Re: aio-core why not using SuS? [Re: [rfc] aio-core for 2.5.29 (Re: async-io API registration for 2.5.29)]

Suparna Bhattacharya (suparna@in.ibm.com)
Fri, 16 Aug 2002 16:58:31 +0530


On Fri, Aug 16, 2002 at 04:53:06PM +0530, Suparna Bhattacharya wrote:
> On Fri, Aug 16, 2002 at 12:03:34PM +0200, Andrea Arcangeli wrote:
> > On Fri, Aug 16, 2002 at 03:09:46PM +0530, Suparna Bhattacharya wrote:
> > > Also, wasn't the fact that the API was designed to support both POSIX
> > > and completion port style semantics, another reason for a different
> > > (lightweight) in-kernel api? The c10k users of aio are likely to find
> > > the latter model (i.e. completion ports) more efficient.
> >
> > if it's handy for you, can you post a link to the API defined by
> > POSIX and completion ports so I can read them too and not only SuS?
>
> Don't have anything handy atm that's any better than what you could
> get through doing a google on "IO Completion ports". (See section at
> the end of this note for some info)

Oh sorry, I should have mentioned Dan Kegel's site which actually
has all the pointers you need. See http://www.kegel.com/c10k.html
(It has pointers to links to both NT and OS/400 completion ports)

Regards
Suparna
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/