Bug 1099025

+++ This bug was initially created as a clone of Bug #1064945 +++

Description of problem:
The configure script in the lrzsz package checks if the ftime() function works correctly and it seems this test hangs (infinite loop) on ppc in the current RHEL7 buildroot. It looks like the time field in the timeb structure doesn't change or the call now fails.

I've patched the lrzsz package to not try ftime(), but I thought this might be worth further investigation.

The last build that passed was with glibc-2.17-4.el7. 

Version-Release number of selected component (if applicable):
glibc-2.17-48.el7

How reproducible:
easily

Steps to Reproduce:
1. compile and run the c code below

Actual results:
hangs

Expected results:
runs for 10 seconds and returns 0.

Additional info:

The test code:

#include <sys/types.h>
#include <sys/timeb.h>
#include <stdio.h>
main ()
{
  struct timeb s, slast;
  int c = 0;
  ftime (&slast);
  while (c < 10)
    {
      ftime (&s);
      if (s.time < slast.time
      || (s.time == slast.time && s.millitm < slast.millitm)) {
      fprintf(stderr,"s: %ld, %ld, slast: %ld, %ld\n",
        s.time,s.millitm, slast.time,slast.millitm);
    exit (1);
        }
      if (s.time != slast.time)
    ++c;
      slast.time = s.time;
      slast.millitm = s.millitm;
    }
  exit (0);
}