiple animation try to change single property, the dirty flag
of PropertyBase would not be works well.

To ensure the dirty flag, let we collect resetter required properties
into single container, and call ResetToBaseValue only one times.

It will ensure that we call the resetter only 1 times per 1 frame.

Change-Id: I1a3792993f2ef681399f72f6f1718731408f2128
Signed-off-by: Eunki Hong <eunkiki.hong@samsung.com>
5 weeks agoFix for GCC14 62/312462/2
Seungho Baek [Mon, 10 Jun 2024 09:06:51 +0000 (18:06 +0900)]
Fix for GCC14

Change-Id: I3a0528f8c6752e2f042a5cf6bfa11b66d875b22a
Signed-off-by: Seungho Baek <sbsh.baek@samsung.com>
5 weeks agoFix Animation with EndAction::DISCARD dont reset properties. 43/312143/7
Eunki, Hong [Tue, 4 Jun 2024 12:38:38 +0000 (21:38 +0900)]
Fix Animation with EndAction::DISCARD dont reset properties.

If Animation::EndAction::DISCARD finisehd normally cases,
The mDirtyFlag was not matched with real world

For example

ResetToBaseValue[0](mDirtyFlag become 1)
Animate and finished (mValue[0] changed.)
(update)
ResetToBaseValue[1](mDirtyFlag become 0)
(update)
(ResetToBaseValue did not called. So mValue[0] is last frame value)
(update)
(ResetToBaseValue did not called. So mValue[1] is BaseValue)

Now, mValue become flickering.

To avoid this problem, let we call ResetToBaseValue at least 2 frames
if finished animation's EndAction is DISCARD.
(Note that we don't consider Stop() call cases, since Stop() will not Animate
Animator, so mValue[0] is BaseValue)

And also, There was some issue that visual renderer property changeness not updated to
the dirty rect infomation.
(Since SG::Renderer don't be mark as updated)

To fix this issue, let we ensure to check the visual renderer property dirty.

And also, for apply Animation::EndAction::DISCARD case,
Let we make visual renderer coefficient use double buffered flags,
and age down every frames.

TODO : UpdateManager need to ResetBaseValue at least 2 frames
if Finished animation is EndAction::DISCARD.

This will need update manager side fix. So just keep this bug and fix as another patch.

Change-Id: Ibf654f723e1f986843cda620bc741b1121ee95d7
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
5 weeks agoMake InternalState for animation. So ensure Stop() Play() make the playstate is Playing 79/312079/4
Eunki, Hong [Tue, 4 Jun 2024 02:26:51 +0000 (11:26 +0900)]
Make InternalState for animation. So ensure Stop() Play() make the playstate is Playing

Since there is no way to determine that Animation::HasFinished() function called
due to we call Stop() API, or Animation finished normally.

To seperate the state, let we make new internal state
: STOPPING, PLAYING_DURING_STOPPING, PAUSED_DURING_STOPPING

STOPPING will be changed as STOPPED after Animation::HasFinished() function called.

And also, make CLEARED state so we can easly control Clear() function call.

Change-Id: I2acccfaaa034dd1edc956e4a23b0726b5d1b65e7
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
5 weeks agoDALi Version 2.3.27 79/312379/1 dali_2.3.27
Adeel Kazmi [Fri, 7 Jun 2024 17:25:58 +0000 (18:25 +0100)]
DALi Version 2.3.27

Change-Id: Icdbaecd822027f41547003ee98b716227d07306d

6 weeks agoEnsure cleared animation not emit finished signal during finished signal emit 21/312021/1
Eunki, Hong [Mon, 3 Jun 2024 01:51:16 +0000 (10:51 +0900)]
Ensure cleared animation not emit finished signal during finished signal emit

Since we collect all finished animations and emit after,
It is possible that finisehd signal emitted what we should not
(e.g. Animation::Clear() called during another animation's finished signal)

To avoid it, let we ensure to check whether given animation is ignored or not.

