blob: a2104352b99a6e9c760efd5ce01ed29d2697a5d6 [file] [log] [blame]
/*
* 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.
*/
#include "rsContext.h"
#include "rsAllocation.h"
#include "rs_hal.h"
#ifndef RS_COMPATIBILITY_LIB
#include "rsGrallocConsumer.h"
#endif
namespace android {
namespace renderscript {
Allocation::Allocation(Context *rsc, const Type *type, uint32_t usages,
RsAllocationMipmapControl mc, void * ptr)
: ObjectBase(rsc) {
memset(&mHal, 0, sizeof(mHal));
mHal.state.mipmapControl = RS_ALLOCATION_MIPMAP_NONE;
mHal.state.usageFlags = usages;
mHal.state.mipmapControl = mc;
mHal.state.userProvidedPtr = ptr;
setType(type);
updateCache();
}
Allocation::Allocation(Context *rsc, const Allocation *alloc, const Type *type)
: ObjectBase(rsc) {
memset(&mHal, 0, sizeof(mHal));
mHal.state.baseAlloc = alloc;
mHal.state.usageFlags = alloc->mHal.state.usageFlags;
mHal.state.mipmapControl = RS_ALLOCATION_MIPMAP_NONE;
setType(type);
updateCache();
}
void Allocation::operator delete(void* ptr) {
if (ptr) {
Allocation *a = (Allocation*) ptr;
a->getContext()->mHal.funcs.freeRuntimeMem(ptr);
}
}
Allocation * Allocation::createAllocationStrided(Context *rsc, const Type *type, uint32_t usages,
RsAllocationMipmapControl mc, void * ptr,
size_t requiredAlignment) {
// Allocation objects must use allocator specified by the driver
void* allocMem = rsc->mHal.funcs.allocRuntimeMem(sizeof(Allocation), 0);
if (!allocMem) {
rsc->setError(RS_ERROR_FATAL_DRIVER, "Couldn't allocate memory for Allocation");
return nullptr;
}
bool success = false;
Allocation *a = nullptr;
if (usages & RS_ALLOCATION_USAGE_OEM) {
if (rsc->mHal.funcs.allocation.initOem != nullptr) {
a = new (allocMem) Allocation(rsc, type, usages, mc, nullptr);
success = rsc->mHal.funcs.allocation.initOem(rsc, a, type->getElement()->getHasReferences(), ptr);
} else {
rsc->setError(RS_ERROR_FATAL_DRIVER, "Allocation Init called with USAGE_OEM but driver does not support it");
return nullptr;
}
#ifdef RS_COMPATIBILITY_LIB
} else if (usages & RS_ALLOCATION_USAGE_INCREMENTAL_SUPPORT){
a = new (allocMem) Allocation(rsc, type, usages, mc, ptr);
success = rsc->mHal.funcs.allocation.initStrided(rsc, a, type->getElement()->getHasReferences(), requiredAlignment);
#endif
} else {
a = new (allocMem) Allocation(rsc, type, usages, mc, ptr);
success = rsc->mHal.funcs.allocation.init(rsc, a, type->getElement()->getHasReferences());
}
if (!success) {
rsc->setError(RS_ERROR_FATAL_DRIVER, "Allocation::Allocation, alloc failure");
delete a;
return nullptr;
}
return a;
}
Allocation * Allocation::createAllocation(Context *rsc, const Type *type, uint32_t usages,
RsAllocationMipmapControl mc, void * ptr) {
return Allocation::createAllocationStrided(rsc, type, usages, mc, ptr, kMinimumRSAlignment);
}
Allocation * Allocation::createAdapter(Context *rsc, const Allocation *alloc, const Type *type) {
// Allocation objects must use allocator specified by the driver
void* allocMem = rsc->mHal.funcs.allocRuntimeMem(sizeof(Allocation), 0);
if (!allocMem) {
rsc->setError(RS_ERROR_FATAL_DRIVER, "Couldn't allocate memory for Allocation");
return nullptr;
}
Allocation *a = new (allocMem) Allocation(rsc, alloc, type);
if (!rsc->mHal.funcs.allocation.initAdapter(rsc, a)) {
rsc->setError(RS_ERROR_FATAL_DRIVER, "Allocation::Allocation, alloc failure");
delete a;
return nullptr;
}
return a;
}
void Allocation::adapterOffset(Context *rsc, const uint32_t *offsets, size_t len) {
if (len >= sizeof(uint32_t) * 9) {
mHal.state.originX = offsets[0];
mHal.state.originY = offsets[1];
mHal.state.originZ = offsets[2];
mHal.state.originLOD = offsets[3];
mHal.state.originFace = offsets[4];
