python-docs-fa/c-api/memory.html at gh-pages · python/python-docs-fa

History

1270 lines (1186 loc) · 115 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

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

810

811

812

813

814

815

816

817

818

819

820

821

822

823

824

825

826

827

828

829

830

831

832

833

834

835

836

837

838

839

840

841

842

843

844

845

846

847

848

849

850

851

852

853

854

855

856

857

858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

878

879

880

881

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

899

900

901

902

903

904

905

906

907

908

909

910

911

912

913

914

915

916

917

918

919

920

921

922

923

924

925

926

927

928

929

930

931

932

933

934

935

936

937

938

939

940

941

942

943

944

945

946

947

948

949

950

951

952

953

954

955

956

957

958

959

960

961

962

963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985

986

987

988

989

990

991

992

993

994

995

996

997

998

999

1000

<!DOCTYPE html>

<head>

<title>Memory Management — مستندات Python3.14.6</title><meta name="viewport" content="width=device-width, initial-scale=1.0">

title="جستجو در مستندات Python3.14.6"

href="../_static/opensearch.xml"/>

<style>

@media only screen {

table.full-width-table {

width: 100%;

}

</style>

</head>

<body>

<input type="checkbox" id="menuToggler" class="toggler__input" aria-controls="navigation"

aria-pressed="false" aria-expanded="false" role="button" aria-label="Menu">

</label>

</a>

</svg>

</form>

</nav>

Theme

<option value="light">Light</option>

</select>

</label>

<div>

<h3><a href="../contents.html">فهرست عناوین</a></h3>

<ul>

<li><a class="reference internal" href="#">Memory Management</a><ul>

<li><a class="reference internal" href="#overview">Overview</a></li>

<li><a class="reference internal" href="#allocator-domains">Allocator Domains</a></li>

<li><a class="reference internal" href="#raw-memory-interface">Raw Memory Interface</a></li>

<li><a class="reference internal" href="#memory-interface">Memory Interface</a><ul>

<li><a class="reference internal" href="#deprecated-aliases">Deprecated aliases</a></li>

</ul>

</li>

<li><a class="reference internal" href="#object-allocators">Object allocators</a></li>

<li><a class="reference internal" href="#default-memory-allocators">Default Memory Allocators</a></li>

<li><a class="reference internal" href="#customize-memory-allocators">Customize Memory Allocators</a></li>

<li><a class="reference internal" href="#debug-hooks-on-the-python-memory-allocators">Debug hooks on the Python memory allocators</a></li>

<li><a class="reference internal" href="#the-pymalloc-allocator">The pymalloc allocator</a><ul>

<li><a class="reference internal" href="#customize-pymalloc-arena-allocator">Customize pymalloc Arena Allocator</a></li>

</ul>

</li>

<li><a class="reference internal" href="#the-mimalloc-allocator">The mimalloc allocator</a></li>

<li><a class="reference internal" href="#tracemalloc-c-api">tracemalloc C API</a></li>

<li><a class="reference internal" href="#examples">Examples</a></li>

</ul>

</li>

</ul>

</div>

<div>

<h4>موضوع قبلی</h4>

<a href="init_config.html"

title="فصل قبلی">Python Initialization Configuration</a>

</div>

<div>

<h4>موضوع بعدی</h4>

<a href="objimpl.html"

title="فصل بعدی">Object Implementation Support</a>

</div>

document.addEventListener('DOMContentLoaded', () => {

const title = document.querySelector('meta[property="og:title"]').content;

const elements = document.querySelectorAll('.improvepage');

const pageurl = window.location.href.split('?')[0];

elements.forEach(element => {

const url = new URL(element.href.split('?')[0].replace("-nojs", ""));

url.searchParams.set('pagetitle', title);

url.searchParams.set('pageurl', pageurl);

url.searchParams.set('pagesource', "c-api/memory.rst");

element.href = url.toString();

});

</script>

<li><a href="../bugs.html">گزارش یک اشکال</a></li>

<li><a class="improvepage" href="../improve-page-nojs.html">بهبود این صفحه</a></li>

<li>

<a href="https://github.com/python/cpython/blob/main/Doc/c-api/memory.rst?plain=1"

rel="nofollow">نمایش منبع

</a>

</li>

<li>

<a href="https://github.com/python/python-docs-fa/blob/3.14/c-api/memory.po?plain=1"

rel="nofollow">نمایش منبع ترجمه</a>

</li>

</ul>

</div>

</nav>

</div>

<h3>ناوبری</h3>

<ul>

<a href="../genindex.html" title="فهرست کلی"

accesskey="I">فهرست</a></li>

<a href="../py-modindex.html" title="نمایه ی ماژول های پایتون"

>ماژول ها</a> |</li>

<a href="objimpl.html" title="Object Implementation Support"

accesskey="N">بعدی</a> |</li>

<a href="init_config.html" title="Python Initialization Configuration"

accesskey="P">قبلی</a> |</li>

<li><a href="https://www.python.org/">Python</a> »</li>

</li>

<li>

</li>

<a href="../index.html">3.14.6 Documentation</a> »

</li>

<li class="nav-item nav-item-1"><a href="index.html" accesskey="U">Python/C API reference manual</a> »</li>

<li class="nav-item nav-item-this"><a href="">Memory Management</a></li>

</form>

</div>

</li>

Theme

<option value="light">Light</option>

</select>

</label> |</li>

</ul>

</div>

<h1>Memory Management<a class="headerlink" href="#memory-management" title="Link to this heading">¶</a></h1>

<h2>Overview<a class="headerlink" href="#overview" title="Link to this heading">¶</a></h2>

Memory management in Python involves a private heap containing all Python

objects and data structures. The management of this private heap is ensured

internally by the Python memory manager. The Python memory manager has

different components which deal with various dynamic storage management aspects,

like sharing, segmentation, preallocation or caching.

At the lowest level, a raw memory allocator ensures that there is enough room in

the private heap for storing all Python-related data by interacting with the

memory manager of the operating system. On top of the raw memory allocator,

several object-specific allocators operate on the same heap and implement

distinct memory management policies adapted to the peculiarities of every object

type. For example, integer objects are managed differently within the heap than

strings, tuples or dictionaries because integers imply different storage

requirements and speed/space tradeoffs. The Python memory manager thus delegates

some of the work to the object-specific allocators, but ensures that the latter

operate within the bounds of the private heap.

It is important to understand that the management of the Python heap is

performed by the interpreter itself and that the user has no control over it,

even if they regularly manipulate object pointers to memory blocks inside that

heap. The allocation of heap space for Python objects and other internal

buffers is performed on demand by the Python memory manager through the Python/C

API functions listed in this document.

