New x86 failures

Mark Mitchell mark@markmitchell.com
Sat Oct 10 00:16:00 GMT 1998


I'm seeing a few new failures on x86-linux-gnu with the current CVS
tree.  In particular,

  FAIL: gcc.c-torture/compile/930217-1.c,  -O1  
  FAIL: gcc.c-torture/execute/comp-goto-1.c compilation,  -O2
        -fomit-frame-pointe
  FAIL: gcc.c-torture/special/981006-1.c 

Since the last of these was added very recently, I'll assumed it's
expected to fail.  It should probably be an XFAIL, then, right?  (If
new tests just FAIL, it's very hard to tell whether or not you're
breaking the compiler.)

Richard, I'm guessing this change

Fri Oct  9 15:49:29 1998  Richard Henderson  <rth@cygnus.com>

	* expmed.c (store_bit_field): Pun non-integral str_rtx modes.
	Take extra care for op0 now possibly being a subreg.
	(extract_bit_field): Likewise.

could be at the root of at least some of these problems, since
930217-1.c contains:

  double g ();
  typedef union {
    struct {
      unsigned s:1, e:8, f:23;
    } u;
    float f;
  } s;

  f(x, n)
       float x;
  {
    ((s *)&x)->u.e -= n;
    x = g((double)x, -n);
  }

which looks like fun with bitfields.

-- 
Mark Mitchell 			mark@markmitchell.com
Mark Mitchell Consulting	http://www.markmitchell.com



More information about the Gcc-bugs mailing list