mHal.state.originArray[0] = offsets[5];
mHal.state.originArray[1] = offsets[6];
mHal.state.originArray[2] = offsets[7];
mHal.state.originArray[3] = offsets[8];
}
rsc->mHal.funcs.allocation.adapterOffset(rsc, this);
}
void Allocation::updateCache() {
const Type *type = mHal.state.type;
mHal.state.yuv = type->getDimYuv();
mHal.state.hasFaces = type->getDimFaces();
mHal.state.hasMipmaps = type->getDimLOD();
mHal.state.elementSizeBytes = type->getElementSizeBytes();
mHal.state.hasReferences = mHal.state.type->getElement()->getHasReferences();
}
Allocation::~Allocation() {
#ifndef RS_COMPATIBILITY_LIB
if (mGrallocConsumer) {
mGrallocConsumer->releaseIdx(mCurrentIdx);
if (!mGrallocConsumer->isActive()) {
delete mGrallocConsumer;
}
mGrallocConsumer = nullptr;
}
#endif
freeChildrenUnlocked();
mRSC->mHal.funcs.allocation.destroy(mRSC, this);
}
void Allocation::syncAll(Context *rsc, RsAllocationUsageType src) {
rsc->mHal.funcs.allocation.syncAll(rsc, this, src);
}
void * Allocation::getPointer(const Context *rsc, uint32_t lod, RsAllocationCubemapFace face,
uint32_t z, uint32_t array, size_t *stride) {
if ((lod >= mHal.drvState.lodCount) ||
(z && (z >= mHal.drvState.lod[lod].dimZ)) ||
((face != RS_ALLOCATION_CUBEMAP_FACE_POSITIVE_X) && !mHal.state.hasFaces) ||
(array != 0)) {
return nullptr;
}
if (mRSC->mHal.funcs.allocation.getPointer != nullptr) {
// Notify the driver, if present that the user is mapping the buffer
mRSC->mHal.funcs.allocation.getPointer(rsc, this, lod, face, z, array);
}
size_t s = 0;
if ((stride != nullptr) && mHal.drvState.lod[0].dimY) {
*stride = mHal.drvState.lod[lod].stride;
}
return mHal.drvState.lod[lod].mallocPtr;
}
void Allocation::data(Context *rsc, uint32_t xoff, uint32_t lod,
uint32_t count, const void *data, size_t sizeBytes) {
const size_t eSize = mHal.state.type->getElementSizeBytes();
if ((count * eSize) != sizeBytes) {
char buf[1024];
snprintf(buf, sizeof(buf),
"Allocation::subData called with mismatched size expected %zu, got %zu",
(count * eSize), sizeBytes);
rsc->setError(RS_ERROR_BAD_VALUE, buf);
mHal.state.type->dumpLOGV("type info");
return;
}
rsc->mHal.funcs.allocation.data1D(rsc, this, xoff, lod, count, data, sizeBytes);
sendDirty(rsc);
}
void Allocation::data(Context *rsc, uint32_t xoff, uint32_t yoff, uint32_t lod, RsAllocationCubemapFace face,
uint32_t w, uint32_t h, const void *data, size_t sizeBytes, size_t stride) {
rsc->mHal.funcs.allocation.data2D(rsc, this, xoff, yoff, lod, face, w, h, data, sizeBytes, stride);
sendDirty(rsc);
}
void Allocation::data(Context *rsc, uint32_t xoff, uint32_t yoff, uint32_t zoff,
uint32_t lod,
uint32_t w, uint32_t h, uint32_t d, const void *data, size_t sizeBytes, size_t stride) {
rsc->mHal.funcs.allocation.data3D(rsc, this, xoff, yoff, zoff, lod, w, h, d, data, sizeBytes, stride);
sendDirty(rsc);
}
void Allocation::read(Context *rsc, uint32_t xoff, uint32_t lod,
uint32_t count, void *data, size_t sizeBytes) {
const size_t eSize = mHal.state.type->getElementSizeBytes();
if ((count * eSize) != sizeBytes) {
char buf[1024];
snprintf(buf, sizeof(buf),
"Allocation::read called with mismatched size expected %zu, got %zu",
(count * eSize), sizeBytes);
rsc->setError(RS_ERROR_BAD_VALUE, buf);
mHal.state.type->dumpLOGV("type info");
return;
}
rsc->mHal.funcs.allocation.read1D(rsc, this, xoff, lod, count, data, sizeBytes);
}
void Allocation::read(Context *rsc, uint32_t xoff, uint32_t yoff, uint32_t lod, RsAllocationCubemapFace face,
uint32_t w, uint32_t h, void *data, size_t sizeBytes, size_t stride) {
const size_t eSize = mHal.state.elementSizeBytes;
const size_t lineSize = eSize * w;
if (!stride) {
