services/surfaceflinger/CompositionEngine/src/OutputLayer.cpp

/*
 * Copyright 2019 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.
 */
#include <DisplayHardware/Hal.h>
#include <android-base/stringprintf.h>
#include <compositionengine/DisplayColorProfile.h>
#include <compositionengine/LayerFECompositionState.h>
#include <compositionengine/Output.h>
#include <compositionengine/impl/HwcBufferCache.h>
#include <compositionengine/impl/OutputCompositionState.h>
#include <compositionengine/impl/OutputLayer.h>
#include <compositionengine/impl/OutputLayerCompositionState.h>
#include <cstdint>
#include "system/graphics-base-v1.0.h"
#include "ui/FloatRect.h"

#include <ui/HdrRenderTypeUtils.h>

// TODO(b/129481165): remove the #pragma below and fix conversion issues
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wconversion"

#include "DisplayHardware/HWComposer.h"

// TODO(b/129481165): remove the #pragma below and fix conversion issues
#pragma clang diagnostic pop // ignored "-Wconversion"

using aidl::android::hardware::graphics::composer3::Composition;
using aidl::android::hardware::graphics::composer3::Luts;

namespace android::compositionengine {

OutputLayer::~OutputLayer() = default;

namespace impl {

namespace {

FloatRect reduce(const FloatRect& win, const Region& exclude) {
    if (CC_LIKELY(exclude.isEmpty())) {
        return win;
    }
    // Convert through Rect (by rounding) for lack of FloatRegion
    return Region(Rect{win}).subtract(exclude).getBounds().toFloatRect();
}

} // namespace

std::unique_ptr<OutputLayer> createOutputLayer(const compositionengine::Output& output,
                                               const sp<compositionengine::LayerFE>& layerFE) {
    return createOutputLayerTemplated<OutputLayer>(output, layerFE);
}

OutputLayer::~OutputLayer() = default;

void OutputLayer::setHwcLayer(std::shared_ptr<HWC2::Layer> hwcLayer) {
    auto& state = editState();
    if (hwcLayer) {
        state.hwc.emplace(std::move(hwcLayer));
    } else {
        state.hwc.reset();
    }
}

Rect OutputLayer::calculateInitialCrop() const {
    const auto& layerState = *getLayerFE().getCompositionState();

    // apply the projection's clipping to the window crop in
    // layerstack space, and convert-back to layer space.
    // if there are no window scaling involved, this operation will map to full
    // pixels in the buffer.

    FloatRect activeCropFloat =
            reduce(layerState.geomLayerBounds, layerState.transparentRegionHint);

    const Rect& viewport = getOutput().getState().layerStackSpace.getContent();
    const ui::Transform& layerTransform = layerState.geomLayerTransform;
    const ui::Transform& inverseLayerTransform = layerState.geomInverseLayerTransform;
    // Transform to screen space.
    activeCropFloat = layerTransform.transform(activeCropFloat);
    activeCropFloat = activeCropFloat.intersect(viewport.toFloatRect());
    // Back to layer space to work with the content crop.
    activeCropFloat = inverseLayerTransform.transform(activeCropFloat);

    // This needs to be here as transform.transform(Rect) computes the
    // transformed rect and then takes the bounding box of the result before
    // returning. This means
    // transform.inverse().transform(transform.transform(Rect)) != Rect
    // in which case we need to make sure the final rect is clipped to the
    // display bounds.
    Rect activeCrop{activeCropFloat};
    if (!activeCrop.intersect(layerState.geomBufferSize, &activeCrop)) {
        activeCrop.clear();
    }
    return activeCrop;
}

FloatRect OutputLayer::calculateOutputSourceCrop(uint32_t internalDisplayRotationFlags) const {
    const auto& layerState = *getLayerFE().getCompositionState();

    if (!layerState.geomUsesSourceCrop) {
        return {};
    }

    // the content crop is the area of the content that gets scaled to the
    // layer's size. This is in buffer space.
    FloatRect crop = layerState.geomContentCrop.toFloatRect();

    // In addition there is a WM-specified crop we pull from our drawing state.
    Rect activeCrop = calculateInitialCrop();
    const Rect& bufferSize = layerState.geomBufferSize;

    int winWidth = bufferSize.getWidth();
    int winHeight = bufferSize.getHeight();

    // The bufferSize for buffer state layers can be unbounded ([0, 0, -1, -1])
    // if display frame hasn't been set and the parent is an unbounded layer.
    if (winWidth < 0 && winHeight < 0) {
        return crop;
    }

    // Transform the window crop to match the buffer coordinate system,
    // which means using the inverse of the current transform set on the
    // SurfaceFlingerConsumer.
    uint32_t invTransform = layerState.geomBufferTransform;
    if (layerState.geomBufferUsesDisplayInverseTransform) {
        /*
         * the code below applies the primary display's inverse transform to the
         * buffer
         */
        uint32_t invTransformOrient = internalDisplayRotationFlags;
        // calculate the inverse transform
        if (invTransformOrient & HAL_TRANSFORM_ROT_90) {
            invTransformOrient ^= HAL_TRANSFORM_FLIP_V | HAL_TRANSFORM_FLIP_H;
        }
        // and apply to the current transform
        invTransform =
                (ui::Transform(invTransformOrient) * ui::Transform(invTransform)).getOrientation();
    }

    if (invTransform & HAL_TRANSFORM_ROT_90) {
        // If the activeCrop has been rotate the ends are rotated but not
        // the space itself so when transforming ends back we can't rely on
        // a modification of the axes of rotation. To account for this we
        // need to reorient the inverse rotation in terms of the current
        // axes of rotation.
        bool isHFlipped = (invTransform & HAL_TRANSFORM_FLIP_H) != 0;
        bool isVFlipped = (invTransform & HAL_TRANSFORM_FLIP_V) != 0;
        if (isHFlipped == isVFlipped) {
            invTransform ^= HAL_TRANSFORM_FLIP_V | HAL_TRANSFORM_FLIP_H;
        }
        std::swap(winWidth, winHeight);
    }
    const Rect winCrop =
            activeCrop.transform(invTransform, bufferSize.getWidth(), bufferSize.getHeight());

