1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650
|
/*
Stockfish, a UCI chess playing engine derived from Glaurung 2.1
Copyright (C) 2004-2020 The Stockfish developers (see AUTHORS file)
Stockfish is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Stockfish is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#ifdef _WIN32
#if _WIN32_WINNT < 0x0601
#undef _WIN32_WINNT
#define _WIN32_WINNT 0x0601 // Force to include needed API prototypes
#endif
#ifndef NOMINMAX
#define NOMINMAX
#endif
#include <windows.h>
// The needed Windows API for processor groups could be missed from old Windows
// versions, so instead of calling them directly (forcing the linker to resolve
// the calls at compile time), try to load them at runtime. To do this we need
// first to define the corresponding function pointers.
extern "C" {
typedef bool(*fun1_t)(LOGICAL_PROCESSOR_RELATIONSHIP,
PSYSTEM_LOGICAL_PROCESSOR_INFORMATION_EX, PDWORD);
typedef bool(*fun2_t)(USHORT, PGROUP_AFFINITY);
typedef bool(*fun3_t)(HANDLE, CONST GROUP_AFFINITY*, PGROUP_AFFINITY);
}
#endif
#include <fstream>
#include <iomanip>
#include <iostream>
#include <sstream>
#include <vector>
#include <cstdlib>
#if defined(__linux__) && !defined(__ANDROID__)
#include <stdlib.h>
#include <sys/mman.h>
#endif
#if defined(__APPLE__) || defined(__ANDROID__) || defined(__OpenBSD__) || (defined(__GLIBCXX__) && !defined(_GLIBCXX_HAVE_ALIGNED_ALLOC) && !defined(_WIN32))
#define POSIXALIGNEDALLOC
#include <stdlib.h>
#endif
#include "misc.h"
#include "thread.h"
using namespace std;
namespace {
/// Version number. If Version is left empty, then compile date in the format
/// DD-MM-YY and show in engine_info.
const string Version = "12";
/// Our fancy logging facility. The trick here is to replace cin.rdbuf() and
/// cout.rdbuf() with two Tie objects that tie cin and cout to a file stream. We
/// can toggle the logging of std::cout and std:cin at runtime whilst preserving
/// usual I/O functionality, all without changing a single line of code!
/// Idea from http://groups.google.com/group/comp.lang.c++/msg/1d941c0f26ea0d81
struct Tie: public streambuf { // MSVC requires split streambuf for cin and cout
Tie(streambuf* b, streambuf* l) : buf(b), logBuf(l) {}
int sync() override { return logBuf->pubsync(), buf->pubsync(); }
int overflow(int c) override { return log(buf->sputc((char)c), "<< "); }
int underflow() override { return buf->sgetc(); }
int uflow() override { return log(buf->sbumpc(), ">> "); }
streambuf *buf, *logBuf;
int log(int c, const char* prefix) {
static int last = '\n'; // Single log file
if (last == '\n')
logBuf->sputn(prefix, 3);
return last = logBuf->sputc((char)c);
}
};
class Logger {
Logger() : in(cin.rdbuf(), file.rdbuf()), out(cout.rdbuf(), file.rdbuf()) {}
~Logger() { start(""); }
ofstream file;
Tie in, out;
public:
static void start(const std::string& fname) {
static Logger l;
if (!fname.empty() && !l.file.is_open())
{
l.file.open(fname, ifstream::out);
if (!l.file.is_open())
{
cerr << "Unable to open debug log file " << fname << endl;
exit(EXIT_FAILURE);
}
cin.rdbuf(&l.in);
cout.rdbuf(&l.out);
}
else if (fname.empty() && l.file.is_open())
{
cout.rdbuf(l.out.buf);
cin.rdbuf(l.in.buf);
l.file.close();
}
}
};
} // namespace
/// engine_info() returns the full name of the current Stockfish version. This
/// will be either "Stockfish <Tag> DD-MM-YY" (where DD-MM-YY is the date when
/// the program was compiled) or "Stockfish <Version>", depending on whether
/// Version is empty.
