> -Current is incredibly stable. Commits that break the tree are
> fairly rare and get repaired quickly. Before using a -current
> system in something important you want to test it, but you
> want to test a stock -stable system before using it for your
> applications, too. Most of my systems that I use for real are
> -current. I've not had a problem in 12 years--but I test stuff
> before using it!
>
> I've screwed myself twice with -current: once when I didn't pay
> attention to a flag day and gotten myself a broken compiler, and
> once during a hackathon when commits were coming in every 90
> seconds or so, and my eagerness to try something new made a
> system which went BOOM. But these kinds of errors are good
> practice for dealing with a real-world broken system. ;-)
>
> If you want to use -current you *really* want to pay close
> attention to the source change mailing list. Your mirror of
> choice updates itself every N hours. There will be times when
> the latest update to the mirror will not include all the physical
> parts of a logical commit, so attempting to compile stuff will
> result in an error (think code change that uses a new #define
> in a .h that you didn't get), or really subtle stuff because you
> got *part* of a change.
>
> And of course, caution during a hackathon. I'll update my tree
> during one, but will stay away from using it, as the tree of 45
> minutes ago might not be right by the time your mirror gets
> its update.
That was the 2nd best email of my day. The following is the best
email of my day:
CVSROOT: /cvs
Module name: src
Changes by: t...@cvs.openbsd.org 2014/05/07 15:06:05
Modified files:
lib/libssl/src/crypto/bn: bn_lib.c
Log message:
in BN_clear_free, don't cleanse the data if the static data flag is set.
much debugging work done by otto. ok miod otto.
side note: BN_FLG_STATIC_DATA doesn't actually mean the data is static.
it's also used to indicate the data may be secretly shared behind your back
as a sort of poor man's refcounting, but without the refcounting.
