summaryrefslogtreecommitdiffstats
path: root/libs/hwui/PathCache.cpp
Commit message (Collapse)AuthorAgeFilesLines
* Fix HWUI Path Cache dangling pointerDigish Pandya2015-11-111-0/+7
| | | | | | | | | | | | | When precache, PathTexture is added to PathCache, and it is released after drawn if we want to clean it. But the PathCache LRU still holds the entry of the PathTexture object. When trim the cache in the end of each frame, LRU finds that its mListener is not NULL and invoke the functor, however, mListerer points to the released PathTexture object and is a dangling pointer, thus leads to crash. Smart pointer don't help here since they only manage scopes, while PathTexture is also controled by its cleanup field. The fix is to also remove the LRU entry of PathTexture*, it will also release the texture object and there won't be texture leaks. Change-Id: Iaa0621df5dc71532e9e75b38ad94384353930b95
* Remove all usage of fmin and fmaxChris Craik2015-07-071-3/+3
| | | | | | | | bug:22208220 Removes needless call, and upconversion to doubles in multiple places. Change-Id: I1b949fa5f206446ac34de800154c0147d6bd8034
* Add tracing for path texture uploadChris Craik2015-05-131-0/+1
| | | | Change-Id: I4918e5a3c7e01b321d428b20babd48bf69a708e4
* Remove usage of Texture::cleanup in PathCacheChris Craik2015-05-111-25/+12
| | | | | | | | bug:21039605 Was causing double-deletes, is unneeded. Change-Id: I5c0a21d076fae7039d777ca09b21492f97fea98b
* Cleanup propertiesChris Craik2015-05-051-1/+1
| | | | | | | | | | | | bug:19967854 Separate properties from Caches, into static, RenderThread-only class. Also rewrites the means for java to set properties to correctly handle threading, and adds an override for profile bars so that SysUi doesn't clutter the screen with them. Change-Id: I6e21a96065f52b9ecc49d1a126244804ba106fa9
* Simplify TaskManager fallback pathChris Craik2015-04-201-3/+1
| | | | Change-Id: Ie3833449f7fe6aa69b9f71eb8bab1bef0fe434aa
* Use glops for text renderingChris Craik2015-03-181-30/+19
| | | | Change-Id: I5e155c8baf3149f0ff231ec3c89dbff6bb8eae92
* Refactor DisplayList path caching.Derek Sollenberger2015-03-131-48/+13
| | | | | | | | | | | This removes dependence on SkPath ptrs that HWUI does not control the lifecycle of. This clears up some errors where the paths are not generated from Java, but rather the Skia test suites. Cherry-pick of a change that originally landed in master-skia and is dependent on a skia merge (ag/655422). Change-Id: I41b9797a2b0af5d6b4ea51891565469d4f1d832d
* resolved conflicts for merge of c7e6b93a to masterJohn Reck2015-02-041-1/+3
|\ | | | | | | Change-Id: I4b791f37193727a0306214645f2f98cdf218cad0
| * am af6eaf65: Merge "Fix ANR caused by hwuiTask thread" automerge: 5ad36d4John Reck2015-02-041-1/+3
| |\ | | | | | | | | | | | | * commit 'af6eaf656ec0de5fcbea03968daffe2c845e2089': Fix ANR caused by hwuiTask thread
| | * Fix ANR caused by hwuiTask threadSangkyu Lee2015-01-121-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If hwuiTask thread is exited while HWUI renders something, some tasks can remain unfinished forever. This can make ANR problem if RenderThread waits this kind of tasks. According to the current implementation, hwuiTask threads are exited when HWUI receives trimMemory() callback with level >= 20 and some applications such as SystemUI can receive trimMemory() with level >= 20 even though they renders something yet. (For instance, when RecentsActivity in SystemUI is finished, HWUI receives trimMemory() callback with level >= 20 but SystemUI should still render the status bar and navigation bar.) This patch prevents the tasks from remaining unfinished and make the tasks executed immediately if they cannot be added to their TaskProcessors. Change-Id: I5bd26439aa5f183b1a7c1ce466362e27554b4d16
| | * Revert "Frameworks/base: Unused parameters in hwui"John Reck2014-11-221-1/+1
