httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ryan Bloom <>
Subject Re: unixd_accept
Date Mon, 24 Dec 2001 21:41:48 GMT
On Monday 24 December 2001 10:01 am, Justin Erenkrantz wrote:
> On Mon, Dec 24, 2001 at 09:00:09AM -0800, Ryan Bloom wrote:
> > On Monday 24 December 2001 03:58 am, David Reid wrote:
> > > This is defined, but how about we rename it to something without the unixd_
> > > portion so that it's simply a more generic os function, something like
> > > ap_os_accept?  I ask as I added a beosd_accept and that seems a little
> > > clumsy as in the worker MPM I have to then add a series of #if's to get the
> > > function used, so simply having an ap_os_accept would solve the problem and
> > > seems like an easier way to go.
> <snip, snip>
> > I have no problem with doing this, but we need to keep this logic out of the
> > main server.  The whole point of the unixd_* functions is that they don't make
> > any sense for Windows, and possibly for OS/2.  If you can use all or even
> > most of those functions, then why don't you just rename your platform
> > to unixd, and abstract out the differences.
> Silly question, but why isn't this a hook that various platforms
> can implement?  I kind of thought that was the purpose of hooks.
> (And, I believe that hooks may not be there, so nothing will run -
> perfect for Win32 or OS/2).  -- justin

Because it doesn't make any sense as a hook.  Each socket type only
has one way to accept a connection.  The point of hooks is to allow
multiple modules a chance to intercept a call, so that the best one can be
used.  The point of this code is to allow the module that created the socket
to add a function that knows how to accept a connection.  The code that
creates a socket must also know how to accept a connection, or the
socket is useless.  Remember that hooks have performance implications
that means we don't always want them.


Ryan Bloom
Covalent Technologies

View raw message