mirror of https://github.com/alibaba/MNN.git
812 lines
37 KiB
C++
812 lines
37 KiB
C++
//
|
|
// ConvInt8TiledExecutor.cpp
|
|
// MNN
|
|
//
|
|
// Created by MNN on 2019/5/17.
|
|
// Copyright © 2018, Alibaba Group Holding Limited
|
|
//
|
|
|
|
#include "ConvInt8TiledExecutor.hpp"
|
|
#include "ConvolutionTiledExecutor.hpp"
|
|
#include "core/Macro.h"
|
|
#include "core/BufferAllocator.hpp"
|
|
|
|
#include <math.h>
|
|
#include "backend/cpu/CPUBackend.hpp"
|
|
#include "core/Concurrency.h"
|
|
#include "core/TensorUtils.hpp"
|
|
|
|
namespace MNN {
|
|
|
|
ConvInt8TiledExecutor::ConvInt8TiledExecutor(Backend* backend, const Convolution2D* convOp, std::shared_ptr<ResourceInt8> res): CPUConvolution(convOp->common(), backend), mResourceInt8(res), mMutableResource(res, backend) {
|
|
mValid = mMutableResource.mValid;
|
|
}
|
|
|
|
ConvInt8TiledExecutor::~ConvInt8TiledExecutor() {
|
|
// Do nothing
|
|
}
|
|
|
|
bool ConvInt8TiledExecutor::onClone(Backend* bn, const Op* op, Execution** dst) {
|
|
return false;
|
|
}
|
|
|
|
ErrorCode ConvInt8TiledExecutor::onResize(const std::vector<Tensor*>& inputs, const std::vector<Tensor*>& outputs) {
|
|
mMutableResource.updateInputOutputScale(TensorUtils::getQuantInfo(inputs[0]), TensorUtils::getQuantInfo(outputs[0]));
|
|
CPUConvolution::onResize(inputs, outputs);
|
|
ConvolutionTiledExecutor::setIm2ColParameter(mIm2ColParamter, mCommon, inputs[0], outputs[0], mPadX, mPadY, static_cast<CPUBackend*>(backend())->functions(), static_cast<CPUBackend*>(backend())->int8Functions());
|
|
return NO_ERROR;
|
|
}
|
|
|
|
void ConvInt8TiledExecutor::reorderWeight(Tensor* weight, const uint8_t* weightSrc, int SRC_UNIT, int UNIT, int ic, int oc, int kernelCount, int pack) {
|
|
auto weightDst = weight->host<uint8_t>();
|
|
memset(weightDst, 0, weight->size());
|
|
if (SRC_UNIT > pack) {
|
|
auto icDivU = UP_DIV(ic, pack);
|
|
for (int k = 0; k < kernelCount; ++k) {
|
|
const auto srcK = weightSrc + k;
|
|
for (int y = 0; y < ic; ++y) {
|
|
const int yOutSide = y / pack;
|
|
const int yInSide = y % pack;
|
|
const int yIndex = yOutSide + k * icDivU;
|
|
const int ySubOutSide = yIndex / (SRC_UNIT / pack);
|
|
const int ySubInSide = yIndex % (SRC_UNIT / pack);
|
|
|
|
auto dstY = weightDst + ySubOutSide * weight->stride(1) + ySubInSide * pack + yInSide;
|
|
const auto srcY = srcK + y * kernelCount;
|
|
for (int x = 0; x < oc; ++x) {
|
|
const int xOutSide = x / UNIT;
|
|
const int xInSide = x % UNIT;
|
|
const int dstIndex = xOutSide * weight->stride(0) + xInSide * SRC_UNIT;
|
|
const int srcIndex = x * kernelCount * ic;
|
|
dstY[dstIndex] = srcY[srcIndex];
|
|
}
|
|
}
|
|
}
|
|
} else {
|
|
for (int k = 0; k < kernelCount; ++k) {
|
|
auto icDivU = UP_DIV(ic, SRC_UNIT);
|
|
const auto srcK = weightSrc + k;
|
|
for (int y = 0; y < ic; ++y) {
|
|
const int yOutSide = y / SRC_UNIT;
|
|
const int yInSide = y % SRC_UNIT;
|
|
|
|
auto dstY = weightDst + (yOutSide + k * icDivU) * weight->stride(1) + yInSide;
|
|
const auto srcY = srcK + y * kernelCount;
|
|
for (int x = 0; x < oc; ++x) {
|
|
const int xOutSide = x / UNIT;
|
|
const int xInSide = x % UNIT;
|
|
const int dstIndex = xOutSide * weight->stride(0) + xInSide * SRC_UNIT;
|
|
const int srcIndex = x * kernelCount * ic;
|
|
dstY[dstIndex] = srcY[srcIndex];
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
static bool _reorderWeightInside(Backend* bn, const Convolution2DCommon* common,
|
|
const std::shared_ptr<Tensor>& weightOrigin,
|
|
std::shared_ptr<Tensor>& weight) {
|
|
auto core = static_cast<CPUBackend*>(bn)->int8Functions();
|
|
auto gcore = static_cast<CPUBackend*>(bn)->functions();
|
|
int UNIT, SRC_UNIT, DST_XUNIT;
|
|
core->MNNGetGemmUnit(&UNIT, &SRC_UNIT, &DST_XUNIT);
|
|
// reorder weight, [oc, ic, k^2] => [oc/unit, ((ic/unit)*k^2)/(src_unit/unit), unit(oc), (src_unit/unit), unit(ic)]
|
|
int oc = common->outputCount(), ic = common->inputCount(), kernelCount = common->kernelX() * common->kernelY();
|
|
std::vector<int> shape;
|
|
int pack = gcore->pack;
|
|
if (gcore->bytes == 2 && gcore->pack == 8) {
|
|
pack = 4;
|
|
}
|
|
if (SRC_UNIT > pack) {
|
|
MNN_ASSERT(SRC_UNIT % UNIT == 0);
|
|
shape = {UP_DIV(oc, UNIT), UP_DIV(UP_DIV(ic, pack) * kernelCount, SRC_UNIT / pack), UNIT, SRC_UNIT};
|
|
} else {
|
|
shape = {UP_DIV(oc, UNIT), UP_DIV(ic, SRC_UNIT) * kernelCount, UNIT, SRC_UNIT};
|
|
}
|
|
|
|
weight.reset(Tensor::createDevice<int8_t>(shape));
|
|
|
|
bool succ = bn->onAcquireBuffer(weight.get(), Backend::STATIC);
|
|
if (!succ) {
|
|
MNN_ERROR("Memory not enough");
|
|
return false;
|
|
}
|
|
ConvInt8TiledExecutor::reorderWeight(weight.get(), weightOrigin->host<uint8_t>(), SRC_UNIT, UNIT, ic, oc, kernelCount, pack);
|
|
return true;
|
|
}
|
|
|
|
static void Getfp32Info (std::shared_ptr<CPUConvolution::Resource> resource, std::shared_ptr<Tensor> weightOrigin, const Convolution2D* conv2d, std::shared_ptr<ConvolutionCommon::Int8Common> quantCommon) {
|
|
// common parameters
|
|
int outputCount = conv2d->common()->outputCount();
|
|
auto core = static_cast<CPUBackend*>(resource->backend)->functions();
|
|
int LSize = conv2d->common()->inputCount() * conv2d->common()->kernelX() * conv2d->common()->kernelY();
|
|
int ocUp4 = ROUND_UP(outputCount, core->pack);
|
|
|
|
int dequantCnt = quantCommon->alpha.size();
|
|
if (quantCommon->asymmetric) {
|
|
dequantCnt /= 2;
|
|
}
|
|
int blockNum = dequantCnt / outputCount;
|
|
int scaleSize = blockNum * ocUp4; // pack size.
