From: Bill Findlay <yaldnif.w@blueyonder.co.uk>
Subject: Re: Anti-Ada FUD (rant)
Date: Mon, 23 Sep 2013 13:54:43 +0100
Date: 2013-09-23T13:54:43+01:00 [thread overview]
Message-ID: <CE65F8A3.37211%yaldnif.w@blueyonder.co.uk> (raw)
In-Reply-To: 51a27fa6-027d-4655-a78c-6a582811e23f@googlegroups.com
On 23/09/2013 13:16, in article
51a27fa6-027d-4655-a78c-6a582811e23f@googlegroups.com, "Peter Brooks"
<peter.h.m.brooks@gmail.com> wrote:
> On Monday, 23 September 2013 11:39:09 UTC+2, Bill Findlay wrote:
>>
>>> Fork/exec represent a unique combination of total disregard to efficiency
>>> with violation of practically any known principle of good software design.
>>
>> Now there I have to agree with the UNIX critics.
>>
> Why? I worked with a number of operating systems before encountering unix, and
> I've seen a few since. The setup and tear-down work involved with launching
> processes was a massive overhead that slowed things down and added very little
> value. Fork-exec, on the other hand, is quick because the child process
> inherits the environment, which has many other advantages - security, for
> example, is resolved at login and then inherited, similarly with I/O.
Inheriting (some subset of) the parent context is fine.
> What's your objection to it? What dispatching system works better?
It duplicates the parent program. It should allow the program to be run to
be specified as part of the fork. Postponing that to a following exec
imposes costs that are very often unnecessary, and it can lead to race
conditions in establishing the new running program's context.
> What particular principles of software design does the mechanism violate?
Unnecessary coupling.
--
Bill Findlay
with blueyonder.co.uk;
use surname & forename;
next prev parent reply other threads:[~2013-09-23 12:54 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-20 22:18 Anti-Ada FUD (rant) krfkeith
2013-09-20 22:47 ` Adam Beneschan
2013-09-20 23:32 ` John B. Matthews
2013-09-21 12:06 ` Georg Bauhaus
2013-09-22 0:47 ` gautier_niouzes
2013-09-22 21:24 ` Shark8
2013-09-23 4:43 ` Peter Brooks
2013-09-23 5:14 ` Nasser M. Abbasi
2013-09-23 6:05 ` Shark8
2013-09-23 23:19 ` Nasser M. Abbasi
2013-09-24 7:10 ` Bill Richards
2013-09-23 7:00 ` Bill Richards
2013-09-23 7:55 ` Bill Findlay
2013-09-23 8:31 ` Peter Brooks
2013-09-23 8:37 ` Bill Richards
2013-09-23 8:52 ` Bill Findlay
2013-09-23 8:59 ` Georg Bauhaus
2013-09-23 11:31 ` Bill Richards
2013-09-29 1:04 ` Shmuel Metz
2013-09-24 1:40 ` Paul Rubin
2013-09-24 7:05 ` Bill Richards
2013-09-24 7:34 ` Bill Findlay
2013-09-24 7:49 ` Bill Richards
2013-09-24 10:47 ` Georg Bauhaus
2013-09-24 10:38 ` Georg Bauhaus
2013-09-24 15:39 ` Paul Rubin
2013-09-28 23:28 ` Shmuel Metz
2013-09-23 5:55 ` Shark8
2013-09-23 6:12 ` Peter Brooks
2013-09-23 7:34 ` Dmitry A. Kazakov
2013-09-23 8:36 ` Peter Brooks
2013-09-23 9:36 ` Dmitry A. Kazakov
2013-09-23 9:39 ` Bill Findlay
2013-09-23 12:16 ` Peter Brooks
2013-09-23 12:54 ` Bill Findlay [this message]
2013-09-23 14:32 ` Peter Brooks
2013-09-23 22:06 ` Peter C. Chapin
2013-09-29 1:00 ` Shmuel Metz
2013-09-23 7:26 ` Peter Brooks
2013-09-24 6:03 ` Shark8
2013-09-24 7:51 ` Peter Brooks
2013-09-26 20:58 ` erlo
2013-09-24 6:38 ` krfkeith
2013-09-24 23:18 ` Dennis Lee Bieber
2013-09-28 6:21 ` Shark8
2013-09-28 14:40 ` Per Sandberg
2013-09-28 21:54 ` Dennis Lee Bieber
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox