[Libevent-users] Re: Linux, epoll, libevent, regressions, stability, and 2.0.9

[Nick Mathewson <nickm@xxxxxxxxxxxxx>]

On Sat, Nov 20, 2010 at 3:25 AM, Nick Mathewson <nickm@xxxxxxxxxxxxx> wrote:
[...]
>  * Include both backends; make the non-changelist backend on by
> default.  The problems here are that a) the non-changelist backend is
> slower, and most people won't do whatever is necessary to activate the
> faster one, and b) the non-changelist backend has had not nearly so
> much testing as the current changelist-based backend.  If we do this,
> the lack of testing means we cannot possibly call 2.0.9
> "2.0.9-stable"; we'll need to call it "-rc" instead. :/
>
> I am currently leaning towards the last option.  Efficiency is
> important, but even more important is knowing that if you wrote a
> program using Libevent version N, your program will still work when
> Libevent N+1 is released.  Setting an option to enable extra
> performance is more important than setting an option to enable
> backward compatibility.

Based on comments here, I'm reasonably sure that this is the right
answer.  It's implemented in the git master branch.  2.0.9-rc should
be out really soon; I'd appreciate anybody who uses the epoll backed
to look for regressions now if they can spare the time... or at the
very least to check out 2.0.9-rc on its release to make sure that bugs
get found before -stable.

Also, I'm going to be light on email for the rest of this week: I'm
taking a break for the US holiday, and trying not to spend the whole
time online.

peace,
-- 
Nick
***********************************************************************
To unsubscribe, send an e-mail to majordomo@xxxxxxxxxxxxx with
unsubscribe libevent-users    in the body.