Change-Id: I681c2c8e0f1624e0846409eb862862e5392ab7ec
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
6 weeks agoMerge "Ignore AnimationFinishedCallback when we call Animation::Clear() hard + Resolv...
Eunki Hong [Mon, 3 Jun 2024 01:35:17 +0000 (01:35 +0000)]
Merge "Ignore AnimationFinishedCallback when we call Animation::Clear() hard + Resolve bug when we call Play multiple times" into devel/master

7 weeks agoDALi Version 2.3.26 86/311986/1 dali_2.3.26
Adeel Kazmi [Fri, 31 May 2024 10:45:04 +0000 (11:45 +0100)]
DALi Version 2.3.26

Change-Id: I8aa7325ed7b1a0ea7980c47e6c9f062db0c3060b

7 weeks agoIgnore AnimationFinishedCallback when we call Animation::Clear() hard + Resolve bug... 25/311925/3
Eunki, Hong [Thu, 30 May 2024 11:18:50 +0000 (20:18 +0900)]
Ignore AnimationFinishedCallback when we call Animation::Clear() hard + Resolve bug when we call Play multiple times

There was several bugs at AnimationPlayList.

1. Core::ProcessCoreEvents sequence is like this :
     SceneEvent -> NotificationManager -> Relayout -> Flush.
   If we call Animation::Clear() at SceneEvent timing, Notfication will send animation finished signal
   what we must not send.

   To avoid this system, let we store the cleared animations at AnimationPlayList.

   Since AnimationFinished callback required at least 1 frame times after flush,
   we can ignore whole notify events before Flush done.

2. Scene mPlayList keep the reference of Animation whenever we call Play, there
   were several problems. For example, If we call Play() 2 times and Clear(),
   that Dali::Animation never be removed.

Change-Id: Ie9d5298311b9e5318fecad21b1124c896fcf3a8e
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
7 weeks agoRemove render once trace log 91/311891/1
Eunki, Hong [Thu, 30 May 2024 04:18:09 +0000 (13:18 +0900)]
Remove render once trace log

Since it print too much logs, let we remove it

Change-Id: I8987d8fe6b184938780f268f0349702456dd0433
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
7 weeks agoMerge "Add FittingMode for visual fittingMode" into devel/master
sunghyun kim [Wed, 29 May 2024 03:59:11 +0000 (03:59 +0000)]
Merge "Add FittingMode for visual fittingMode" into devel/master

7 weeks agoMerge "Apply RenderUntil() on sorted RenderList" into devel/master
Jeongmin Lee [Tue, 28 May 2024 14:32:29 +0000 (14:32 +0000)]
Merge "Apply RenderUntil() on sorted RenderList" into devel/master

7 weeks agoApply RenderUntil() on sorted RenderList 66/311366/14
jmm [Fri, 19 Apr 2024 11:55:52 +0000 (20:55 +0900)]
Apply RenderUntil() on sorted RenderList

Change-Id: I64a3b567984494d1f284095d68e44b7e318a04d5

7 weeks agoImplement ProcessorOnce 95/311795/2
Eunki, Hong [Tue, 28 May 2024 04:06:18 +0000 (13:06 +0900)]
Implement ProcessorOnce

Since unregister processor during Process will give overhead,
we'd better make another container to process only once time
during ProcessCoreEvent.

Change-Id: I0b38490ccdbd7da4f7ec1acf80253a1aee8c42ad
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
7 weeks agoOptimize Add/Remove SG::RenderTask from SG::RenderTaskList 04/308204/3
Eunki Hong [Tue, 19 Mar 2024 14:28:22 +0000 (23:28 +0900)]
Optimize Add/Remove SG::RenderTask from SG::RenderTaskList

Let we make render task list use OrderedSet instead of vector container.

Vector container is slow if we call RemoveRenderTask.

TODO : Optimize event side render task list too.
But since it is quite complex, we'd better optimize that points
in another patch

Change-Id: Icd6f2e7a24964eb05c92fa9a5ce0e306814445ee
Signed-off-by: Eunki Hong <eunkiki.hong@samsung.com>
8 weeks agoDALi Version 2.3.25 18/311718/1 dali_2.3.25
Richard Huang [Fri, 24 May 2024 11:32:58 +0000 (12:32 +0100)]
DALi Version 2.3.25

Change-Id: Ib32a2c7cb1d61bff3955047cb03be7142d14693f

8 weeks agoControl core policy as one flag 71/311671/1
Eunki, Hong [Thu, 23 May 2024 08:27:59 +0000 (17:27 +0900)]
Control core policy as one flag

Let we collect core creation status, instead of parameter increasement.

It will be useful when we add more features of Core.

Change-Id: I746bdb714eac5dcbd792a1a33a93abd2b96fe4cb
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
8 weeks agoPerformance trace for some bottleneck of update/render manager 48/311648/3
Eunki, Hong [Thu, 23 May 2024 05:13:06 +0000 (14:13 +0900)]
Performance trace for some bottleneck of update/render manager

Let we check trace the top 5 area of update : update itself / animation /
frame update callback / update renderer / transform
and top 2 area of render : render instruction process / flush

TODO : Since the name of environment need to be sync with peformance
trace tool, we'd better change it after the system prepared.