stride = lineSize;
} else {
if ((lineSize * h) != sizeBytes) {
char buf[1024];
snprintf(buf, sizeof(buf), "Allocation size mismatch, expected %zu, got %zu",
(lineSize * h), sizeBytes);
rsc->setError(RS_ERROR_BAD_VALUE, buf);
return;
}
}
rsc->mHal.funcs.allocation.read2D(rsc, this, xoff, yoff, lod, face, w, h, data, sizeBytes, stride);
}
void Allocation::read(Context *rsc, uint32_t xoff, uint32_t yoff, uint32_t zoff, uint32_t lod,
uint32_t w, uint32_t h, uint32_t d, void *data, size_t sizeBytes, size_t stride) {
const size_t eSize = mHal.state.elementSizeBytes;
const size_t lineSize = eSize * w;
if (!stride) {
stride = lineSize;
}
rsc->mHal.funcs.allocation.read3D(rsc, this, xoff, yoff, zoff, lod, w, h, d, data, sizeBytes, stride);
}
void Allocation::elementData(Context *rsc, uint32_t x, uint32_t y, uint32_t z,
const void *data, uint32_t cIdx, size_t sizeBytes) {
if (x >= mHal.drvState.lod[0].dimX) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData X offset out of range.");
return;
}
if (y > 0 && y >= mHal.drvState.lod[0].dimY) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData Y offset out of range.");
return;
}
if (z > 0 && z >= mHal.drvState.lod[0].dimZ) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData Z offset out of range.");
return;
}
if (cIdx >= mHal.state.type->getElement()->getFieldCount()) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData component out of range.");
return;
}
const Element * e = mHal.state.type->getElement()->getField(cIdx);
uint32_t elemArraySize = mHal.state.type->getElement()->getFieldArraySize(cIdx);
if (sizeBytes != e->getSizeBytes() * elemArraySize) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData bad size.");
return;
}
rsc->mHal.funcs.allocation.elementData(rsc, this, x, y, z, data, cIdx, sizeBytes);
sendDirty(rsc);
}
void Allocation::elementRead(Context *rsc, uint32_t x, uint32_t y, uint32_t z,
void *data, uint32_t cIdx, size_t sizeBytes) {
if (x >= mHal.drvState.lod[0].dimX) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData X offset out of range.");
return;
}
if (y > 0 && y >= mHal.drvState.lod[0].dimY) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData Y offset out of range.");
return;
}
if (z > 0 && z >= mHal.drvState.lod[0].dimZ) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData Z offset out of range.");
return;
}
if (cIdx >= mHal.state.type->getElement()->getFieldCount()) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData component out of range.");
return;
}
const Element * e = mHal.state.type->getElement()->getField(cIdx);
uint32_t elemArraySize = mHal.state.type->getElement()->getFieldArraySize(cIdx);
if (sizeBytes != e->getSizeBytes() * elemArraySize) {
rsc->setError(RS_ERROR_BAD_VALUE, "subElementData bad size.");
return;
}
rsc->mHal.funcs.allocation.elementRead(rsc, this, x, y, z, data, cIdx, sizeBytes);
}
void Allocation::addProgramToDirty(const Program *p) {
mToDirtyList.push_back(p);
}
void Allocation::removeProgramToDirty(const Program *p) {
for (size_t ct=0; ct < mToDirtyList.size(); ct++) {
if (mToDirtyList[ct] == p) {
mToDirtyList.erase(mToDirtyList.begin() + ct);
return;
}
}
rsAssert(0);
}
void Allocation::dumpLOGV(const char *prefix) const {
ObjectBase::dumpLOGV(prefix);
char buf[1024];
if ((strlen(prefix) + 10) < sizeof(buf)) {
snprintf(buf, sizeof(buf), "%s type ", prefix);
if (mHal.state.type) {
mHal.state.type->dumpLOGV(buf);
}
}
ALOGV("%s allocation ptr=%p mUsageFlags=0x04%x, mMipmapControl=0x%04x",
prefix, mHal.drvState.lod[0].mallocPtr, mHal.state.usageFlags, mHal.state.mipmapControl);
}
uint32_t Allocation::getPackedSize() const {
