Re: [PATCH] DEFAULT_SIGNED_BITFIELDS Macro

[mark at codesourcery dot com]

>>>>> "Donn" == Donn Terry <donn@interix.com> writes:

    Donn> To (a) reiterate, and (b) conclude:

    Donn> (a) gcc is intended to be ABI conformant for whatever
    Donn> architecture it is on.  To the extent that such conformance
    Donn> changes the semantics of a specific construct, the
    Donn> programmer is responsible to write his application stricutly
    Donn> portably (that is, in a way that doesn't care about that
    Donn> difference).  (In the absence of using whatever fine grained
    Donn> control is available, anyway.)

    Donn> Given the level of debate on this topic originally, does
    Donn> such a statement appear anywhere in the "official"
    Donn> documentation?  If not, it probably belongs there.  It would
    Donn> be useful guideance to future porters.

 Jeff's statements reflect a policy change relative to what has been
in the manual for quite some time, at least with respect to the
particular issue of bitfields.

I believe that GCC has always tried to handle structure layout
compatibly, and so forth, so that user code could be linked to native
libraries.   However, the bitfield issue was clearly documented as
*not* following the native conventions.

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