|
|
int blockSize = LSize / blockNum;
|
|
int originOffset = 0;
|
|
if (quantCommon->canUseInt4) {
|
|
originOffset = -8;
|
|
}
|
|
|
|
// Save weight quant scale and bias: wf=scale*wi+bias
|
|
int bytes = 4;
|
|
resource->mDequantize.mScaleBias.reset(Tensor::createDevice<uint8_t>({2 * scaleSize * bytes}));
|
|
auto success = resource->backend->onAcquireBuffer(resource->mDequantize.mScaleBias.get(), Backend::STATIC);
|
|
if (!success) {
|
|
MNN_ERROR("Alloc denquant scaleBias memory error\n");
|
|
return;
|
|
}
|
|
auto alphaPtr = resource->mDequantize.mScaleBias->host<float>();
|
|
auto biasPtr = reinterpret_cast<float*>(reinterpret_cast<uint8_t*>(alphaPtr) + scaleSize * bytes);
|
|
::memset(alphaPtr, 1, scaleSize * bytes);
|
|
::memset(biasPtr, 0, scaleSize * bytes);
|
|
auto quanInfoPtr = quantCommon->alpha.get();
|
|
int h = quantCommon->alpha.size();
|
|
if (quantCommon->asymmetric) {
|
|
for (int i = 0; i < blockNum; ++i) {
|
|
auto dstAlpha = alphaPtr + i * ocUp4;
|
|
auto dstBias = biasPtr + i * ocUp4;
|
|
for (int j = 0; j < outputCount; ++j) {
|
|
int scaleIndex = j * blockNum + i;
|
|
dstAlpha[j] = quanInfoPtr[2 * scaleIndex + 1];
|
|
dstBias[j] = quanInfoPtr[2 * scaleIndex] + (float)originOffset * dstAlpha[j];
|
|
}
|
|
}
|
|
|
|
} else {
|
|
for (int i = 0; i < blockNum; ++i) {
|
|
auto dstAlpha = alphaPtr + i * ocUp4;
|
|
auto dstBias = biasPtr + i * ocUp4;
|
|
for (int j = 0; j < outputCount; ++j) {
|
|
int scaleIndex = j * blockNum + i;
|
|
dstAlpha[j] = quanInfoPtr[scaleIndex];
|
|
dstBias[j] = (float)originOffset * dstAlpha[j];
|
|
}
|
|
}
|
|
}
|
|
// Save float weight kernel sum
|
|
resource->mWeightKernelSum.reset(Tensor::createDevice<uint8_t>({bytes * ocUp4}));
|
|
success = resource->backend->onAcquireBuffer(resource->mWeightKernelSum.get(), Backend::STATIC);
|
|
if (!success) {
|
|
MNN_ERROR("Alloc denquant mWeightKernelSum memory error\n");
|
|
return;
|
|
}
|
|
auto weightKernelSum = resource->mWeightKernelSum->host<float>();
|
|
auto realWeightData = weightOrigin->host<int8_t>();
|
|
::memset(weightKernelSum, 0, resource->mWeightKernelSum->size());
|
|
for (int j = 0; j < outputCount; ++j) {
|
|
float sum = 0.f;
|
|
for (int k = 0; k < blockNum; ++k) {
|
|
int scaleIndex = k + j * blockNum;
|
|
float scale = 0;
|
|
float bias = 0;
|
|
if (quantCommon->asymmetric) {
|
|
scale = quanInfoPtr[2 * scaleIndex + 1];
|
|
bias = quanInfoPtr[2 * scaleIndex];
|
|
} else {
|
|
scale = quanInfoPtr[scaleIndex];
|
|
bias = 0;
|
|
}
|
|
int tmp = 0;
|
|
for (int i = 0; i < blockSize; ++i) {
|
|
int l_index = k * blockSize + i;
|
|
tmp += (int)realWeightData[j * blockNum * blockSize + l_index];
|
|
}
|
|
sum += (tmp * scale + blockSize * bias);
|
|
}
|
|
weightKernelSum[j] = sum;
|
|
}
|
|
}
|
|
|
|
DenseConvInt8TiledExecutor::DenseConvInt8TiledExecutor(Backend* backend, const Convolution2D* convOp, std::shared_ptr<ResourceInt8> res, bool dynamicQuantExe) : ConvInt8TiledExecutor(backend, convOp, res) {
|
|
std::shared_ptr<Tensor> weightOrigin = mResourceInt8->mWeightInt8;
|
|
std::shared_ptr<MNN::ConvolutionCommon::Int8Common> quanCommon ;
|
|
mDynamicQuantExe = dynamicQuantExe;
|
|
if (dynamicQuantExe) {
|
|
MNN_ASSERT(convOp->quanParameter() != nullptr && convOp->quanParameter()->buffer() != nullptr);
|
|
quanCommon = ConvolutionCommon::load(convOp, backend, false, true);
|
|
// fp32 weightKernelSum
|
|
mResource.reset(new CPUConvolution::Resource);
|
|
mResource->backend = backend;
|
|
Getfp32Info(mResource, weightOrigin, convOp, quanCommon); // Call this before reorder weight.
