cpython/Modules/_remote_debugging/code_objects.c at main · python/cpython · GitHub

History

550 lines (487 loc) · 18.7 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

/******************************************************************************

* Remote Debugging Module - Code Object Functions

* This file contains functions for parsing code objects and line tables

* from remote process memory.

******************************************************************************/

#include "_remote_debugging.h"

#define MAX_LINETABLE_SIZE (64 * 1024)

/* ============================================================================

* TLBC CACHING FUNCTIONS (Py_GIL_DISABLED only)

* ============================================================================ */

#ifdef Py_GIL_DISABLED

void

tlbc_cache_entry_destroy(void *ptr)

{

TLBCCacheEntry *entry = (TLBCCacheEntry *)ptr;

if (entry->tlbc_array) {

PyMem_RawFree(entry->tlbc_array);

}

PyMem_RawFree(entry);

}

TLBCCacheEntry *

get_tlbc_cache_entry(RemoteUnwinderObject *self, uintptr_t code_addr, uint32_t current_generation)

{

void *key = (void *)code_addr;

TLBCCacheEntry *entry = _Py_hashtable_get(self->tlbc_cache, key);

if (entry && entry->generation != current_generation) {

// Entry is stale, remove it by setting to NULL

_Py_hashtable_set(self->tlbc_cache, key, NULL);

entry = NULL;

}

return entry;

}

int

cache_tlbc_array(RemoteUnwinderObject *unwinder, uintptr_t code_addr, uintptr_t tlbc_array_addr, uint32_t generation)

{

uintptr_t tlbc_array_ptr;

void *tlbc_array = NULL;

TLBCCacheEntry *entry = NULL;

// Read the TLBC array pointer

if (read_ptr(unwinder, tlbc_array_addr, &tlbc_array_ptr) != 0) {

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read TLBC array pointer");

return 0; // Read error

}

// Validate TLBC array pointer

if (tlbc_array_ptr == 0) {

PyErr_SetString(PyExc_RuntimeError, "TLBC array pointer is NULL");

return 0; // No TLBC array

}

// Read the TLBC array size

Py_ssize_t tlbc_size;

if (_Py_RemoteDebug_PagedReadRemoteMemory(&unwinder->handle, tlbc_array_ptr, sizeof(tlbc_size), &tlbc_size) != 0) {

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read TLBC array size");

return 0; // Read error

}

// Validate TLBC array size

if (tlbc_size <= 0) {

PyErr_SetString(PyExc_RuntimeError, "Invalid TLBC array size");

return 0; // Invalid size

}

if (tlbc_size > MAX_TLBC_SIZE) {

PyErr_SetString(PyExc_RuntimeError, "TLBC array size exceeds maximum limit");

return 0; // Invalid size

}

assert(tlbc_size > 0 && tlbc_size <= MAX_TLBC_SIZE);

// Allocate and read the entire TLBC array

size_t array_data_size = tlbc_size * sizeof(void*);

tlbc_array = PyMem_RawMalloc(sizeof(Py_ssize_t) + array_data_size);

if (!tlbc_array) {

PyErr_NoMemory();

set_exception_cause(unwinder, PyExc_MemoryError, "Failed to allocate TLBC array");

return 0; // Memory error

}

if (_Py_RemoteDebug_PagedReadRemoteMemory(&unwinder->handle, tlbc_array_ptr, sizeof(Py_ssize_t) + array_data_size, tlbc_array) != 0) {

PyMem_RawFree(tlbc_array);

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read TLBC array data");

return 0; // Read error

}

// Create cache entry

entry = PyMem_RawMalloc(sizeof(TLBCCacheEntry));

if (!entry) {

PyErr_NoMemory();

PyMem_RawFree(tlbc_array);

set_exception_cause(unwinder, PyExc_MemoryError, "Failed to allocate TLBC cache entry");

return 0; // Memory error

}

entry->tlbc_array = tlbc_array;

entry->tlbc_array_size = tlbc_size;

entry->generation = generation;

// Store in cache

void *key = (void *)code_addr;

if (_Py_hashtable_set(unwinder->tlbc_cache, key, entry) < 0) {

tlbc_cache_entry_destroy(entry);

PyErr_NoMemory();

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to store TLBC entry in cache");

return 0; // Cache error

}

return 1; // Success

}

#endif

/* ============================================================================

* LINE TABLE PARSING FUNCTIONS

* ============================================================================ */

// Inline helper for bounds-checked byte reading (no function call overhead)

static inline int

read_byte(const uint8_t **ptr, const uint8_t *end, uint8_t *out)

{

if (*ptr >= end) {

return -1;

}

*out = *(*ptr)++;

return 0;

}

static int

scan_varint(const uint8_t **ptr, const uint8_t *end)

{

uint8_t read;

if (read_byte(ptr, end, &read) < 0) {

return -1;

}

unsigned int val = read & 63;

unsigned int shift = 0;

while (read & 64) {

if (read_byte(ptr, end, &read) < 0) {

return -1;

}

shift += 6;

// Prevent infinite loop on malformed data (shift overflow)

if (shift > 28) {

return -1;

}

val |= (read & 63) << shift;

}

return (int)val;

}

static int

scan_signed_varint(const uint8_t **ptr, const uint8_t *end)

{

int uval = scan_varint(ptr, end);

if (uval < 0) {

return INT_MIN; // Error sentinel (valid signed varints won't be INT_MIN)

}

if (uval & 1) {

return -(int)((unsigned int)uval >> 1);

}

else {

return (int)((unsigned int)uval >> 1);

}

bool

parse_linetable(const uintptr_t addrq, const char* linetable, Py_ssize_t linetable_size,

int firstlineno, LocationInfo* info)

{

// Reject garbage: zero or negative size

if (linetable_size <= 0) {

return false;

}

const uint8_t* ptr = (const uint8_t*)(linetable);

const uint8_t* end = ptr + linetable_size;

uintptr_t addr = 0;

int computed_line = firstlineno; // Running accumulator, separate from output

int val; // Temporary for varint results

uint8_t byte; // Temporary for byte reads

while (ptr < end && *ptr != '\0') {

uint8_t first_byte = *(ptr++);

uint8_t code = (first_byte >> 3) & 15;

size_t length = (first_byte & 7) + 1;

uintptr_t end_addr = addr + length;

switch (code) {

case PY_CODE_LOCATION_INFO_NONE:

info->lineno = info->end_lineno = -1;

info->column = info->end_column = -1;

break;

case PY_CODE_LOCATION_INFO_LONG:

val = scan_signed_varint(&ptr, end);

if (val == INT_MIN) {

return false;

}

computed_line += val;

info->lineno = computed_line;

val = scan_varint(&ptr, end);

if (val < 0) {

return false;

}

info->end_lineno = computed_line + val;

val = scan_varint(&ptr, end);

if (val < 0) {

return false;

}

info->column = val - 1;

val = scan_varint(&ptr, end);

if (val < 0) {

return false;

}

info->end_column = val - 1;

break;

case PY_CODE_LOCATION_INFO_NO_COLUMNS:

val = scan_signed_varint(&ptr, end);

if (val == INT_MIN) {

return false;

}

computed_line += val;

info->lineno = info->end_lineno = computed_line;

info->column = info->end_column = -1;

break;

case PY_CODE_LOCATION_INFO_ONE_LINE0:

case PY_CODE_LOCATION_INFO_ONE_LINE1:

case PY_CODE_LOCATION_INFO_ONE_LINE2:

computed_line += code - 10;

info->lineno = info->end_lineno = computed_line;

if (read_byte(&ptr, end, &byte) < 0) {

return false;

}

info->column = byte;

if (read_byte(&ptr, end, &byte) < 0) {

return false;

}

info->end_column = byte;

break;

default: {

if (read_byte(&ptr, end, &byte) < 0) {

return false;

}

if ((byte & 128) != 0) {

return false;

}

info->lineno = info->end_lineno = computed_line;

info->column = code << 3 | (byte >> 4);

info->end_column = info->column + (byte & 15);

break;

}

if (addr <= addrq && end_addr > addrq) {

return true;

}

addr = end_addr;

}

return false;

}

/* ============================================================================

* CODE OBJECT AND FRAME INFO FUNCTIONS

* ============================================================================ */

PyObject *

make_location_info(RemoteUnwinderObject *unwinder, int lineno, int end_lineno,

int col_offset, int end_col_offset)

{

RemoteDebuggingState *state = RemoteDebugging_GetStateFromObject((PyObject*)unwinder);

PyObject *info = PyStructSequence_New(state->LocationInfo_Type);

if (info == NULL) {

set_exception_cause(unwinder, PyExc_MemoryError, "Failed to create LocationInfo");

return NULL;

}

PyObject *py_lineno = PyLong_FromLong(lineno);

if (py_lineno == NULL) {

Py_DECREF(info);

return NULL;

}

PyStructSequence_SetItem(info, 0, py_lineno); // steals reference

PyObject *py_end_lineno = PyLong_FromLong(end_lineno);

if (py_end_lineno == NULL) {

Py_DECREF(info);

return NULL;

}

PyStructSequence_SetItem(info, 1, py_end_lineno); // steals reference

PyObject *py_col_offset = PyLong_FromLong(col_offset);

if (py_col_offset == NULL) {

Py_DECREF(info);

return NULL;