uint32_t numItems = mHal.state.type->getCellCount();
return numItems * mHal.state.type->getElement()->getSizeBytesUnpadded();
}
void Allocation::writePackedData(Context *rsc, const Type *type,
uint8_t *dst, const uint8_t *src, bool dstPadded) {
const Element *elem = type->getElement();
uint32_t unpaddedBytes = elem->getSizeBytesUnpadded();
uint32_t paddedBytes = elem->getSizeBytes();
uint32_t numItems = type->getPackedSizeBytes() / paddedBytes;
uint32_t srcInc = !dstPadded ? paddedBytes : unpaddedBytes;
uint32_t dstInc = dstPadded ? paddedBytes : unpaddedBytes;
// no sub-elements
uint32_t fieldCount = elem->getFieldCount();
if (fieldCount == 0) {
for (uint32_t i = 0; i < numItems; i ++) {
memcpy(dst, src, unpaddedBytes);
src += srcInc;
dst += dstInc;
}
return;
}
// Cache offsets
uint32_t *offsetsPadded = new uint32_t[fieldCount];
uint32_t *offsetsUnpadded = new uint32_t[fieldCount];
uint32_t *sizeUnpadded = new uint32_t[fieldCount];
for (uint32_t i = 0; i < fieldCount; i++) {
offsetsPadded[i] = elem->getFieldOffsetBytes(i);
offsetsUnpadded[i] = elem->getFieldOffsetBytesUnpadded(i);
sizeUnpadded[i] = elem->getField(i)->getSizeBytesUnpadded();
}
uint32_t *srcOffsets = !dstPadded ? offsetsPadded : offsetsUnpadded;
uint32_t *dstOffsets = dstPadded ? offsetsPadded : offsetsUnpadded;
// complex elements, need to copy subelem after subelem
for (uint32_t i = 0; i < numItems; i ++) {
for (uint32_t fI = 0; fI < fieldCount; fI++) {
memcpy(dst + dstOffsets[fI], src + srcOffsets[fI], sizeUnpadded[fI]);
}
src += srcInc;
dst += dstInc;
}
delete[] offsetsPadded;
delete[] offsetsUnpadded;
delete[] sizeUnpadded;
}
void Allocation::unpackVec3Allocation(Context *rsc, const void *data, size_t dataSize) {
const uint8_t *src = (const uint8_t*)data;
uint8_t *dst = (uint8_t *)rsc->mHal.funcs.allocation.lock1D(rsc, this);
writePackedData(rsc, getType(), dst, src, true);
rsc->mHal.funcs.allocation.unlock1D(rsc, this);
}
void Allocation::packVec3Allocation(Context *rsc, OStream *stream) const {
uint32_t unpaddedBytes = getType()->getElement()->getSizeBytesUnpadded();
uint32_t numItems = mHal.state.type->getCellCount();
const uint8_t *src = (const uint8_t*)rsc->mHal.funcs.allocation.lock1D(rsc, this);
uint8_t *dst = new uint8_t[numItems * unpaddedBytes];
writePackedData(rsc, getType(), dst, src, false);
stream->addByteArray(dst, getPackedSize());
delete[] dst;
rsc->mHal.funcs.allocation.unlock1D(rsc, this);
}
void Allocation::serialize(Context *rsc, OStream *stream) const {
// Need to identify ourselves
stream->addU32((uint32_t)getClassId());
stream->addString(getName());
// First thing we need to serialize is the type object since it will be needed
// to initialize the class
mHal.state.type->serialize(rsc, stream);
uint32_t dataSize = mHal.state.type->getPackedSizeBytes();
// 3 element vectors are padded to 4 in memory, but padding isn't serialized
uint32_t packedSize = getPackedSize();
// Write how much data we are storing
stream->addU32(packedSize);
if (dataSize == packedSize) {
// Now write the data
stream->addByteArray(rsc->mHal.funcs.allocation.lock1D(rsc, this), dataSize);
rsc->mHal.funcs.allocation.unlock1D(rsc, this);
} else {
// Now write the data
packVec3Allocation(rsc, stream);
}
}
Allocation *Allocation::createFromStream(Context *rsc, IStream *stream) {
// First make sure we are reading the correct object
RsA3DClassID classID = (RsA3DClassID)stream->loadU32();
if (classID != RS_A3D_CLASS_ID_ALLOCATION) {
rsc->setError(RS_ERROR_FATAL_DRIVER,
"allocation loading failed due to corrupt file. (invalid id)\n");
return nullptr;
}
const char *name = stream->loadString();
Type *type = Type::createFromStream(rsc, stream);
