#include <sys/types.h>
#include <sys/tls.h>
#include <machine/mcontext.h>
#include <link.h>
#include <stdio.h>

__thread int a;

static struct tls_tcb * ThreadSelfTlsTcb() {
  struct tls_tcb * tcb;
# ifdef __HAVE___LWP_GETTCB_FAST
  tcb = (struct tls_tcb *)__lwp_gettcb_fast();
# elif defined(__HAVE___LWP_GETPRIVATE_FAST)
  tcb = (struct tls_tcb *)__lwp_getprivate_fast();
# endif
  return tcb;
}

int GetSizeFromHdr(struct dl_phdr_info *info, size_t size, void *data) {
  const Elf_Phdr *hdr = info->dlpi_phdr;
  const Elf_Phdr *last_hdr = hdr + info->dlpi_phnum;

  for (; hdr != last_hdr; ++hdr) {
    if (hdr->p_type == PT_TLS && info->dlpi_tls_modid == 1) {
      *(long*)data = hdr->p_memsz;
      break;
    }
  }
  return 0;
}

int
main(int argc, char **argv)
{
  long addr, size;
  struct tls_tcb * const tcb = ThreadSelfTlsTcb();
  addr = 0;
  size = 0;
  if (tcb != 0) {
    // Find size (p_memsz) of dlpi_tls_modid 1 (TLS block of the main program).
    // ld.elf_so hardcodes the index 1.
    dl_iterate_phdr(GetSizeFromHdr, &size);

    if (size != 0) {
      // The block has been found and tcb_dtv[1] contains the base address
      addr = (long)tcb->tcb_dtv[1];
    }
  }

  printf("addr=%#lx size=%ld\n", addr, size);

  return 0;
}