opencv/modules/core/src/cuda_gpu_mat.cpp at master · pythonmjs/opencv · GitHub

History

515 lines (431 loc) · 13.2 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

/*M///////////////////////////////////////////////////////////////////////////////////////

// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.

// By downloading, copying, installing or using the software you agree to this license.

// If you do not agree to this license, do not download, install,

// copy or use the software.

// License Agreement

// For Open Source Computer Vision Library

// Third party copyrights are property of their respective owners.

// Redistribution and use in source and binary forms, with or without modification,

// are permitted provided that the following conditions are met:

// * Redistribution's of source code must retain the above copyright notice,

// this list of conditions and the following disclaimer.

// * Redistribution's in binary form must reproduce the above copyright notice,

// this list of conditions and the following disclaimer in the documentation

// and/or other materials provided with the distribution.

// * The name of the copyright holders may not be used to endorse or promote products

// derived from this software without specific prior written permission.

// This software is provided by the copyright holders and contributors "as is" and

// any express or implied warranties, including, but not limited to, the implied

// warranties of merchantability and fitness for a particular purpose are disclaimed.

// In no event shall the Intel Corporation or contributors be liable for any direct,

// indirect, incidental, special, exemplary, or consequential damages

// (including, but not limited to, procurement of substitute goods or services;

// loss of use, data, or profits; or business interruption) however caused

// and on any theory of liability, whether in contract, strict liability,

// or tort (including negligence or otherwise) arising in any way out of

// the use of this software, even if advised of the possibility of such damage.

//M*/

#include "precomp.hpp"

using namespace cv;

using namespace cv::cuda;

void cv::cuda::GpuMat::updateContinuityFlag()

{

int sz[] = { rows, cols };

size_t steps[] = { step, elemSize() };

flags = cv::updateContinuityFlag(flags, 2, sz, steps);

}

cv::cuda::GpuMat::GpuMat(int rows_, int cols_, int type_, void* data_, size_t step_) :

flags(Mat::MAGIC_VAL + (type_ & Mat::TYPE_MASK)), rows(rows_), cols(cols_),

step(step_), data((uchar*)data_), refcount(0),

datastart((uchar*)data_), dataend((const uchar*)data_),

allocator(defaultAllocator())

{

size_t minstep = cols * elemSize();

if (step == Mat::AUTO_STEP)

{

step = minstep;

}

else

{

if (rows == 1)

step = minstep;

CV_DbgAssert( step >= minstep );

}

dataend += step * (rows - 1) + minstep;

updateContinuityFlag();

}

cv::cuda::GpuMat::GpuMat(Size size_, int type_, void* data_, size_t step_) :

flags(Mat::MAGIC_VAL + (type_ & Mat::TYPE_MASK)), rows(size_.height), cols(size_.width),

step(step_), data((uchar*)data_), refcount(0),

datastart((uchar*)data_), dataend((const uchar*)data_),

allocator(defaultAllocator())

{

size_t minstep = cols * elemSize();

if (step == Mat::AUTO_STEP)

{

step = minstep;

}

else

{

if (rows == 1)

step = minstep;

CV_DbgAssert( step >= minstep );

}

dataend += step * (rows - 1) + minstep;

updateContinuityFlag();

}

cv::cuda::GpuMat::GpuMat(const GpuMat& m, Range rowRange_, Range colRange_)

{

flags = m.flags;

step = m.step; refcount = m.refcount;

data = m.data; datastart = m.datastart; dataend = m.dataend;

allocator = m.allocator;

if (rowRange_ == Range::all())

{

rows = m.rows;

}

else

{

CV_Assert( 0 <= rowRange_.start && rowRange_.start <= rowRange_.end && rowRange_.end <= m.rows );

rows = rowRange_.size();

data += step*rowRange_.start;

}

if (colRange_ == Range::all())

{

cols = m.cols;

}

else

{

CV_Assert( 0 <= colRange_.start && colRange_.start <= colRange_.end && colRange_.end <= m.cols );

cols = colRange_.size();

data += colRange_.start*elemSize();

}

if (refcount)

CV_XADD(refcount, 1);

if (rows <= 0 || cols <= 0)

rows = cols = 0;

updateContinuityFlag();

}

cv::cuda::GpuMat::GpuMat(const GpuMat& m, Rect roi) :

flags(m.flags), rows(roi.height), cols(roi.width),

step(m.step), data(m.data + roi.y*step), refcount(m.refcount),

datastart(m.datastart), dataend(m.dataend),

allocator(m.allocator)

{

data += roi.x * elemSize();

CV_Assert( 0 <= roi.x && 0 <= roi.width &&

roi.x + roi.width <= m.cols &&

0 <= roi.y && 0 <= roi.height &&

roi.y + roi.height <= m.rows );

if (refcount)