|
|
}
|
|
|
|
mValid = _reorderWeightInside(backend, convOp->common(), weightOrigin, mResourceInt8->mWeightInt8);
|
|
if(!mValid) {
|
|
return;
|
|
}
|
|
auto core = static_cast<CPUBackend*>(backend)->int8Functions();
|
|
auto gcore = static_cast<CPUBackend*>(backend)->functions();
|
|
// offline quant
|
|
if (false == dynamicQuantExe) {
|
|
mGemmKernel = core->Int8GemmKernel;
|
|
#ifdef MNN_USE_SSE
|
|
int actBits = convOp->symmetricQuan()->nbits();
|
|
if (actBits <= 7) {
|
|
mGemmKernel = core->Int8GemmKernelFast;
|
|
}
|
|
#else
|
|
if(convOp->symmetricQuan()->method() == QuantizeAlgo_OVERFLOW_AWARE){
|
|
mGemmKernel = core->Int8GemmKernelFast;
|
|
}
|
|
#endif
|
|
mResource.reset(new CPUConvolution::Resource);
|
|
CPUConvolution::makeResource(backend, mResource, convOp, mResourceInt8);
|
|
return;
|
|
}
|
|
|
|
// dynamic quant
|
|
int UNIT, SRC_UNIT, DST_XUNIT;
|
|
core->MNNGetGemmUnit(&UNIT, &SRC_UNIT, &DST_XUNIT);
|
|
bool needPermuteInt4weight = ((UNIT == 8 && SRC_UNIT == 8 && DST_XUNIT ==10) || (UNIT == 4 && SRC_UNIT == 8 && DST_XUNIT ==20) || (UNIT == 64 && SRC_UNIT == 4 && DST_XUNIT ==4));
|
|
mResource->mDequantize.bits = 8;
|
|
if (quanCommon->canUseInt4) {
|
|
mResourceInt8->mWeightAsymmetricQuant = true;
|
|
auto weightLength = mResourceInt8->mWeightInt8->size();
|
|
MNN_ASSERT(weightLength % 2 == 0);
|
|
mResource->mDequantize.bits = 4;
|
|
std::shared_ptr<MNN::Tensor> weightLow(Tensor::createDevice<uint8_t>( mResourceInt8->mWeightInt8->shape()));
|
|
auto res = mResource->backend->onAcquireBuffer(weightLow.get(), Backend::STATIC);
|
|
if (!res) {
|
|
MNN_ERROR("int4 weight acquire buffer error\n");
|
|
return ;
|
|
}
|
|
auto srcPtr = mResourceInt8->mWeightInt8->host<int8_t>();
|
|
auto dstPtr = weightLow->host<uint8_t>();
|
|
// Pack two int4-weight to one int8-weight.
|
|
if (false == needPermuteInt4weight) {
|
|
weightLength = UP_DIV(weightLength, 2);
|
|
for (int i=0; i < weightLength; ++i) {
|
|
int s0 = srcPtr[2 * i + 0];
|
|
int s1 = srcPtr[2 * i + 1];
|
|
int d = (s0 + 8) * 16 + (s1 + 8);
|
|
dstPtr[i] = d;
|
|
}
|
|
} else {
|
|
int permuteUnit = UNIT * SRC_UNIT;
|
|
int halfPermuteStride = static_cast<int32_t>(permuteUnit / 2);
|
|
for (int i = 0; i < weightLength / permuteUnit; ++i) {
|
|
auto src0 = srcPtr + i * permuteUnit;
|
|
auto dst0 = dstPtr + i * halfPermuteStride;
|
|
for (int j = 0; j < halfPermuteStride; ++j) {
|
|
int s0 = src0[j];
|
|
int s1 = src0[j + halfPermuteStride];
|
|
int d = (s0 + 8) * 16 + (s1 + 8);
|
|
dst0[j] = d;
|
|
}
|
|
}
|
|
}
|
|
// Update int4 weight to mWeightInt8.
|
|
mResourceInt8->mWeightInt8 = weightLow;
|
|
}
|
|
// Relu/Relu6 post parameters
|
|
auto postPtr = getPostParameters();
|
|
mResource->mReluThreshold.resize(2);
|
|
mResource->mReluThreshold[0] = postPtr[2];
|
|
mResource->mReluThreshold[1] = postPtr[3];
|
|
if (gcore->bytes == 2) {
|
|
gcore->MNNFp32ToLowp(mResource->mReluThreshold.data(), reinterpret_cast<int16_t*>(mResource->mReluThreshold.data()), 2);
|
|
}
|
|
if (mCommon->relu()) {
|
|
mResource->mReluThreshold[0] = 0.f;
|
|
}
|
|
if (mCommon->relu6()) {
|
|
mResource->mReluThreshold[0] = 0.f;
|
|
mResource->mReluThreshold[1] = 6.f;
|
|
}
|
|
}
|
|
|
|
DenseConvInt8TiledExecutor::DenseConvInt8TiledExecutor(Backend* backend, const Convolution2D* convOp, bool dynamicQuantExe, const DenseConvInt8TiledExecutor& exe)
|
|
: ConvInt8TiledExecutor(backend, convOp, exe.mResourceInt8), mGemmKernel(exe.mGemmKernel), mResource(exe.mResource), mDynamicQuantExe(dynamicQuantExe) {
|
|
}
|
|
|
|
DenseConvInt8TiledExecutor::~DenseConvInt8TiledExecutor() {
|
|
// Do nothing
|
|
}
|
|
|
|
bool DenseConvInt8TiledExecutor::onClone(Backend* bn, const Op* op, Execution** dst) {
|
|
if (nullptr == dst) {
|
|
return true;
|
|
}
|
|
auto exe = new DenseConvInt8TiledExecutor(bn, op->main_as_Convolution2D(), mDynamicQuantExe, *this);
|
|
if (!exe->valid()) {
|
|
return false;
|
|
}
|
|
*dst = exe;
|
|
return true;
|
|
}
|
|
|
|
void DenseConvInt8TiledExecutor::getPackParameter(int* Unit, int* srcUnit, int* DestUnit, const CoreInt8Functions* core) {
|
|
core->MNNGetGemmUnit(Unit, srcUnit, DestUnit);
|
|
}
|
|
|
|
|
|
ErrorCode DenseConvInt8TiledExecutor::onResize(const std::vector<Tensor*>& inputs, const std::vector<Tensor*>& outputs) {
|
|
mUseBatchQuan = (static_cast<CPUBackend*>(backend())->getRuntime()->hint().dynamicQuantOption == 1);
|
|
mUseBatchQuan &= mCommon->kernelY() == 1 && mCommon->kernelX() == 1
|
|
&& outputs[0]->width() == inputs[0]->width() && outputs[0]->height() == inputs[0]->height()
|
|
&& mCommon->strideX() == 1 && mCommon->strideY() == 1 && mCommon->padX() == 0 && mCommon->padY() == 0
|
|