const string engine_info(bool to_uci) {
const string months("Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec");
string month, day, year;
stringstream ss, date(__DATE__); // From compiler, format is "Sep 21 2008"
ss << "Stockfish " << Version << setfill('0');
if (Version.empty())
{
date >> month >> day >> year;
ss << setw(2) << day << setw(2) << (1 + months.find(month) / 4) << year.substr(2);
}
ss << (to_uci ? "\nid author ": " by ")
<< "the Stockfish developers (see AUTHORS file)";
return ss.str();
}
/// compiler_info() returns a string trying to describe the compiler we use
const std::string compiler_info() {
#define stringify2(x) #x
#define stringify(x) stringify2(x)
#define make_version_string(major, minor, patch) stringify(major) "." stringify(minor) "." stringify(patch)
/// Predefined macros hell:
///
/// __GNUC__ Compiler is gcc, Clang or Intel on Linux
/// __INTEL_COMPILER Compiler is Intel
/// _MSC_VER Compiler is MSVC or Intel on Windows
/// _WIN32 Building on Windows (any)
/// _WIN64 Building on Windows 64 bit
std::string compiler = "\nCompiled by ";
#ifdef __clang__
compiler += "clang++ ";
compiler += make_version_string(__clang_major__, __clang_minor__, __clang_patchlevel__);
#elif __INTEL_COMPILER
compiler += "Intel compiler ";
compiler += "(version ";
compiler += stringify(__INTEL_COMPILER) " update " stringify(__INTEL_COMPILER_UPDATE);
compiler += ")";
#elif _MSC_VER
compiler += "MSVC ";
compiler += "(version ";
compiler += stringify(_MSC_FULL_VER) "." stringify(_MSC_BUILD);
compiler += ")";
#elif __GNUC__
compiler += "g++ (GNUC) ";
compiler += make_version_string(__GNUC__, __GNUC_MINOR__, __GNUC_PATCHLEVEL__);
#else
compiler += "Unknown compiler ";
compiler += "(unknown version)";
#endif
#if defined(__APPLE__)
compiler += " on Apple";
#elif defined(__CYGWIN__)
compiler += " on Cygwin";
#elif defined(__MINGW64__)
compiler += " on MinGW64";
#elif defined(__MINGW32__)
compiler += " on MinGW32";
#elif defined(__ANDROID__)
compiler += " on Android";
#elif defined(__linux__)
compiler += " on Linux";
#elif defined(_WIN64)
compiler += " on Microsoft Windows 64-bit";
#elif defined(_WIN32)
compiler += " on Microsoft Windows 32-bit";
#else
compiler += " on unknown system";
#endif
compiler += "\nCompilation settings include: ";
compiler += (Is64Bit ? " 64bit" : " 32bit");
#if defined(USE_VNNI)
compiler += " VNNI";
#endif
#if defined(USE_AVX512)
compiler += " AVX512";
#endif
compiler += (HasPext ? " BMI2" : "");
#if defined(USE_AVX2)
compiler += " AVX2";
#endif
#if defined(USE_SSE41)
compiler += " SSE41";
#endif
#if defined(USE_SSSE3)
compiler += " SSSE3";
#endif
#if defined(USE_SSE2)
compiler += " SSE2";
#endif
compiler += (HasPopCnt ? " POPCNT" : "");
#if defined(USE_MMX)
compiler += " MMX";
#endif
#if defined(USE_NEON)
compiler += " NEON";
#endif
#if !defined(NDEBUG)
compiler += " DEBUG";
#endif
compiler += "\n__VERSION__ macro expands to: ";
#ifdef __VERSION__
compiler += __VERSION__;
#else
compiler += "(undefined macro)";
#endif
compiler += "\n";
return compiler;
}
/// Debug functions used mainly to collect run-time statistics
static std::atomic<int64_t> hits[2], means[2];
void dbg_hit_on(bool b) { ++hits[0]; if (b) ++hits[1]; }
void dbg_hit_on(bool c, bool b) { if (c) dbg_hit_on(b); }
void dbg_mean_of(int v) { ++means[0]; means[1] += v; }
void dbg_print() {
if (hits[0])
cerr << "Total " << hits[0] << " Hits " << hits[1]
<< " hit rate (%) " << 100 * hits[1] / hits[0] << endl;
if (means[0])
cerr << "Total " << means[0] << " Mean "
<< (double)means[1] / means[0] << endl;
}
/// Used to serialize access to std::cout to avoid multiple threads writing at
/// the same time.