    // below, crop is intersected with winCrop expressed in crop's coordinate space
    const float xScale = crop.getWidth() / float(winWidth);
    const float yScale = crop.getHeight() / float(winHeight);

    const float insetLeft = winCrop.left * xScale;
    const float insetTop = winCrop.top * yScale;
    const float insetRight = (winWidth - winCrop.right) * xScale;
    const float insetBottom = (winHeight - winCrop.bottom) * yScale;

    crop.left += insetLeft;
    crop.top += insetTop;
    crop.right -= insetRight;
    crop.bottom -= insetBottom;

    return crop;
}

Rect OutputLayer::calculateOutputDisplayFrame() const {
    const auto& layerState = *getLayerFE().getCompositionState();
    const auto& outputState = getOutput().getState();

    // Convert from layer space to layerStackSpace
    // apply the layer's transform, followed by the display's global transform
    // here we're guaranteed that the layer's transform preserves rects
    const ui::Transform& layerTransform = layerState.geomLayerTransform;
    Region activeTransparentRegion = layerTransform.transform(layerState.transparentRegionHint);
    if (!layerState.geomCrop.isEmpty() && layerState.geomBufferSize.isValid()) {
        FloatRect activeCrop = layerTransform.transform(layerState.geomCrop);
        activeCrop = activeCrop.intersect(outputState.layerStackSpace.getContent().toFloatRect());
        const FloatRect& bufferSize =
                layerTransform.transform(layerState.geomBufferSize.toFloatRect());
        activeCrop = activeCrop.intersect(bufferSize);

        // mark regions outside the crop as transparent
        Rect topRegion = Rect(layerTransform.transform(
                FloatRect(0, 0, layerState.geomBufferSize.getWidth(), layerState.geomCrop.top)));
        Rect bottomRegion = Rect(layerTransform.transform(
                FloatRect(0, layerState.geomCrop.bottom, layerState.geomBufferSize.getWidth(),
                          layerState.geomBufferSize.getHeight())));
        Rect leftRegion = Rect(layerTransform.transform(FloatRect(0, layerState.geomCrop.top,
                                                                 layerState.geomCrop.left,
                                                                 layerState.geomCrop.bottom)));
        Rect rightRegion = Rect(layerTransform.transform(
                FloatRect(layerState.geomCrop.right, layerState.geomCrop.top,
                          layerState.geomBufferSize.getWidth(), layerState.geomCrop.bottom)));

        activeTransparentRegion.orSelf(topRegion);
        activeTransparentRegion.orSelf(bottomRegion);
        activeTransparentRegion.orSelf(leftRegion);
        activeTransparentRegion.orSelf(rightRegion);
    }

    // reduce uses a FloatRect to provide more accuracy during the
    // transformation. We then round upon constructing 'frame'.
    FloatRect geomLayerBounds = layerState.geomLayerBounds;

    // Some HWCs may clip client composited input to its displayFrame. Make sure
    // that this does not cut off the shadow.
    if (layerState.forceClientComposition && layerState.shadowSettings.length > 0.0f) {
        // RenderEngine currently blurs shadows to smooth out edges, so outset by
        // 2x the length instead of 1x to compensate
        const auto outset = layerState.shadowSettings.length * 2;
        geomLayerBounds.left -= outset;
        geomLayerBounds.top -= outset;
        geomLayerBounds.right += outset;
        geomLayerBounds.bottom += outset;
    }

    geomLayerBounds = layerTransform.transform(geomLayerBounds);
    FloatRect frame = reduce(geomLayerBounds, activeTransparentRegion);
    frame = frame.intersect(outputState.layerStackSpace.getContent().toFloatRect());

    // convert from layerStackSpace to displaySpace
    const ui::Transform displayTransform{outputState.transform};
    return Rect(displayTransform.transform(frame));
}

uint32_t OutputLayer::calculateOutputRelativeBufferTransform(
        uint32_t internalDisplayRotationFlags) const {
    const auto& layerState = *getLayerFE().getCompositionState();
    const auto& outputState = getOutput().getState();

    /*
     * Transformations are applied in this order:
     * 1) buffer orientation/flip/mirror
     * 2) state transformation (window manager)
     * 3) layer orientation (screen orientation)
     * (NOTE: the matrices are multiplied in reverse order)
     */
    const ui::Transform& layerTransform = layerState.geomLayerTransform;
    const ui::Transform displayTransform{outputState.transform};
    const ui::Transform bufferTransform{layerState.geomBufferTransform};
    ui::Transform transform(displayTransform * layerTransform * bufferTransform);

    if (layerState.geomBufferUsesDisplayInverseTransform) {
        /*
         * We must apply the internal display's inverse transform to the buffer
         * transform, and not the one for the output this layer is on.
         */
        uint32_t invTransform = internalDisplayRotationFlags;