CV_XADD(refcount, 1);

if (rows <= 0 || cols <= 0)

rows = cols = 0;

updateContinuityFlag();

}

GpuMat cv::cuda::GpuMat::reshape(int new_cn, int new_rows) const

{

GpuMat hdr = *this;

int cn = channels();

if (new_cn == 0)

new_cn = cn;

int total_width = cols * cn;

if ((new_cn > total_width || total_width % new_cn != 0) && new_rows == 0)

new_rows = rows * total_width / new_cn;

if (new_rows != 0 && new_rows != rows)

{

int total_size = total_width * rows;

if (!isContinuous())

CV_Error(cv::Error::BadStep, "The matrix is not continuous, thus its number of rows can not be changed");

if ((unsigned)new_rows > (unsigned)total_size)

CV_Error(cv::Error::StsOutOfRange, "Bad new number of rows");

total_width = total_size / new_rows;

if (total_width * new_rows != total_size)

CV_Error(cv::Error::StsBadArg, "The total number of matrix elements is not divisible by the new number of rows");

hdr.rows = new_rows;

hdr.step = total_width * elemSize1();

}

int new_width = total_width / new_cn;

if (new_width * new_cn != total_width)

CV_Error(cv::Error::BadNumChannels, "The total width is not divisible by the new number of channels");

hdr.cols = new_width;

hdr.flags = (hdr.flags & ~CV_MAT_CN_MASK) | ((new_cn - 1) << CV_CN_SHIFT);

return hdr;

}

void cv::cuda::GpuMat::locateROI(Size& wholeSize, Point& ofs) const

{

CV_DbgAssert( step > 0 );

size_t esz = elemSize();

ptrdiff_t delta1 = data - datastart;

ptrdiff_t delta2 = dataend - datastart;

if (delta1 == 0)

{

ofs.x = ofs.y = 0;

}

else

{

ofs.y = static_cast<int>(delta1 / step);

ofs.x = static_cast<int>((delta1 - step * ofs.y) / esz);

CV_DbgAssert( data == datastart + ofs.y * step + ofs.x * esz );

}

size_t minstep = (ofs.x + cols) * esz;

wholeSize.height = std::max(static_cast<int>((delta2 - minstep) / step + 1), ofs.y + rows);

wholeSize.width = std::max(static_cast<int>((delta2 - step * (wholeSize.height - 1)) / esz), ofs.x + cols);

}

GpuMat& cv::cuda::GpuMat::adjustROI(int dtop, int dbottom, int dleft, int dright)

{

Size wholeSize;

Point ofs;

locateROI(wholeSize, ofs);

size_t esz = elemSize();

int row1 = std::max(ofs.y - dtop, 0);

int row2 = std::min(ofs.y + rows + dbottom, wholeSize.height);

int col1 = std::max(ofs.x - dleft, 0);

int col2 = std::min(ofs.x + cols + dright, wholeSize.width);

data += (row1 - ofs.y) * step + (col1 - ofs.x) * esz;

rows = row2 - row1;

cols = col2 - col1;

updateContinuityFlag();

return *this;

}

namespace

{

template <class ObjType>

void createContinuousImpl(int rows, int cols, int type, ObjType& obj)

{

const int area = rows * cols;

if (obj.empty() || obj.type() != type || !obj.isContinuous() || obj.size().area() != area)

obj.create(1, area, type);

obj = obj.reshape(obj.channels(), rows);

}

void cv::cuda::createContinuous(int rows, int cols, int type, OutputArray arr)

{

switch (arr.kind())

{

case _InputArray::MAT:

::createContinuousImpl(rows, cols, type, arr.getMatRef());

break;

case _InputArray::CUDA_GPU_MAT:

::createContinuousImpl(rows, cols, type, arr.getGpuMatRef());

break;

case _InputArray::CUDA_HOST_MEM:

::createContinuousImpl(rows, cols, type, arr.getHostMemRef());

break;

default:

arr.create(rows, cols, type);

}

namespace

{

template <class ObjType>

void ensureSizeIsEnoughImpl(int rows, int cols, int type, ObjType& obj)

{

if (obj.empty() || obj.type() != type || obj.data != obj.datastart)

{

obj.create(rows, cols, type);

}

else

{

const size_t esz = obj.elemSize();

const ptrdiff_t delta2 = obj.dataend - obj.datastart;

const size_t minstep = obj.cols * esz;

Size wholeSize;

wholeSize.height = std::max(static_cast<int>((delta2 - minstep) / static_cast<size_t>(obj.step) + 1), obj.rows);

wholeSize.width = std::max(static_cast<int>((delta2 - static_cast<size_t>(obj.step) * (wholeSize.height - 1)) / esz), obj.cols);

if (wholeSize.height < rows || wholeSize.width < cols)

{

obj.create(rows, cols, type);

}

else

{

obj.cols = cols;

obj.rows = rows;