std::ostream& operator<<(std::ostream& os, SyncCout sc) {
static std::mutex m;
if (sc == IO_LOCK)
m.lock();
if (sc == IO_UNLOCK)
m.unlock();
return os;
}
/// Trampoline helper to avoid moving Logger to misc.h
void start_logger(const std::string& fname) { Logger::start(fname); }
/// prefetch() preloads the given address in L1/L2 cache. This is a non-blocking
/// function that doesn't stall the CPU waiting for data to be loaded from memory,
/// which can be quite slow.
#ifdef NO_PREFETCH
void prefetch(void*) {}
#else
void prefetch(void* addr) {
# if defined(__INTEL_COMPILER)
// This hack prevents prefetches from being optimized away by
// Intel compiler. Both MSVC and gcc seem not be affected by this.
__asm__ ("");
# endif
# if defined(__INTEL_COMPILER) || defined(_MSC_VER)
_mm_prefetch((char*)addr, _MM_HINT_T0);
# else
__builtin_prefetch(addr);
# endif
}
#endif
/// std_aligned_alloc() is our wrapper for systems where the c++17 implementation
/// does not guarantee the availability of aligned_alloc(). Memory allocated with
/// std_aligned_alloc() must be freed with std_aligned_free().
void* std_aligned_alloc(size_t alignment, size_t size) {
#if defined(POSIXALIGNEDALLOC)
void *mem;
return posix_memalign(&mem, alignment, size) ? nullptr : mem;
#elif defined(_WIN32)
return _mm_malloc(size, alignment);
#else
return std::aligned_alloc(alignment, size);
#endif
}
void std_aligned_free(void* ptr) {
#if defined(POSIXALIGNEDALLOC)
free(ptr);
#elif defined(_WIN32)
_mm_free(ptr);
#else
free(ptr);
#endif
}
/// aligned_ttmem_alloc() will return suitably aligned memory, if possible using large pages.
/// The returned pointer is the aligned one, while the mem argument is the one that needs
/// to be passed to free. With c++17 some of this functionality could be simplified.
#if defined(__linux__) && !defined(__ANDROID__)
void* aligned_ttmem_alloc(size_t allocSize, void*& mem) {
constexpr size_t alignment = 2 * 1024 * 1024; // assumed 2MB page sizes
size_t size = ((allocSize + alignment - 1) / alignment) * alignment; // multiple of alignment
if (posix_memalign(&mem, alignment, size))
mem = nullptr;
#if defined(MADV_HUGEPAGE)
madvise(mem, allocSize, MADV_HUGEPAGE);
#endif
return mem;
}
#elif defined(_WIN64)
static void* aligned_ttmem_alloc_large_pages(size_t allocSize) {
HANDLE hProcessToken { };
LUID luid { };
void* mem = nullptr;
const size_t largePageSize = GetLargePageMinimum();
if (!largePageSize)
return nullptr;
// We need SeLockMemoryPrivilege, so try to enable it for the process
if (!OpenProcessToken(GetCurrentProcess(), TOKEN_ADJUST_PRIVILEGES | TOKEN_QUERY, &hProcessToken))
return nullptr;
if (LookupPrivilegeValue(NULL, SE_LOCK_MEMORY_NAME, &luid))
{
TOKEN_PRIVILEGES tp { };
TOKEN_PRIVILEGES prevTp { };
DWORD prevTpLen = 0;
tp.PrivilegeCount = 1;
tp.Privileges[0].Luid = luid;
tp.Privileges[0].Attributes = SE_PRIVILEGE_ENABLED;
// Try to enable SeLockMemoryPrivilege. Note that even if AdjustTokenPrivileges() succeeds,
// we still need to query GetLastError() to ensure that the privileges were actually obtained.