To avoid memory corruption, extension writers should never try to operate on

Python objects with the functions exported by the C library: <code class="xref c c-func docutils literal notranslate">malloc()</code>,

<code class="xref c c-func docutils literal notranslate">calloc()</code>, <code class="xref c c-func docutils literal notranslate">realloc()</code> and <code class="xref c c-func docutils literal notranslate">free()</code>. This will result in mixed

calls between the C allocator and the Python memory manager with fatal

consequences, because they implement different algorithms and operate on

different heaps. However, one may safely allocate and release memory blocks

with the C library allocator for individual purposes, as shown in the following

example:

<div class="highlight-c notranslate"><div class="highlight"><pre>PyObject *res;

char *buf = (char *) malloc(BUFSIZ); /* for I/O */

if (buf == NULL)

return PyErr_NoMemory();

...Do some I/O operation involving buf...

res = PyBytes_FromString(buf);

free(buf); /* malloc'ed */

return res;

</pre></div>

</div>

In this example, the memory request for the I/O buffer is handled by the C

library allocator. The Python memory manager is involved only in the allocation

of the bytes object returned as a result.

In most situations, however, it is recommended to allocate memory from the

Python heap specifically because the latter is under control of the Python

memory manager. For example, this is required when the interpreter is extended

with new object types written in C. Another reason for using the Python heap is

the desire to inform the Python memory manager about the memory needs of the

extension module. Even when the requested memory is used exclusively for

internal, highly specific purposes, delegating all memory requests to the Python

memory manager causes the interpreter to have a more accurate image of its

memory footprint as a whole. Consequently, under certain circumstances, the

Python memory manager may or may not trigger appropriate actions, like garbage

collection, memory compaction or other preventive procedures. Note that by using

the C library allocator as shown in the previous example, the allocated memory

for the I/O buffer escapes completely the Python memory manager.

همچنین ملاحظه نمائید

The <a class="reference internal" href="../using/cmdline.html#envvar-PYTHONMALLOC"><code class="xref std std-envvar docutils literal notranslate">PYTHONMALLOC</code></a> environment variable can be used to configure

the memory allocators used by Python.

The <a class="reference internal" href="../using/cmdline.html#envvar-PYTHONMALLOCSTATS"><code class="xref std std-envvar docutils literal notranslate">PYTHONMALLOCSTATS</code></a> environment variable can be used to print

statistics of the <a class="reference internal" href="#pymalloc">pymalloc memory allocator</a> every time a

new pymalloc object arena is created, and on shutdown.

</div>

</section>

<h2>Allocator Domains<a class="headerlink" href="#allocator-domains" title="Link to this heading">¶</a></h2>

All allocating functions belong to one of three different "domains" (see also

<a class="reference internal" href="#c.PyMemAllocatorDomain" title="PyMemAllocatorDomain"><code class="xref c c-type docutils literal notranslate">PyMemAllocatorDomain</code></a>). These domains represent different allocation

strategies and are optimized for different purposes. The specific details on

how every domain allocates memory or what internal functions each domain calls

is considered an implementation detail, but for debugging purposes a simplified

table can be found at <a class="reference internal" href="#default-memory-allocators">Default Memory Allocators</a>.

The APIs used to allocate and free a block of memory must be from the same domain.

For example, <a class="reference internal" href="#c.PyMem_Free" title="PyMem_Free"><code class="xref c c-func docutils literal notranslate">PyMem_Free()</code></a> must be used to free memory allocated using <a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>.

The three allocation domains are:

<li>Raw domain: intended for allocating memory for general-purpose memory

buffers where the allocation must go to the system allocator or where the

allocator can operate without an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a>. The memory

is requested directly from the system. See <a class="reference internal" href="#raw-memoryinterface">Raw Memory Interface</a>.</li>

<li>"Mem" domain: intended for allocating memory for Python buffers and

general-purpose memory buffers where the allocation must be performed with

an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a>. The memory is taken from the Python private heap.

See <a class="reference internal" href="#memoryinterface">Memory Interface</a>.</li>

<li>Object domain: intended for allocating memory for Python objects. The

memory is taken from the Python private heap. See <a class="reference internal" href="#objectinterface">Object allocators</a>.</li>

</ul>

توجه

The <a class="reference internal" href="../glossary.html#term-free-threading">free-threaded</a> build requires that only Python objects are allocated using the "object" domain

and that all Python objects are allocated using that domain. This differs from the prior Python versions,

where this was only a best practice and not a hard requirement.

For example, buffers (non-Python objects) should be allocated using <a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>,

<a class="reference internal" href="#c.PyMem_RawMalloc" title="PyMem_RawMalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawMalloc()</code></a>, or <code class="xref c c-func docutils literal notranslate">malloc()</code>, but not <a class="reference internal" href="#c.PyObject_Malloc" title="PyObject_Malloc"><code class="xref c c-func docutils literal notranslate">PyObject_Malloc()</code></a>.

See <a class="reference internal" href="../howto/free-threading-extensions.html#free-threaded-memory-allocation">Memory Allocation APIs</a>.

</div>

</section>

<h2>Raw Memory Interface<a class="headerlink" href="#raw-memory-interface" title="Link to this heading">¶</a></h2>

The following function sets are wrappers to the system allocator. These

functions are thread-safe, so a <a class="reference internal" href="../glossary.html#term-thread-state">thread state</a> does not

need to be <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached</a>.

The <a class="reference internal" href="#default-memory-allocators">default raw memory allocator</a> uses

the following functions: <code class="xref c c-func docutils literal notranslate">malloc()</code>, <code class="xref c c-func docutils literal notranslate">calloc()</code>, <code class="xref c c-func docutils literal notranslate">realloc()</code>

and <code class="xref c c-func docutils literal notranslate">free()</code>; call <code class="docutils literal notranslate">malloc(1)</code> (or <code class="docutils literal notranslate">calloc(1, 1)</code>) when requesting

zero bytes.

Added in version 3.4.

</div>

void *PyMem_RawMalloc(size_t n)<a class="headerlink" href="#c.PyMem_RawMalloc" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a> از نسخه‌ی 3.13.Allocates n bytes and returns a pointer of type void* to the

allocated memory, or <code class="docutils literal notranslate">NULL</code> if the request fails.

Requesting zero bytes returns a distinct non-<code class="docutils literal notranslate">NULL</code> pointer if possible, as

if <code class="docutils literal notranslate">PyMem_RawMalloc(1)</code> had been called instead. The memory will not have

been initialized in any way.

</dd></dl>

void *PyMem_RawCalloc(size_t nelem, size_t elsize)<a class="headerlink" href="#c.PyMem_RawCalloc" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a> از نسخه‌ی 3.13.Allocates nelem elements each whose size in bytes is elsize and returns

a pointer of type void* to the allocated memory, or <code class="docutils literal notranslate">NULL</code> if the

request fails. The memory is initialized to zeros.

Requesting zero elements or elements of size zero bytes returns a distinct

non-<code class="docutils literal notranslate">NULL</code> pointer if possible, as if <code class="docutils literal notranslate">PyMem_RawCalloc(1, 1)</code> had been

called instead.