if (!type) {
return nullptr;
}
type->compute();
Allocation *alloc = Allocation::createAllocation(rsc, type, RS_ALLOCATION_USAGE_SCRIPT);
type->decUserRef();
// Number of bytes we wrote out for this allocation
uint32_t dataSize = stream->loadU32();
// 3 element vectors are padded to 4 in memory, but padding isn't serialized
uint32_t packedSize = alloc->getPackedSize();
if (dataSize != type->getPackedSizeBytes() &&
dataSize != packedSize) {
rsc->setError(RS_ERROR_FATAL_DRIVER,
"allocation loading failed due to corrupt file. (invalid size)\n");
ObjectBase::checkDelete(alloc);
ObjectBase::checkDelete(type);
return nullptr;
}
alloc->assignName(name);
if (dataSize == type->getPackedSizeBytes()) {
uint32_t count = dataSize / type->getElementSizeBytes();
// Read in all of our allocation data
alloc->data(rsc, 0, 0, count, stream->getPtr() + stream->getPos(), dataSize);
} else {
alloc->unpackVec3Allocation(rsc, stream->getPtr() + stream->getPos(), dataSize);
}
stream->reset(stream->getPos() + dataSize);
return alloc;
}
void Allocation::sendDirty(const Context *rsc) const {
#ifndef RS_COMPATIBILITY_LIB
for (size_t ct=0; ct < mToDirtyList.size(); ct++) {
mToDirtyList[ct]->forceDirty();
}
#endif
mRSC->mHal.funcs.allocation.markDirty(rsc, this);
}
void Allocation::incRefs(const void *ptr, size_t ct, size_t startOff) const {
mHal.state.type->incRefs(ptr, ct, startOff);
}
void Allocation::decRefs(const void *ptr, size_t ct, size_t startOff) const {
if (!mHal.state.hasReferences || !getIsScript()) {
return;
}
mHal.state.type->decRefs(ptr, ct, startOff);
}
void Allocation::callUpdateCacheObject(const Context *rsc, void *dstObj) const {
if (rsc->mHal.funcs.allocation.updateCachedObject != nullptr) {
rsc->mHal.funcs.allocation.updateCachedObject(rsc, this, (rs_allocation *)dstObj);
} else {
*((const void **)dstObj) = this;
}
}
void Allocation::freeChildrenUnlocked () {
void *ptr = mRSC->mHal.funcs.allocation.lock1D(mRSC, this);
decRefs(ptr, mHal.state.type->getCellCount(), 0);
mRSC->mHal.funcs.allocation.unlock1D(mRSC, this);
}
bool Allocation::freeChildren() {
if (mHal.state.hasReferences) {
incSysRef();
freeChildrenUnlocked();
return decSysRef();
}
return false;
}
void Allocation::copyRange1D(Context *rsc, const Allocation *src, int32_t srcOff, int32_t destOff, int32_t len) {
}
void Allocation::resize1D(Context *rsc, uint32_t dimX) {
uint32_t oldDimX = mHal.drvState.lod[0].dimX;
if (dimX == oldDimX) {
return;
}
ObjectBaseRef<Type> t = mHal.state.type->cloneAndResize1D(rsc, dimX);
if (dimX < oldDimX) {
decRefs(rsc->mHal.funcs.allocation.lock1D(rsc, this), oldDimX - dimX, dimX);
rsc->mHal.funcs.allocation.unlock1D(rsc, this);
}
rsc->mHal.funcs.allocation.resize(rsc, this, t.get(), mHal.state.hasReferences);
setType(t.get());
updateCache();
}
void Allocation::resize2D(Context *rsc, uint32_t dimX, uint32_t dimY) {
rsc->setError(RS_ERROR_FATAL_DRIVER, "resize2d not implemented");
}
void Allocation::setupGrallocConsumer(const Context *rsc, uint32_t numAlloc) {
#ifndef RS_COMPATIBILITY_LIB
// Configure GrallocConsumer to be in asynchronous mode
if (numAlloc > MAX_NUM_ALLOC || numAlloc <= 0) {
rsc->setError(RS_ERROR_FATAL_DRIVER, "resize2d not implemented");
return;
}
mGrallocConsumer = new GrallocConsumer(rsc, this, numAlloc);
mCurrentIdx = 0;
mBufferQueueInited = true;
#endif
}
void * Allocation::getSurface(const Context *rsc) {
#ifndef RS_COMPATIBILITY_LIB
// Configure GrallocConsumer to be in asynchronous mode
if (!mBufferQueueInited) {
// This case is only used for single frame processing,
// since we will always call setupGrallocConsumer first in
// multi-frame case.
