PATCH: New libiberty sorting routine

Mark Mitchell mark@codesourcery.com
Sun Apr 23 22:40:00 GMT 2000


>>>>> "Richard" == Richard Henderson <rth@cygnus.com> writes:

    Richard> On Sun, Apr 23, 2000 at 06:01:30PM -0700, Mark Mitchell
    Richard> wrote:
    >> + /* Figure out the endianness of the machine.  */ + for (i =
    >> 0; i < sizeof (size_t); ++i) + ((char *)&j)[i] = i; +
    >> big_endian_p = (((char *)&j)[0] == 0);

    Richard> This is always false.

Yes.  This, and the other bug you found, show what happens when you
only test on a little-endian machine.  Thanks; fix coming.

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com


More information about the Gcc-patches mailing list