All of lore.kernel.org
 help / color / mirror / Atom feed
From: "H. Peter Anvin" <hpa@zytor.com>
To: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Futex Asynchronous Interface
Date: 10 Jun 2002 13:57:22 -0700	[thread overview]
Message-ID: <ae33ri$b8b$1@cesium.transmeta.com> (raw)
In-Reply-To: <8QXHSZTXw-B@khms.westfalen.de> <Pine.LNX.4.44.0206091056550.13459-100000@home.transmeta.com>

Followup to:  <Pine.LNX.4.44.0206091056550.13459-100000@home.transmeta.com>
By author:    Linus Torvalds <torvalds@transmeta.com>
In newsgroup: linux.dev.kernel
> 
> But what would you _do_ with them? What would be the advantage as compared
> to the current situation?
> 
> Now, to configure a device, you get a fd to the device the same way you
> get a fd _anyway_ - with "socket()".
> 
> And anybody who says that "socket()" is utterly unnatural to the UNIX way
> is quite far out to lunch. It may be unnatural to the Plan-9 way of
> "everything is a namespace", but that was never the UNIX way. The UNIX way
> is "everything is a file descriptor or a process", but that was never
> about namespaces.
> 
> Yes, some old-timers could argue that original UNIX didn't have sockets,
> and that the BSD interface is ugly and an abomination and that it _should_
> have been a namespace thing, but that argument falls flat on its face when
> you realize that the "pipe()" system call _was_ in original UNIX, and has
> all the same issues.
> 

The biggest problem with the socket API is that it makes it impossible
to use in a shell script context or other context where it isn't
practical to operate on binary data.  The second-biggest problem with
the API is that it at least historically makes it basically impossible
to write address-family-independent code... something like "telnet"
shouldn't need to know if it's operating over TCP on IPv4, TCP on
IPv6, SPX on IPX...; it should be able to resolve a string to a name and
then create a connection, and the address family stuff should be
encasulated in the library.

The first problem a namespace-like solution would deal with... then
one could open /dev/sock/stream/ipv4/206.189.222.9/23 using a normal
open() call.  The second is to resolve "poetry.vogon.net","telnet" to
the above string, but that's a user-space issue.

At least it would be nice if open() on a Unix domain socket would
connect to that socket.

	-hpa

-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt	<amsp@zytor.com>

  parent reply	other threads:[~2002-06-10 20:57 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-06-06  7:26 [PATCH] Futex Asynchronous Interface Rusty Russell
2002-06-02  0:10 ` Pavel Machek
2002-06-10  6:57   ` Rusty Russell
2002-06-06 16:36 ` Linus Torvalds
2002-06-06 19:27   ` Alan Cox
2002-06-06 23:21   ` Rusty Russell
2002-06-07  8:33     ` Peter Wächtler
2002-06-08 22:28       ` Linus Torvalds
2002-06-09  9:49         ` Kai Henningsen
2002-06-09 18:09           ` Linus Torvalds
2002-06-09 19:06             ` Thunder from the hill
2002-06-10  6:39             ` Kai Henningsen
2002-06-10  7:55             ` Helge Hafting
2002-06-10 14:10               ` Thunder from the hill
2002-06-10 20:46                 ` Kai Henningsen
2002-06-11 14:14                   ` john slee
2002-06-10 15:11               ` Linus Torvalds
2002-06-11 15:06                 ` Eric W. Biederman
2002-06-10 20:57             ` H. Peter Anvin [this message]
2002-06-09 10:07         ` Peter Wächtler
2002-06-09 17:49           ` Linus Torvalds
2002-06-07  9:06   ` Rusty Russell
2002-06-08 22:42     ` Linus Torvalds
2002-06-11  9:15       ` Rusty Russell
2002-06-11 16:53         ` Linus Torvalds
2002-06-12  5:32           ` Rusty Russell
2002-06-12  9:16             ` Peter Wächtler
2002-06-12 14:19               ` Hubertus Franke
2002-06-12 16:50                 ` Peter Wächtler
2002-06-12 18:15                   ` Vladimir Zidar
2002-06-12 15:39               ` Linus Torvalds
2002-06-12 16:29                 ` Peter Wächtler
2002-06-12 16:52                   ` Linus Torvalds
2002-06-12 17:07                     ` Peter Wächtler
2002-06-12 18:32                     ` Saurabh Desai
2002-06-12 20:05                     ` Oliver Xymoron
2002-06-12 20:16                       ` Linus Torvalds
2002-06-13  2:57                     ` Rusty Russell
2002-06-13  9:37                       ` Peter Wächtler
2002-06-13  9:55                         ` Rusty Russell
2002-06-13 16:38                     ` Gabriel Paubert
2002-06-13 16:40                       ` Linus Torvalds
2002-06-13  1:32               ` Rusty Russell
2002-06-06 16:08 Martin Wirth
2002-06-06 22:59 ` Rusty Russell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='ae33ri$b8b$1@cesium.transmeta.com' \
    --to=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.