>On Mon, 16 Feb 1998, Dan Jacobowitz wrote:
>
>> Oddly, on FreeBSD-i386, gcc compiled perfectly fine.  ksh, however did not;
>> don't expect it to work on anything but NetBSD-*.  You might want to
>> comment it out.
>
>About gcc, libcc1 (for which all my gcc patches are needed) isn't built
>for powerpc. So it'll be fine. Actually, it looks like we don't really
>compile gcc on powerpc. At least it doesn't seem to compile as a result of
>compiling in usr.bin. We make man pages, but that's it.

We'll just have to see about this once we get a usable system.

>> I also noticed several warnings, probably erroneous, from gas about
>> possible unsafe uses of mktemp().  gcc-2.8 completely refused to compile
>> libg++ because of stricter ANSI compliance; binutils warned me several
>> times about "operand out of range: 32 is not between 0 and 31" on a swli
>> instruction which had no 32 in it at all.  Go figure.
>
>Weird.

Aint it though...try upgrading to binutils-2.8.1 in your toolchain sometime
and see where it dies.  2.8.1 considers this a fatal error but 2.7 only a
warning.


Oddly, gcc for me is refusing to automatically define __powerpc__!  The
symbol seems to be __PPC__.  Just remember to set it in the environment
variable CFLAGS  (CPPFLAGS?) in advance.

Dan

---------------------------------------------------------------------
| Dan Jacobowitz                 | drow@drow.net                    |
|  Administrator Extraordinaire  | Web site coming someday          |
| My opinions are my own -       | Day Job: http://www.wwwcomm.com  |
| My mistakes are someone else's |                                  |
---------------------------------------------------------------------