Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 28 Oct 2007 00:28:57 GMT
From:      John Birrell <jb@FreeBSD.org>
To:        Perforce Change Reviews <perforce@freebsd.org>
Subject:   PERFORCE change 128216 for review
Message-ID:  <200710280028.l9S0SvXU081548@repoman.freebsd.org>

next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=128216

Change 128216 by jb@jb_freebsd1 on 2007/10/28 00:28:14

	The comment "Solaris defines _LITTLE_ENDIAN or _BIG_ENDIAN, but
	never both and decides which architecture it is based on which
	thing is defined." is true, but undefining _LITTLE_ENDIAN or
	_BIG_ENDIAN breaks lots of FreeBSD header files which all
	assume that both _LITTLE_ENDIAN and _BIG_ENDIAN are defined
	and that BYTE_ORDER is set to the one that matches the host.
	
	This was a hack to get ZFS working on FreeBSD, but it prevents
	DTrace code from using FreeBSD headers which include _BIG_ENDIAN
	and _LITTLE_ENDIAN.
	
	ZFS (and other Solaris) code will _have _to be changed to do what
	FreeBSD does: e.g. if BYTE_ORDER == _LITTLE_ENDIAN rather than
	ifdef _LITTLE_ENDIAN.
	
	Without this change, I can't share the OpenSolaris compatibility
	headers that pjd@ has created.

Affected files ...

.. //depot/projects/dtrace/src/sys/compat/opensolaris/machine/endian.h#2 delete

Differences ...



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200710280028.l9S0SvXU081548>