}

PyStructSequence_SetItem(info, 2, py_col_offset); // steals reference

PyObject *py_end_col_offset = PyLong_FromLong(end_col_offset);

if (py_end_col_offset == NULL) {

Py_DECREF(info);

return NULL;

}

PyStructSequence_SetItem(info, 3, py_end_col_offset); // steals reference

return info;

}

PyObject *

make_frame_info(RemoteUnwinderObject *unwinder, PyObject *file, PyObject *location,

PyObject *func, PyObject *opcode)

{

RemoteDebuggingState *state = RemoteDebugging_GetStateFromObject((PyObject*)unwinder);

PyObject *info = PyStructSequence_New(state->FrameInfo_Type);

if (info == NULL) {

set_exception_cause(unwinder, PyExc_MemoryError, "Failed to create FrameInfo");

return NULL;

}

Py_INCREF(file);

Py_INCREF(location);

Py_INCREF(func);

Py_INCREF(opcode);

PyStructSequence_SetItem(info, 0, file);

PyStructSequence_SetItem(info, 1, location);

PyStructSequence_SetItem(info, 2, func);

PyStructSequence_SetItem(info, 3, opcode);

return info;

}

int

parse_code_object(RemoteUnwinderObject *unwinder,

PyObject **result,

const CodeObjectContext *ctx)

{

void *key = (void *)ctx->code_addr;

CachedCodeMetadata *meta = NULL;

PyObject *func = NULL;

PyObject *file = NULL;

PyObject *linetable = NULL;

#ifdef Py_GIL_DISABLED

// In free threading builds, code object addresses might have the low bit set

// as a flag, so we need to mask it off to get the real address

uintptr_t real_address = ctx->code_addr & (~1);

#else

uintptr_t real_address = ctx->code_addr;

#endif

if (unwinder && unwinder->code_object_cache != NULL) {

meta = _Py_hashtable_get(unwinder->code_object_cache, key);

if (meta) {

STATS_INC(unwinder, code_object_cache_hits);

} else {

STATS_INC(unwinder, code_object_cache_misses);

}

if (meta == NULL) {

char code_object[SIZEOF_CODE_OBJ];

if (_Py_RemoteDebug_PagedReadRemoteMemory(

&unwinder->handle, real_address, SIZEOF_CODE_OBJ, code_object) < 0)

{

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read code object");

goto error;

}

func = read_py_str(unwinder,

GET_MEMBER(uintptr_t, code_object, unwinder->debug_offsets.code_object.qualname), 1024);

if (!func) {

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read function name from code object");

goto error;

}

file = read_py_str(unwinder,

GET_MEMBER(uintptr_t, code_object, unwinder->debug_offsets.code_object.filename), 1024);

if (!file) {

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read filename from code object");

goto error;

}

linetable = read_py_bytes(unwinder,

GET_MEMBER(uintptr_t, code_object, unwinder->debug_offsets.code_object.linetable),

MAX_LINETABLE_SIZE);

if (!linetable) {

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to read linetable from code object");

goto error;

}

meta = PyMem_RawMalloc(sizeof(CachedCodeMetadata));

if (!meta) {

PyErr_NoMemory();

set_exception_cause(unwinder, PyExc_MemoryError, "Failed to allocate cached code metadata");

goto error;

}

meta->func_name = func;

meta->file_name = file;

meta->linetable = linetable;

meta->last_frame_info = NULL;

meta->last_addrq = -1;

meta->first_lineno = GET_MEMBER(int, code_object, unwinder->debug_offsets.code_object.firstlineno);

meta->addr_code_adaptive = real_address + (uintptr_t)unwinder->debug_offsets.code_object.co_code_adaptive;

if (unwinder && unwinder->code_object_cache && _Py_hashtable_set(unwinder->code_object_cache, key, meta) < 0) {

// Ownership of func/file/linetable was transferred to meta,

// so NULL them before destroying meta to prevent double-free

// in the error label's Py_XDECREF calls.

func = NULL;

file = NULL;

linetable = NULL;

cached_code_metadata_destroy(meta);

PyErr_NoMemory();

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to cache code metadata");

goto error;

}

// Ownership transferred to meta

func = NULL;

file = NULL;

linetable = NULL;

}

uintptr_t ip = ctx->instruction_pointer;

ptrdiff_t addrq;

#ifdef Py_GIL_DISABLED

// Handle thread-local bytecode (TLBC) in free threading builds

if (ctx->tlbc_index == 0 || unwinder == NULL || unwinder->debug_offsets.code_object.co_tlbc == 0) {