&& outputs[0]->height() == 1 && outputs[0]->width() == 1;
|
|
mUseBatchQuan &= mDynamicQuantExe;
|
|
mUseBatchQuan &= (inputs[0]->batch() > 1);
|
|
auto core = static_cast<CPUBackend*>(backend())->int8Functions();
|
|
auto gcore =static_cast<CPUBackend*>(backend())->functions();
|
|
int UNIT, SRC_UNIT, DST_XUNIT;
|
|
core->MNNGetGemmUnit(&UNIT, &SRC_UNIT, &DST_XUNIT);
|
|
|
|
if (mDynamicQuantExe == false) {
|
|
mMutableResource.updateInputOutputScale(TensorUtils::getQuantInfo(inputs[0]), TensorUtils::getQuantInfo(outputs[0]));
|
|
CPUConvolution::onResize(inputs, outputs);
|
|
ConvolutionTiledExecutor::setIm2ColParameter(mIm2ColParamter, mCommon, inputs[0], outputs[0], mPadX, mPadY, gcore, core);
|
|
mBlockNum = 1;
|
|
} else { // Dynamic Quant kernels
|
|
CPUConvolution::onResize(inputs, outputs);
|
|
// Gemm Kernel
|
|
mGemmKernel = core->Int8GemmKernel;
|
|
if (mResource->mDequantize.bits == 4) {
|
|
mGemmKernel = core->Int8GemmKernel_W4;
|
|
}
|
|
mQuantFunc = core->MNNFloat2Int8;
|
|
if (gcore->bytes == 2 && gcore->pack == 8) {
|
|
mGemmKernel = core->MNNGemmInt8AddBiasScale_Unit_FP16;
|
|
if (mResource->mDequantize.bits == 4) {
|
|
mGemmKernel = core->MNNGemmInt8AddBiasScale_w4_Unit_FP16;
|
|
}
|
|
mQuantFunc = core->DynamicQuanInput_ARM82;
|
|
mQuantAndReorderFunc = core->DynamicQuanInputAndReorder_ARM82;
|
|
|
|
}
|
|
// A axisSum kernel
|
|
mSumByAxisLFunc = gcore->MNNSumByAxisLForMatmul_A;
|
|
if (gcore->bytes == 2 && gcore->pack == 8) { // use fp16
|
|
ConvolutionTiledExecutor::setIm2ColParameter(mIm2ColParamter, mCommon, inputs[0], outputs[0], mPadX, mPadY, gcore, core, 4);
|
|
} else {
|
|
ConvolutionTiledExecutor::setIm2ColParameter(mIm2ColParamter, mCommon, inputs[0], outputs[0], mPadX, mPadY, gcore, core);
|
|
}
|
|
int ocUp4 = ROUND_UP(outputs[0]->channel(), gcore->pack);
|
|
int alphaSize = mResource->mDequantize.mScaleBias->size() / (4 * 2);
|
|
mBlockNum = alphaSize / ocUp4;
|
|
}
|
|
|
|
// input scale buffer
|
|
int batch = inputs[0]->batch();
|
|
// mTempIm2ColBuffer.reset(Tensor::createDevice<int8_t>({mThreadNums, DST_XUNIT * mIm2ColCount * mResourceInt8->mWeightInt8->length(1) * SRC_UNIT}));
|
|
mInputDeqScales.reset(Tensor::createDevice<int8_t>({batch * 4}));
|
|
bool success = backend()->onAcquireBuffer(mInputDeqScales.get(), Backend::DYNAMIC);
|
|
|
|
// Im2col info
|
|
auto output = outputs[0];
|
|
const int threads = static_cast<CPUBackend*>(backend())->threadNumber();
|
|
auto planeSize = output->width() * output->height() * output->batch();
|
|
const int L2Size = 2048;
|
|
const int tileLimitByC = UP_DIV(L2Size, mIm2ColParamter.kernelCountUnit * SRC_UNIT);
|
|
int tileLimit = 0;
|
|
int outC = output->channel();
|
|
int outC4 = UP_DIV(outC, gcore->pack);
|
|
|
|
if (threads < planeSize) { // Thread split by output nhw.
|
|
tileLimit = ALIMIN(tileLimitByC, UP_DIV(planeSize, threads));
|
|
mSplitByOc = false;
|
|
} else {
|
|
tileLimit = ALIMIN(tileLimitByC, planeSize);
|
|
auto ocPerThread = UP_DIV(outC4, threads);
|
|
auto threadNeed = UP_DIV(outC4, ocPerThread);
|
|
if (UNIT > gcore->pack) { // AVX512:UNIT=64,pack=16
|
|
MNN_ASSERT(UNIT % gcore->pack == 0);
|
|
int ocDivUnit = UP_DIV(outC4 * gcore->pack, UNIT);
|
|
ocPerThread = UP_DIV(ocDivUnit, threads);
|
|
threadNeed = UP_DIV(ocDivUnit, ocPerThread);
|
|
}
|
|
mThreadNums = ALIMIN(threads, threadNeed);
|
|
mSplitByOc = true;
|
|
|
|
mDivides.resize(threads+1);
|
|
mDivides[0] = 0;
|
|
static_cast<const CPURuntime*>(backend()->getRuntime())->computeDivideSizes(outC4, mDivides.data() + 1);
|
|
}
|
|
mIm2ColCount = UP_DIV(tileLimit, DST_XUNIT);
|
|
auto DynamicDestUnit = DST_XUNIT * mIm2ColCount;
|
|
mTileCount = UP_DIV(planeSize, DynamicDestUnit);
|
|
|
|
if (threads < planeSize) {
|
|
mThreadNums = ALIMIN(threads, mTileCount);
|
|
mDivides.resize(threads+1);
|
|
mDivides[0] = 0;
|
|
static_cast<const CPURuntime*>(backend()->getRuntime())->computeDivideSizes(mTileCount, mDivides.data() + 1);
|
|
}
|
|
int ocUp4 = ROUND_UP(outC, gcore->pack);
|
|
int alphaSize = mResource->mDequantize.mScaleBias->size() / (4 * 2);
|
|
|
|
auto bufferAlloc = static_cast<CPUBackend*>(backend())->getBufferAllocator();
|
|
auto blitInfoSize = ConvolutionTiledExecutor::computeBlitInfoSize(DST_XUNIT * mIm2ColCount, mIm2ColParamter.ow, mIm2ColParamter.kernelX * mIm2ColParamter.kernelY, mThreadNums);
|
|
mBlitInfoStride = blitInfoSize.second;
|
|
mBlitInfo = bufferAlloc->alloc(blitInfoSize.first);
|
|
mTempIm2ColBuffer.reset(Tensor::createDevice<int8_t>({mThreadNums, DST_XUNIT * mIm2ColCount * mResourceInt8->mWeightInt8->length(1) * SRC_UNIT}));
|
|
mTempSrcSum.resize(mThreadNums * mBlockNum * DST_XUNIT * mIm2ColCount * 4); // Use 4 bytes to save kernel sum.
