This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Patch (rs6000): fix TFmode addressing


On Jun 24, 2004, at 9:22 PM, Alan Modra wrote:

On Sat, Jun 19, 2004 at 11:54:39AM -0700, Dale Johannesen wrote: 
--- 3355,3364 ----
           || XEXP (x, 0) == arg_pointer_rtx)
        && GET_CODE (XEXP (x, 1)) == CONST_INT)
      return 1;
    if (legitimate_offset_address_p (mode, x, reg_ok_strict))
      return 1;
    if (mode != TImode
+       && mode != TFmode
        && ((TARGET_HARD_FLOAT && TARGET_FPRS)
          || TARGET_POWERPC64
          || (mode != DFmode && mode != TFmode))

Shouldn't that be

&& !(mode == TFmode && TARGET_LONG_DOUBLE_128)

I don't think so, I was seeing DFmode for long double without -mlong-double-128.
Is that not true on other OSs?

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]