Re: i370 port

["Ulrich Weigand" <uweigand@de.ibm.com>]

Paul Edwards wrote:

> I have reviewed the 'W' code in PRINT_OPERAND:
> 
> else if (CODE == 'W')
>   {
>     /* hand-built sign-extension of signed 32-bit to 64-bit */
>     mvs_page_lit += 8;
>     if (0 <=  INTVAL (XV)) {
>        fprintf (FILE, "=XL8'00000000");
>     } else {
>        fprintf (FILE, "=XL8'FFFFFFFF");
>     }
>     fprintf (FILE, "%08X'", INTVAL (XV));
>   }
> 
> and it looks to me like it is already correct. If movdi is given a
> const_int as a parameter, then sign-extending to 64-bit is
> exactly what needs to happen, isn't it?
> 
> I'm only expecting to compile programs as 32-bit, so I'm not
> expecting more than 32-bit integers. The IFOX assembler
> won't do more than that. In case that's the issue.

Well, even on 32-bit you may get 64-bit integer constants,
e.g. via the "long long" data type:

  long long x = 0x123456789abcdefLL;

However, the real question in your case is whether those are
represented as CONST_INT.  This is only true if HOST_WIDE_INT
is a 64-bit type; otherwise, such constants would be represented
as a CONST_DOUBLE.

Whether or not HOST_WIDE_INT is a 64-bit type now depends on
which *host* you're building GCC as a cross-compiler on.  If
you only ever support 32-bit hosts, then HOST_WIDE_INT will
always be a 32-bit type, and the code above should be fine.

If you want to support 64-bit hosts as well, however, you
will need to handle 64-bit CONST_INT values too.

> But regardless I don't know how to make this code:
> 
> mvs_check_page (0, 6, 8);
> return \"MVC^I%O0(8,%R0),%1\";
> 
> make use of that 'W' operand.
> 
> Do I change that %1 to %W1 perhaps?

Yes, exactly.

Bye,
Ulrich

-- 
  Dr. Ulrich Weigand
  GNU Toolchain for Linux on System z and Cell BE
  Ulrich.Weigand@de.ibm.com