Change-Id: I4a9e38b6b957508138737ee225c0cfa9ad94a8f9
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
8 weeks agoMerge changes Ib3f51bd8,Ibfbeca64 into devel/master
Eunki Hong [Thu, 23 May 2024 06:20:51 +0000 (06:20 +0000)]
Merge changes Ib3f51bd8,Ibfbeca64 into devel/master

* changes:
  Pint log if animation play / stop / pause / clear / destroy
  Print top5 longest time spend AnimationFinished & FrameUpdate callbacks

8 weeks agoPint log if animation play / stop / pause / clear / destroy 46/311646/1
Eunki, Hong [Thu, 23 May 2024 04:53:40 +0000 (13:53 +0900)]
Pint log if animation play / stop / pause / clear / destroy

Let we know the animation's state by id, if user want to know.

Change-Id: Ib3f51bd8035ae6958f37f2340b1dbbe10cfaf029
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
8 weeks agoPrint top5 longest time spend AnimationFinished & FrameUpdate callbacks 43/311643/1
Eunki, Hong [Tue, 20 Feb 2024 04:55:34 +0000 (13:55 +0900)]
Print top5 longest time spend AnimationFinished & FrameUpdate callbacks

Very rarely, we got some report from performance team that
"DALI_NOTIFICATION_NOTIFY_COMPLETED" and "DALI_FRAME_CALLBACK_UPDATE" is bottleneck.

To escape the false accusation, let we collect the App side callback
execute time and print it.

This time checker already merged at tizen_8.0 branch.

Change-Id: Ibfbeca6440f1b9c46ebb2f71200139de611a2b4a
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
8 weeks agoEven when multi-touching, the captured actor is used without a hittest every time. 87/311487/5
joogab.yun [Wed, 22 May 2024 05:07:38 +0000 (14:07 +0900)]
Even when multi-touching, the captured actor is used without a hittest every time.

Change-Id: Ib934b4b5626747a9ed7036d7489f82c94e7b4b16

8 weeks agoMerge "Remove an unneccessary argument at Process()" into devel/master
Jeongmin Lee [Tue, 21 May 2024 04:49:53 +0000 (04:49 +0000)]
Merge "Remove an unneccessary argument at Process()" into devel/master

8 weeks agoRemove an unneccessary argument at Process() 70/311370/1
jmm [Mon, 20 May 2024 07:05:36 +0000 (16:05 +0900)]
Remove an unneccessary argument at Process()

Change-Id: I62c8ab7fe11b28c67fa3fe5e277142d969af8698

8 weeks agoAdd FittingMode for visual fittingMode 29/311429/1
sunghyun kim [Mon, 20 May 2024 06:15:05 +0000 (15:15 +0900)]
Add FittingMode for visual fittingMode

Add new type for visual fittingMode

Change-Id: I17cb51564117a1c0426052dac547d0beb944e1c2

2 months agoDALi Version 2.3.24 02/311302/1 dali_2.3.24
David Steele [Fri, 17 May 2024 11:10:57 +0000 (12:10 +0100)]
DALi Version 2.3.24

Change-Id: Ifbf8d83b68753dea68479978185e75da8fb81b49

2 months agoMerge "(Partial Update) Refactorize PartialRenderingData check logic" into devel...
Eunki Hong [Tue, 14 May 2024 05:38:24 +0000 (05:38 +0000)]
Merge "(Partial Update) Refactorize PartialRenderingData check logic" into devel/master

2 months agoMerge "Add compile time option to memory management for low spec devices (Dali::Vecto...
David Steele [Mon, 13 May 2024 16:45:22 +0000 (16:45 +0000)]
Merge "Add compile time option to memory management for low spec devices (Dali::Vector, OrderedSet and memory pool)" into devel/master

2 months agoMerge "Use some observer list use OrderedSet instead of Dali::Vector" into devel...
David Steele [Mon, 13 May 2024 16:10:36 +0000 (16:10 +0000)]
Merge "Use some observer list use OrderedSet instead of Dali::Vector" into devel/master

2 months agoMerge "Assert to add / remove observer during notifying" into devel/master
David Steele [Mon, 13 May 2024 16:07:58 +0000 (16:07 +0000)]
Merge "Assert to add / remove observer during notifying" into devel/master

2 months ago(Partial Update) Refactorize PartialRenderingData check logic 00/310900/5
Eunki, Hong [Thu, 9 May 2024 07:51:46 +0000 (16:51 +0900)]
(Partial Update) Refactorize PartialRenderingData check logic

Since we check node's partial update data only 1 times and then update directly
when we add RenderItem.

