From owner-freebsd-hackers  Mon Dec 15 22:08:59 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id WAA27861
          for hackers-outgoing; Mon, 15 Dec 1997 22:08:59 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from noether.blah.org (slmel12p25.ozemail.com.au [203.108.200.113])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA27829
          for <hackers@FreeBSD.ORG>; Mon, 15 Dec 1997 22:08:43 -0800 (PST)
          (envelope-from ada@not-enough.bandwidth.org)
Received: (from ada@localhost)
          by noether.blah.org (8.8.7/8.8.8) id RAA03776
          for hackers@FreeBSD.ORG; Tue, 16 Dec 1997 17:08:18 +1100 (EST)
From: Ada <ada@not-enough.bandwidth.org>
Message-Id: <199712160608.RAA03776@noether.blah.org>
Subject: Re: Why FIONREAD has no dual for write ?
In-Reply-To: <199712091520.HAA09154@hub.freebsd.org> from "owner-hackers-digest@FreeBSD.ORG" at "Dec 9, 97 07:20:23 am"
To: hackers@FreeBSD.ORG
Date: Tue, 16 Dec 1997 17:08:18 +1100 (EST)
Reply-To: ada@bsd.org
X-Mailer: ELM [version 2.4ME+ PL32 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> From: Bakul Shah <bakul@torrentnet.com>
> Date: Mon, 08 Dec 1997 19:45:05 -0500
> Subject: Re: Why FIONREAD has no dual for write ? 
> 
> Perhaps one solution is to add an ioctl to set high/low
> watermarks on a device and add new event bits POLLIN_WM and
> POLLLOUT_WM for the poll() syscall.
> 
> In pollfd->events you set these bits instead of (or in
> addition to) POLLIN/POLLOUT.  On return the corresponding bit
> in pollfd->revents is set only if there are _greater than low
> watermark_ bytes on input and _less than high watermark_
> bytes on output.  This guarantees that you can transfer some
> minimum number of bytes on read/write (provided you use the
> O_EXCL mode).
> 
> This is a more general solution that allows finer control
> over when to schedule IO and is useful in all sorts of
> situations.

How about a counterpart to select() which can request a minimum number
of bytes {free,ready} in the buffer?  This could also be useful in
situations where one does not wish to implement buffering; instead the
kernel buffers could be used.

Is this feasable in light of how select() works?

Ada.