Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 05:48:47 +01:00
Code Issues Packages Projects Releases Wiki Activity

common : better default number of threads (#934)

Browse Source 
* commit

* fix

* try-catch

* apply code review

* improve

* improve

* add macos headers

* done

* remove color

* fix windows

* minor

* fix

* Apply suggestions from code review

Co-authored-by: DannyDaemonic <DannyDaemonic@gmail.com>

* remove

* minor

* minor

---------

Co-authored-by: jon-chuang <jon-chuang@users.noreply.github.com>
Co-authored-by: DannyDaemonic <DannyDaemonic@gmail.com>
This commit is contained in:
jon-chuang 2023-04-30 14:41:35 -04:00 committed by GitHub
parent 76a884920a
commit a5d30b1f53
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 42 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

49
examples/common.cpp
View File

@ -1,13 +1,18 @@
#include "common.h" #include "common.h"
#include <cassert> #include <cassert>
#include <iostream>
#include <cstring> #include <cstring>
#include <fstream> #include <fstream>
#include <string> #include <string>
#include <iterator> #include <iterator>
#include <algorithm> #include <algorithm>
#include <sstream> #include <sstream>
#include <iostream>
#if defined(__APPLE__) && defined(__MACH__)
#include <sys/types.h>
#include <sys/sysctl.h>
#endif
#if defined (_WIN32) #if defined (_WIN32)
#include <fcntl.h> #include <fcntl.h>
@ -25,19 +30,43 @@ extern "C" __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int
#define CP_UTF8 65001 #define CP_UTF8 65001
#endif #endif
bool gpt_params_parse(int argc, char ** argv, gpt_params & params) { int32_t get_num_physical_cores() {
// determine sensible default number of threads.
// std::thread::hardware_concurrency may not be equal to the number of cores, or may return 0.
#ifdef __linux__ #ifdef __linux__
std::ifstream cpuinfo("/proc/cpuinfo"); std::ifstream cpuinfo("/proc/cpuinfo");
params.n_threads = std::count(std::istream_iterator<std::string>(cpuinfo), std::string line;
std::istream_iterator<std::string>(), while (std::getline(cpuinfo, line)) {
std::string("processor")); std::size_t pos = line.find("cpu cores");
#endif if (pos != std::string::npos) {
if (params.n_threads == 0) { pos = line.find(": ", pos);
params.n_threads = std::max(1, (int32_t) std::thread::hardware_concurrency()); if (pos != std::string::npos) {
try {
// Extract the number and return it
return static_cast<int32_t>(std::stoul(line.substr(pos + 2)));
} catch (const std::invalid_argument &) {
// Ignore if we could not parse
}
}
}
} }
#elif defined(__APPLE__) && defined(__MACH__)
int32_t num_physical_cores;
size_t len = sizeof(num_physical_cores);
int result = sysctlbyname("hw.perflevel0.physicalcpu", &num_physical_cores, &len, NULL, 0);
if (result == 0) {
return num_physical_cores;
}
result = sysctlbyname("hw.physicalcpu", &num_physical_cores, &len, NULL, 0);
if (result == 0) {
return num_physical_cores;
}
#elif defined(_WIN32)
//TODO: Implement
#endif
unsigned int n_threads = std::thread::hardware_concurrency();
return n_threads > 0 ? (n_threads <= 4 ? n_threads : n_threads / 2) : 4;
}
bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
bool invalid_param = false; bool invalid_param = false;
std::string arg; std::string arg;
gpt_params default_params; gpt_params default_params;

5
examples/common.h
View File

@ -13,11 +13,12 @@
// //
// CLI argument parsing // CLI argument parsing
// //
int32_t get_num_physical_cores();
struct gpt_params { struct gpt_params {
int32_t seed = -1; // RNG seed int32_t seed = -1; // RNG seed
int32_t n_threads = std::min(4, (int32_t) std::thread::hardware_concurrency()); int32_t n_threads = get_num_physical_cores();
int32_t n_predict = -1; // new tokens to predict int32_t n_predict = -1; // new tokens to predict
int32_t n_parts = -1; // amount of model parts (-1 = determine from model dimensions) int32_t n_parts = -1; // amount of model parts (-1 = determine from model dimensions)
int32_t n_ctx = 512; // context size int32_t n_ctx = 512; // context size
int32_t n_batch = 512; // batch size for prompt processing (must be >=32 to use BLAS) int32_t n_batch = 512; // batch size for prompt processing (must be >=32 to use BLAS)