if (AdjustTokenPrivileges(
hProcessToken, FALSE, &tp, sizeof(TOKEN_PRIVILEGES), &prevTp, &prevTpLen) &&
GetLastError() == ERROR_SUCCESS)
{
// Round up size to full pages and allocate
allocSize = (allocSize + largePageSize - 1) & ~size_t(largePageSize - 1);
mem = VirtualAlloc(
NULL, allocSize, MEM_RESERVE | MEM_COMMIT | MEM_LARGE_PAGES, PAGE_READWRITE);
// Privilege no longer needed, restore previous state
AdjustTokenPrivileges(hProcessToken, FALSE, &prevTp, 0, NULL, NULL);
}
}
CloseHandle(hProcessToken);
return mem;
}
void* aligned_ttmem_alloc(size_t allocSize, void*& mem) {
static bool firstCall = true;
// Try to allocate large pages
mem = aligned_ttmem_alloc_large_pages(allocSize);
// Suppress info strings on the first call. The first call occurs before 'uci'
// is received and in that case this output confuses some GUIs.
if (!firstCall)
{
if (mem)
sync_cout << "info string Hash table allocation: Windows large pages used." << sync_endl;
else
sync_cout << "info string Hash table allocation: Windows large pages not used." << sync_endl;
}
firstCall = false;
// Fall back to regular, page aligned, allocation if necessary
if (!mem)
mem = VirtualAlloc(NULL, allocSize, MEM_RESERVE | MEM_COMMIT, PAGE_READWRITE);
return mem;
}
#else
void* aligned_ttmem_alloc(size_t allocSize, void*& mem) {
constexpr size_t alignment = 64; // assumed cache line size
size_t size = allocSize + alignment - 1; // allocate some extra space
mem = malloc(size);
void* ret = reinterpret_cast<void*>((uintptr_t(mem) + alignment - 1) & ~uintptr_t(alignment - 1));
return ret;
}
#endif
/// aligned_ttmem_free() will free the previously allocated ttmem
#if defined(_WIN64)
void aligned_ttmem_free(void* mem) {
if (mem && !VirtualFree(mem, 0, MEM_RELEASE))
{
DWORD err = GetLastError();
std::cerr << "Failed to free transposition table. Error code: 0x" <<
std::hex << err << std::dec << std::endl;
exit(EXIT_FAILURE);
}
}
#else
void aligned_ttmem_free(void *mem) {
free(mem);
}
#endif
namespace WinProcGroup {
#ifndef _WIN32
void bindThisThread(size_t) {}
#else
/// best_group() retrieves logical processor information using Windows specific
/// API and returns the best group id for the thread with index idx. Original
/// code from Texel by Peter Ă–sterlund.
