Wish for 2002 ...

[Kaz Kylheku]

On 10 Jan 2002, Thomas Bushnell, BSG wrote:

> Date: 10 Jan 2002 16:56:19 -0800
> From: "Thomas Bushnell, BSG" <tb@becket.net>
> To: Kaz Kylheku <kaz@ashi.footprints.net>
> Cc: Francois Leclerc <leclerc@austin.sns.slb.com>,
>      Security Audit <security-audit@ferret.lmh.ox.ac.uk>,
>      Andrew Josey <a.josey@opengroup.org>, Tiemann <tiemann@redhat.com>,
>      libc-alpha@sources.redhat.com,
>      Robust Open Source <open-source@csl.sri.com>
> Subject: Re: Wish for 2002 ...
> 
> Kaz Kylheku <kaz@ashi.footprints.net> writes:
> 
> > So you are saying this software doesn't run on GNU/Linux?
> 
> No, but it requires the use of a suboptimal autoconf version of the
> function.

So in fact this is not an example of a program which needs these entry
points to be present in the library, which tends to explain why it
actually runs.

Is there any profiling data available that would reveal how much time
OpenSSH spends executing these functions? 

Imagine that you have magic versions of these functions which run in
zero time, regardless of input length. How much faster would OpenSSH be
with these magic functions, based on the profiling data?

Are any GNU/Linux users of OpenSSH complaining about performance problems?

Lastly, whose job is it really to optimize OpenSSH?

These are questions that any competent development manager at a software
company might ask you when you come to him or her with dreams about
gold-plating.  And you'd be immediately assigned to some work, seeing
that you have lots of time on your hands.