Added in version 3.5.

</div>

</dd></dl>

void *PyMem_RawRealloc(void *p, size_t n)<a class="headerlink" href="#c.PyMem_RawRealloc" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a> از نسخه‌ی 3.13.Resizes the memory block pointed to by p to n bytes. The contents will

be unchanged to the minimum of the old and the new sizes.

If p is <code class="docutils literal notranslate">NULL</code>, the call is equivalent to <code class="docutils literal notranslate">PyMem_RawMalloc(n)</code>; else if

n is equal to zero, the memory block is resized but is not freed, and the

returned pointer is non-<code class="docutils literal notranslate">NULL</code>.

Unless p is <code class="docutils literal notranslate">NULL</code>, it must have been returned by a previous call to

<a class="reference internal" href="#c.PyMem_RawMalloc" title="PyMem_RawMalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawMalloc()</code></a>, <a class="reference internal" href="#c.PyMem_RawRealloc" title="PyMem_RawRealloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawRealloc()</code></a> or

<a class="reference internal" href="#c.PyMem_RawCalloc" title="PyMem_RawCalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawCalloc()</code></a>.

If the request fails, <a class="reference internal" href="#c.PyMem_RawRealloc" title="PyMem_RawRealloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawRealloc()</code></a> returns <code class="docutils literal notranslate">NULL</code> and p

remains a valid pointer to the previous memory area.

</dd></dl>

void PyMem_RawFree(void *p)<a class="headerlink" href="#c.PyMem_RawFree" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a> از نسخه‌ی 3.13.Frees the memory block pointed to by p, which must have been returned by a

previous call to <a class="reference internal" href="#c.PyMem_RawMalloc" title="PyMem_RawMalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawMalloc()</code></a>, <a class="reference internal" href="#c.PyMem_RawRealloc" title="PyMem_RawRealloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawRealloc()</code></a> or

<a class="reference internal" href="#c.PyMem_RawCalloc" title="PyMem_RawCalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawCalloc()</code></a>. Otherwise, or if <code class="docutils literal notranslate">PyMem_RawFree(p)</code> has been

called before, undefined behavior occurs.

If p is <code class="docutils literal notranslate">NULL</code>, no operation is performed.

</dd></dl>

</section>

<h2>Memory Interface<a class="headerlink" href="#memory-interface" title="Link to this heading">¶</a></h2>

The following function sets, modeled after the ANSI C standard, but specifying

behavior when requesting zero bytes, are available for allocating and releasing

memory from the Python heap.

In the GIL-enabled build (default build) the

<a class="reference internal" href="#default-memory-allocators">default memory allocator</a> uses the

<a class="reference internal" href="#pymalloc">pymalloc memory allocator</a>, whereas in the

<a class="reference internal" href="../glossary.html#term-free-threaded-build">free-threaded build</a>, the default is the

<a class="reference internal" href="#mimalloc">mimalloc memory allocator</a> instead.

هشدار

There must be an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a> when using these functions.

</div>

تغییر داده شده در نسخه 3.6: The default allocator is now pymalloc instead of system <code class="xref c c-func docutils literal notranslate">malloc()</code>.

</div>

تغییر داده شده در نسخه 3.13: In the <a class="reference internal" href="../glossary.html#term-free-threading">free-threaded</a> build, the default allocator

is now <a class="reference internal" href="#mimalloc">mimalloc</a>.

</div>

void *PyMem_Malloc(size_t n)<a class="headerlink" href="#c.PyMem_Malloc" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a>.Allocates n bytes and returns a pointer of type void* to the

allocated memory, or <code class="docutils literal notranslate">NULL</code> if the request fails.

Requesting zero bytes returns a distinct non-<code class="docutils literal notranslate">NULL</code> pointer if possible, as

if <code class="docutils literal notranslate">PyMem_Malloc(1)</code> had been called instead. The memory will not have

been initialized in any way.

</dd></dl>

void *PyMem_Calloc(size_t nelem, size_t elsize)<a class="headerlink" href="#c.PyMem_Calloc" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a> از نسخه‌ی 3.7.Allocates nelem elements each whose size in bytes is elsize and returns

request fails. The memory is initialized to zeros.

Requesting zero elements or elements of size zero bytes returns a distinct

non-<code class="docutils literal notranslate">NULL</code> pointer if possible, as if <code class="docutils literal notranslate">PyMem_Calloc(1, 1)</code> had been called

instead.

Added in version 3.5.

</div>

</dd></dl>

void *PyMem_Realloc(void *p, size_t n)<a class="headerlink" href="#c.PyMem_Realloc" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a>.Resizes the memory block pointed to by p to n bytes. The contents will be

unchanged to the minimum of the old and the new sizes.

is equal to zero, the memory block is resized but is not freed, and the

returned pointer is non-<code class="docutils literal notranslate">NULL</code>.

Unless p is <code class="docutils literal notranslate">NULL</code>, it must have been returned by a previous call to

<a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>, <a class="reference internal" href="#c.PyMem_Realloc" title="PyMem_Realloc"><code class="xref c c-func docutils literal notranslate">PyMem_Realloc()</code></a> or <a class="reference internal" href="#c.PyMem_Calloc" title="PyMem_Calloc"><code class="xref c c-func docutils literal notranslate">PyMem_Calloc()</code></a>.

If the request fails, <a class="reference internal" href="#c.PyMem_Realloc" title="PyMem_Realloc"><code class="xref c c-func docutils literal notranslate">PyMem_Realloc()</code></a> returns <code class="docutils literal notranslate">NULL</code> and p remains

a valid pointer to the previous memory area.

</dd></dl>

void PyMem_Free(void *p)<a class="headerlink" href="#c.PyMem_Free" title="Link to this definition">¶</a> </dt>

<dd> قسمتی از <a class="reference internal" href="stable.html#stable">ABI پایدار</a>.Frees the memory block pointed to by p, which must have been returned by a

previous call to <a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>, <a class="reference internal" href="#c.PyMem_Realloc" title="PyMem_Realloc"><code class="xref c c-func docutils literal notranslate">PyMem_Realloc()</code></a> or

<a class="reference internal" href="#c.PyMem_Calloc" title="PyMem_Calloc"><code class="xref c c-func docutils literal notranslate">PyMem_Calloc()</code></a>. Otherwise, or if <code class="docutils literal notranslate">PyMem_Free(p)</code> has been called

before, undefined behavior occurs.

If p is <code class="docutils literal notranslate">NULL</code>, no operation is performed.

</dd></dl>

The following type-oriented macros are provided for convenience. Note that

TYPE refers to any C type.

PyMem_New(TYPE, n)<a class="headerlink" href="#c.PyMem_New" title="Link to this definition">¶</a> </dt>

<dd>Same as <a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>, but allocates <code class="docutils literal notranslate">(n * sizeof(TYPE))</code> bytes of

memory. Returns a pointer cast to <code class="docutils literal notranslate">TYPE*</code>. The memory will not have

been initialized in any way.