|
|
|
|
success &= backend()->onAcquireBuffer(mTempIm2ColBuffer.get(), Backend::DYNAMIC);
|
|
if (!success || mBlitInfo.invalid()) {
|
|
return OUT_OF_MEMORY;
|
|
}
|
|
if (false == mDynamicQuantExe) {
|
|
bufferAlloc->free(mBlitInfo);
|
|
backend()->onReleaseBuffer(mInputDeqScales.get(), Backend::DYNAMIC);
|
|
backend()->onReleaseBuffer(mTempIm2ColBuffer.get(), Backend::DYNAMIC);
|
|
return NO_ERROR;
|
|
}
|
|
|
|
|
|
int inC = inputs[0]->channel();
|
|
// set im2col tensor info
|
|
mQuantInput.reset((Tensor::createDevice<int8_t>({batch, mIm2ColParamter.ih, mIm2ColParamter.iw, ROUND_UP(inC, gcore->pack)})));
|
|
// set dynamic quant buffer
|
|
mTempMaxMinValueBuffer.reset(Tensor::createDevice<uint8_t>({mThreadNums, 2 * gcore->bytes}));
|
|
// set compute buffer
|
|
mDynamicBias.reset(Tensor::createDevice<uint8_t>({ocUp4 * 4}));
|
|
mScaleFuse.reset(Tensor::createDevice<uint8_t>({alphaSize * 4}));
|
|
|
|
success &= backend()->onAcquireBuffer(mQuantInput.get(), Backend::DYNAMIC);
|
|
success &= backend()->onAcquireBuffer(mDynamicBias.get(), Backend::DYNAMIC);
|
|
success &= backend()->onAcquireBuffer(mTempMaxMinValueBuffer.get(), Backend::DYNAMIC);
|
|
success &= backend()->onAcquireBuffer(mScaleFuse.get(), Backend::DYNAMIC);
|
|
|
|
if (mUseBatchQuan) {
|
|
int infobytes = 4; // use float32 to save dequant scale and quant scale.
|
|
int size = mThreadNums * batch * gcore->bytes + 2 * batch * infobytes;
|
|
mBatchQuantInfo.reset(Tensor::createDevice<int8_t>({size}));
|
|
success &= backend()->onAcquireBuffer(mBatchQuantInfo.get(), Backend::DYNAMIC);
|
|
}
|
|
if (!success) {
|
|
return OUT_OF_MEMORY;
|
|
}
|
|
bufferAlloc->free(mBlitInfo);
|
|
backend()->onReleaseBuffer(mInputDeqScales.get(), Backend::DYNAMIC);
|
|
backend()->onReleaseBuffer(mTempIm2ColBuffer.get(), Backend::DYNAMIC);
|
|
backend()->onReleaseBuffer(mQuantInput.get(), Backend::DYNAMIC);
|
|
backend()->onReleaseBuffer(mDynamicBias.get(), Backend::DYNAMIC);
|
|
backend()->onReleaseBuffer(mTempMaxMinValueBuffer.get(), Backend::DYNAMIC);
|
|
backend()->onReleaseBuffer(mScaleFuse.get(), Backend::DYNAMIC);
|
|
if (mUseBatchQuan) {
|
|
backend()->onReleaseBuffer(mBatchQuantInfo.get(), Backend::DYNAMIC);
|
|
}
|
|
return NO_ERROR;
|
|
}
|
|
|
|
ErrorCode DenseConvInt8TiledExecutor::onExecute(const std::vector<Tensor*>& inputs, const std::vector<Tensor*>& outputs) {
|
|
// Timer kernelTimer;
|
|
const auto input = inputs[0];
|
|
auto output = outputs[0];
|
|
auto core = static_cast<CPUBackend*>(backend())->int8Functions();
|
|
auto gcore = static_cast<CPUBackend*>(backend())->functions();
|
|
|
|
int UNIT__, SRC_UNIT, DST_XUNIT;
|
|
core->MNNGetGemmUnit(&UNIT__, &SRC_UNIT, &DST_XUNIT);
|
|
auto blitProc = core->MNNPackC4Int8ForMatMul_A;
|
|
if ( mDynamicQuantExe && gcore->bytes == 2 && core->MNNPackC4Int8ForMatMul_A_ARM86FP16) {
|
|
blitProc = core->MNNPackC4Int8ForMatMul_A_ARM86FP16;
|
|
}
|
|
const int plane = output->batch() * mIm2ColParamter.oh * mIm2ColParamter.ow;
|
|
const int batch = input->batch();
|
|
const int PackUnit = gcore->pack;
|
|
const int dstZStep = plane * PackUnit;
|
|
const int ocDiv4 = UP_DIV(output->channel(), PackUnit);
|
|
const int ocUp4 = ROUND_UP(output->channel(), PackUnit);
|
|
const auto kernelCountUnitDouble = mIm2ColParamter.kernelCountUnit;
|
|
const auto col_buffer_unit_size = kernelCountUnitDouble * DST_XUNIT * SRC_UNIT * sizeof(int8_t);
|
|
const auto col_buffer_size = col_buffer_unit_size * mIm2ColCount;
|
|
const int dstBytes = static_cast<CPUBackend*>(backend())->getBytes(backend(), output);
|
|
const int alphaSize = mResource->mDequantize.mScaleBias->size() / (4 * 2);
|
|
const int blockL = kernelCountUnitDouble / mBlockNum; // source depthQuad for each block.
