Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 19 Jun 2000 18:35:01 -0500
From:      Jonathan Lemon <jlemon@flugsvamp.com>
To:        Alfred Perlstein <bright@wintelcom.net>
Cc:        Jonathan Lemon <jlemon@flugsvamp.com>, Mike Smith <msmith@FreeBSD.ORG>, arch@FreeBSD.ORG
Subject:   Re: kblob discussion.
Message-ID:  <20000619183501.M37084@prism.flugsvamp.com>
In-Reply-To: <20000619160518.F17420@fw.wintelcom.net>
References:  <20000619164329.F37084@prism.flugsvamp.com> <200006192156.OAA09767@mass.osd.bsdi.com> <20000619172041.G37084@prism.flugsvamp.com> <20000619153325.D17420@fw.wintelcom.net> <20000619175805.J37084@prism.flugsvamp.com> <20000619160518.F17420@fw.wintelcom.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jun 19, 2000 at 04:05:19PM -0700, Alfred Perlstein wrote:
> * Jonathan Lemon <jlemon@flugsvamp.com> [000619 15:53] wrote:
> > On Mon, Jun 19, 2000 at 03:33:25PM -0700, Alfred Perlstein wrote:
> > 
> > For which it is a very specific problem, and I'm trying to expand
> > it a little bit so it can be used for other things as well.
> 
> Ok, here's what I'll do, I'll also provide for a way to create a kblob
> and tell the kernel to read the data from an already open file.

Okay.


> > > So then use my soon to be committed accept filters to accomplish this.
> > 
> > How is this applicable?  The accept filters just delay telling the
> > user when data is available, not redirect the data.
> 
> None of the patches you've present do either.

I didn't know that you had to write the code in advance in order to
have an architectural discussion.  But as a matter of fact, I *DO*
have working code, but I don't feel that it's in any form to be 
released at the moment, as its still evolving a bit.

Put another way, I'm not waving my hands here and talking theory,
I'm speaking from my own experience.


> > > > Exactly.  You're looking at kblob as a "static send-only buffer which
> > > > comes from user space".  I'm looking at kblob as an "in-kernel object
> > > > cache".  The two views are _NOT_ that far apart, and with a little API
> > > > help, the current code can be made much more flexible.  Why are you 
> > > > opposed to that?
> > > 
> > > I'm not opposed to it at all, feel free to work on the code as much
> > > as you want as long as you don't kill the fastpath it offers.
> > 
> > Sure, as long as you don't mind me ripping out your API later, and
> > replacing it with something which accomplishes the same thing but
> > is more flexible.
> 
> If it's flexible enough then you won't need to rip out my API, you'll
> find it obnoxiously easy to create compatibility shims for it.

True.  But if the syscalls change, the app needs to be recompiled.


> > > I'm not writing this code for your application Johnathan, I'm
> > > writing it for myself and for the other people that need this
> > > functionality.
> > 
> > Fine.  I'm not complaining about your functionality, I'm complaining
> > about your API.   In one of my last messages, I presented a list of 
> > things I wanted to accomplish with in-kernel buffers.  Now, I don't 
> > explect you to implement the code for it, but I also don't expect
> > you to stick with such a narrow API that makes those aims impossible to 
> > accomplish in the future.
> 
> I never said that the code will be set in stone.  I won't stand in the
> way of your enhancements.

Fair enough.
--
Jonathan


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000619183501.M37084>