setupGrallocConsumer(rsc, 1);
}
return mGrallocConsumer->getNativeWindow();
#else
return nullptr;
#endif
}
void Allocation::shareBufferQueue(const Context *rsc, const Allocation *alloc) {
#ifndef RS_COMPATIBILITY_LIB
mGrallocConsumer = alloc->mGrallocConsumer;
mCurrentIdx = mGrallocConsumer->getNextAvailableIdx(this);
if (mCurrentIdx >= mGrallocConsumer->mNumAlloc) {
rsc->setError(RS_ERROR_DRIVER, "Maximum allocations attached to a BufferQueue");
return;
}
mBufferQueueInited = true;
#endif
}
void Allocation::setSurface(const Context *rsc, RsNativeWindow sur) {
ANativeWindow *nw = (ANativeWindow *)sur;
rsc->mHal.funcs.allocation.setSurface(rsc, this, nw);
}
void Allocation::ioSend(const Context *rsc) {
rsc->mHal.funcs.allocation.ioSend(rsc, this);
}
void Allocation::ioReceive(const Context *rsc) {
void *ptr = nullptr;
size_t stride = 0;
#ifndef RS_COMPATIBILITY_LIB
if (mHal.state.usageFlags & RS_ALLOCATION_USAGE_SCRIPT) {
media_status_t ret = mGrallocConsumer->lockNextBuffer(mCurrentIdx);
if (ret == AMEDIA_OK) {
rsc->mHal.funcs.allocation.ioReceive(rsc, this);
} else if (ret == AMEDIA_IMGREADER_NO_BUFFER_AVAILABLE) {
// No new frame, don't do anything
} else {
rsc->setError(RS_ERROR_DRIVER, "Error receiving IO input buffer.");
}
}
#endif
}
bool Allocation::hasSameDims(const Allocation *other) const {
const Type *type0 = this->getType(),
*type1 = other->getType();
return (type0->getCellCount() == type1->getCellCount()) &&
(type0->getDimLOD() == type1->getDimLOD()) &&
(type0->getDimFaces() == type1->getDimFaces()) &&
(type0->getDimYuv() == type1->getDimYuv()) &&
(type0->getDimX() == type1->getDimX()) &&
(type0->getDimY() == type1->getDimY()) &&
(type0->getDimZ() == type1->getDimZ());
}
/////////////////
//
void rsi_AllocationSyncAll(Context *rsc, RsAllocation va, RsAllocationUsageType src) {
Allocation *a = static_cast<Allocation *>(va);
a->sendDirty(rsc);
a->syncAll(rsc, src);
}
void rsi_AllocationGenerateMipmaps(Context *rsc, RsAllocation va) {
Allocation *alloc = static_cast<Allocation *>(va);
rsc->mHal.funcs.allocation.generateMipmaps(rsc, alloc);
}
void rsi_AllocationCopyToBitmap(Context *rsc, RsAllocation va, void *data, size_t sizeBytes) {
Allocation *a = static_cast<Allocation *>(va);
const Type * t = a->getType();
a->read(rsc, 0, 0, 0, RS_ALLOCATION_CUBEMAP_FACE_POSITIVE_X,
t->getDimX(), t->getDimY(), data, sizeBytes, 0);
}
void rsi_Allocation1DData(Context *rsc, RsAllocation va, uint32_t xoff, uint32_t lod,
uint32_t count, const void *data, size_t sizeBytes) {
Allocation *a = static_cast<Allocation *>(va);
a->data(rsc, xoff, lod, count, data, sizeBytes);
}
void rsi_Allocation1DElementData(Context *rsc, RsAllocation va, uint32_t x,
uint32_t lod, const void *data, size_t sizeBytes, size_t eoff) {
Allocation *a = static_cast<Allocation *>(va);
a->elementData(rsc, x, 0, 0, data, eoff, sizeBytes);
}
void rsi_AllocationElementData(Context *rsc, RsAllocation va, uint32_t x, uint32_t y, uint32_t z,
uint32_t lod, const void *data, size_t sizeBytes, size_t eoff) {
Allocation *a = static_cast<Allocation *>(va);
a->elementData(rsc, x, y, z, data, eoff, sizeBytes);
}
void rsi_Allocation2DData(Context *rsc, RsAllocation va, uint32_t xoff, uint32_t yoff, uint32_t lod, RsAllocationCubemapFace face,
uint32_t w, uint32_t h, const void *data, size_t sizeBytes, size_t stride) {
Allocation *a = static_cast<Allocation *>(va);
a->data(rsc, xoff, yoff, lod, face, w, h, data, sizeBytes, stride);
}
void rsi_Allocation3DData(Context *rsc, RsAllocation va, uint32_t xoff, uint32_t yoff, uint32_t zoff, uint32_t lod,
uint32_t w, uint32_t h, uint32_t d, const void *data, size_t sizeBytes, size_t stride) {
Allocation *a = static_cast<Allocation *>(va);
a->data(rsc, xoff, yoff, zoff, lod, w, h, d, data, sizeBytes, stride);
}