|
|
float weightBytes = 1.f;
|
|
int weight_step_Y = weightBytes * (UNIT__ * SRC_UNIT);
|
|
int src_step_Y = DST_XUNIT * SRC_UNIT;
|
|
|
|
auto inputDataPtr = input->host<int8_t>();
|
|
auto im2colPtr = mTempIm2ColBuffer->host<int8_t>();
|
|
const auto weightDataPtr = mResourceInt8->mWeightInt8->host<int8_t>();
|
|
auto srcKernelSumPtr = mTempSrcSum.data();
|
|
auto weightDequantBias = mResource->mDequantize.mScaleBias->host<uint8_t>() + alphaSize * 4;
|
|
|
|
auto outputDataPtr = output->host<int8_t>();
|
|
auto biasPtr = mMutableResource.mBiasFloat->host<uint8_t>();
|
|
auto scalePtr = mMutableResource.mScaleFloat->host<uint8_t>();
|
|
|
|
auto inputZeroPoint = mMutableResource.mInputZeroPoint;
|
|
auto inputScalePtr = mInputDeqScales->host<uint8_t>();
|
|
(reinterpret_cast<float*>(inputScalePtr))[0] = mMutableResource.mInputScale;
|
|
|
|
auto SingleDynamicQuant = [&] () {
|
|
const auto floatptr = input->host<float>();
|
|
auto int8ptr = mQuantInput->host<int8_t>();
|
|
auto inputsize = static_cast<CPUBackend*>(backend())->getTensorSize(inputs[0]);
|
|
float quantscale = 0.f;
|
|
float dequantscale = 0.f;
|
|
int zeropoint = 0;
|
|
|
|
/* Count max and min value to compute input scale and zeropoint */
|
|
auto maxMinValPtr = mTempMaxMinValueBuffer->host<uint8_t>();
|
|
int threadNeed = mThreadNums;
|
|
auto inputSizeCount = UP_DIV(inputsize, mThreadNums);
|
|
if (inputSizeCount < 9) {
|
|
threadNeed = 1;
|
|
inputSizeCount = inputsize;
|
|
} else {
|
|
threadNeed = ALIMIN(UP_DIV(inputsize, inputSizeCount), mThreadNums);
|
|
inputSizeCount = UP_DIV(inputsize, threadNeed);
|
|
}
|
|
auto findMaxMinValueFunction = [&](int tId) {
|
|
auto perThreadWorkCount = ALIMIN(inputSizeCount, inputsize - tId * inputSizeCount);
|
|
auto minValPtrTid = reinterpret_cast<float*>(maxMinValPtr + tId * mTempMaxMinValueBuffer->stride(0));
|
|
auto maxValPtrTid = reinterpret_cast<float*>(maxMinValPtr + tId * mTempMaxMinValueBuffer->stride(0) + gcore->bytes);
|
|
auto inputDataPtrTid = reinterpret_cast<float*>(reinterpret_cast<uint8_t*>(floatptr) + tId * inputSizeCount * gcore->bytes);
|
|
gcore->MNNCountMaxMinValue(inputDataPtrTid, minValPtrTid, maxValPtrTid, perThreadWorkCount);
|
|
};
|
|
MNN_CONCURRENCY_BEGIN(tId, threadNeed) {
|
|
findMaxMinValueFunction((int)tId);
|
|
}
|
|
MNN_CONCURRENCY_END();
|
|
if (threadNeed > 1) {
|
|
gcore->MNNCountMaxMinValue(reinterpret_cast<float*>(maxMinValPtr),reinterpret_cast<float*>(maxMinValPtr), reinterpret_cast<float*>(maxMinValPtr + gcore->bytes), 2 * mThreadNums);
|
|
}
|
|
float maxVal = 0;
|
|
float minVal = 0;
|
|
if (gcore->bytes == 4) {
|
|
maxVal = (reinterpret_cast<float*>(maxMinValPtr))[1];
|
|
minVal = (reinterpret_cast<float*>(maxMinValPtr))[0];
|
|
}
|
|
if (gcore->bytes == 2) {
|
|
std::vector<float> _mVal(2);
|
|
gcore->MNNLowpToFp32(reinterpret_cast<int16_t*>(maxMinValPtr), _mVal.data(), 2);
|
|
maxVal = _mVal[1];
|
|
minVal = _mVal[0];
|
|
}
|
|
|
|
/* Dynamic quant */
|
|
float range = maxVal - minVal;
|
|
quantscale = 255.0f / range;
|
|
dequantscale = range / 255.0f;
|
|
zeropoint = static_cast<int32_t>(roundf(-minVal * 255.f / range) - 128.0f);
|
|
std::vector<float>qsVec(PackUnit, quantscale);
|
|
auto sizeDiv = UP_DIV(inputsize, PackUnit);
|
|
int inputPlane = input->batch() * mIm2ColParamter.iw * mIm2ColParamter.ih;
|
|
if (gcore->bytes == 2 && gcore->pack == 8 && inputPlane > 1) { // C8->C4
|
|
mQuantAndReorderFunc(floatptr, int8ptr, inputPlane, qsVec.data(), -128, 127, (ssize_t)zeropoint, UP_DIV(input->channel(), PackUnit), 4 * inputPlane);
|
|
} else {
|
|
mQuantFunc(floatptr, int8ptr, sizeDiv, qsVec.data(), -128, 127, (ssize_t)zeropoint);
|
|
}
|
|
|
|
/* bias float */
|
|
#ifdef MNN_USE_SSE
|
|
int offset = 128;
|
|
#else
|
|
int offset = 0;
|
|
#endif
|
|
