android_frameworks_base/libs/hwui/DeferredDisplayList.h
Romain Guy 7f6d6b0370 Split assets atlas batches
Bug #10185769

The assets atlas contains assets that need to be blended and assets
that do not need to be blended. With a single merge id, currently
set to be the pointer to the atlas itself, draw ops merging could
generate batches of commands containing both opaque and translucent
assets. The blend state was chosen from only one of the assets in
the batch, leading either to inefficiencies (blending large opaque
assets) or incorrect behaviors (not blending translucent assets.)

This change introduces two new merge ids in the atlas: an opaque
key and a blend key. These keys are simple booleans set to false
and true respectively (the values do not matter really.) Their
memory addresses are used as the merge ids when createing draw ops
batches, allowing all opaque ops to be batched together and all
translucent ops to be batched together.

Change-Id: I114dba0533c44987e53864b471ccb28c811f2025
2013-08-06 18:35:01 -07:00

158 lines
4.6 KiB
C++

/*
* Copyright (C) 2013 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#ifndef ANDROID_HWUI_DEFERRED_DISPLAY_LIST_H
#define ANDROID_HWUI_DEFERRED_DISPLAY_LIST_H
#include <utils/Errors.h>
#include <utils/Vector.h>
#include "Matrix.h"
#include "Rect.h"
#include "utils/TinyHashMap.h"
class SkBitmap;
namespace android {
namespace uirenderer {
class ClipOp;
class DrawOp;
class SaveOp;
class SaveLayerOp;
class StateOp;
class OpenGLRenderer;
class Batch;
class DrawBatch;
class MergingDrawBatch;
typedef const void* mergeid_t;
class DeferredDisplayList {
public:
DeferredDisplayList(const Rect& bounds, bool avoidOverdraw = true) :
mBounds(bounds), mAvoidOverdraw(avoidOverdraw) {
clear();
}
~DeferredDisplayList() { clear(); }
void reset(const Rect& bounds) { mBounds.set(bounds); }
enum OpBatchId {
kOpBatch_None = 0, // Don't batch
kOpBatch_Bitmap,
kOpBatch_Patch,
kOpBatch_AlphaVertices,
kOpBatch_Vertices,
kOpBatch_AlphaMaskTexture,
kOpBatch_Text,
kOpBatch_ColorText,
kOpBatch_Count, // Add other batch ids before this
};
bool isEmpty() { return mBatches.isEmpty(); }
/**
* Plays back all of the draw ops recorded into batches to the renderer.
* Adjusts the state of the renderer as necessary, and restores it when complete
*/
status_t flush(OpenGLRenderer& renderer, Rect& dirty);
void addClip(OpenGLRenderer& renderer, ClipOp* op);
void addSaveLayer(OpenGLRenderer& renderer, SaveLayerOp* op, int newSaveCount);
void addSave(OpenGLRenderer& renderer, SaveOp* op, int newSaveCount);
void addRestoreToCount(OpenGLRenderer& renderer, StateOp* op, int newSaveCount);
/**
* Add a draw op into the DeferredDisplayList, reordering as needed (for performance) if
* disallowReorder is false, respecting draw order when overlaps occur.
*/
void addDrawOp(OpenGLRenderer& renderer, DrawOp* op);
private:
/**
* Resets the batching back-pointers, creating a barrier in the operation stream so that no ops
* added in the future will be inserted into a batch that already exist.
*/
void resetBatchingState();
void clear();
void storeStateOpBarrier(OpenGLRenderer& renderer, StateOp* op);
void storeRestoreToCountBarrier(OpenGLRenderer& renderer, StateOp* op, int newSaveCount);
bool recordingComplexClip() const { return mComplexClipStackStart >= 0; }
int getStateOpDeferFlags() const;
int getDrawOpDeferFlags() const;
void discardDrawingBatches(const unsigned int maxIndex);
// layer space bounds of rendering
Rect mBounds;
const bool mAvoidOverdraw;
/**
* At defer time, stores the *defer time* savecount of save/saveLayer ops that were deferred, so
* that when an associated restoreToCount is deferred, it can be recorded as a
* RestoreToCountBatch
*/
Vector<int> mSaveStack;
int mComplexClipStackStart;
Vector<Batch*> mBatches;
// Maps batch ids to the most recent *non-merging* batch of that id
Batch* mBatchLookup[kOpBatch_Count];
// Points to the index after the most recent barrier
int mEarliestBatchIndex;
// Points to the first index that may contain a pure drawing batch
int mEarliestUnclearedIndex;
/**
* Maps the mergeid_t returned by an op's getMergeId() to the most recently seen
* MergingDrawBatch of that id. These ids are unique per draw type and guaranteed to not
* collide, which avoids the need to resolve mergeid collisions.
*/
TinyHashMap<mergeid_t, DrawBatch*> mMergingBatches[kOpBatch_Count];
};
/**
* Struct containing information that instructs the defer
*/
struct DeferInfo {
public:
DeferInfo() :
batchId(DeferredDisplayList::kOpBatch_None),
mergeId((mergeid_t) -1),
mergeable(false),
opaqueOverBounds(false) {
};
int batchId;
mergeid_t mergeId;
bool mergeable;
bool opaqueOverBounds; // opaque over bounds in DeferredDisplayState - can skip ops below
};
}; // namespace uirenderer
}; // namespace android
#endif // ANDROID_HWUI_DEFERRED_DISPLAY_LIST_H