Linux-Development-Sys Digest #790, Volume #6 Sun, 6 Jun 99 18:14:31 EDT
Contents:
compiling for Hitachi SH-3 with gcc ([EMAIL PROTECTED])
Re: uid and gid assignments as distributed (Kai Henningsen)
Re: Linux daemons ("Ashutosh S. Rajekar")
Re: system call ("Stefan Monnier "
<[EMAIL PROTECTED]>)
Re: Reliable (!) nic for 2.2 kernel? (bryan)
Re: Linux development tools - new : download freeware compiler with IDE and
GuiDesigner for Linux & Windows (Martin Maney)
Re: Proposal for XML Configuration files (Keith Wright)
Re: TAO: the ultimate OS (David Magda)
Re: the ultimate OS ("FM")
Re: the ultimate OS ("FM")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED]
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.anwers
Subject: compiling for Hitachi SH-3 with gcc
Date: Sun, 06 Jun 1999 14:59:25 +0200
I'm actually looking for the libraries (or modules; whatever needed) to
compile programs for a Hitachi SH-3 CPU with gcc running on i386 linux.
Thanks, Peter [[EMAIL PROTECTED]]
------------------------------
Date: 06 Jun 1999 15:48:00 +0200
From: [EMAIL PROTECTED] (Kai Henningsen)
Subject: Re: uid and gid assignments as distributed
[EMAIL PROTECTED] (Phil Howard) wrote on 03.06.99 in
<3756f6c4$0$[EMAIL PROTECTED]>:
> On 3 Jun 1999 09:38:09 +0200 Villy Kruse
> <[EMAIL PROTECTED]> wrote:
>
> | The xfs user id is actually created by an install script by the useradd
> | command and will get the next available user id. So if user id 100 is
> | already taken it gets user id 101, and if that is taken as well 102, and
> | so on. If you inustall the package Xfree86-xfs completely and remove the
> | user xfs from the password file. the install procedure might very
> | well assign a different number if you reinstall that package.
> |
> | check: 'rpm -q --scripts Xfree86-xfs'
>
> This sounds reasonable. What is needed is a class of uid/gid numbers for
> services and applications. Install should ask me if I want override the
> uid/gid class definitions if it detects any packages that need uid/gid
> assignments. Then I can get the install to conform to my installation
> standards. A global standard would be the default, and what I would do
> is just adapt to that global standard.
This is what's in /etc/adduser.conf on a Debian system:
# /etc/adduser.conf: `adduser' configuration.
# See adduser(8) and adduser.conf(5) for full documentation.
# The DSHELL variable specifies the default login shell on your
# system.
DSHELL=/bin/bash
# The DHOME variable specifies the directory containing users' home
# directories.
DHOME=/home
# If GROUPHOMES is "yes", then the home directories will be created as
# /home/groupname/user.
GROUPHOMES=no
# If LETTERHOMES is "yes", then the created home directories will have
# an extra directory - the first letter of the user name. For example:
# /home/u/user.
LETTERHOMES=no
# The SKEL variable specifies the directory containing "skeletal" user
# files; in other words, files such as a sample .profile that will be
# copied to the new user's home directory when it is created.
SKEL=/etc/skel
# FIRST_SYSTEM_UID to LAST_SYSTEM_UID inclusive is the range for UIDs
# for dynamically allocated administrative and system accounts.
FIRST_SYSTEM_UID=100
LAST_SYSTEM_UID=999
# FIRST_UID to LAST_UID inclusive is the range of UIDs of dynamically
# allocated user accounts.
FIRST_UID=1000
LAST_UID=29999
# The USERGROUPS variable can be either "yes" or "no". If "yes" each
# created user will be given their own group to use as a default, and
# their home directories will be g+s. If "no", each created user will
# be placed in the group whose gid is USERS_GID (see below).
USERGROUPS=yes
# If USERGROUPS is "no", then USERS_GID should be the GID of the group
# `users' (or the equivalent group) on your system.
USERS_GID=100
# If QUOTAUSER is set, a default quota will be set from that user with
# `edquota -p QUOTAUSER newuser'
QUOTAUSER=""
Kai
--
http://www.westfalen.de/private/khms/
"... by God I *KNOW* what this network is for, and you can't have it."