        // calculate the inverse transform
        if (invTransform & HAL_TRANSFORM_ROT_90) {
            invTransform ^= HAL_TRANSFORM_FLIP_V | HAL_TRANSFORM_FLIP_H;
        }

        /*
         * Here we cancel out the orientation component of the WM transform.
         * The scaling and translate components are already included in our bounds
         * computation so it's enough to just omit it in the composition.
         * See comment in BufferLayer::prepareClientLayer with ref to b/36727915 for why.
         */
        transform = ui::Transform(invTransform) * displayTransform * bufferTransform;
    }

    // this gives us only the "orientation" component of the transform
    return transform.getOrientation();
}

void OutputLayer::updateLuts(
        std::shared_ptr<gui::DisplayLuts> layerFEStateLut,
        const std::optional<std::vector<std::optional<LutProperties>>>& properties) {
    auto& state = editState();

    if (!properties) {
        // GPU composition if no Hwc Luts
        state.forceClientComposition = true;
        return;
    }

    std::vector<LutProperties> hwcLutProperties;
    for (auto& p : *properties) {
        if (p) {
            hwcLutProperties.emplace_back(*p);
        }
    }

    for (const auto& inputLut : layerFEStateLut->lutProperties) {
        bool foundInHwcLuts = false;
        for (const auto& hwcLut : hwcLutProperties) {
            if (static_cast<int32_t>(hwcLut.dimension) ==
                        static_cast<int32_t>(inputLut.dimension) &&
                hwcLut.size == inputLut.size &&
                std::find(hwcLut.samplingKeys.begin(), hwcLut.samplingKeys.end(),
                          static_cast<LutProperties::SamplingKey>(inputLut.samplingKey)) !=
                        hwcLut.samplingKeys.end()) {
                foundInHwcLuts = true;
                break;
            }
        }
        // if any lut properties of layerFEStateLut can not be found in hwcLutProperties,
        // GPU composition instead
        if (!foundInHwcLuts) {
            state.forceClientComposition = true;
            return;
        }
    }
}

void OutputLayer::updateCompositionState(
        bool includeGeometry, bool forceClientComposition,
        ui::Transform::RotationFlags internalDisplayRotationFlags,
        const std::optional<std::vector<std::optional<LutProperties>>> properties) {
    const auto* layerFEState = getLayerFE().getCompositionState();
    if (!layerFEState) {
        return;
    }

    const auto& outputState = getOutput().getState();
    const auto& profile = *getOutput().getDisplayColorProfile();
    auto& state = editState();

    if (includeGeometry) {
        // Clear the forceClientComposition flag before it is set for any
        // reason. Note that since it can be set by some checks below when
        // updating the geometry state, we only clear it when updating the
        // geometry since those conditions for forcing client composition won't
        // go away otherwise.
        state.forceClientComposition = false;

        state.displayFrame = calculateOutputDisplayFrame();
        state.sourceCrop = calculateOutputSourceCrop(internalDisplayRotationFlags);
        state.bufferTransform = static_cast<Hwc2::Transform>(
                calculateOutputRelativeBufferTransform(internalDisplayRotationFlags));

        if ((layerFEState->isSecure && !outputState.isSecure) ||
            (state.bufferTransform & ui::Transform::ROT_INVALID)) {
            state.forceClientComposition = true;
        }
    }

    auto pixelFormat = layerFEState->buffer ? std::make_optional(static_cast<ui::PixelFormat>(
                                                      layerFEState->buffer->getPixelFormat()))
                                            : std::nullopt;

    auto hdrRenderType =
            getHdrRenderType(outputState.dataspace, pixelFormat, layerFEState->desiredHdrSdrRatio);

    // Determine the output dependent dataspace for this layer. If it is
    // colorspace agnostic, it just uses the dataspace chosen for the output to
    // avoid the need for color conversion.
    // For now, also respect the colorspace agnostic flag if we're drawing to HDR, to avoid drastic
    // luminance shift. TODO(b/292162273): we should check if that's true though.
    state.dataspace = layerFEState->isColorspaceAgnostic && hdrRenderType == HdrRenderType::SDR
            ? outputState.dataspace
            : layerFEState->dataspace;

    // Override the dataspace transfer from 170M to sRGB if the device configuration requests this.
    // We do this here instead of in buffer info so that dumpsys can still report layers that are
    // using the 170M transfer. Also we only do this if the colorspace is not agnostic for the
    // layer, in case the color profile uses a 170M transfer function.
    if (outputState.treat170mAsSrgb && !layerFEState->isColorspaceAgnostic &&
        (state.dataspace & HAL_DATASPACE_TRANSFER_MASK) == HAL_DATASPACE_TRANSFER_SMPTE_170M) {
        state.dataspace = static_cast<ui::Dataspace>(
                (state.dataspace & HAL_DATASPACE_STANDARD_MASK) |
                (state.dataspace & HAL_DATASPACE_RANGE_MASK) | HAL_DATASPACE_TRANSFER_SRGB);
    }

    // re-get HdrRenderType after the dataspace gets changed.
    hdrRenderType =
            getHdrRenderType(state.dataspace, pixelFormat, layerFEState->desiredHdrSdrRatio);