auto biasfp32 = mMutableResource.mResource->mOriginBias->host<float>();
|
|
auto weightDequantScale = mResource->mDequantize.mScaleBias->host<float>();
|
|
float zerofp32 = (zeropoint + offset) * dequantscale;
|
|
|
|
gcore->MNNDynamicUpdateConvBiasScale(mDynamicBias->host<float>(), mScaleFuse->host<float>(), biasfp32, weightDequantScale, &dequantscale, mResource->mWeightKernelSum->host<float>(), &zerofp32, UP_DIV(output->channel(), 4), alphaSize);
|
|
// Move step for A and B for each block computing
|
|
|
|
inputZeroPoint = zeropoint;
|
|
(reinterpret_cast<float*>(inputScalePtr))[0] = dequantscale;
|
|
biasPtr = mDynamicBias->host<uint8_t>();
|
|
scalePtr = mScaleFuse->host<uint8_t>();
|
|
inputDataPtr = int8ptr;
|
|
};
|
|
|
|
auto BatchDynamicQuant = [&]() {
|
|
// Allocate input max/sum/dequant/quant buffer
|
|
auto infobytes = 4;
|
|
auto dequantPtr = mBatchQuantInfo->host<uint8_t>();
|
|
auto quantPtr = dequantPtr + batch * infobytes;
|
|
auto maxPtr = mBatchQuantInfo->host<uint8_t>() + 2 * batch * infobytes;
|
|
|
|
// compute sum and absmax
|
|
int icDiv4 = UP_DIV(input->channel(), PackUnit);
|
|
int threadwork = UP_DIV(icDiv4, mThreadNums);
|
|
int threadNeed = UP_DIV(icDiv4, threadwork);
|
|
int threadTmp = ALIMIN(mThreadNums, threadNeed);
|
|
threadwork = UP_DIV(icDiv4, threadTmp);
|
|
MNN_CONCURRENCY_BEGIN(tId, threadTmp) {
|
|
int workCount = threadwork;
|
|
if (tId == threadTmp - 1) {
|
|
workCount = icDiv4 - tId * threadwork;
|
|
}
|
|
int icIndex = tId * threadwork;
|
|
auto inputData = reinterpret_cast<const float*>(input->host<uint8_t>() + icIndex * batch * PackUnit * gcore->bytes);
|
|
auto batchMax = reinterpret_cast<float*>(maxPtr + tId * batch * gcore->bytes);
|
|
gcore->MNNAbsMax(inputData, batchMax, workCount, batch, PackUnit);
|
|
}
|
|
MNN_CONCURRENCY_END();
|
|
|
|
// Compute quant scale
|
|
gcore->MNNQuantScale((float*)maxPtr, (float*)quantPtr, (float*)dequantPtr, threadTmp, batch);
|
|
|
|
// quant
|
|
MNN_CONCURRENCY_BEGIN(tId, threadTmp) {
|
|
int workCount = threadwork;
|
|
if (tId == threadTmp - 1) {
|
|
workCount = icDiv4 - tId * threadwork;
|
|
}
|
|
auto icIndex = tId * threadwork;
|
|
auto inputData = reinterpret_cast<float*>(input->host<uint8_t>() + icIndex * batch * PackUnit * gcore->bytes);
|
|
auto int8ptr = mQuantInput->host<int8_t>() + icIndex * batch * PackUnit;
|
|
auto scale_ptr = reinterpret_cast<float*>(quantPtr);
|
|
gcore->MNNDynamicQuant(inputData, int8ptr, scale_ptr, workCount, batch, PackUnit);
|
|
}
|
|
MNN_CONCURRENCY_END();
|
|
|
|
inputZeroPoint = 0;
|
|
inputScalePtr = (uint8_t*)dequantPtr;
|
|
inputDataPtr = mQuantInput->host<int8_t>();
|
|
biasPtr = mMutableResource.mResource->mOriginBias->host<uint8_t>();
|
|
scalePtr = mResource->mDequantize.mScaleBias->host<uint8_t>();
|
|
};
|
|
ssize_t oneScale = 1;
|
|
if (mUseBatchQuan) {
|
|
BatchDynamicQuant();
|
|
oneScale = 0;
|
|
} else if (mDynamicQuantExe) {
|
|
SingleDynamicQuant();
|
|
} else {
|
|
// offline quant.
|
|
}
|
|
|
|
if (mResource->mDequantize.bits == 4) {
|
|
weightBytes = 0.5;
|
|
weight_step_Y *= 0.5;
|
|
}
|
|
|
|
SumByAxisParams sumParams;
|
|
sumParams.oneScale = oneScale;
|
|
sumParams.SRC_UNIT = SRC_UNIT;
|
|
sumParams.blockNum = mBlockNum;
|
|
sumParams.DST_XUNIT = DST_XUNIT;
|
|
sumParams.col_buffer_unit_size = col_buffer_unit_size;
|
|
sumParams.kernelCountUnitDouble = kernelCountUnitDouble;
|
|
|
|
auto ThreadFunction = [&](int tId, int eStartIndex, int eEndIndex, int estep, int ocIndex) {
|
|
auto ocDivThread = ocDiv4;
|
|
if (mSplitByOc) { // Thread split by OC
|
|
ocDivThread = ALIMIN(mDivides[tId + 1] - mDivides[tId], ocDiv4 - mDivides[tId]);
|
|
}
|
|
float* reluPtr = mResource->mReluThreshold.data();
|
|
uint8_t* extraScale = nullptr; // input scale for batch dynamic quant.