</dd></dl>

PyMem_Resize(p, TYPE, n)<a class="headerlink" href="#c.PyMem_Resize" title="Link to this definition">¶</a> </dt>

<dd>Same as <a class="reference internal" href="#c.PyMem_Realloc" title="PyMem_Realloc"><code class="xref c c-func docutils literal notranslate">PyMem_Realloc()</code></a>, but the memory block is resized to <code class="docutils literal notranslate">(n *

sizeof(TYPE))</code> bytes. Returns a pointer cast to <code class="docutils literal notranslate">TYPE*</code>. On return,

p will be a pointer to the new memory area, or <code class="docutils literal notranslate">NULL</code> in the event of

failure.

This is a C preprocessor macro; p is always reassigned. Save the original

value of p to avoid losing memory when handling errors.

</dd></dl>

void PyMem_Del(void *p)<a class="headerlink" href="#c.PyMem_Del" title="Link to this definition">¶</a> </dt>

<dd>Same as <a class="reference internal" href="#c.PyMem_Free" title="PyMem_Free"><code class="xref c c-func docutils literal notranslate">PyMem_Free()</code></a>.

</dd></dl>

<h3>Deprecated aliases<a class="headerlink" href="#deprecated-aliases" title="Link to this heading">¶</a></h3>

These are <a class="reference internal" href="../glossary.html#term-soft-deprecated">soft deprecated</a> aliases to existing functions and macros.

They exist solely for backwards compatibility.

<thead>

<tr class="row-odd"><th class="head">Deprecated alias</th>

<th class="head">Corresponding function or macro</th>

</tr>

</thead>

<tbody>

PyMem_MALLOC(size)<a class="headerlink" href="#c.PyMem_MALLOC" title="Link to this definition">¶</a> </dt>

</td>

<td><a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a></td>

</tr>

PyMem_NEW(type, size)<a class="headerlink" href="#c.PyMem_NEW" title="Link to this definition">¶</a> </dt>

</td>

<td><a class="reference internal" href="#c.PyMem_New" title="PyMem_New"><code class="xref c c-macro docutils literal notranslate">PyMem_New</code></a></td>

</tr>

PyMem_REALLOC(ptr, size)<a class="headerlink" href="#c.PyMem_REALLOC" title="Link to this definition">¶</a> </dt>

</td>

<td><a class="reference internal" href="#c.PyMem_Realloc" title="PyMem_Realloc"><code class="xref c c-func docutils literal notranslate">PyMem_Realloc()</code></a></td>

</tr>

PyMem_RESIZE(ptr, type, size)<a class="headerlink" href="#c.PyMem_RESIZE" title="Link to this definition">¶</a> </dt>

</td>

<td><a class="reference internal" href="#c.PyMem_Resize" title="PyMem_Resize"><code class="xref c c-macro docutils literal notranslate">PyMem_Resize</code></a></td>

</tr>

PyMem_FREE(ptr)<a class="headerlink" href="#c.PyMem_FREE" title="Link to this definition">¶</a> </dt>

</td>

<td><a class="reference internal" href="#c.PyMem_Free" title="PyMem_Free"><code class="xref c c-func docutils literal notranslate">PyMem_Free()</code></a></td>

</tr>

PyMem_DEL(ptr)<a class="headerlink" href="#c.PyMem_DEL" title="Link to this definition">¶</a> </dt>

</td>

<td><a class="reference internal" href="#c.PyMem_Free" title="PyMem_Free"><code class="xref c c-func docutils literal notranslate">PyMem_Free()</code></a></td>

</tr>

</tbody>

</table>

تغییر داده شده در نسخه 3.4: The macros are now aliases of the corresponding functions and macros.

Previously, their behavior was the same, but their use did not necessarily

preserve binary compatibility across Python versions.

</div>

منسوخ شده از نسخه 2.0.

</div>

</section>

<h2>Object allocators<a class="headerlink" href="#object-allocators" title="Link to this heading">¶</a></h2>

The following function sets, modeled after the ANSI C standard, but specifying

behavior when requesting zero bytes, are available for allocating and releasing

memory from the Python heap.

توجه

There is no guarantee that the memory returned by these allocators can be

successfully cast to a Python object when intercepting the allocating

functions in this domain by the methods described in

the <a class="reference internal" href="#customize-memory-allocators">Customize Memory Allocators</a> section.

</div>

The <a class="reference internal" href="#default-memory-allocators">default object allocator</a> uses the

<a class="reference internal" href="#pymalloc">pymalloc memory allocator</a>. In the

<a class="reference internal" href="../glossary.html#term-free-threading">free-threaded</a> build, the default is the

<a class="reference internal" href="#mimalloc">mimalloc memory allocator</a> instead.

هشدار

There must be an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a> when using these functions.

</div>

void *PyObject_Malloc(size_t n)<a class="headerlink" href="#c.PyObject_Malloc" title="Link to this definition">¶</a> </dt>

allocated memory, or <code class="docutils literal notranslate">NULL</code> if the request fails.

Requesting zero bytes returns a distinct non-<code class="docutils literal notranslate">NULL</code> pointer if possible, as

if <code class="docutils literal notranslate">PyObject_Malloc(1)</code> had been called instead. The memory will not have

been initialized in any way.

</dd></dl>

void *PyObject_Calloc(size_t nelem, size_t elsize)<a class="headerlink" href="#c.PyObject_Calloc" title="Link to this definition">¶</a> </dt>

request fails. The memory is initialized to zeros.

Requesting zero elements or elements of size zero bytes returns a distinct

non-<code class="docutils literal notranslate">NULL</code> pointer if possible, as if <code class="docutils literal notranslate">PyObject_Calloc(1, 1)</code> had been called

instead.

Added in version 3.5.

</div>

</dd></dl>

void *PyObject_Realloc(void *p, size_t n)<a class="headerlink" href="#c.PyObject_Realloc" title="Link to this definition">¶</a> </dt>

unchanged to the minimum of the old and the new sizes.

is equal to zero, the memory block is resized but is not freed, and the

returned pointer is non-<code class="docutils literal notranslate">NULL</code>.

Unless p is <code class="docutils literal notranslate">NULL</code>, it must have been returned by a previous call to

<a class="reference internal" href="#c.PyObject_Malloc" title="PyObject_Malloc"><code class="xref c c-func docutils literal notranslate">PyObject_Malloc()</code></a>, <a class="reference internal" href="#c.PyObject_Realloc" title="PyObject_Realloc"><code class="xref c c-func docutils literal notranslate">PyObject_Realloc()</code></a> or <a class="reference internal" href="#c.PyObject_Calloc" title="PyObject_Calloc"><code class="xref c c-func docutils literal notranslate">PyObject_Calloc()</code></a>.

