Mailing List Archive

(Fwd) Re: Any Cygwin compile success????
Moving the conversation here.

------- Forwarded message follows -------

On 25 Jul 2002 at 17:27, Renaud Deraison wrote:

> Now, cygwin does have a fork() function. "So what's the deal" you may
> ask ? Well, they implemented fork() in an ingenious yet totally
> slow way - basically, it copies the whole amount of memory (5mb at each
> fork() call). That's slow. Horribly slow.

:) Yes I'm all too familiar with this problem. It's the road-block that
I hit on my afformentioned programming tool. Acutally I ran into
the road-block that a fork() doesn't exist or is a royal pain to
implement (correctly/safely) if you aren't using cygwin (I was using
mingw for the project as the target users have an aversion to unix
inside windows environments).

> Honnestly, I'm not sure this is a good thing. I'm having a hard time
> with "unix users" already (for what is worth, I receive so many
> requests that I sometimes forget that my mailbox may receive
> personal messages too. And of course, people refer to Nessus as
> "your software" which make this kind of mail impossible to filter
> properly). The Windows user base is too large for me to handle.
>
>
> So the bottom line is: no windows support at this time. No no no no.

Understandable. What if someone else stepped up to the plate and
handled the Windows support/users and 'proxied' for them? I
understand the it's slow and I don't like giving that impression (what
would we have if not for a pride in our work?)

------- End of forwarded message ---------
George Boutwell,
Programmer II - Valley Hope Association
gboutwel@valleyhope.com
--
George Boutwell,
Programmer II - Valley Hope Association
gboutwel@valleyhope.com
Re: (Fwd) Re: Any Cygwin compile success???? [ In reply to ]
On Thu, Jul 25, 2002 at 11:54:36AM -0500, gboutwel@valleyhope.com wrote:
> :) Yes I'm all too familiar with this problem. It's the road-block that
> I hit on my afformentioned programming tool. Acutally I ran into
> the road-block that a fork() doesn't exist or is a royal pain to
> implement (correctly/safely) if you aren't using cygwin (I was using
> mingw for the project as the target users have an aversion to unix
> inside windows environments).


Alas, cygwin's implementation is "better than nothing", but for Nessus,
with hundreds of calls to fork, that's probably not enough...

> > Honnestly, I'm not sure this is a good thing. I'm having a hard time
> > with "unix users" already (for what is worth, I receive so many
> > requests that I sometimes forget that my mailbox may receive
> > personal messages too. And of course, people refer to Nessus as
> > "your software" which make this kind of mail impossible to filter
> > properly). The Windows user base is too large for me to handle.
> >
> >
> > So the bottom line is: no windows support at this time. No no no no.
>
> Understandable. What if someone else stepped up to the plate and
> handled the Windows support/users and 'proxied' for them? I

Then people would continue to send me their requests, as usual ;)

> understand the it's slow and I don't like giving that impression (what
> would we have if not for a pride in our work?)

You can try to port it and distribute binaries. As worst, they will be
labelled as "unsupported" or "buggybuggydontuse", hence not disgracing
the project too much.


-- Renaud