void rsi_AllocationRead(Context *rsc, RsAllocation va, void *data, size_t sizeBytes) {
Allocation *a = static_cast<Allocation *>(va);
const Type * t = a->getType();
if(t->getDimZ()) {
a->read(rsc, 0, 0, 0, 0, t->getDimX(), t->getDimY(), t->getDimZ(),
data, sizeBytes, 0);
} else if(t->getDimY()) {
a->read(rsc, 0, 0, 0, RS_ALLOCATION_CUBEMAP_FACE_POSITIVE_X,
t->getDimX(), t->getDimY(), data, sizeBytes, 0);
} else {
a->read(rsc, 0, 0, t->getDimX(), data, sizeBytes);
}
}
void rsi_AllocationResize1D(Context *rsc, RsAllocation va, uint32_t dimX) {
Allocation *a = static_cast<Allocation *>(va);
a->resize1D(rsc, dimX);
}
void rsi_AllocationResize2D(Context *rsc, RsAllocation va, uint32_t dimX, uint32_t dimY) {
Allocation *a = static_cast<Allocation *>(va);
a->resize2D(rsc, dimX, dimY);
}
RsAllocation rsi_AllocationCreateTyped(Context *rsc, RsType vtype,
RsAllocationMipmapControl mipmaps,
uint32_t usages, uintptr_t ptr) {
Allocation * alloc = Allocation::createAllocation(rsc, static_cast<Type *>(vtype), usages, mipmaps, (void*)ptr);
if (!alloc) {
return nullptr;
}
alloc->incUserRef();
return alloc;
}
RsAllocation rsi_AllocationCreateStrided(Context *rsc, RsType vtype,
RsAllocationMipmapControl mipmaps,
uint32_t usages, uintptr_t ptr,
size_t requiredAlignment) {
Allocation * alloc = Allocation::createAllocationStrided(rsc, static_cast<Type *>(vtype), usages, mipmaps,
(void*)ptr, requiredAlignment);
if (!alloc) {
return nullptr;
}
alloc->incUserRef();
return alloc;
}
RsAllocation rsi_AllocationCreateFromBitmap(Context *rsc, RsType vtype,
RsAllocationMipmapControl mipmaps,
const void *data, size_t sizeBytes, uint32_t usages) {
Type *t = static_cast<Type *>(vtype);
RsAllocation vTexAlloc = rsi_AllocationCreateTyped(rsc, vtype, mipmaps, usages, 0);
Allocation *texAlloc = static_cast<Allocation *>(vTexAlloc);
if (texAlloc == nullptr) {
ALOGE("Memory allocation failure");
return nullptr;
}
texAlloc->data(rsc, 0, 0, 0, RS_ALLOCATION_CUBEMAP_FACE_POSITIVE_X,
t->getDimX(), t->getDimY(), data, sizeBytes, 0);
if (mipmaps == RS_ALLOCATION_MIPMAP_FULL) {
rsc->mHal.funcs.allocation.generateMipmaps(rsc, texAlloc);
}
texAlloc->sendDirty(rsc);
return texAlloc;
}
RsAllocation rsi_AllocationCubeCreateFromBitmap(Context *rsc, RsType vtype,
RsAllocationMipmapControl mipmaps,
const void *data, size_t sizeBytes, uint32_t usages) {
Type *t = static_cast<Type *>(vtype);
// Cubemap allocation's faces should be Width by Width each.
// Source data should have 6 * Width by Width pixels
// Error checking is done in the java layer
RsAllocation vTexAlloc = rsi_AllocationCreateTyped(rsc, vtype, mipmaps, usages, 0);
Allocation *texAlloc = static_cast<Allocation *>(vTexAlloc);
if (texAlloc == nullptr) {
ALOGE("Memory allocation failure");
return nullptr;
}
uint32_t faceSize = t->getDimX();
uint32_t strideBytes = faceSize * 6 * t->getElementSizeBytes();
uint32_t copySize = faceSize * t->getElementSizeBytes();
uint8_t *sourcePtr = (uint8_t*)data;
for (uint32_t face = 0; face < 6; face ++) {
for (uint32_t dI = 0; dI < faceSize; dI ++) {
texAlloc->data(rsc, 0, dI, 0, (RsAllocationCubemapFace)face,
t->getDimX(), 1, sourcePtr + strideBytes * dI, copySize, 0);
}
// Move the data pointer to the next cube face
sourcePtr += copySize;
}
if (mipmaps == RS_ALLOCATION_MIPMAP_FULL) {
rsc->mHal.funcs.allocation.generateMipmaps(rsc, texAlloc);
}
texAlloc->sendDirty(rsc);
return texAlloc;
}
void rsi_AllocationCopy2DRange(Context *rsc,
RsAllocation dstAlloc,
uint32_t dstXoff, uint32_t dstYoff,
uint32_t dstMip, uint32_t dstFace,
uint32_t width, uint32_t height,
RsAllocation srcAlloc,
uint32_t srcXoff, uint32_t srcYoff,
uint32_t srcMip, uint32_t srcFace) {
Allocation *dst = static_cast<Allocation *>(dstAlloc);
Allocation *src= static_cast<Allocation *>(srcAlloc);
rsc->mHal.funcs.allocation.allocData2D(rsc, dst, dstXoff, dstYoff, dstMip,
