Increase stack size in Linux with setrlimit

Question

reading information about how to increase stack size for a c++ application compiled with gnu, at compilation time, I understood that it can be done with setrlimit at the beginning of the program. Nevertheless I could not find any successful example on how to use it and in which part of the program apply it in order to get a 64M stack size for a c++ program, could anybody help me?

Thanlks

Answer 1

To get beyond the hard limit in setrlimit (on OSX its only 64MB by default), create a new thread using pthreads with a stack size of your choice. Here's a C snippet:

    // Call function f with a 256MB stack.
    static int bigstack(void *(*f)(void *), void* userdata) {

      pthread_t thread;
      pthread_attr_t attr;

      // allocate a 256MB region for the stack.
      size_t stacksize = 256*1024*1024;
      pthread_attr_init(&attr);
      pthread_attr_setstacksize(&attr, stacksize);

      int rc = pthread_create(&thread, &attr, f, userdata);
      if (rc){
        printf("ERROR: return code from pthread_create() is %d\n", rc);
        return 0;
      }
      pthread_join(thread, NULL);
      return 1;

    }

Answer 2

See if the runtime execution maximum is limiting it:

[wally@zf conf]$  ulimit -all
core file size          (blocks, -c) 0
data seg size           (kbytes, -d) unlimited
scheduling priority             (-e) 0
file size               (blocks, -f) unlimited
pending signals                 (-i) 16114
max locked memory       (kbytes, -l) 32
max memory size         (kbytes, -m) unlimited
open files                      (-n) 1024
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) 819200
real-time priority              (-r) 0
stack size              (kbytes, -s) 10240
cpu time               (seconds, -t) unlimited
max user processes              (-u) 16114
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

Note that the stack size, by default, is limited to 10 MiB. So to increase it to 64 MiB:

[wally@zf conf]$ ulimit -s 64M
-bash: ulimit: 64M: invalid number
[wally@zf conf]$ ulimit -s 65536
[wally@zf conf]$ ulimit -all
core file size          (blocks, -c) 0
data seg size           (kbytes, -d) unlimited
scheduling priority             (-e) 0
file size               (blocks, -f) unlimited
pending signals                 (-i) 16114
max locked memory       (kbytes, -l) 32
max memory size         (kbytes, -m) unlimited
open files                      (-n) 1024
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) 819200
real-time priority              (-r) 0
stack size              (kbytes, -s) 65536
cpu time               (seconds, -t) unlimited
max user processes              (-u) 16114
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

Answer 3

Normally you would set the stack size early on, e,g, at the start of main(), before calling any other functions. Typically the logic would be:

• call getrlimit to get current stack size
• if current size < required stack size then
  • call setrlimit to increase stack size to required size

In C that might be coded something like this:

#include <sys/resource.h>
#include <stdio.h>

int main (int argc, char **argv)
{
    const rlim_t kStackSize = 64L * 1024L * 1024L;   // min stack size = 64 Mb
    struct rlimit rl;
    int result;

    result = getrlimit(RLIMIT_STACK, &rl);
    if (result == 0)
    {
        if (rl.rlim_cur < kStackSize)
        {
            rl.rlim_cur = kStackSize;
            result = setrlimit(RLIMIT_STACK, &rl);
            if (result != 0)
            {
                fprintf(stderr, "setrlimit returned result = %d\n", result);
            }
        }
    }

    // ...

    return 0;
}