patches/gcc/4.2.2/270-soft-float.patch
author "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Fri Aug 19 22:53:48 2011 +0200 (2011-08-19)
changeset 2620 d9c5b3b75864
parent 616 6e78f7085707
permissions -rw-r--r--
Makefile: introduce PROG_NAME to contain the 'ct-ng' executable name

With the upcomming --program-{prefix,suffix,transform-name} options,
we'll have a need to centralise the actual executable name, that is
now no longer be a constant.

Rather than spread the prefix/suffix through-out the code, just
centralise the name setting in one place. Beside, transform-name
would not be possible without setting the name at a single location.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
     1 diff -uNpr gcc-4.2.1_orig/gcc/config/rs6000/darwin-ldouble.c gcc-4.2.1/gcc/config/rs6000/darwin-ldouble.c
     2 --- gcc-4.2.1_orig/gcc/config/rs6000/darwin-ldouble.c	2007-03-05 11:54:00.000000000 -0500
     3 +++ gcc-4.2.1/gcc/config/rs6000/darwin-ldouble.c	2008-01-31 17:51:24.000000000 -0500
     4 @@ -70,6 +70,8 @@ Software Foundation, 51 Franklin Street,
     5     but GCC currently generates poor code when a union is used to turn
     6     a long double into a pair of doubles.  */
     7  
     8 +#if defined (_SOFT_FLOAT) && defined (__LONG_DOUBLE_128__)
     9 +
    10  long double __gcc_qadd (double, double, double, double);
    11  long double __gcc_qsub (double, double, double, double);
    12  long double __gcc_qmul (double, double, double, double);
    13 @@ -219,8 +221,6 @@ __gcc_qdiv (double a, double b, double c
    14    return z.ldval;
    15  }
    16  
    17 -#if defined (_SOFT_FLOAT) && defined (__LONG_DOUBLE_128__)
    18 -
    19  long double __gcc_qneg (double, double);
    20  int __gcc_qeq (double, double, double, double);
    21  int __gcc_qne (double, double, double, double);