Unfortunatly, VisualRenderer has there own update area calculation logic
(like offset and size) and Node can add multiple renderer,
The partial update data said that it is not updated for the 'second' Renderer.

So it was possible that second renderer's update area could not be applied
when we check dirty rects.

===

To fix this issue, we reset the partial update data flags for all nodes every begin of
update. And keep + calculate the partial data only 1 times.

If we check the updateness at this frame, keep this infomation and reuse it.
It will avoid useless Hash calculation.

Change-Id: Ib01712b56131b5813c8629c013f60223ed850577
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 months agoDALi Version 2.3.23 69/310969/1 dali_2.3.23
Adeel Kazmi [Fri, 10 May 2024 10:01:56 +0000 (11:01 +0100)]
DALi Version 2.3.23

Change-Id: Ie50bf0ceb4c85aafd71a46b2497e30e419e99d33

2 months agoAdd compile time option to memory management for low spec devices (Dali::Vector,... 69/308569/20
Eunki Hong [Wed, 27 Mar 2024 12:20:24 +0000 (21:20 +0900)]
Add compile time option to memory management for low spec devices (Dali::Vector, OrderedSet and memory pool)

Let we give some option to reduce memory capacity for low spec devices.

If app create a lots of Actors and release all of them,
the memory might not be reduced, since we don't release capacity of continous array.

Most of memory bottleneck occured at Transform::Data and Node memory pool, and
Graphics::CommandBuffer.

CommandBuffer pool control at dali-adaptor. So let we control what dali-core
can reduce capacities first.

Note : 64bit OS will always use memory pool feature even if we give compile option
, due to KeyType is uint32_t.

USAGE :

 * gbs build -A armv7l --include-all --define "%enable_low_spec_memory_management 1"

 * cmake -DCMAKE_INSTALL_PREFIX=$DESKTOP_PREFIX -DENABLE_LOW_SPEC_MEMORY_MANAGEMENT=ON

Change-Id: Ifeb2e81a5e56fd31b7a61e55dc311f1205b8dfa8
Signed-off-by: Eunki Hong <eunkiki.hong@samsung.com>
2 months agoUse some observer list use OrderedSet instead of Dali::Vector 72/309872/5
Eunki, Hong [Wed, 17 Apr 2024 12:40:40 +0000 (21:40 +0900)]
Use some observer list use OrderedSet instead of Dali::Vector

Let we use Integration::OrderedSet instead of Dali::Vector.

Integration::OrderedSet has good point if we try to remove the items.

For BaseObject::Observer case, it will be called only 1 times per
each object's lifetime. So we have low priority of iteration overhead.

Change-Id: I227e8ecb2021448d892fcaeae78b814d52e09739
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 months agoAssert to add / remove observer during notifying 26/310026/7
Eunki, Hong [Fri, 19 Apr 2024 09:25:16 +0000 (18:25 +0900)]
Assert to add / remove observer during notifying

Let we assert the cases when observer list try to be changed during notifying.

But, allow to remove observer during notifying, only for Object cases.

Until now, we just change the observer list directly, so if notifying, then
memory corruption occured.

To avoid this case, follow as ProcessorInterface did, copy the observer lists
and check whether it is valid or not, only if somebody touch original list.

Note : For BaseObject::Impl case, it is not opened
and it didn't send notify when scene on/off.
The only observer for BaseObject::Impl is WeakHandle. and OnDestroy() didnt'
send callback that timing.

Note2 : PropertyOwner case, it is also only for update-render thread for DALi.
We can control that non of callback change the observers for this thing.

So let we assume BaseObject::Impl::RemoveObserver and
PropertyOwner::RemoveObserver will not be called
during desturct object.

TODO : Should we also control Object::RemoveObserver case?
TODO : Should we remove observer copy?

Change-Id: Ia373243bdaae83613b1b91680a77b86409977028
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 months agoMerge "Fix several hit issue for offscreen rendering" into devel/master
Seungho BAEK [Tue, 7 May 2024 02:48:19 +0000 (02:48 +0000)]
Merge "Fix several hit issue for offscreen rendering" into devel/master

2 months agoSupport for recognizing legacy shaders. 79/310179/10
Adam Bialogonski [Fri, 3 May 2024 13:57:38 +0000 (14:57 +0100)]
Support for recognizing legacy shaders.