| | | | | | | | | | | | | | | | | | This reverts commit 42ddc18d108f789705ad4eb697ce9599ad322507. Change-Id: I3574a936a39a96314db6437d0a1eb58d260d893d
| * | Fix styleJohn Reck2014-12-121-1/+1
| | | | | | | | | | | | Change-Id: I7227b0eac126bf470ed50249a7809b845872983b
* | | Refactor blending and texture gl stateChris Craik2015-01-301-2/+2
| | | | | | | | | | | | Change-Id: Ia6b3c8b2afd3dfcee7f3ce401d846b789612054a
* | | Move more GL state management to RenderState and its directoryChris Craik2015-01-281-1/+0
| | | | | | | | | | | | Change-Id: Ic68584e1c08dc64be2ad43450cb6caa1de834fdc
* | | Add overrides and switch to nullptr keyword for all filesChris Craik2015-01-051-12/+12
| | | | | | | | | | | | | | | | | | | | | Adds remaining missing overrides and nullptr usages, missed due to an extreme failure in tool usage. Change-Id: I56abd72975a3999ad13330003c348db40f59aebf
* | | Revert "resolved conflicts for merge of 220c3f4f to master"Andreas Gampe2014-11-221-1/+1
|/ / | | | | | | | | | | | | | | Reverted as hwui doesn't agree. This reverts commit 8a902d9f24e83c87b054adb5836b4a5b8a257be9. Change-Id: I109e7b02bee2921e2155ded6df36f52e6f574b5a
* | resolved conflicts for merge of 99377df1 to lmp-mr1-dev-plus-aospAndreas Gampe2014-11-211-1/+1
|\ \ | |/ | | | | Change-Id: I3a98f55832ac447b1ed0dd129c7a93d088025943
| * Frameworks/base: Unused parameters in hwuiAndreas Gampe2014-11-211-1/+1
| | | | | | | | | | | | | | | | | | Remove Clang cutout for unused parameters. Fix warnings. Remove Clang cutout for deprecated Skia function usage. Has been fixed in the L push. Change-Id: I7ea073ff67127cc1e14e798b655e2c50615fe8e7
* | resolve merge conflicts of 57b641a to lmp-mr1-dev-plus-aosp.Chris Craik2014-11-201-1/+1
|\ \ | |/ |/| | | Change-Id: I1be708f5a0cd4d64021a57f1063d54330f1d2d7e
| * Improve logging around performance critical eventsChris Craik2014-11-191-1/+1
| | | | | | | | | | | | | | | | | | | | bug:17702227 Add details useful to developers (such as layer size/View name), and switch away from logging implementation names/details, since they are generally not relevant to developers. Change-Id: Iee605d182f241450f7e75a6d0c283d51fa1312f5
* | Frameworks/base: Wall Werror in libs/hwuiAndreas Gampe2014-11-101-1/+1
|/ | | | | | Turn on -Wall -Werror in libs/hwui. Fix errors. Change-Id: I74962d08c889712dacbd0d86d6760fc10802b6bd
* Minor cleanups around color usageChris Craik2014-07-171-1/+2
| | | | Change-Id: I56733cbe933a58d5977c032f056abb35265ee252
* Deep copy SkPath parameter to PathCache taskChris Craik2014-06-301-2/+2
| | | | | bug:15440706 Change-Id: I2b5b25f620df838cb1155cc8502d86ad3644c212
* stop using (deprecated) SkBitmap::ConfigMike Reed2014-06-201-2/+1
| | | | Change-Id: Ic75b5fc6996578e9d95bd3a220439ec1541d7c3b
* am b1c76a1a: Merge "Tessellate on worker threads" into lmp-preview-devChris Craik2014-06-101-12/+5
|\ | | | | | | | | * commit 'b1c76a1abcfb2f33e12fef37ec71d20724863b5e': Tessellate on worker threads
| * Tessellate on worker threadsChris Craik2014-06-101-12/+5
| | | | | | | | | | | | Tessellate and cache (where possible) shadow and round rect tessellation tasks. Change-Id: I2cfda8e11d83d51ea74af871235cf26e8f831d40
* | resolved conflicts for merge of b68f5f09 to masterKenny Root2014-05-301-2/+2
|\ \ | |/ |/| | | Change-Id: Ib8dd3a353f40a3357c8dc5ac591cd5e3ab4d0a4b
| * Fix a resource race bug in PathCacheyuyang2014-05-301-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When enabled defer rendering, it will do precache for DrawPathOp. The paint used for creating PathTask in precache just get the address of mFilteredPaint of OpenGLRenderer. So for the following defer operation like DrawTextOp has possibility change the mFilteredPaint by getPaint while another WorkerThread in PathCache is using the paint which pointed to the same address of mFilteredPaint to generate bitmap. As a result, it will generate a wrong bitmap for generateTexture in PathCache. To fix it, do a copy of paint when creating PathTask. CRs-Fixed: 664244 Change-Id: I5516f5b143458b88d3573d15b7ebb34f688800c7
