cvs commit: src/sys/alpha/alpha support.s src/sys/i386/i386 identcpu.c support.s src/sys/i386/include md_var.h src/sys/i386/isa npx.c src/sys/ia64/ia64 support.s src/sys/powerpc/powerpc bcopy.csr

[John Baldwin]

On 05-Apr-2003 Scott Long wrote:
> On Sat, Apr 05, 2003 at 01:31:03PM +0200, Dag-Erling Sm?rgrav wrote:
>> Bruce Evans <bde at zeta.org.au> writes:
>> > The main problem with the commit is not mentioned in its log message:
>> > changing bzero from a function pointer back to a function breaks
>> > binary compatibility of many modules.
>> 
>> I am aware of that...
>> 
>> >                                        In defence of this, 5.0 breaks
>> > binary compatibility of most modules for other reasons.
>> 
>> which is precisely why I took the liberty of doing it now, before 5.x
>> goes -STABLE.
>> 
> 
> As I've scolded others, changes to HEAD that change the ABI and/or API
> should be preceded by an email to re at .  Please re-read the many emails
> about the semi-frozen state of HEAD for futher guidance.
> 
> I'm not terribly thrilled by this change, but I'm not going to ask for a
> backout right now.

Eh?  We agreed on arch@ a while back that ABI breakage in 5.0 post
5.0-RELEASE was ok and that the ABI "freeze" wouldn't be until
RELENG_5 was branched.  The topic was actually posed by Robert Watson
as he wants to change MAC over to using m_tag before RELENG_5 IIRC.

-- 

John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/