Legacy shaders should be bypassed through the shader processing
pipeline as they are.

Change-Id: I04b2d15c5e95edaaf78ddff3e5c3ca99096fcc2e

2 months agoMerge "DALi Version 2.3.22" into devel/master
Adeel Kazmi [Fri, 3 May 2024 09:07:47 +0000 (09:07 +0000)]
Merge "DALi Version 2.3.22" into devel/master

2 months agoDALi Version 2.3.22 41/310641/1 dali_2.3.22
Adam Bialogonski [Fri, 3 May 2024 08:26:37 +0000 (09:26 +0100)]
DALi Version 2.3.22

Change-Id: I2cf0620aa085c2c9de12bc47519570164f6d2e19

2 months agoFix several hit issue for offscreen rendering 58/310258/6
Seungho Baek [Wed, 24 Apr 2024 12:21:39 +0000 (21:21 +0900)]
Fix several hit issue for offscreen rendering

 - Previous implementation
   - MappingActor is not hittable by default, but it was used to check
     whether it is hitted from OnScreen RenderTask or not in HitTestRenderTaskList.
   - There is nothing to check the layer including MappingActor is consuming hit.

 - Current implementation
   - For the OnScreen hit result, check whether there is a mappingActor of OffScreen
     hit results that can be hit in front of the OnScreen hit result.
     If it is, returns the OffScreen hit results.
   - If the OnScreen hit result is layer and the layer consumes hit(the layer must
     not be hittable), returns the OffScreen hit results.
   - If there is no hit in OnScreen but there are hit results from OffScreen
     RenderTask, returns the top OffScreen hit results.

Change-Id: I2788ed90dbe0145b4c263c371353b201c65c2a80
Signed-off-by: Seungho Baek <sbsh.baek@samsung.com>
2 months agoPrint logs if dali skip rendering 83/310583/3
Eunki, Hong [Thu, 2 May 2024 06:58:19 +0000 (15:58 +0900)]
Print logs if dali skip rendering

Change-Id: I9ddca01a172b742c60b4d01c3d997306b0710d5f
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 months agoDo not allow to add duplicated renderer into actor 06/310606/1
Eunki, Hong [Thu, 2 May 2024 13:19:52 +0000 (22:19 +0900)]
Do not allow to add duplicated renderer into actor

Internal::Actor allow to add same renderer, but SceneGraph::Node doesn't allow.

If we do something like this, the logic might be breakdown.

actor.AddRenderer(renderer);
actor.AddRenderer(renderer);
actor.RemoveRenderer(renderer);

///< actor will have renderer, but actor's node doesn't have renderer.

To match with node's logic, let we don't allow to push-back the duplicated renderer.
Instead, just check the duplication and return the index.

Change-Id: I689896171b3df5d140089c3b99f44e1d3f0f8722
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
2 months agoAdd RenderTask::RenderUntil() 44/310044/10
jmm [Fri, 19 Apr 2024 11:52:19 +0000 (20:52 +0900)]
Add RenderTask::RenderUntil()

Change-Id: I264e29303deedaf0f8539137b26e7d0c3e8454ce

2 months agoAdd Inherited Visibility Changed signal 84/310084/12
Seungho Baek [Mon, 22 Apr 2024 07:43:13 +0000 (16:43 +0900)]
Add Inherited Visibility Changed signal

Change-Id: I1d592e6a5c9f17b77b8ee9f8828586f424985699
Signed-off-by: Seungho Baek <sbsh.baek@samsung.com>
2 months agoMerge "DALi Version 2.3.21" into devel/master
David Steele [Fri, 26 Apr 2024 09:48:47 +0000 (09:48 +0000)]
Merge "DALi Version 2.3.21" into devel/master

2 months agoDALi Version 2.3.21 91/310391/1 dali_2.3.21
Richard Huang [Fri, 26 Apr 2024 09:42:44 +0000 (10:42 +0100)]
DALi Version 2.3.21

Change-Id: I9ebe010c86446d8cae0e243144673306e0235ac9

2 months agoMerge "Add log in EmitSingleTap" into devel/master
joogab yun [Fri, 26 Apr 2024 09:02:07 +0000 (09:02 +0000)]
Merge "Add log in EmitSingleTap" into devel/master