    // For hdr content, treat the white point as the display brightness - HDR content should not be
    // boosted or dimmed.
    // If the layer explicitly requests to disable dimming, then don't dim either.
    if (hdrRenderType == HdrRenderType::GENERIC_HDR ||
        getOutput().getState().displayBrightnessNits == getOutput().getState().sdrWhitePointNits ||
        getOutput().getState().displayBrightnessNits == 0.f || !layerFEState->dimmingEnabled) {
        state.dimmingRatio = 1.f;
        state.whitePointNits = getOutput().getState().displayBrightnessNits;
    } else {
        float layerBrightnessNits = getOutput().getState().sdrWhitePointNits;
        // RANGE_EXTENDED can "self-promote" to HDR, but is still rendered for a particular
        // range that we may need to re-adjust to the current display conditions
        if (hdrRenderType == HdrRenderType::DISPLAY_HDR) {
            layerBrightnessNits *= layerFEState->currentHdrSdrRatio;
        }
        state.dimmingRatio =
                std::clamp(layerBrightnessNits / getOutput().getState().displayBrightnessNits, 0.f,
                           1.f);
        state.whitePointNits = layerBrightnessNits;
    }

    const auto& layerFEStateLut = layerFEState->luts;
    if (layerFEStateLut) {
        updateLuts(layerFEStateLut, properties);
    }

    // These are evaluated every frame as they can potentially change at any
    // time.
    if (layerFEState->forceClientComposition || !profile.isDataspaceSupported(state.dataspace) ||
        forceClientComposition) {
        state.forceClientComposition = true;
    }
}

void OutputLayer::writeStateToHWC(bool includeGeometry, bool skipLayer, uint32_t z,
                                  bool zIsOverridden, bool isPeekingThrough) {
    const auto& state = getState();
    // Skip doing this if there is no HWC interface
    if (!state.hwc) {
        return;
    }

    auto& hwcLayer = (*state.hwc).hwcLayer;
    if (!hwcLayer) {
        ALOGE("[%s] failed to write composition state to HWC -- no hwcLayer for output %s",
              getLayerFE().getDebugName(), getOutput().getName().c_str());
        return;
    }

    const auto* outputIndependentState = getLayerFE().getCompositionState();
    if (!outputIndependentState) {
        return;
    }

    auto requestedCompositionType = outputIndependentState->compositionType;

    if (requestedCompositionType == Composition::SOLID_COLOR && state.overrideInfo.buffer) {
        requestedCompositionType = Composition::DEVICE;
    }

    // TODO(b/181172795): We now update geometry for all flattened layers. We should update it
    // only when the geometry actually changes
    const bool isOverridden =
            state.overrideInfo.buffer != nullptr || isPeekingThrough || zIsOverridden;
    const bool prevOverridden = state.hwc->stateOverridden;
    if (isOverridden || prevOverridden || skipLayer || includeGeometry) {
        writeOutputDependentGeometryStateToHWC(hwcLayer.get(), requestedCompositionType, z);
        writeOutputIndependentGeometryStateToHWC(hwcLayer.get(), *outputIndependentState,
                                                 skipLayer);
    }

    writeOutputDependentPerFrameStateToHWC(hwcLayer.get());
    writeOutputIndependentPerFrameStateToHWC(hwcLayer.get(), *outputIndependentState,
                                             requestedCompositionType, skipLayer);

    writeCompositionTypeToHWC(hwcLayer.get(), requestedCompositionType, isPeekingThrough,
                              skipLayer);

    writeLutToHWC(hwcLayer.get(), *outputIndependentState);

    if (requestedCompositionType == Composition::SOLID_COLOR) {
        writeSolidColorStateToHWC(hwcLayer.get(), *outputIndependentState);
    }

    editState().hwc->stateOverridden = isOverridden;
    editState().hwc->layerSkipped = skipLayer;
}

void OutputLayer::writeOutputDependentGeometryStateToHWC(HWC2::Layer* hwcLayer,
                                                         Composition requestedCompositionType,
                                                         uint32_t z) {
    const auto& outputDependentState = getState();

    Rect displayFrame = outputDependentState.displayFrame;
    FloatRect sourceCrop = outputDependentState.sourceCrop;

    if (outputDependentState.overrideInfo.buffer != nullptr) {
        displayFrame = outputDependentState.overrideInfo.displayFrame;
        sourceCrop =
                FloatRect(0.f, 0.f,
                          static_cast<float>(outputDependentState.overrideInfo.buffer->getBuffer()
                                                     ->getWidth()),
                          static_cast<float>(outputDependentState.overrideInfo.buffer->getBuffer()
                                                     ->getHeight()));
    }

    ALOGV("Writing display frame [%d, %d, %d, %d]", displayFrame.left, displayFrame.top,
          displayFrame.right, displayFrame.bottom);

    if (auto error = hwcLayer->setDisplayFrame(displayFrame); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set display frame [%d, %d, %d, %d]: %s (%d)",
              getLayerFE().getDebugName(), displayFrame.left, displayFrame.top, displayFrame.right,
              displayFrame.bottom, to_string(error).c_str(), static_cast<int32_t>(error));
    }

    if (auto error = hwcLayer->setSourceCrop(sourceCrop); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set source crop [%.3f, %.3f, %.3f, %.3f]: "
              "%s (%d)",
              getLayerFE().getDebugName(), sourceCrop.left, sourceCrop.top, sourceCrop.right,
              sourceCrop.bottom, to_string(error).c_str(), static_cast<int32_t>(error));
    }

    if (auto error = hwcLayer->setZOrder(z); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set Z %u: %s (%d)", getLayerFE().getDebugName(), z,
              to_string(error).c_str(), static_cast<int32_t>(error));
    }