* | am 328b7ee6: am aecb8c43: am 7b4cce68: am c3bac8a0: Merge "Fix graphics ↵Chris Craik2014-02-261-1/+3
|\ \ | |/ | | | | | | | | | | corruption caused by HWUI caches" * commit '328b7ee6a2ba2dc8f162fa57dbafd4bfc89bb72e': Fix graphics corruption caused by HWUI caches
| * Fix graphics corruption caused by HWUI cachesSangkyu Lee2014-02-261-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some caches(PatchCache, TextureCache, PathCache) for HWUI uses deferred removal for their cache entries even though actual resource objects are immediately freed by ResourceCache. For this reason, the uniqueness of a resource address in the caches is not guaranteed in specific cases. (Because malloc() can return the same address when malloc() and free() called very frequently.) So it can be possible the cache have two cache entries for two different resources but the same memory address. (Of course one of the resources is already freed.) It also can be possible mGarbage vector in PatchCache has duplicated addresses and this can lead to duplicated free blocks in the free block list and graphics corruption. (Deferred removal was implmeneted based on an assumption of unique resource addresses.) So this patch makes sure resource objects are freed after the resources are removed from the caches to guarantee the uniqueness of a resource address and prevent graphics corruption. Change-Id: I040f033a4fc783d2c4bc04b113589657c36fb15b Signed-off-by: Sangkyu Lee <sk82.lee@lge.com>
* | Use const where possible for drawing parametersChris Craik2014-01-031-10/+10
|/ | | | | | They should never be modified by a Renderer, only read and copied. Change-Id: I9d8d55dca19115ee9dfeb2bb3f092ba2fb327cd4
* Second attempt at avoiding infinite loop in PathCache::trim()Romain Guy2013-08-211-1/+16
| | | | | | Bug #10347089 Change-Id: I70f5a3933e848632473acc6636c88be5dc6ac430
* Properly account for created paths in the cacheRomain Guy2013-08-211-0/+5
| | | | Change-Id: I47b89b3085cefab6daac9194e7bfd3c140b37fa2
* Refcount 9-patches and properly handle GC eventsRomain Guy2013-06-261-11/+14
| | | | | | | | | | | | | | | | | This change adds refcounting of Res_png_9patch instances, the native data structure used to represent 9-patches. The Dalvik NinePatch class now holds a native pointer instead of a Dalvik byte[]. This pointer is used whenever we need to draw the 9-patch (software or hardware.) Since we are now tracking garbage collection of NinePatch objects libhwui's PatchCache must keep a list of free blocks in the VBO used to store the meshes. This change also removes unnecessary instances tracking from GLES20DisplayList. Bitmaps and 9-patches are refcounted at the native level and do not need to be tracked by the Dalvik layer. Change-Id: Ib8682d573a538aaf1945f8ec5a9bd5da5d16f74b
* Assume a texture is unbound after deleting itRomain Guy2013-06-061-1/+1
| | | | | | | | | | | Bug #9316260 The GL specification indicates that deleting a bound texture has the side effect of binding the default texture (name=0). This change replaces all calls to glDeleteTextures() by Caches::deleteTexture() to properly keep track of texture bindings. Change-Id: Ifbc60ef433e0f9776a668dd5bd5f0adbc65a77a0
* Introduce Caches::bindTexture() to reduce glBindTexture callsRomain Guy2013-06-041-2/+2
| | | | Change-Id: Ic345422567c020c0a9035ff51dcf2ae2a1fc59f4
* Path precaching creates duplicate cache entriesRomain Guy2013-03-261-4/+9
| | | | | | Bug #8478275 Change-Id: Ib541ea051e42e01cc0d277790e9c09de38ef72ee
* Stop worker threads on memory trim & fix bad pointer accessRomain Guy2013-03-201-3/+4
| | | | Change-Id: I6fe7e31aeb6dd41fa65ab952caed97bc2da510d7
* Merge all shapes/paths caches to PathCacheRomain Guy2013-03-181-19/+424
