/* { dg-do compile { target float128 } } */
/* { dg-options "-Wabsolute-value" } */

#include <stdlib.h>
#include <inttypes.h>
#include <math.h>

void
tst_float128_size (_Float128 *pf128)
{
  *pf128 = fabsl (*pf128); /* { dg-warning "may cause truncation of value" } */
}