    // Solid-color layers and overridden buffers should always use an identity transform.
    const auto bufferTransform = (requestedCompositionType != Composition::SOLID_COLOR &&
                                  getState().overrideInfo.buffer == nullptr)
            ? outputDependentState.bufferTransform
            : static_cast<hal::Transform>(0);
    if (auto error = hwcLayer->setTransform(static_cast<hal::Transform>(bufferTransform));
        error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set transform %s: %s (%d)", getLayerFE().getDebugName(),
              toString(outputDependentState.bufferTransform).c_str(), to_string(error).c_str(),
              static_cast<int32_t>(error));
    }
}

void OutputLayer::writeOutputIndependentGeometryStateToHWC(
        HWC2::Layer* hwcLayer, const LayerFECompositionState& outputIndependentState,
        bool skipLayer) {
    // If there is a peekThroughLayer, then this layer has a hole in it. We need to use
    // PREMULTIPLIED so it will peek through.
    const auto& overrideInfo = getState().overrideInfo;
    const auto blendMode = overrideInfo.buffer || overrideInfo.peekThroughLayer
            ? hardware::graphics::composer::hal::BlendMode::PREMULTIPLIED
            : outputIndependentState.blendMode;
    if (auto error = hwcLayer->setBlendMode(blendMode); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set blend mode %s: %s (%d)", getLayerFE().getDebugName(),
              toString(blendMode).c_str(), to_string(error).c_str(), static_cast<int32_t>(error));
    }

    const float alpha = skipLayer
            ? 0.0f
            : (getState().overrideInfo.buffer ? 1.0f : outputIndependentState.alpha);
    ALOGV("Writing alpha %f", alpha);

    if (auto error = hwcLayer->setPlaneAlpha(alpha); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set plane alpha %.3f: %s (%d)", getLayerFE().getDebugName(), alpha,
              to_string(error).c_str(), static_cast<int32_t>(error));
    }

    for (const auto& [name, entry] : outputIndependentState.metadata) {
        if (auto error = hwcLayer->setLayerGenericMetadata(name, entry.mandatory, entry.value);
            error != hal::Error::NONE) {
            ALOGE("[%s] Failed to set generic metadata %s %s (%d)", getLayerFE().getDebugName(),
                  name.c_str(), to_string(error).c_str(), static_cast<int32_t>(error));
        }
    }
}

void OutputLayer::writeLutToHWC(HWC2::Layer* hwcLayer,
                                const LayerFECompositionState& outputIndependentState) {
    if (!outputIndependentState.luts) {
        return;
    }
    auto& lutFileDescriptor = outputIndependentState.luts->getLutFileDescriptor();
    auto lutOffsets = outputIndependentState.luts->offsets;
    auto& lutProperties = outputIndependentState.luts->lutProperties;

    std::vector<LutProperties> aidlProperties;
    aidlProperties.reserve(lutProperties.size());
    for (size_t i = 0; i < lutOffsets.size(); i++) {
        LutProperties properties;
        properties.dimension = static_cast<LutProperties::Dimension>(lutProperties[i].dimension);
        properties.size = lutProperties[i].size;
        properties.samplingKeys = {
                static_cast<LutProperties::SamplingKey>(lutProperties[i].samplingKey)};
        aidlProperties.emplace_back(properties);
    }

    Luts luts;
    luts.pfd = ndk::ScopedFileDescriptor(dup(lutFileDescriptor.get()));
    luts.offsets = lutOffsets;
    luts.lutProperties = std::move(aidlProperties);

    switch (auto error = hwcLayer->setLuts(luts)) {
        case hal::Error::NONE:
            break;
        default:
            ALOGE("[%s] Failed to set Luts: %s (%d)", getLayerFE().getDebugName(),
                  to_string(error).c_str(), static_cast<int32_t>(error));
    }
}

void OutputLayer::writeOutputDependentPerFrameStateToHWC(HWC2::Layer* hwcLayer) {
    const auto& outputDependentState = getState();

    // TODO(lpique): b/121291683 outputSpaceVisibleRegion is output-dependent geometry
    // state and should not change every frame.
    Region visibleRegion = outputDependentState.overrideInfo.buffer
            ? Region(outputDependentState.overrideInfo.visibleRegion)
            : outputDependentState.outputSpaceVisibleRegion;
    if (auto error = hwcLayer->setVisibleRegion(visibleRegion); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set visible region: %s (%d)", getLayerFE().getDebugName(),
              to_string(error).c_str(), static_cast<int32_t>(error));
        visibleRegion.dump(LOG_TAG);
    }

    if (auto error =
                hwcLayer->setBlockingRegion(outputDependentState.outputSpaceBlockingRegionHint);
        error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set blocking region: %s (%d)", getLayerFE().getDebugName(),
              to_string(error).c_str(), static_cast<int32_t>(error));
        outputDependentState.outputSpaceBlockingRegionHint.dump(LOG_TAG);
    }

    const auto dataspace = outputDependentState.overrideInfo.buffer
            ? outputDependentState.overrideInfo.dataspace
            : outputDependentState.dataspace;

    if (auto error = hwcLayer->setDataspace(dataspace); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set dataspace %d: %s (%d)", getLayerFE().getDebugName(), dataspace,
              to_string(error).c_str(), static_cast<int32_t>(error));
    }