If the request fails, <a class="reference internal" href="#c.PyObject_Realloc" title="PyObject_Realloc"><code class="xref c c-func docutils literal notranslate">PyObject_Realloc()</code></a> returns <code class="docutils literal notranslate">NULL</code> and p remains

a valid pointer to the previous memory area.

</dd></dl>

void PyObject_Free(void *p)<a class="headerlink" href="#c.PyObject_Free" title="Link to this definition">¶</a> </dt>

previous call to <a class="reference internal" href="#c.PyObject_Malloc" title="PyObject_Malloc"><code class="xref c c-func docutils literal notranslate">PyObject_Malloc()</code></a>, <a class="reference internal" href="#c.PyObject_Realloc" title="PyObject_Realloc"><code class="xref c c-func docutils literal notranslate">PyObject_Realloc()</code></a> or

<a class="reference internal" href="#c.PyObject_Calloc" title="PyObject_Calloc"><code class="xref c c-func docutils literal notranslate">PyObject_Calloc()</code></a>. Otherwise, or if <code class="docutils literal notranslate">PyObject_Free(p)</code> has been called

before, undefined behavior occurs.

If p is <code class="docutils literal notranslate">NULL</code>, no operation is performed.

Do not call this directly to free an object's memory; call the type's

<a class="reference internal" href="typeobj.html#c.PyTypeObject.tp_free" title="PyTypeObject.tp_free"><code class="xref c c-member docutils literal notranslate">tp_free</code></a> slot instead.

Do not use this for memory allocated by <a class="reference internal" href="gcsupport.html#c.PyObject_GC_New" title="PyObject_GC_New"><code class="xref c c-macro docutils literal notranslate">PyObject_GC_New</code></a> or

<a class="reference internal" href="gcsupport.html#c.PyObject_GC_NewVar" title="PyObject_GC_NewVar"><code class="xref c c-macro docutils literal notranslate">PyObject_GC_NewVar</code></a>; use <a class="reference internal" href="gcsupport.html#c.PyObject_GC_Del" title="PyObject_GC_Del"><code class="xref c c-func docutils literal notranslate">PyObject_GC_Del()</code></a> instead.

همچنین ملاحظه نمائید

<li><a class="reference internal" href="gcsupport.html#c.PyObject_GC_Del" title="PyObject_GC_Del"><code class="xref c c-func docutils literal notranslate">PyObject_GC_Del()</code></a> is the equivalent of this function for memory

allocated by types that support garbage collection.</li>

<li><a class="reference internal" href="#c.PyObject_Malloc" title="PyObject_Malloc"><code class="xref c c-func docutils literal notranslate">PyObject_Malloc()</code></a></li>

<li><a class="reference internal" href="#c.PyObject_Realloc" title="PyObject_Realloc"><code class="xref c c-func docutils literal notranslate">PyObject_Realloc()</code></a></li>

<li><a class="reference internal" href="#c.PyObject_Calloc" title="PyObject_Calloc"><code class="xref c c-func docutils literal notranslate">PyObject_Calloc()</code></a></li>

<li><a class="reference internal" href="allocation.html#c.PyObject_New" title="PyObject_New"><code class="xref c c-macro docutils literal notranslate">PyObject_New</code></a></li>

<li><a class="reference internal" href="allocation.html#c.PyObject_NewVar" title="PyObject_NewVar"><code class="xref c c-macro docutils literal notranslate">PyObject_NewVar</code></a></li>

<li><a class="reference internal" href="type.html#c.PyType_GenericAlloc" title="PyType_GenericAlloc"><code class="xref c c-func docutils literal notranslate">PyType_GenericAlloc()</code></a></li>

</ul>

</div>

</dd></dl>

</section>

<h2>Default Memory Allocators<a class="headerlink" href="#default-memory-allocators" title="Link to this heading">¶</a></h2>

Default memory allocators:

<thead>

<tr class="row-odd"><th class="head">Configuration</th>

<th class="head">PyMem_RawMalloc</th>

<th class="head">PyMem_Malloc</th>

<th class="head">PyObject_Malloc</th>

</tr>

</thead>

<tbody>

<tr class="row-even"><td>Release build</td>

<td><code class="docutils literal notranslate">"pymalloc"</code></td>

<td><code class="docutils literal notranslate">malloc</code></td>

<td><code class="docutils literal notranslate">pymalloc</code></td>

</tr>

<tr class="row-odd"><td>Debug build</td>

<td><code class="docutils literal notranslate">"pymalloc_debug"</code></td>

<td><code class="docutils literal notranslate">malloc</code> + debug</td>

<td><code class="docutils literal notranslate">pymalloc</code> + debug</td>

</tr>

<tr class="row-even"><td>Release build, without pymalloc</td>

<td><code class="docutils literal notranslate">"malloc"</code></td>

<td><code class="docutils literal notranslate">malloc</code></td>

</tr>

<tr class="row-odd"><td>Debug build, without pymalloc</td>

<td><code class="docutils literal notranslate">"malloc_debug"</code></td>

<td><code class="docutils literal notranslate">malloc</code> + debug</td>

</tr>

<tr class="row-even"><td>Free-threaded build</td>

<td><code class="docutils literal notranslate">"mimalloc"</code></td>

<td><code class="docutils literal notranslate">mimalloc</code></td>

</tr>

<tr class="row-odd"><td>Free-threaded debug build</td>

<td><code class="docutils literal notranslate">"mimalloc_debug"</code></td>

<td><code class="docutils literal notranslate">mimalloc</code> + debug</td>

</tr>

</tbody>

</table>

Legend:

<li>Name: value for <a class="reference internal" href="../using/cmdline.html#envvar-PYTHONMALLOC"><code class="xref std std-envvar docutils literal notranslate">PYTHONMALLOC</code></a> environment variable.</li>

<li><code class="docutils literal notranslate">malloc</code>: system allocators from the standard C library, C functions:

<code class="xref c c-func docutils literal notranslate">malloc()</code>, <code class="xref c c-func docutils literal notranslate">calloc()</code>, <code class="xref c c-func docutils literal notranslate">realloc()</code> and <code class="xref c c-func docutils literal notranslate">free()</code>.</li>

<li><code class="docutils literal notranslate">pymalloc</code>: <a class="reference internal" href="#pymalloc">pymalloc memory allocator</a>.</li>

<li><code class="docutils literal notranslate">mimalloc</code>: <a class="reference internal" href="#mimalloc">mimalloc memory allocator</a>.</li>

<li>"+ debug": with <a class="reference internal" href="#pymem-debug-hooks">debug hooks on the Python memory allocators</a>.</li>

<li>"Debug build": <a class="reference internal" href="../using/configure.html#debug-build">Python build in debug mode</a>.</li>

</ul>

</section>

<h2>Customize Memory Allocators<a class="headerlink" href="#customize-memory-allocators" title="Link to this heading">¶</a></h2>

Added in version 3.4.

