httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexei Kosut <ako...@leland.Stanford.EDU>
Subject Re: apache-nspr-01.tar.gz
Date Mon, 27 Apr 1998 16:38:47 GMT
On Mon, 27 Apr 1998, Dean Gaudet wrote:

> On Mon, 27 Apr 1998, Alexei Kosut wrote:
> 
> > On Mon, 27 Apr 1998, Dean Gaudet wrote:
> > 
> > > - buffered i/o:
> > >     They don't have any buffered I/O... I'm trying to figure out the
> > >     cleanest way to put BUFFs on top of the layers.  I'll probably make
> > >     the buffering a layer itself... but there's no flush function to
> > >     run down the layers.
> [...] 
> > Take the NSPR layers, and put our BUFF code on top of them. i.e., make all
> > layers (optionally) buffered, and define mechanisms to flush and do other
> > such things to them.
> 
> You've said almost exactly what I said, except my solution is probably
> more general. 

Oh, okay :)

I had the impression you were saying "I think we should put buffering on
the top-most layer, and maybe chunking, but the rest of the server doesn't
need it." Which I think is wrong. Sorry about the misconception.

> To use your terminology, buffering is a filter.  When you
> don't need it you don't push it on the stack.  When you need it, you push
> a buffering layer underneath your code.  By default the top of the stack
> will always have a buffer (but maybe we want to change that eventually,
> right now I'm not changing that until I get a working server).

Well, if we do chunking, then we definitely need buffering at the top;
otherwise, as you point out, we get little chunks, which aren't desirable.

> Buffering is not required on static-only servers, and proxy caches for
> example.  So it's not strictly necessary.  But it is desirable for
> flexibility.

-- Alexei Kosut <akosut@stanford.edu> <http://www.stanford.edu/~akosut/>
   Stanford University, Class of 2001 * Apache <http://www.apache.org> *



Mime
View raw message