    // Cached layers are not dimmed, which means that composer should attempt to dim.
    // Note that if the dimming ratio is large, then this may cause the cached layer
    // to kick back into GPU composition :(
    // Also note that this assumes that there are no HDR layers that are able to be cached.
    // Otherwise, this could cause HDR layers to be dimmed twice.
    const auto dimmingRatio = outputDependentState.overrideInfo.buffer
            ? (getOutput().getState().displayBrightnessNits != 0.f
                       ? std::clamp(getOutput().getState().sdrWhitePointNits /
                                            getOutput().getState().displayBrightnessNits,
                                    0.f, 1.f)
                       : 1.f)
            : outputDependentState.dimmingRatio;

    if (auto error = hwcLayer->setBrightness(dimmingRatio); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set brightness %f: %s (%d)", getLayerFE().getDebugName(),
              dimmingRatio, to_string(error).c_str(), static_cast<int32_t>(error));
    }
}

void OutputLayer::writeOutputIndependentPerFrameStateToHWC(
        HWC2::Layer* hwcLayer, const LayerFECompositionState& outputIndependentState,
        Composition compositionType, bool skipLayer) {
    switch (auto error = hwcLayer->setColorTransform(outputIndependentState.colorTransform)) {
        case hal::Error::NONE:
            break;
        case hal::Error::UNSUPPORTED:
            editState().forceClientComposition = true;
            break;
        default:
            ALOGE("[%s] Failed to set color transform: %s (%d)", getLayerFE().getDebugName(),
                  to_string(error).c_str(), static_cast<int32_t>(error));
    }

    const Region& surfaceDamage = getState().overrideInfo.buffer
            ? getState().overrideInfo.damageRegion
            : (getState().hwc->stateOverridden ? Region::INVALID_REGION
                                               : outputIndependentState.surfaceDamage);

    if (auto error = hwcLayer->setSurfaceDamage(surfaceDamage); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set surface damage: %s (%d)", getLayerFE().getDebugName(),
              to_string(error).c_str(), static_cast<int32_t>(error));
        outputIndependentState.surfaceDamage.dump(LOG_TAG);
    }

    // Content-specific per-frame state
    switch (compositionType) {
        case Composition::SOLID_COLOR:
            // For compatibility, should be written AFTER the composition type.
            break;
        case Composition::SIDEBAND:
            writeSidebandStateToHWC(hwcLayer, outputIndependentState);
            break;
        case Composition::CURSOR:
        case Composition::DEVICE:
        case Composition::DISPLAY_DECORATION:
        case Composition::REFRESH_RATE_INDICATOR:
            writeBufferStateToHWC(hwcLayer, outputIndependentState, skipLayer);
            break;
        case Composition::INVALID:
        case Composition::CLIENT:
            // Ignored
            break;
    }
}

void OutputLayer::writeSolidColorStateToHWC(HWC2::Layer* hwcLayer,
                                            const LayerFECompositionState& outputIndependentState) {
    aidl::android::hardware::graphics::composer3::Color color = {outputIndependentState.color.r,
                                                                 outputIndependentState.color.g,
                                                                 outputIndependentState.color.b,
                                                                 1.0f};

    if (auto error = hwcLayer->setColor(color); error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set color: %s (%d)", getLayerFE().getDebugName(),
              to_string(error).c_str(), static_cast<int32_t>(error));
    }
}

void OutputLayer::writeSidebandStateToHWC(HWC2::Layer* hwcLayer,
                                          const LayerFECompositionState& outputIndependentState) {
    if (auto error = hwcLayer->setSidebandStream(outputIndependentState.sidebandStream->handle());
        error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set sideband stream %p: %s (%d)", getLayerFE().getDebugName(),
              outputIndependentState.sidebandStream->handle(), to_string(error).c_str(),
              static_cast<int32_t>(error));
    }
}

void OutputLayer::uncacheBuffers(const std::vector<uint64_t>& bufferIdsToUncache) {
    auto& state = editState();
    // Skip doing this if there is no HWC interface
    if (!state.hwc) {
        return;
    }

    // Uncache the active buffer last so that it's the first buffer to be purged from the cache
    // next time a buffer is sent to this layer.
    bool uncacheActiveBuffer = false;

    std::vector<uint32_t> slotsToClear;
    for (uint64_t bufferId : bufferIdsToUncache) {
        if (bufferId == state.hwc->activeBufferId) {
            uncacheActiveBuffer = true;
        } else {
            uint32_t slot = state.hwc->hwcBufferCache.uncache(bufferId);
            if (slot != UINT32_MAX) {
                slotsToClear.push_back(slot);
            }
        }
    }
    if (uncacheActiveBuffer) {
        slotsToClear.push_back(state.hwc->hwcBufferCache.uncache(state.hwc->activeBufferId));
    }

    hal::Error error =
            state.hwc->hwcLayer->setBufferSlotsToClear(slotsToClear, state.hwc->activeBufferSlot);
    if (error != hal::Error::NONE) {
        ALOGE("[%s] Failed to clear buffer slots: %s (%d)", getLayerFE().getDebugName(),
              to_string(error).c_str(), static_cast<int32_t>(error));
    }
}

void OutputLayer::writeBufferStateToHWC(HWC2::Layer* hwcLayer,
                                        const LayerFECompositionState& outputIndependentState,
                                        bool skipLayer) {
    if (skipLayer && outputIndependentState.buffer == nullptr) {
        return;
    }
    auto supportedPerFrameMetadata =
            getOutput().getDisplayColorProfile()->getSupportedPerFrameMetadata();
    if (auto error = hwcLayer->setPerFrameMetadata(supportedPerFrameMetadata,
                                                   outputIndependentState.hdrMetadata);
        error != hal::Error::NONE && error != hal::Error::UNSUPPORTED) {
        ALOGE("[%s] Failed to set hdrMetadata: %s (%d)", getLayerFE().getDebugName(),
              to_string(error).c_str(), static_cast<int32_t>(error));
    }