// No TLBC or no unwinder - use main bytecode directly

addrq = (uint16_t *)ip - (uint16_t *)meta->addr_code_adaptive;

goto done_tlbc;

}

// Try to get TLBC data from cache (we'll get generation from the caller)

TLBCCacheEntry *tlbc_entry = get_tlbc_cache_entry(unwinder, real_address, unwinder->tlbc_generation);

if (!tlbc_entry) {

// Cache miss - try to read and cache TLBC array

if (!cache_tlbc_array(unwinder, real_address, real_address + unwinder->debug_offsets.code_object.co_tlbc, unwinder->tlbc_generation)) {

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to cache TLBC array");

goto error;

}

tlbc_entry = get_tlbc_cache_entry(unwinder, real_address, unwinder->tlbc_generation);

}

// Validate tlbc_index and check TLBC cache

if (tlbc_entry) {

// Validate index bounds (also catches negative values since tlbc_index is signed)

if (ctx->tlbc_index < 0 || ctx->tlbc_index >= tlbc_entry->tlbc_array_size) {

PyErr_Format(PyExc_RuntimeError,

"Invalid tlbc_index %d (array size %zd, corrupted remote memory)",

ctx->tlbc_index, tlbc_entry->tlbc_array_size);

set_exception_cause(unwinder, PyExc_RuntimeError,

"Invalid tlbc_index (corrupted remote memory)");

goto error;

}

assert(tlbc_entry->tlbc_array_size > 0);

// Use cached TLBC data

uintptr_t *entries = (uintptr_t *)((char *)tlbc_entry->tlbc_array + sizeof(Py_ssize_t));

uintptr_t tlbc_bytecode_addr = entries[ctx->tlbc_index];

if (tlbc_bytecode_addr != 0) {

// Calculate offset from TLBC bytecode

addrq = (uint16_t *)ip - (uint16_t *)tlbc_bytecode_addr;

goto done_tlbc;

}

// Fall back to main bytecode (no tlbc_entry or tlbc_bytecode_addr was 0)

addrq = (uint16_t *)ip - (uint16_t *)meta->addr_code_adaptive;

done_tlbc:

#else

// Non-free-threaded build, always use the main bytecode

addrq = (uint16_t *)ip - (uint16_t *)meta->addr_code_adaptive;

#endif

; // Empty statement to avoid C23 extension warning

if (!unwinder->opcodes && meta->last_frame_info != NULL && meta->last_addrq == addrq) {

*result = Py_NewRef(meta->last_frame_info);

return 0;

}

LocationInfo info = {0};

bool ok = parse_linetable(addrq, PyBytes_AS_STRING(meta->linetable),

PyBytes_GET_SIZE(meta->linetable),

meta->first_lineno, &info);

if (!ok) {

info.lineno = -1;

info.end_lineno = -1;

info.column = -1;

info.end_column = -1;

}

// Create the LocationInfo structseq: (lineno, end_lineno, col_offset, end_col_offset)

PyObject *location = make_location_info(unwinder,

info.lineno,

info.end_lineno,

info.column,

info.end_column);

if (!location) {

goto error;

}

// Read the instruction opcode from target process if opcodes flag is set

PyObject *opcode_obj = NULL;

if (unwinder->opcodes) {

uint16_t instruction_word = 0;

if (_Py_RemoteDebug_PagedReadRemoteMemory(&unwinder->handle, ip,

sizeof(uint16_t), &instruction_word) == 0) {

opcode_obj = PyLong_FromLong(instruction_word & 0xFF);

if (!opcode_obj) {

Py_DECREF(location);

set_exception_cause(unwinder, PyExc_RuntimeError, "Failed to create opcode object");

goto error;

}

} else {

// Opcode read failed - clear the exception since opcode is optional

PyErr_Clear();

}

PyObject *tuple = make_frame_info(unwinder, meta->file_name, location,

meta->func_name, opcode_obj ? opcode_obj : Py_None);

Py_DECREF(location);

Py_XDECREF(opcode_obj);

if (!tuple) {

goto error;

}

if (!unwinder->opcodes) {

Py_XSETREF(meta->last_frame_info, Py_NewRef(tuple));

meta->last_addrq = addrq;

}

*result = tuple;

return 0;

error:

Py_XDECREF(func);

Py_XDECREF(file);

Py_XDECREF(linetable);

return -1;

}

cpython/Modules/_remote_debugging/code_objects.c at main · python/cpython · GitHub

Uh oh!

Uh oh!

cpython/Modules/_remote_debugging/code_objects.c at main · python/cpython · GitHub

Uh oh!

Uh oh!

Expand file tree

Breadcrumbs

code_objects.c

Latest commit

History

Breadcrumbs

code_objects.c

File metadata and controls