</div>

type PyMemAllocatorEx<a class="headerlink" href="#c.PyMemAllocatorEx" title="Link to this definition">¶</a> </dt>

<dd>Structure used to describe a memory block allocator. The structure has

the following fields:

<thead>

<tr class="row-odd"><th class="head">Field</th>

<th class="head">Meaning</th>

</tr>

</thead>

<tbody>

<td>user context passed as first argument</td>

</tr>

<tr class="row-odd"><td><code class="docutils literal notranslate">void* malloc(void *ctx, size_t size)</code></td>

<td>allocate a memory block</td>

</tr>

<tr class="row-even"><td><code class="docutils literal notranslate">void* calloc(void *ctx, size_t nelem, size_t elsize)</code></td>

<td>allocate a memory block initialized

with zeros</td>

</tr>

<tr class="row-odd"><td><code class="docutils literal notranslate">void* realloc(void *ctx, void *ptr, size_t new_size)</code></td>

<td>allocate or resize a memory block</td>

</tr>

<td>free a memory block</td>

</tr>

</tbody>

</table>

تغییر داده شده در نسخه 3.5: The <code class="xref c c-type docutils literal notranslate">PyMemAllocator</code> structure was renamed to

<a class="reference internal" href="#c.PyMemAllocatorEx" title="PyMemAllocatorEx"><code class="xref c c-type docutils literal notranslate">PyMemAllocatorEx</code></a> and a new <code class="docutils literal notranslate">calloc</code> field was added.

</div>

</dd></dl>

type PyMemAllocatorDomain<a class="headerlink" href="#c.PyMemAllocatorDomain" title="Link to this definition">¶</a> </dt>

<dd>Enum used to identify an allocator domain. Domains:

PYMEM_DOMAIN_RAW<a class="headerlink" href="#c.PYMEM_DOMAIN_RAW" title="Link to this definition">¶</a> </dt>

<dd>Functions:

<li><a class="reference internal" href="#c.PyMem_RawMalloc" title="PyMem_RawMalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawMalloc()</code></a></li>

<li><a class="reference internal" href="#c.PyMem_RawRealloc" title="PyMem_RawRealloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawRealloc()</code></a></li>

<li><a class="reference internal" href="#c.PyMem_RawCalloc" title="PyMem_RawCalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawCalloc()</code></a></li>

<li><a class="reference internal" href="#c.PyMem_RawFree" title="PyMem_RawFree"><code class="xref c c-func docutils literal notranslate">PyMem_RawFree()</code></a></li>

</ul>

</dd></dl>

PYMEM_DOMAIN_MEM<a class="headerlink" href="#c.PYMEM_DOMAIN_MEM" title="Link to this definition">¶</a> </dt>

<dd>Functions:

<li><a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>,</li>

<li><a class="reference internal" href="#c.PyMem_Realloc" title="PyMem_Realloc"><code class="xref c c-func docutils literal notranslate">PyMem_Realloc()</code></a></li>

<li><a class="reference internal" href="#c.PyMem_Calloc" title="PyMem_Calloc"><code class="xref c c-func docutils literal notranslate">PyMem_Calloc()</code></a></li>

<li><a class="reference internal" href="#c.PyMem_Free" title="PyMem_Free"><code class="xref c c-func docutils literal notranslate">PyMem_Free()</code></a></li>

</ul>

</dd></dl>

PYMEM_DOMAIN_OBJ<a class="headerlink" href="#c.PYMEM_DOMAIN_OBJ" title="Link to this definition">¶</a> </dt>

<dd>Functions:

<li><a class="reference internal" href="#c.PyObject_Free" title="PyObject_Free"><code class="xref c c-func docutils literal notranslate">PyObject_Free()</code></a></li>

</ul>

</dd></dl>

void PyMem_GetAllocator(<a class="reference internal" href="#c.PyMemAllocatorDomain" title="PyMemAllocatorDomain">PyMemAllocatorDomain</a> domain, <a class="reference internal" href="#c.PyMemAllocatorEx" title="PyMemAllocatorEx">PyMemAllocatorEx</a> *allocator)<a class="headerlink" href="#c.PyMem_GetAllocator" title="Link to this definition">¶</a> </dt>

<dd>Get the memory block allocator of the specified domain.

</dd></dl>

void PyMem_SetAllocator(<a class="reference internal" href="#c.PyMemAllocatorDomain" title="PyMemAllocatorDomain">PyMemAllocatorDomain</a> domain, <a class="reference internal" href="#c.PyMemAllocatorEx" title="PyMemAllocatorEx">PyMemAllocatorEx</a> *allocator)<a class="headerlink" href="#c.PyMem_SetAllocator" title="Link to this definition">¶</a> </dt>

<dd>Set the memory block allocator of the specified domain.

The new allocator must return a distinct non-<code class="docutils literal notranslate">NULL</code> pointer when requesting

zero bytes.

For the <a class="reference internal" href="#c.PYMEM_DOMAIN_RAW" title="PYMEM_DOMAIN_RAW"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_RAW</code></a> domain, the allocator must be

thread-safe: a <a class="reference internal" href="../glossary.html#term-thread-state">thread state</a> is not <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached</a>

when the allocator is called.

For the remaining domains, the allocator must also be thread-safe:

the allocator may be called in different interpreters that do not

share a <a class="reference internal" href="../glossary.html#term-GIL">GIL</a>.

If the new allocator is not a hook (does not call the previous allocator),

the <a class="reference internal" href="#c.PyMem_SetupDebugHooks" title="PyMem_SetupDebugHooks"><code class="xref c c-func docutils literal notranslate">PyMem_SetupDebugHooks()</code></a> function must be called to reinstall the

debug hooks on top on the new allocator.

See also <a class="reference internal" href="init_config.html#c.PyPreConfig.allocator" title="PyPreConfig.allocator"><code class="xref c c-member docutils literal notranslate">PyPreConfig.allocator</code></a> and <a class="reference internal" href="init_config.html#c-preinit">Preinitialize Python

with PyPreConfig</a>.

هشدار

<a class="reference internal" href="#c.PyMem_SetAllocator" title="PyMem_SetAllocator"><code class="xref c c-func docutils literal notranslate">PyMem_SetAllocator()</code></a> does have the following contract:

<li>It can be called after <a class="reference internal" href="init_config.html#c.Py_PreInitialize" title="Py_PreInitialize"><code class="xref c c-func docutils literal notranslate">Py_PreInitialize()</code></a> and before

<a class="reference internal" href="interp-lifecycle.html#c.Py_InitializeFromConfig" title="Py_InitializeFromConfig"><code class="xref c c-func docutils literal notranslate">Py_InitializeFromConfig()</code></a> to install a custom memory

allocator. There are no restrictions over the installed allocator

other than the ones imposed by the domain (for instance, the Raw

Domain allows the allocator to be called without an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a>).