| | | | | | | | | This change will greatly simplify the multi-threading of all shape types. This change also uses PathTessellator to render convex paths. Change-Id: I4e65bc95c9d24ecae2183b72204de5c2dfb6ada4
* Prevent infinite loop when trimming the path cacheRomain Guy2013-03-131-2/+3
| | | | Change-Id: I04b5fa498336068f997c68d8613b35a99f67adbe
* Add TaskManager APIRomain Guy2013-03-121-68/+41
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This API can be used to run arbitrary tasks on a pool of worker threads. The number of threads is calculated based on the number of CPU cores available. The API is made of 3 classes: TaskManager Creates and manages the worker threads. Task Describes the work to be done and the type of the output. A task contains a future used to wait for the worker thread to be done computing the result of the task. TaskProcessor The processor dispatches tasks to the TaskManager and is responsible for performing the computation required by each task. A processor will only be asked to process tasks sent to the manager through the processor. A typical use case: class MyTask: Task<MyType> class MyProcessor: TaskProcessor<MyType> TaskManager m = new TaskManager(); MyProcessor p = new MyProcessor(m); MyTask t = new MyTask(); p.add(t); // Waits until the result is available MyType result = t->getResult(); Change-Id: I1fe845ba4c49bb0e1b0627ab147f9a861c8e0749
* Precache paths from a worker threadRomain Guy2013-03-111-23/+139
| | | | Change-Id: I3e7b53d67e0e03e403beaf55c39350ead7f1e309
* Add plumbing for better text scalingRomain Guy2013-01-081-3/+0
| | | | | | | | Fonts are now described by a transform matrix. This lead to switching from a vector to a hashmap. This change therefore adds new comparators and hash computations to Font. Change-Id: I2daffa7d6287c18554c606b8bfa06640d28b4530
* Use LruCache instead of GenerationCache in libhwuiRomain Guy2012-11-291-11/+8
| | | | Change-Id: Ic26ddc7151eb5462bcd243b21daf7187ed6d3bec
* Work-around for a Skia rasterization bugRomain Guy2012-04-271-1/+6
| | | | | | | | | Bug #6411457 Skia does not generates the bottom right pixel of a rect when drawing a rect as an SkPath into an alpha8 bitmap. Change-Id: Ifb5286ae67745c9e44ee387b6d6ad607a9a2e6ce
* Only recreate path textures when necessaryRomain Guy2012-02-231-0/+5
| | | | | | | | | | | | | | When a drawPath command is recorded in a display list, a copy of the source path is made to preserve against possible modifications of the said source path. Copies are discarded when a display list is cleared, which usually happens on invalidate(). This means that even if a path is never modified, the texture generated to draw it on screen is destroyed every time an invalidate() is issued. This change fixes this problem by introducing a reference to the source path in the copy. If both the copy and the source path have the same genID, they are the same path and can share the same texture. Change-Id: I34849311c183e06336a1391d2d1568a087f973f6
* Record possible clip rejects when recording display listsRomain Guy2012-02-171-0/+20
| | | | | | | | | | | This optimization allows us to quickly skip operations that lie entirely outside of the known bounds of a display list. Because of ViewGroup.setClipChildren, we must keep the operations recorded in the display list. setClipChildren(false) is however a very uncommon operation and we will therefore often benefit from this new optimization. Change-Id: I0942c864e55298e6dccd9977d15adefbce3ba3ad
* Clean up GenerationCache.Jeff Brown2011-11-111-2/+2
| | | | | | | | | | | Use const references to keys and values where appropriate to avoid copying them unnecessarily. Deleted some dead code. Simplified a few pieces that were doing unnecessary redundant work. Change-Id: Ib2145b7094a40db2d679e05dafe050fe1e87b846
* Remove the right entries from the cache when a GC happens.Romain Guy2011-02-241-1/+3
| | | | Change-Id: I4649b18c4721a5511a404299e771e675c843407b