    HwcSlotAndBuffer hwcSlotAndBuffer;
    sp<Fence> hwcFence;
    {
        // Editing the state only because we update the HWC buffer cache and active buffer.
        auto& state = editState();
        // Override buffers use a special cache slot so that they don't evict client buffers.
        if (state.overrideInfo.buffer != nullptr && !skipLayer) {
            hwcSlotAndBuffer = state.hwc->hwcBufferCache.getOverrideHwcSlotAndBuffer(
                    state.overrideInfo.buffer->getBuffer());
            hwcFence = state.overrideInfo.acquireFence;
            // Keep track of the active buffer ID so when it's discarded we uncache it last so its
            // slot will be used first, allowing the memory to be freed as soon as possible.
            state.hwc->activeBufferId = state.overrideInfo.buffer->getBuffer()->getId();
        } else {
            hwcSlotAndBuffer =
                    state.hwc->hwcBufferCache.getHwcSlotAndBuffer(outputIndependentState.buffer);
            hwcFence = outputIndependentState.acquireFence;
            // Keep track of the active buffer ID so when it's discarded we uncache it last so its
            // slot will be used first, allowing the memory to be freed as soon as possible.
            state.hwc->activeBufferId = outputIndependentState.buffer->getId();
        }
        // Keep track of the active buffer slot, so we can restore it after clearing other buffer
        // slots.
        state.hwc->activeBufferSlot = hwcSlotAndBuffer.slot;
    }

    if (auto error = hwcLayer->setBuffer(hwcSlotAndBuffer.slot, hwcSlotAndBuffer.buffer, hwcFence);
        error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set buffer %p: %s (%d)", getLayerFE().getDebugName(),
              hwcSlotAndBuffer.buffer->handle, to_string(error).c_str(),
              static_cast<int32_t>(error));
    }
}

void OutputLayer::writeCompositionTypeToHWC(HWC2::Layer* hwcLayer,
                                            Composition requestedCompositionType,
                                            bool isPeekingThrough, bool skipLayer) {
    auto& outputDependentState = editState();

    if (isClientCompositionForced(isPeekingThrough)) {
        // If we are forcing client composition, we need to tell the HWC
        requestedCompositionType = Composition::CLIENT;
    }

    // Set the requested composition type with the HWC whenever it changes
    // We also resend the composition type when this layer was previously skipped, to ensure that
    // the composition type is up-to-date.
    if (outputDependentState.hwc->hwcCompositionType != requestedCompositionType ||
        (outputDependentState.hwc->layerSkipped && !skipLayer)) {
        outputDependentState.hwc->hwcCompositionType = requestedCompositionType;

        if (auto error = hwcLayer->setCompositionType(requestedCompositionType);
            error != hal::Error::NONE) {
            ALOGE("[%s] Failed to set composition type %s: %s (%d)", getLayerFE().getDebugName(),
                  to_string(requestedCompositionType).c_str(), to_string(error).c_str(),
                  static_cast<int32_t>(error));
        }
    }
}

void OutputLayer::writeCursorPositionToHWC() const {
    // Skip doing this if there is no HWC interface
    auto hwcLayer = getHwcLayer();
    if (!hwcLayer) {
        return;
    }

    const auto* layerFEState = getLayerFE().getCompositionState();
    if (!layerFEState) {
        return;
    }

    const auto& outputState = getOutput().getState();

    Rect frame = layerFEState->cursorFrame;
    frame.intersect(outputState.layerStackSpace.getContent(), &frame);
    Rect position = outputState.transform.transform(frame);

    if (auto error = hwcLayer->setCursorPosition(position.left, position.top);
        error != hal::Error::NONE) {
        ALOGE("[%s] Failed to set cursor position to (%d, %d): %s (%d)",
              getLayerFE().getDebugName(), position.left, position.top, to_string(error).c_str(),
              static_cast<int32_t>(error));
    }
}

HWC2::Layer* OutputLayer::getHwcLayer() const {
    const auto& state = getState();
    return state.hwc ? state.hwc->hwcLayer.get() : nullptr;
}

bool OutputLayer::requiresClientComposition() const {
    const auto& state = getState();
    return !state.hwc || state.hwc->hwcCompositionType == Composition::CLIENT;
}

bool OutputLayer::isHardwareCursor() const {
    const auto& state = getState();
    return state.hwc && state.hwc->hwcCompositionType == Composition::CURSOR;
}

void OutputLayer::detectDisallowedCompositionTypeChange(Composition from, Composition to) const {
    bool result = false;
    switch (from) {
        case Composition::INVALID:
        case Composition::CLIENT:
            result = false;
            break;

        case Composition::DEVICE:
        case Composition::SOLID_COLOR:
            result = (to == Composition::CLIENT);
            break;

