bug in GCC 4.4 on sparc64 (was: Re: GNU PSPP 0.6.2pre3 now available)

[Ben Pfaff]

Ben Pfaff <address@hidden> writes:

> I now suspect that this is a problem caused by changes in
> optimization, since float-format.c does some things that should
> be OK but I can imagine a compiler getting wrong (and maybe there
> is some technicality that I have missed so far).  Can you try
> building with "-O0" instead of the default "-O2"?

I found out where GCC 4.4 was installed on the GCC compile farm
and managed to isolate the problem, which indeed appears to be a
bug in GCC.  When the following test program is built with "gcc
-O2 -m64" or "gcc -O3 -fno-inline -m64" using GCC 4.4.0 on
sparc64, it prints 0, but it should print 12345678.

I have filed this in GCC Bugzilla:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40668

----------------------------------------------------------------------

#include <stdint.h>
#include <stdio.h>
#include <string.h>

/* Stores 32-bit unsigned integer X at P,
   which need not be aligned. */
static void
put_uint32 (uint32_t x, void *p)
{
  memcpy (p, &x, sizeof x);
}

void
store_12345678 (int type, void *number)
{
  switch (type)
    {
    case 1:
      printf ("got here\n");
      put_uint32 (0x12345678, number);
      break;

    case 7:
      put_uint32 (0, number);
      break;
    case 8:
      put_uint32 (0, number);
      break;
    case 9:
      put_uint32 (0, number);
      break;
    }
}

int
main (void)
{
  uint32_t x;
  store_12345678 (1, &x);
  printf ("%x\n", (unsigned int) x);
  return 0;
}


-- 
Ben Pfaff 
http://benpfaff.org