Re: [hybi] Background info: Properties of sendfile()

Thus my statement of "theoretical" benefits.
I've not seen the benefits of either splice or sendfile yet.
-=R

On Thu, Aug 5, 2010 at 6:28 PM, Jamie Lokier <jamie@shareable.org> wrote:

> Roy T. Fielding wrote:
> > On Aug 5, 2010, at 5:49 PM, Jamie Lokier wrote:
> >
> > > It's time to explain sendfile() better.
> > >
> > > 0. Nearly all OSes used for web serving have something called
> > >   sendfile() or under a different name.  It has become a standard
> > >   method used in high performance static servers, and is beginning to
> > >   be used in dynamic servers that do smart caching.  It is given an
> > >   open file handle and range, and writes content of that file to a
> > >   TCP socket as efficiently as possible.  Some OSes allow contents of
> > >   process memory to be sent as well.
> >
> > Unfortunately, you should also be aware that nearly all OSes have
> > severe kernel bugs in their implementations of sendfile and seem
> > to be getting progressively worse over time, to the point where
> > Apache httpd needs to disable it by default.  I don't know why.
>
> Thanks.  Ugh.  (I would like to know more, if you can point me at the
> details.)
>
> For WebSocket, details of specific current implementations aren't so
> important, as acknowledging the principles that hardware may continue
> to imply for some years, in those rare cases of OSes / kernels that
> aren't too buggy to take advantage :-)
>
> I.e. it just means "don't mandate that chunks must be small".
> It's not asking much.
>
> Who knows, we might end up with WebSocket multiplexing implemented in
> a kernel, which is likely to avoid sendfile() bugs but still be
> subject to NIC constraints.
>
> -- Jamie
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>