Re: More SSE infrastructure

[Richard Henderson <rth at cygnus dot com>]

On Mon, Jul 03, 2000 at 07:08:34PM +0100, Bernd Schmidt wrote:
> ... the only place in the compiler I've found so far that relies
> on it is debugging output (where TImode constants are used for
> TYPE_{MIN,MAX}_VALUE of 128 bit integers.

I wonder if we can just bail on that?  I do see code like

  if (TYPE_MAX_VALUE (type) != 0
      && host_integerp (TYPE_MAX_VALUE (type), 0))
    {
      fputc (';', asmfile);
      fprintf (asmfile, HOST_WIDE_INT_PRINT_DEC,
               tree_low_cst (TYPE_MAX_VALUE (type), 0));
      fputc (';', asmfile);
    }
  else
    fprintf (asmfile, ";-1;");

in dbxout.c...

> + #ifndef HAVE_LONG_LONG_HOST_WIDE_INT
> +   if (TARGET_SSE)
> +     fatal ("SSE is not supported in this build of gcc.");
> + #endif

You should instead check HOST_BITS_PER_WIDE_INT.  There's
no reason why a 64-bit host should have to support long long
in order to run this code.


r~