Re: Comments on accept-mutex/single-listen patch ??

[Marc Slemko]

On Wed, 29 Aug 2001, Marc Slemko wrote:

> On Wed, 29 Aug 2001, Jim Jagielski wrote:
> 
> > Marc Slemko wrote:
> > > 
> > > So I don't see how "NONE" is viable on _ANY_ platform in the multiple
> > > listener case.  It may seem to "mostly" work, but it is not reliable and
> > > can not be permitted.
> > > 
> > 
> > threaded
> 
> ???
> 
> First off, simply being threaded doesn't avoid the issue... it depends on
> the particular model being used (ie. the MPM in 2.x...).  

...and if you are threaded using a MPM that doesn't need it (eg. a
separate accept thread), then why would what OS you are using have
anything to do with if it should be used?  ie. why would it be
configurable.  Why isn't the MPM simply not using serialized accepts
period?

The only thing that is a function of what OS it is on is which methods
are available when serialization is necessary, and if serialization
is necessary in the multiple-processes/threads blocked in accept()
on the same socket situation or not.  (although any OS with a usable
threading model probably supports multiple threads blocked in
accept() on the same socket anyway).