How to know on which physical processor and on which physical core my code is running

Question

How can I know in a C program, on which physical processor and core my code is running? I'm using Linux and gcc 4.4.3.

Eridanus

Why do you need or want to know that? – Boaz

It can help me to understand the behavior of my program on a SMT Architecture – Eridanus

@Boaz software.intel.com/en-us/articles/… – Renegado

Non-programmatically, htop(1) (a more modern version of top(1)) has as PROCESSOR column, which shows you which core a process last ran on. (press F2, columns, find PROCESSOR and add it, F10(done)). – Avera

related: stackoverflow.com/questions/491520/… – Phosphorism

Answer 1

You may check /proc//stat file system, according to http://www.kernel.org/doc/Documentation/filesystems/proc.txt, you should just check task_cpu flag.

As example without correct types and error checking:

struct pstat
{
  int  pid;       //process id
  char tcomm[256];//filename of the executable
  char state[2];  //state (R is running, S is sleeping, D is sleeping in an
                  //uninterruptible wait, Z is zombie, T is traced or stopped)
  int ppid;//          process id of the parent process
  int pgrp;//          pgrp of the process
  int sid;//           session id
  int tty_nr;//        tty the process uses
  int tty_pgrp;//      pgrp of the tty
  int flags;//         task flags
  int min_flt;//       number of minor faults
  int cmin_flt;//      number of minor faults with child's
  int maj_flt;//       number of major faults
  int cmaj_flt;//      number of major faults with child's
  int utime;//         user mode jiffies
  int stime;//         kernel mode jiffies
  int cutime;//        user mode jiffies with child's
  int cstime;//        kernel mode jiffies with child's
  int priority;//      priority level
  int nice;//          nice level
  int num_threads;//   number of threads
  int it_real_value;//  (obsolete, always 0)
  int start_time;//    time the process started after system boot
  int vsize;//         virtual memory size
  int rss;//           resident set memory size
  int rsslim;//        current limit in bytes on the rss
  int start_code;//    address above which program text can run
  int end_code;//      address below which program text can run
  int start_stack;//   address of the start of the stack
  int esp;//           current value of ESP
  int eip;//           current value of EIP
  int pending;//       bitmap of pending signals
  int blocked;//       bitmap of blocked signals
  int sigign;//        bitmap of ignored signals
  int sigcatch;//      bitmap of catched signals
  int wchan;//         address where process went to sleep
  int i0;//             (place holder)
  int i1;//             (place holder)
  int exit_signal;//   signal to send to parent thread on exit
  int task_cpu;//      which CPU the task is scheduled on
  int rt_priority;//   realtime priority
  int policy;//        scheduling policy (man sched_setscheduler)
  int blkio_ticks;//   time spent waiting for block IO
  int gtime;//         guest time of the task in jiffies
  int cgtime;//        guest time of the task children in jiffies
} p ;

int main()
{
    char name[256];
    char state[8];
    FILE* f = fopen("/proc/self/stat", "r");

    fscanf(f,  "%d%s%s%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d",
           &p.pid, &p.tcomm, &p.state, &p.ppid, &p.pgrp, &p.sid, &p.tty_nr, &p.tty_pgrp, &p.flags,
           &p.min_flt, &p.cmin_flt, &p.maj_flt, &p.cmaj_flt, &p.utime, &p.stime,  &p.cutime, &p.cstime,
           &p.priority, &p.nice, &p.num_threads, &p.it_real_value, &p.start_time,  &p.vsize, &p.rss,
           &p.rsslim, &p.start_code, &p.end_code, &p.start_stack, &p.esp, &p.eip,  &p.pending, &p.blocked,
           &p.sigign, &p.sigcatch, &p.wchan, &p.i0, &p.i1, &p.exit_signal,  &p.task_cpu, &p.rt_priority, &p.policy,
           &p.blkio_ticks, &p.gtime, &p.cgtime);

     printf("CPU %d\n", p.task_cpu);
 return 0;  
 }

Eutherian

Answer 2

sched_getcpu()

call returns virtual CPU number. Mapping of virtual CPU to real CPU info is in /proc/cpuinfo.

If your system supports VDSO, then sched_getcpu() is relatively fast.

CPU number can be also obtained using CPUID instruction, but it is slower than sched_getcpu().

Butterscotch

Here is a concrete example with sched_setaffinity: stackoverflow.com/questions/10490756/… – Phosphorism

Answer 3

See http://en.wikipedia.org/wiki/CPUID#Accessing_the_id_from_other_languages

What you want is the APIC ID... basically:

    cpuid
    shr     ebx, 24

Terpene

cpuid is slow. sched_getcpu() can be used instead. It is fast if VDSO is enabled. – Butterscotch

Answer 4

You can get the affinity of the process to a processor using GCC. The CPU affinity APIs might help you out. Are you trying to use this information to ensure your process is not interrupted or such high priority task?

Gleda

Answer 5

By and large it's hard to find this out in a meaningful way. Your thread will often run on many different processors during its life. You could call a function to ask which processor you are on and get a context switch whilst the function is executing. What should the function return?

Maidel

Practical uses do come up... for example - with some potential spinning on detecting such context switches, you can (at worst, after a couple attempts) get a reliable clock-cycles-since-startup reading and know which core it's from. As part of a process startup-procedure, you can take clock readings with affinity set to each core in turn, allowing you to have a range of possible values for drift between specific pairs of cores. It's fiddly, but when applicable can be the best approach to high-res timing on typical modern x86 hardware.... – Terpene

Which function can I use in my program to know on which processor (core) my program at a given time ? and how can I get the context switch ? – Eridanus

CPUID will identify the process or x86, as Tony says. – Maidel