See <a class="reference internal" href="#id1">the section on allocator domains</a> for more

information.</li>

<li>If called after Python has finish initializing (after

must wrap the existing allocator. Substituting the current

allocator for some other arbitrary one is not supported.</li>

</ul>

</div>

تغییر داده شده در نسخه 3.12: All allocators must be thread-safe.

</div>

</dd></dl>

void PyMem_SetupDebugHooks(void)<a class="headerlink" href="#c.PyMem_SetupDebugHooks" title="Link to this definition">¶</a> </dt>

<dd>Setup <a class="reference internal" href="#pymem-debug-hooks">debug hooks in the Python memory allocators</a>

to detect memory errors.

</dd></dl>

</section>

<h2>Debug hooks on the Python memory allocators<a class="headerlink" href="#debug-hooks-on-the-python-memory-allocators" title="Link to this heading">¶</a></h2>

When <a class="reference internal" href="../using/configure.html#debug-build">Python is built in debug mode</a>, the

<a class="reference internal" href="#c.PyMem_SetupDebugHooks" title="PyMem_SetupDebugHooks"><code class="xref c c-func docutils literal notranslate">PyMem_SetupDebugHooks()</code></a> function is called at the <a class="reference internal" href="init_config.html#c-preinit">Python

preinitialization</a> to setup debug hooks on Python memory allocators

to detect memory errors.

The <a class="reference internal" href="../using/cmdline.html#envvar-PYTHONMALLOC"><code class="xref std std-envvar docutils literal notranslate">PYTHONMALLOC</code></a> environment variable can be used to install debug

hooks on a Python compiled in release mode (ex: <code class="docutils literal notranslate">PYTHONMALLOC=debug</code>).

The <a class="reference internal" href="#c.PyMem_SetupDebugHooks" title="PyMem_SetupDebugHooks"><code class="xref c c-func docutils literal notranslate">PyMem_SetupDebugHooks()</code></a> function can be used to set debug hooks

after calling <a class="reference internal" href="#c.PyMem_SetAllocator" title="PyMem_SetAllocator"><code class="xref c c-func docutils literal notranslate">PyMem_SetAllocator()</code></a>.

These debug hooks fill dynamically allocated memory blocks with special,

recognizable bit patterns. Newly allocated memory is filled with the byte

<code class="docutils literal notranslate">0xCD</code> (<code class="docutils literal notranslate">PYMEM_CLEANBYTE</code>), freed memory is filled with the byte <code class="docutils literal notranslate">0xDD</code>

(<code class="docutils literal notranslate">PYMEM_DEADBYTE</code>). Memory blocks are surrounded by "forbidden bytes"

filled with the byte <code class="docutils literal notranslate">0xFD</code> (<code class="docutils literal notranslate">PYMEM_FORBIDDENBYTE</code>). Strings of these bytes

are unlikely to be valid addresses, floats, or ASCII strings.

Runtime checks:

<li>Detect API violations. For example, detect if <a class="reference internal" href="#c.PyObject_Free" title="PyObject_Free"><code class="xref c c-func docutils literal notranslate">PyObject_Free()</code></a> is

called on a memory block allocated by <a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>.</li>

<li>Detect write before the start of the buffer (buffer underflow).</li>

<li>Detect write after the end of the buffer (buffer overflow).</li>

<li>Check that there is an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a> when

allocator functions of <a class="reference internal" href="#c.PYMEM_DOMAIN_OBJ" title="PYMEM_DOMAIN_OBJ"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_OBJ</code></a> (ex:

</ul>

On error, the debug hooks use the <a class="reference internal" href="../library/tracemalloc.html#module-tracemalloc" title="tracemalloc: Trace memory allocations."><code class="xref py py-mod docutils literal notranslate">tracemalloc</code></a> module to get the

traceback where a memory block was allocated. The traceback is only displayed

if <code class="xref py py-mod docutils literal notranslate">tracemalloc</code> is tracing Python memory allocations and the memory block

was traced.

Let S = <code class="docutils literal notranslate">sizeof(size_t)</code>. <code class="docutils literal notranslate">2*S</code> bytes are added at each end of each block

of N bytes requested. The memory layout is like so, where p represents the

address returned by a malloc-like or realloc-like function (<code class="docutils literal notranslate">p[i:j]</code> means

the slice of bytes from <code class="docutils literal notranslate">*(p+i)</code> inclusive up to <code class="docutils literal notranslate">*(p+j)</code> exclusive; note

that the treatment of negative indices differs from a Python slice):

<dl>

<dt><code class="docutils literal notranslate">p[-2*S:-S]</code></dt><dd>Number of bytes originally asked for. This is a size_t, big-endian (easier

to read in a memory dump).

</dd>

<dt><code class="docutils literal notranslate">p[-S]</code></dt><dd>API identifier (ASCII character):

<li><code class="docutils literal notranslate">'r'</code> for <a class="reference internal" href="#c.PYMEM_DOMAIN_RAW" title="PYMEM_DOMAIN_RAW"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_RAW</code></a>.</li>

<li><code class="docutils literal notranslate">'m'</code> for <a class="reference internal" href="#c.PYMEM_DOMAIN_MEM" title="PYMEM_DOMAIN_MEM"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_MEM</code></a>.</li>

<li><code class="docutils literal notranslate">'o'</code> for <a class="reference internal" href="#c.PYMEM_DOMAIN_OBJ" title="PYMEM_DOMAIN_OBJ"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_OBJ</code></a>.</li>

</ul>

</dd>

<dt><code class="docutils literal notranslate">p[-S+1:0]</code></dt><dd>Copies of PYMEM_FORBIDDENBYTE. Used to catch under- writes and reads.

</dd>

<dt><code class="docutils literal notranslate">p[0:N]</code></dt><dd>The requested memory, filled with copies of PYMEM_CLEANBYTE, used to catch

reference to uninitialized memory. When a realloc-like function is called

requesting a larger memory block, the new excess bytes are also filled with

PYMEM_CLEANBYTE. When a free-like function is called, these are

overwritten with PYMEM_DEADBYTE, to catch reference to freed memory. When

a realloc- like function is called requesting a smaller memory block, the

excess old bytes are also filled with PYMEM_DEADBYTE.

</dd>

<dt><code class="docutils literal notranslate">p[N:N+S]</code></dt><dd>Copies of PYMEM_FORBIDDENBYTE. Used to catch over- writes and reads.

</dd>

<dt><code class="docutils literal notranslate">p[N+S:N+2*S]</code></dt><dd>Only used if the <code class="docutils literal notranslate">PYMEM_DEBUG_SERIALNO</code> macro is defined (not defined by

default).

A serial number, incremented by 1 on each call to a malloc-like or

realloc-like function. Big-endian <code class="xref c c-type docutils literal notranslate">size_t</code>. If "bad memory" is detected

later, the serial number gives an excellent way to set a breakpoint on the

next run, to capture the instant at which this block was passed out. The

static function bumpserialno() in obmalloc.c is the only place the serial

number is incremented, and exists so you can set such a breakpoint easily.

