M. Warner Losh
imp at bsdimp.com
Mon Aug 23 11:58:31 PDT 2004
In message: <20040823134827.A0D6A43D3F at mx1.FreeBSD.org>
Danny Braniss <danny at cs.huji.ac.il> writes:
: > On 2004-08-23 16:21, Danny Braniss <danny at cs.huji.ac.il> wrote:
: > > > On 2004-08-23 10:46, Danny Braniss <danny at cs.huji.ac.il> wrote:
: > > > > shouldn't
: > > > > NOTE TO PEOPLE WHO THINK THAT FreeBSD 5.x IS SLOW:
: > > > > now be 6.x ...
: > > >
: > > > Would something like this seem ok to you? It only mentions FreeBSD-CURRENT
: > > > instead of a specific version, so it will also work for FreeBSD 7.X when
: > > > that comes along.
: > >
: > > fine, but what if FreeBSD 7.X is actually faster ... :-)
: > The word "slower" refers to the reduced speed of a system that runs with
: > a lot of debugging/test options enabled. The relative speed of the
: > release to the previous is (probably) of lesser importance for this
: > particular quote.
: i agree, it's that once you take out the 'milestone', it might just become
: superflous, on the other hand 5.x is bad, so just current is fine - one less
: thing to check for.
I believe that we won't have the debugging kernel options turned on by
default over the long haul. Once we're happy with the locking, they
will likely be turned off in GENERIC.
More information about the freebsd-current