David S. Miller davem@dm.cobaltmicro.com
Mon Oct 5 23:04:00 GMT 1998

   Date: Tue, 06 Oct 1998 11:06:58 +1300
   From: Bill Currie <bcurrie@tssc.co.nz>

   David S. Miller wrote:
   > If you just want the register number offset, use SUBREG_REGNO_OFFSET.
   > Both of these must be given a hard register, it will abort if you do
   > not.

   What does this imply for mixed endian chips such as the i860 in
   big-endian mode (ie mem big-endian, regs little-endian)?  Will your
   changes make it easier, harder, or indifferent?

It should be indifferent.

Let me put it another way, I believe that if a target has defined
WORDS_BIG_ENDIAN and BYTES_BIG_ENDIAN correctly, the compiler will act
precisely how it did before.  It should generate the same code sans
bug fixes.  The only case of the latter I know of are the jump.c
changes because rtx_renumbered_equal_p() previous to my patch handled
the 'x' and 'y' parts, if subregs, in an inconsistant manner.

David S. Miller

More information about the Gcc-patches mailing list