(RsAllocationCubemapFace)dstFace,
width, height,
src, srcXoff, srcYoff,srcMip,
(RsAllocationCubemapFace)srcFace);
}
void rsi_AllocationCopy3DRange(Context *rsc,
RsAllocation dstAlloc,
uint32_t dstXoff, uint32_t dstYoff, uint32_t dstZoff,
uint32_t dstMip,
uint32_t width, uint32_t height, uint32_t depth,
RsAllocation srcAlloc,
uint32_t srcXoff, uint32_t srcYoff, uint32_t srcZoff,
uint32_t srcMip) {
Allocation *dst = static_cast<Allocation *>(dstAlloc);
Allocation *src= static_cast<Allocation *>(srcAlloc);
rsc->mHal.funcs.allocation.allocData3D(rsc, dst, dstXoff, dstYoff, dstZoff, dstMip,
width, height, depth,
src, srcXoff, srcYoff, srcZoff, srcMip);
}
void rsi_AllocationSetupBufferQueue(Context *rsc, RsAllocation valloc, uint32_t numAlloc) {
Allocation *alloc = static_cast<Allocation *>(valloc);
alloc->setupGrallocConsumer(rsc, numAlloc);
}
void * rsi_AllocationGetSurface(Context *rsc, RsAllocation valloc) {
Allocation *alloc = static_cast<Allocation *>(valloc);
void *s = alloc->getSurface(rsc);
return s;
}
void rsi_AllocationShareBufferQueue(Context *rsc, RsAllocation valloc1, RsAllocation valloc2) {
Allocation *alloc1 = static_cast<Allocation *>(valloc1);
Allocation *alloc2 = static_cast<Allocation *>(valloc2);
alloc1->shareBufferQueue(rsc, alloc2);
}
void rsi_AllocationSetSurface(Context *rsc, RsAllocation valloc, RsNativeWindow sur) {
Allocation *alloc = static_cast<Allocation *>(valloc);
alloc->setSurface(rsc, sur);
}
void rsi_AllocationIoSend(Context *rsc, RsAllocation valloc) {
Allocation *alloc = static_cast<Allocation *>(valloc);
alloc->ioSend(rsc);
}
int64_t rsi_AllocationIoReceive(Context *rsc, RsAllocation valloc) {
Allocation *alloc = static_cast<Allocation *>(valloc);
alloc->ioReceive(rsc);
return alloc->getTimeStamp();
}
void *rsi_AllocationGetPointer(Context *rsc, RsAllocation valloc,
uint32_t lod, RsAllocationCubemapFace face,
uint32_t z, uint32_t array, size_t *stride, size_t strideLen) {
Allocation *alloc = static_cast<Allocation *>(valloc);
rsAssert(strideLen == sizeof(size_t));
return alloc->getPointer(rsc, lod, face, z, array, stride);
}
void rsi_Allocation1DRead(Context *rsc, RsAllocation va, uint32_t xoff, uint32_t lod,
uint32_t count, void *data, size_t sizeBytes) {
Allocation *a = static_cast<Allocation *>(va);
rsc->mHal.funcs.allocation.read1D(rsc, a, xoff, lod, count, data, sizeBytes);
}
void rsi_AllocationElementRead(Context *rsc, RsAllocation va, uint32_t x, uint32_t y, uint32_t z,
uint32_t lod, void *data, size_t sizeBytes, size_t eoff) {
Allocation *a = static_cast<Allocation *>(va);
a->elementRead(rsc, x, y, z, data, eoff, sizeBytes);
}
void rsi_Allocation2DRead(Context *rsc, RsAllocation va, uint32_t xoff, uint32_t yoff,
uint32_t lod, RsAllocationCubemapFace face, uint32_t w,
uint32_t h, void *data, size_t sizeBytes, size_t stride) {
Allocation *a = static_cast<Allocation *>(va);
a->read(rsc, xoff, yoff, lod, face, w, h, data, sizeBytes, stride);
}
void rsi_Allocation3DRead(Context *rsc, RsAllocation va,
uint32_t xoff, uint32_t yoff, uint32_t zoff,
uint32_t lod, uint32_t w, uint32_t h, uint32_t d,
void *data, size_t sizeBytes, size_t stride) {
Allocation *a = static_cast<Allocation *>(va);
a->read(rsc, xoff, yoff, zoff, lod, w, h, d, data, sizeBytes, stride);
}
RsAllocation rsi_AllocationAdapterCreate(Context *rsc, RsType vwindow, RsAllocation vbase) {
Allocation * alloc = Allocation::createAdapter(rsc,
static_cast<Allocation *>(vbase), static_cast<Type *>(vwindow));
if (!alloc) {
return nullptr;
}
alloc->incUserRef();
return alloc;
}
void rsi_AllocationAdapterOffset(Context *rsc, RsAllocation va, const uint32_t *offsets, size_t len) {
Allocation *a = static_cast<Allocation *>(va);
a->adapterOffset(rsc, offsets, len);
}
} // namespace renderscript
} // namespace android