}

void cv::cuda::ensureSizeIsEnough(int rows, int cols, int type, OutputArray arr)

{

switch (arr.kind())

{

case _InputArray::MAT:

::ensureSizeIsEnoughImpl(rows, cols, type, arr.getMatRef());

break;

case _InputArray::CUDA_GPU_MAT:

::ensureSizeIsEnoughImpl(rows, cols, type, arr.getGpuMatRef());

break;

case _InputArray::CUDA_HOST_MEM:

::ensureSizeIsEnoughImpl(rows, cols, type, arr.getHostMemRef());

break;

default:

arr.create(rows, cols, type);

}

GpuMat cv::cuda::getInputMat(InputArray _src, Stream& stream)

{

#ifndef HAVE_CUDA

(void) _src;

(void) stream;

throw_no_cuda();

#else

GpuMat src;

if (_src.kind() == _InputArray::CUDA_GPU_MAT)

{

src = _src.getGpuMat();

}

else if (!_src.empty())

{

BufferPool pool(stream);

src = pool.getBuffer(_src.size(), _src.type());

src.upload(_src, stream);

}

return src;

#endif

}

GpuMat cv::cuda::getOutputMat(OutputArray _dst, int rows, int cols, int type, Stream& stream)

{

#ifndef HAVE_CUDA

(void) _dst;

(void) rows;

(void) cols;

(void) type;

(void) stream;

throw_no_cuda();

#else

GpuMat dst;

if (_dst.kind() == _InputArray::CUDA_GPU_MAT)

{

_dst.create(rows, cols, type);

dst = _dst.getGpuMat();

}

else

{

BufferPool pool(stream);

dst = pool.getBuffer(rows, cols, type);

}

return dst;

#endif

}

void cv::cuda::syncOutput(const GpuMat& dst, OutputArray _dst, Stream& stream)

{

#ifndef HAVE_CUDA

(void) dst;

(void) _dst;

(void) stream;

throw_no_cuda();

#else

if (_dst.kind() != _InputArray::CUDA_GPU_MAT)

{

if (stream)

dst.download(_dst, stream);

else

dst.download(_dst);

}

#endif

}

#ifndef HAVE_CUDA

GpuMat::Allocator* cv::cuda::GpuMat::defaultAllocator()

{

return 0;

}

void cv::cuda::GpuMat::setDefaultAllocator(Allocator* allocator)

{

(void) allocator;

throw_no_cuda();

}

void cv::cuda::GpuMat::create(int _rows, int _cols, int _type)

{

(void) _rows;

(void) _cols;

(void) _type;

throw_no_cuda();

}

void cv::cuda::GpuMat::release()

{

}

void cv::cuda::GpuMat::upload(InputArray arr)

{

(void) arr;

throw_no_cuda();

}

void cv::cuda::GpuMat::upload(InputArray arr, Stream& _stream)

{

(void) arr;

(void) _stream;

throw_no_cuda();

}

void cv::cuda::GpuMat::download(OutputArray _dst) const

{

(void) _dst;

throw_no_cuda();

}

void cv::cuda::GpuMat::download(OutputArray _dst, Stream& _stream) const

{

(void) _dst;

(void) _stream;

throw_no_cuda();

}

void cv::cuda::GpuMat::copyTo(OutputArray _dst) const

{

(void) _dst;

throw_no_cuda();

}

void cv::cuda::GpuMat::copyTo(OutputArray _dst, Stream& _stream) const

{

(void) _dst;

(void) _stream;

throw_no_cuda();

}

void cv::cuda::GpuMat::copyTo(OutputArray _dst, InputArray _mask, Stream& _stream) const

{

(void) _dst;

(void) _mask;

(void) _stream;

throw_no_cuda();

}

GpuMat& cv::cuda::GpuMat::setTo(Scalar s, Stream& _stream)

{

(void) s;

(void) _stream;

throw_no_cuda();

}

GpuMat& cv::cuda::GpuMat::setTo(Scalar s, InputArray _mask, Stream& _stream)

{

(void) s;

(void) _mask;

(void) _stream;

throw_no_cuda();

}

void cv::cuda::GpuMat::convertTo(OutputArray _dst, int rtype, Stream& _stream) const

{

(void) _dst;

(void) rtype;

(void) _stream;

throw_no_cuda();

}

void cv::cuda::GpuMat::convertTo(OutputArray _dst, int rtype, double alpha, double beta, Stream& _stream) const

{

(void) _dst;

(void) rtype;

(void) alpha;

(void) beta;

(void) _stream;

throw_no_cuda();

}

#endif

opencv/modules/core/src/cuda_gpu_mat.cpp at master · pythonmjs/opencv · GitHub

Expand file tree

Breadcrumbs

cuda_gpu_mat.cpp

Latest commit

History

Breadcrumbs

cuda_gpu_mat.cpp

File metadata and controls