int best_group(size_t idx) {
int threads = 0;
int nodes = 0;
int cores = 0;
DWORD returnLength = 0;
DWORD byteOffset = 0;
// Early exit if the needed API is not available at runtime
HMODULE k32 = GetModuleHandle("Kernel32.dll");
auto fun1 = (fun1_t)(void(*)())GetProcAddress(k32, "GetLogicalProcessorInformationEx");
if (!fun1)
return -1;
// First call to get returnLength. We expect it to fail due to null buffer
if (fun1(RelationAll, nullptr, &returnLength))
return -1;
// Once we know returnLength, allocate the buffer
SYSTEM_LOGICAL_PROCESSOR_INFORMATION_EX *buffer, *ptr;
ptr = buffer = (SYSTEM_LOGICAL_PROCESSOR_INFORMATION_EX*)malloc(returnLength);
// Second call, now we expect to succeed
if (!fun1(RelationAll, buffer, &returnLength))
{
free(buffer);
return -1;
}
while (byteOffset < returnLength)
{
if (ptr->Relationship == RelationNumaNode)
nodes++;
else if (ptr->Relationship == RelationProcessorCore)
{
cores++;
threads += (ptr->Processor.Flags == LTP_PC_SMT) ? 2 : 1;
}
assert(ptr->Size);
byteOffset += ptr->Size;
ptr = (SYSTEM_LOGICAL_PROCESSOR_INFORMATION_EX*)(((char*)ptr) + ptr->Size);
}
free(buffer);
std::vector<int> groups;
// Run as many threads as possible on the same node until core limit is
// reached, then move on filling the next node.
for (int n = 0; n < nodes; n++)
for (int i = 0; i < cores / nodes; i++)
groups.push_back(n);
// In case a core has more than one logical processor (we assume 2) and we
// have still threads to allocate, then spread them evenly across available
// nodes.
for (int t = 0; t < threads - cores; t++)
groups.push_back(t % nodes);
// If we still have more threads than the total number of logical processors
// then return -1 and let the OS to decide what to do.
return idx < groups.size() ? groups[idx] : -1;
}
/// bindThisThread() set the group affinity of the current thread
void bindThisThread(size_t idx) {
// Use only local variables to be thread-safe
int group = best_group(idx);
if (group == -1)
return;
// Early exit if the needed API are not available at runtime
HMODULE k32 = GetModuleHandle("Kernel32.dll");
auto fun2 = (fun2_t)(void(*)())GetProcAddress(k32, "GetNumaNodeProcessorMaskEx");
auto fun3 = (fun3_t)(void(*)())GetProcAddress(k32, "SetThreadGroupAffinity");
if (!fun2 || !fun3)
return;
GROUP_AFFINITY affinity;
if (fun2(group, &affinity))
fun3(GetCurrentThread(), &affinity, nullptr);
}
#endif
} // namespace WinProcGroup
#ifdef _WIN32
#include <direct.h>
#define GETCWD _getcwd
#else
#include <unistd.h>
#define GETCWD getcwd
#endif
namespace CommandLine {
string argv0; // path+name of the executable binary, as given by argv[0]
string binaryDirectory; // path of the executable directory
string workingDirectory; // path of the working directory
string pathSeparator; // Separator for our current OS
void init(int argc, char* argv[]) {
(void)argc;
string separator;
// extract the path+name of the executable binary
argv0 = argv[0];
#ifdef _WIN32
pathSeparator = "\\";
#ifdef _MSC_VER
// Under windows argv[0] may not have the extension. Also _get_pgmptr() had
// issues in some windows 10 versions, so check returned values carefully.
char* pgmptr = nullptr;
if (!_get_pgmptr(&pgmptr) && pgmptr != nullptr && *pgmptr)
argv0 = pgmptr;
#endif
#else
pathSeparator = "/";
#endif
// extract the working directory
workingDirectory = "";
char buff[40000];
char* cwd = GETCWD(buff, 40000);
if (cwd)
workingDirectory = cwd;
// extract the binary directory path from argv0
binaryDirectory = argv0;
size_t pos = binaryDirectory.find_last_of("\\/");
if (pos == std::string::npos)
binaryDirectory = "." + pathSeparator;
else
binaryDirectory.resize(pos + 1);
// pattern replacement: "./" at the start of path is replaced by the working directory
if (binaryDirectory.find("." + pathSeparator) == 0)
binaryDirectory.replace(0, 1, workingDirectory);
}
} // namespace CommandLine
|