Rewrite JROUND macro to multiply by 1e6 rather than divide by 1e-6

and vica versa for the next operation.
This is reputed to fix inline math optimization troubles in glibc-2.0.x.
Regression tests still pass on libc/i686 machine.  Patch suggested by Matt.

src/include/utils/dt.h

@@ -8,7 +8,7 @@
  *
  * Copyright (c) 1994, Regents of the University of California
  *
- * $Id: dt.h,v 1.29 1998/05/09 22:43:50 thomas Exp $
+ * $Id: dt.h,v 1.30 1998/05/31 17:08:35 thomas Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -258,8 +258,8 @@ extern int	datetime_is_epoch(double j);
 #endif
 #define TIMESPAN_NOT_FINITE(j)	TIMESPAN_IS_INVALID(j)
 
-#define TIME_PREC 1e-6
-#define JROUND(j) (rint(((double) (j))/TIME_PREC)*TIME_PREC)
+#define TIME_PREC_INV 1000000.0
+#define JROUND(j) (rint(((double) (j))*TIME_PREC_INV)/TIME_PREC_INV)
 
 /*
  * dt.c prototypes