|
|
QuanPostTreatParameters quanParam;
|
|
quanParam.blockNum = mBlockNum;
|
|
if (mUseBatchQuan) {
|
|
extraScale = inputScalePtr;
|
|
}
|
|
#ifdef MNN_USE_SSE
|
|
quanParam.extraBias = mResource->mWeightKernelSum->host<float>() + ocIndex;
|
|
#endif
|
|
if (dstBytes != 1) {
|
|
quanParam.useInt8 = 0;
|
|
quanParam.fp32minmax = reluPtr;
|
|
} else {
|
|
quanParam.maxValue = mMutableResource.mClampMax;
|
|
if (mResourceInt8->mRelu) {
|
|
quanParam.minValue = mMutableResource.mOutputZeroPoint;
|
|
} else {
|
|
quanParam.minValue = mMutableResource.mClampMin;
|
|
}
|
|
}
|
|
auto outputTid = outputDataPtr + ocIndex * plane * dstBytes;
|
|
const auto biasFloatTid = reinterpret_cast<float*>(biasPtr + ocIndex * 4);
|
|
const auto scaleFloatTid = reinterpret_cast<float*>(scalePtr + ocIndex * 4);
|
|
const auto weightDequanBiasTid = reinterpret_cast<float*>(weightDequantBias + ocIndex * 4);
|
|
const auto weightPtrTid = weightDataPtr + static_cast<int32_t>(ocIndex * kernelCountUnitDouble * SRC_UNIT * weightBytes);
|
|
if (mBlockNum == 1) {
|
|
quanParam.biasFloat = biasFloatTid;
|
|
quanParam.scale = scaleFloatTid;
|
|
quanParam.weightQuanBias = weightDequanBiasTid;
|
|
}
|
|
|
|
auto colAddr = im2colPtr + tId * mTempIm2ColBuffer->stride(0);
|
|
auto srcPtr = (int8_t const **)(mBlitInfo.ptr() + tId * mBlitInfoStride.first);
|
|
auto el = (int32_t *)(srcPtr + mBlitInfoStride.second);
|
|
auto xKernelSumPtrTid = reinterpret_cast<float*>(srcKernelSumPtr + tId * mBlockNum * DST_XUNIT * mIm2ColCount * 4);
|
|
|
|
int32_t info[6];
|
|
info[1] = mIm2ColParamter.iw * mIm2ColParamter.ih * batch;
|
|
info[2] = static_cast<int32_t>(col_buffer_unit_size);
|
|
info[3] = mIm2ColParamter.strideX;
|
|
info[5] = kernelCountUnitDouble;
|
|
for (int tIndex = eStartIndex; tIndex < eEndIndex; tIndex += estep) {
|
|
const int xIndexStart = tIndex * DST_XUNIT * mIm2ColCount;
|
|
int realDstCount = ALIMIN(plane - xIndexStart, DST_XUNIT * mIm2ColCount);
|
|
auto ptrExtraScale = extraScale != nullptr ? (extraScale + xIndexStart * 4) : nullptr;
|
|
auto ptrInputscale = mUseBatchQuan == true ? (inputScalePtr + xIndexStart * 4) : inputScalePtr;
|
|
// im2col
|
|
auto res = ConvolutionTiledExecutor::turnIm2ColToBlitInfo((const float**)srcPtr, el, xIndexStart, realDstCount, mIm2ColParamter, (const uint8_t*)inputDataPtr, 1);
|
|
int number = res.first;
|
|
bool needZero = res.second;
|
|
if (needZero) {
|
|
#ifdef MNN_USE_SSE
|
|
::memset(colAddr, inputZeroPoint + 128, col_buffer_size);
|
|
#else
|
|
::memset(colAddr, inputZeroPoint, col_buffer_size);
|
|
#endif
|
|
}
|
|
info[0] = number;
|
|
info[4] = realDstCount;
|
|
if (number > 0) {
|
|
blitProc(colAddr, srcPtr, info, el);
|
|
}
|
|
if (mResourceInt8->mWeightAsymmetricQuant) {
|
|
mSumByAxisLFunc(xKernelSumPtrTid, colAddr, (float*)ptrInputscale, realDstCount, sumParams);
|
|
}
|
|
auto outputInTilePtr = outputTid + xIndexStart * PackUnit * dstBytes;
|
|
auto colAddrTemp = colAddr;
|
|
auto ptrX = xKernelSumPtrTid;
|
|
if (mBlockNum == 1) {
|
|
do {
|
|
int step = ALIMIN(DST_XUNIT, realDstCount);
|
|
quanParam.srcKernelSum = ptrX;
|
|
quanParam.extraScale = extraScale != nullptr ? (float*)ptrExtraScale : nullptr;
|
|
mGemmKernel(outputInTilePtr, colAddrTemp, weightPtrTid, kernelCountUnitDouble, dstZStep * dstBytes, ocDivThread, &quanParam, step);
|
|
ptrX += step;
|
|
realDstCount-=step;
|
|
outputInTilePtr += DST_XUNIT * PackUnit * dstBytes;
|
|
colAddrTemp += col_buffer_unit_size;
|
|
ptrExtraScale = extraScale != nullptr ? (ptrExtraScale + step * 4) : nullptr;
|
|
} while(realDstCount > 0);
|
|
} else { // Now offline quant do not run into.
|
|
do {
|
|
int step = ALIMIN(DST_XUNIT, realDstCount);
|
|
quanParam.extraScale = extraScale != nullptr ? (float*)ptrExtraScale : nullptr;
|
|
for (int k = 0; k < mBlockNum; ++k) {
|
|
quanParam.biasFloat = nullptr;
|
|
quanParam.fp32minmax = nullptr;
|
|
if (k == 0) {
|
|
quanParam.biasFloat = (float*)biasFloatTid;
|
|
}
|
|
if (k == mBlockNum - 1) {
|
|
quanParam.fp32minmax = reluPtr;
|
|
}
|
|
quanParam.srcKernelSum = ptrX + k * step;
|
|
quanParam.weightQuanBias = weightDequanBiasTid + k * ocUp4;
|
|
quanParam.scale = (float*)(scaleFloatTid + k * ocUp4);
|
|
|
|
mGemmKernel(outputInTilePtr, colAddrTemp + k * blockL * src_step_Y, weightPtrTid + k * blockL * weight_step_Y, blockL, dstZStep * dstBytes, ocDivThread, &quanParam, step);
|
|
}
|
|
ptrX += (step * mBlockNum);
|
|
realDstCount-=step;
|
|
outputInTilePtr += DST_XUNIT * PackUnit * dstBytes;
|
|
colAddrTemp += col_buffer_unit_size;
|
|
ptrExtraScale = extraScale != nullptr ? (ptrExtraScale + step * 4) : nullptr;
|
|
} while(realDstCount > 0);
|
|
}
|
|
}
|
|
};
|
|
|
|
if (!mSplitByOc) {
|
|
MNN_CONCURRENCY_BEGIN(tId, mThreadNums) {
|
|
ThreadFunction((int)tId, mDivides[tId], mDivides[tId + 1], 1, 0);
|
|
}
|
|
MNN_CONCURRENCY_END();
|
|
} else {
|
|
MNN_CONCURRENCY_BEGIN(tId, mThreadNums) {
|
|
int ocIndex = PackUnit * mDivides[tId];
|
|
ThreadFunction((int)tId, 0, mTileCount,1, ocIndex);
|
|
}
|
|
MNN_CONCURRENCY_END();
|
|
}
|
|
|
|
return NO_ERROR;
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
} // namespace MNN
|