</dd>

</dl>

A realloc-like or free-like function first checks that the PYMEM_FORBIDDENBYTE

bytes at each end are intact. If they've been altered, diagnostic output is

written to stderr, and the program is aborted via Py_FatalError(). The other

main failure mode is provoking a memory error when a program reads up one of

the special bit patterns and tries to use it as an address. If you get in a

debugger then and look at the object, you're likely to see that it's entirely

filled with PYMEM_DEADBYTE (meaning freed memory is getting used) or

PYMEM_CLEANBYTE (meaning uninitialized memory is getting used).

تغییر داده شده در نسخه 3.6: The <a class="reference internal" href="#c.PyMem_SetupDebugHooks" title="PyMem_SetupDebugHooks"><code class="xref c c-func docutils literal notranslate">PyMem_SetupDebugHooks()</code></a> function now also works on Python

compiled in release mode. On error, the debug hooks now use

<a class="reference internal" href="../library/tracemalloc.html#module-tracemalloc" title="tracemalloc: Trace memory allocations."><code class="xref py py-mod docutils literal notranslate">tracemalloc</code></a> to get the traceback where a memory block was allocated.

The debug hooks now also check if there is an <a class="reference internal" href="../glossary.html#term-attached-thread-state">attached thread state</a> when

functions of <a class="reference internal" href="#c.PYMEM_DOMAIN_OBJ" title="PYMEM_DOMAIN_OBJ"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_OBJ</code></a> and <a class="reference internal" href="#c.PYMEM_DOMAIN_MEM" title="PYMEM_DOMAIN_MEM"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_MEM</code></a> domains are

called.

</div>

تغییر داده شده در نسخه 3.8: Byte patterns <code class="docutils literal notranslate">0xCB</code> (<code class="docutils literal notranslate">PYMEM_CLEANBYTE</code>), <code class="docutils literal notranslate">0xDB</code> (<code class="docutils literal notranslate">PYMEM_DEADBYTE</code>)

and <code class="docutils literal notranslate">0xFB</code> (<code class="docutils literal notranslate">PYMEM_FORBIDDENBYTE</code>) have been replaced with <code class="docutils literal notranslate">0xCD</code>,

<code class="docutils literal notranslate">0xDD</code> and <code class="docutils literal notranslate">0xFD</code> to use the same values than Windows CRT debug

<code class="docutils literal notranslate">malloc()</code> and <code class="docutils literal notranslate">free()</code>.

</div>

</section>

<h2>The pymalloc allocator<a class="headerlink" href="#the-pymalloc-allocator" title="Link to this heading">¶</a></h2>

Python has a pymalloc allocator optimized for small objects (smaller or equal

to 512 bytes) with a short lifetime. It uses memory mappings called "arenas"

with a fixed size of either 256 KiB on 32-bit platforms or 1 MiB on 64-bit

platforms. It falls back to <a class="reference internal" href="#c.PyMem_RawMalloc" title="PyMem_RawMalloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawMalloc()</code></a> and

<a class="reference internal" href="#c.PyMem_RawRealloc" title="PyMem_RawRealloc"><code class="xref c c-func docutils literal notranslate">PyMem_RawRealloc()</code></a> for allocations larger than 512 bytes.

pymalloc is the <a class="reference internal" href="#default-memory-allocators">default allocator</a> of the

<a class="reference internal" href="#c.PYMEM_DOMAIN_MEM" title="PYMEM_DOMAIN_MEM"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_MEM</code></a> (ex: <a class="reference internal" href="#c.PyMem_Malloc" title="PyMem_Malloc"><code class="xref c c-func docutils literal notranslate">PyMem_Malloc()</code></a>) and

<a class="reference internal" href="#c.PYMEM_DOMAIN_OBJ" title="PYMEM_DOMAIN_OBJ"><code class="xref c c-macro docutils literal notranslate">PYMEM_DOMAIN_OBJ</code></a> (ex: <a class="reference internal" href="#c.PyObject_Malloc" title="PyObject_Malloc"><code class="xref c c-func docutils literal notranslate">PyObject_Malloc()</code></a>) domains.

The arena allocator uses the following functions:

<li><code class="xref c c-func docutils literal notranslate">VirtualAlloc()</code> and <code class="xref c c-func docutils literal notranslate">VirtualFree()</code> on Windows,</li>

<li><code class="xref c c-func docutils literal notranslate">mmap()</code> and <code class="xref c c-func docutils literal notranslate">munmap()</code> if available,</li>

<li><code class="xref c c-func docutils literal notranslate">malloc()</code> and <code class="xref c c-func docutils literal notranslate">free()</code> otherwise.</li>

</ul>

This allocator is disabled if Python is configured with the

<a class="reference internal" href="../using/configure.html#cmdoption-without-pymalloc"><code class="xref std std-option docutils literal notranslate">--without-pymalloc</code></a> option. It can also be disabled at runtime using

the <a class="reference internal" href="../using/cmdline.html#envvar-PYTHONMALLOC"><code class="xref std std-envvar docutils literal notranslate">PYTHONMALLOC</code></a> environment variable (ex: <code class="docutils literal notranslate">PYTHONMALLOC=malloc</code>).

Typically, it makes sense to disable the pymalloc allocator when building

Python with AddressSanitizer (<a class="reference internal" href="../using/configure.html#cmdoption-with-address-sanitizer"><code class="xref std std-option docutils literal notranslate">--with-address-sanitizer</code></a>) which helps

uncover low level bugs within the C code.

<h3>Customize pymalloc Arena Allocator<a class="headerlink" href="#customize-pymalloc-arena-allocator" title="Link to this heading">¶</a></h3>

Added in version 3.4.

</div>

type PyObjectArenaAllocator<a class="headerlink" href="#c.PyObjectArenaAllocator" title="Link to this definition">¶</a> </dt>

<dd>Structure used to describe an arena allocator. The structure has

three fields:

<thead>

<tr class="row-odd"><th class="head">Field</th>

<th class="head">Meaning</th>

</tr>

</thead>

<tbody>

<td>user context passed as first argument</td>

</tr>

<tr class="row-odd"><td><code class="docutils literal notranslate">void* alloc(void *ctx, size_t size)</code></td>

<td>allocate an arena of size bytes</td>

</tr>

<td>free an arena</td>

</tr>

</tbody>

</table>

</dd></dl>

void PyObject_GetArenaAllocator(<a class="reference internal" href="#c.PyObjectArenaAllocator" title="PyObjectArenaAllocator">PyObjectArenaAllocator</a> *allocator)<a class="headerlink" href="#c.PyObject_GetArenaAllocator" title="Link to this definition">¶</a> </dt>

<dd>Get the arena allocator.

</dd></dl>

View remainder of file in raw view

Navigation Menu

Uh oh!

Uh oh!

FilesExpand file tree

memory.html

Latest commit

History

memory.html

File metadata and controls