On Jul 27 12:02, Bence Szépkúti wrote:
> Hello,
> 
> I was poking around in the fork() code in Cygwin, and was wondering if
> someone could explain to me / point me to a discussion of why the NEWVFORK
> code was abandoned.
> 
> I realise that this code had been disabled as "not working" since 2008,
> however I am interested in why this was the case. Was the concept behind
> "short-circuiting" vfork itself unfeasible in the context of Cygwin? Was a
> compatible implementation too complex / too slow to be worth it compared to
> a regular fork?

This was long ago and the original developer hacking this code isn't
with us anymore.  The only vague recollection I have is that it never
really worked right, but no details.  Maybe searching the mailing list
archives helps unearthing some of the problems.


Sorry,
Corinna

-- 
Corinna Vinschen
Cygwin Maintainer