This is the mail archive of the mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: A real fork() on NT

From: "Christopher Faylor" <>
Subject: Re: A real fork() on NT

> Cygwin fork works just fine.  It's slow but it should work as well as UNIX
> fork for a ported application.  The only exception that I can think of is
> if you use dlopen to load a non-cygwin DLL.  In that case there is a
> with relocation of the DLL after a fork.
> If you consider the number of applications that have been ported to
> it would be pretty amazing if there was some basic problem with fork.

Okay, maybe it isn't an issue with fork() specifically, but that was what I
had understood from reading the list archives.

That aside, how tolerant is the list of helping someone getting code to
function? <g>  Eggdrop source is written to skip over forking when running
under cygwin, but you can run it and then close that window and the eggdrop
will continue to run in background.  I edited the source for it so that it
doesn't skip the fork() routines and it returns the controlling terminal but
yet it doesn't function in that it doesn't connect to IRC nor does it listen
on it's telnet port.  I've looked at other sources for things that I might
try to get this to work (code snipettes) but alas nothing has helped
(granted, C is a foreign language to me, though programming in general is
not).  Interesting to note is that when it's ran this way the eggdrop has a
tty of the current login instead of an ? to indicate that it's not using a
terminal output (terminology here may be incorrect, but you get the idea).

TIA for any ideas,

Unsubscribe info:
Bug reporting:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]