2 months agoMerge "Allows updating MinimumDistance and MinimumPanEvents of pan gestures." into...
joogab yun [Fri, 26 Apr 2024 09:02:01 +0000 (09:02 +0000)]
Merge "Allows updating MinimumDistance and MinimumPanEvents of pan gestures." into devel/master

2 months agoAdd VIRTUAL_REMOCON and VIRTUAL_MOUSE device types 16/310216/2
joogab.yun [Wed, 24 Apr 2024 06:50:45 +0000 (15:50 +0900)]
Add VIRTUAL_REMOCON and VIRTUAL_MOUSE device types

Change-Id: I67f3ae35d364253b22a1d218df6c8c1bda848e97

2 months agoAdd log in EmitSingleTap 46/310146/2
joogab.yun [Tue, 23 Apr 2024 06:18:59 +0000 (15:18 +0900)]
Add log in EmitSingleTap

Change-Id: Ic5c4831daa355195c272c298d478b7d1519a20b4

2 months agoAllows updating MinimumDistance and MinimumPanEvents of pan gestures. 39/310139/3
joogab.yun [Tue, 23 Apr 2024 05:45:38 +0000 (14:45 +0900)]
Allows updating MinimumDistance and MinimumPanEvents of pan gestures.

Change-Id: Ibe985928ae9e68f70fc4faf573284ce1ee956fb5

2 months agoMerge "Minor reduce IntrusivePtr<Actor>'s meanless Reference and Unreference" into...
Eunki Hong [Sat, 20 Apr 2024 07:40:47 +0000 (07:40 +0000)]
Merge "Minor reduce IntrusivePtr<Actor>'s meanless Reference and Unreference" into devel/master

2 months agoMerge "Move OrderedSet as integration-api, so dali-toolkit can use it." into devel...
Eunki Hong [Fri, 19 Apr 2024 12:43:16 +0000 (12:43 +0000)]
Merge "Move OrderedSet as integration-api, so dali-toolkit can use it." into devel/master

2 months agoDALi Version 2.3.20 27/310027/1 dali_2.3.20
David Steele [Fri, 19 Apr 2024 09:47:06 +0000 (10:47 +0100)]
DALi Version 2.3.20

Change-Id: Ib80a171ab80256a0a5cdb5e18484c79fdbbe138e

3 months agoMinor reduce IntrusivePtr<Actor>'s meanless Reference and Unreference 13/310013/2
Eunki, Hong [Fri, 19 Apr 2024 06:03:11 +0000 (15:03 +0900)]
Minor reduce IntrusivePtr<Actor>'s meanless Reference and Unreference

During Actor.Remove() API call, we try to increase whole child actor's
reference and unreference again while iteration.

It might give "minor" overhead. So let we don't do it.

Change-Id: If572a06ce72eae2dd2ce2bf4b1b4db29f69038a0
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
3 months agoMove OrderedSet as integration-api, so dali-toolkit can use it. 44/309944/1
Eunki, Hong [Thu, 18 Apr 2024 07:00:56 +0000 (16:00 +0900)]
Move OrderedSet as integration-api, so dali-toolkit can use it.

Let we move the OrderedSet class as integration-api so dali-toolkit can use it.

Change-Id: I2dbbff14db048d52c5a02bc967b379defcd2c03e
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
3 months agoDo not call RemoveObserver when we don't add it for PropertyResetter 98/309898/2
Eunki, Hong [Thu, 18 Apr 2024 03:01:01 +0000 (12:01 +0900)]
Do not call RemoveObserver when we don't add it for PropertyResetter

It is possible that PropertyResetter destructed before Initialize()
when app terminate case.

If then, it is possible to got some crash when we try to access to
'destroyed' property owner, and call RemoveObserver();

To avoid this case, let we make ensure that we call initalized so
it is possible to call RemoveObserver().

Change-Id: I24b74f1d963090d867e1d047fe739132f9465ee3
Signed-off-by: Eunki, Hong <eunkiki.hong@samsung.com>
3 months agoMerge "Seperate KeepRendering reason more detail" into devel/master
David Steele [Mon, 15 Apr 2024 10:49:51 +0000 (10:49 +0000)]
Merge "Seperate KeepRendering reason more detail" into devel/master

3 months agoDALi Version 2.3.19 89/309589/1 dali_2.3.19
Richard Huang [Fri, 12 Apr 2024 09:52:03 +0000 (10:52 +0100)]
DALi Version 2.3.19

Change-Id: I569d4803c7ca83bc4f955b35545665f3ec04c4e6