        case Composition::CURSOR:
        case Composition::SIDEBAND:
        case Composition::DISPLAY_DECORATION:
        case Composition::REFRESH_RATE_INDICATOR:
            result = (to == Composition::CLIENT || to == Composition::DEVICE);
            break;
    }

    if (!result) {
        ALOGE("[%s] Invalid device requested composition type change: %s (%d) --> %s (%d)",
              getLayerFE().getDebugName(), to_string(from).c_str(), static_cast<int>(from),
              to_string(to).c_str(), static_cast<int>(to));
    }
}

bool OutputLayer::isClientCompositionForced(bool isPeekingThrough) const {
    return getState().forceClientComposition ||
            (!isPeekingThrough && getLayerFE().hasRoundedCorners());
}

void OutputLayer::applyDeviceCompositionTypeChange(Composition compositionType) {
    auto& state = editState();
    LOG_FATAL_IF(!state.hwc);
    auto& hwcState = *state.hwc;

    // Only detected disallowed changes if this was not a skip layer, because the
    // validated composition type may be arbitrary (usually DEVICE, to reflect that there were
    // fewer GPU layers)
    if (!hwcState.layerSkipped) {
        detectDisallowedCompositionTypeChange(hwcState.hwcCompositionType, compositionType);
    }

    hwcState.hwcCompositionType = compositionType;
}

void OutputLayer::prepareForDeviceLayerRequests() {
    auto& state = editState();
    state.clearClientTarget = false;
}

void OutputLayer::applyDeviceLayerRequest(hal::LayerRequest request) {
    auto& state = editState();
    switch (request) {
        case hal::LayerRequest::CLEAR_CLIENT_TARGET:
            state.clearClientTarget = true;
            break;

        default:
            ALOGE("[%s] Unknown device layer request %s (%d)", getLayerFE().getDebugName(),
                  toString(request).c_str(), static_cast<int>(request));
            break;
    }
}

void OutputLayer::applyDeviceLayerLut(
        ndk::ScopedFileDescriptor lutFileDescriptor,
        std::vector<std::pair<int, LutProperties>> lutOffsetsAndProperties) {
    auto& state = editState();
    LOG_FATAL_IF(!state.hwc);
    auto& hwcState = *state.hwc;
    std::vector<int32_t> offsets;
    std::vector<int32_t> dimensions;
    std::vector<int32_t> sizes;
    std::vector<int32_t> samplingKeys;
    for (const auto& [offset, properties] : lutOffsetsAndProperties) {
        // The Lut(s) that comes back through CommandResultPayload should be
        // only one sampling key.
        if (properties.samplingKeys.size() == 1) {
            offsets.emplace_back(offset);
            dimensions.emplace_back(static_cast<int32_t>(properties.dimension));
            sizes.emplace_back(static_cast<int32_t>(properties.size));
            samplingKeys.emplace_back(static_cast<int32_t>(properties.samplingKeys[0]));
        }
    }
    hwcState.luts = std::make_shared<gui::DisplayLuts>(base::unique_fd(lutFileDescriptor.release()),
                                                       std::move(offsets), std::move(dimensions),
                                                       std::move(sizes), std::move(samplingKeys));
}

bool OutputLayer::needsFiltering() const {
    const auto& state = getState();
    const auto& sourceCrop = state.sourceCrop;
    auto displayFrameWidth = static_cast<float>(state.displayFrame.getWidth());
    auto displayFrameHeight = static_cast<float>(state.displayFrame.getHeight());

    if (state.bufferTransform & HAL_TRANSFORM_ROT_90) {
        std::swap(displayFrameWidth, displayFrameHeight);
    }

    return sourceCrop.getHeight() != displayFrameHeight ||
            sourceCrop.getWidth() != displayFrameWidth;
}

std::optional<LayerFE::LayerSettings> OutputLayer::getOverrideCompositionSettings() const {
    if (getState().overrideInfo.buffer == nullptr) {
        return {};
    }

    // Compute the geometry boundaries in layer stack space: we need to transform from the
    // framebuffer space of the override buffer to layer space.
    const ProjectionSpace& layerSpace = getOutput().getState().layerStackSpace;
    const ui::Transform transform = getState().overrideInfo.displaySpace.getTransform(layerSpace);
    const Rect boundaries = transform.transform(getState().overrideInfo.displayFrame);

    LayerFE::LayerSettings settings;
    settings.geometry = renderengine::Geometry{
            .boundaries = boundaries.toFloatRect(),
    };
    settings.bufferId = getState().overrideInfo.buffer->getBuffer()->getId();
    settings.source = renderengine::PixelSource{
            .buffer = renderengine::Buffer{
                    .buffer = getState().overrideInfo.buffer,
                    .fence = getState().overrideInfo.acquireFence,
                    // If the transform from layer space to display space contains a rotation, we
                    // need to undo the rotation in the texture transform
                    .textureTransform =
                            ui::Transform(transform.inverse().getOrientation(), 1, 1).asMatrix4(),
            }};
    settings.sourceDataspace = getState().overrideInfo.dataspace;
    settings.alpha = 1.0f;
    settings.whitePointNits = getOutput().getState().sdrWhitePointNits;

    return settings;
}

void OutputLayer::dump(std::string& out) const {
    using android::base::StringAppendF;

    StringAppendF(&out, "  - Output Layer %p(%s)\n", this, getLayerFE().getDebugName());
    dumpState(out);
}

} // namespace impl
} // namespace android::compositionengine