- Russ Allbery ([EMAIL PROTECTED])
------------------------------
From: "Ashutosh S. Rajekar" <[EMAIL PROTECTED]>
Subject: Re: Linux daemons
Date: Sun, 6 Jun 1999 19:40:31 -0500
Hi Baldbass,
The book on UNIX networking written by W. Richard. Stevens: UNIX
Network programming, and the new 3 book series on TCP/IP for UNIX by the
same author are more than enough for learning how to write daemons, in
fact you'll become a master in UNIX network programming if you read these
books. Source code given in the above mentioned books can be found at
http://www.kohala.com/~rstevens (Almost all the code runs without any
change on almost all UNIX systems). Check out http://www.awl.com/cseng,
and do a serach on the above books/author.
On Fri, 4 Jun 1999, Baldbass wrote:
> Are there any places I can find information on writing a daemon for a Linux
> newbie?
==================
Ashutosh Rajekar
http://i.am/asr
==================
------------------------------
From: "Stefan Monnier <[EMAIL PROTECTED]>"
<[EMAIL PROTECTED]>
Subject: Re: system call
Date: 06 Jun 1999 12:19:33 -0400
>>>>> "NTVO910" == NTVO910 <[EMAIL PROTECTED]> writes:
> Am I a complete idiot?
I'm afraid so.
Why don't you tell us what you *really* want to do ?
Stefan
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: Reliable (!) nic for 2.2 kernel?
Crossposted-To: comp.os.linux.networking
Date: Sun, 06 Jun 1999 16:21:29 GMT
In comp.os.linux.development.system Peter Samuelson <[EMAIL PROTECTED]>
wrote:
: [bryan <[EMAIL PROTECTED]>]
: > irix has an ifconfig switch to set speed and duplex. why doesn't
: > linux?
: > even if its not acceptable to some kernel folks, at least 2 people
: > think this is the right place for it (ifconfig).. ;-)
I got email from a developer and he said that this feature is
in-progress! so others do agree that ifconfig is the right
card-neutral place for interface config options.
--
Bryan
------------------------------
From: Martin Maney <[EMAIL PROTECTED]>
Subject: Re: Linux development tools - new : download freeware compiler with IDE and
GuiDesigner for Linux & Windows
Date: 6 Jun 1999 16:04:40 GMT
Max Reason <[EMAIL PROTECTED]> wrote:
> If you can stand to program in something other than C++, there is a complete
> integrated development environment that just became available for download
> as freeware.
A little lie, Max: your demoware is free, but useless as anything other than
a demo for the commercial version. Why don't you post the prices for the
real product while you're spamming us about your crippled demo?
> You can ask me pointed questions about XBasic if you like. :-)
What's it cost for a useful version? I guess that's the "pro" release - if
you can't make standalone executables it's just a toy.
------------------------------
From: Keith Wright <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.programmer,comp.os.linux.development.apps
Subject: Re: Proposal for XML Configuration files
Date: 06 Jun 1999 16:49:00 -0400
[EMAIL PROTECTED] (Kaz Kylheku) writes:
> Get real, config files rule. The more syntaxes the merrier.
You lost me there. If there is a reason for another syntax, then
go for it, but it's hard for me to believe there are programs so
novel that one of the existing syntices won't work.
> >Is there anybody in the web thinking about a standard on this topic,
> >maybe based on XML?
>
> Good luck convincing everyone to use whatever scheme you come up with.
As you have (probably accidentally) pointed out, the best solution
in many cases will be to use Guile to embed a Scheme interpreter,
thus making your config file a real programming language, instead
of a random pile of hacks.
What does XML, (a document format description language) have to do
with it? I want to be able to configure like this:
(set! startup-hook
(lambda argv (if (null? argv)
(display-splash "file"))))
--
-- Keith Wright <[EMAIL PROTECTED]>
Programmer in Chief, Free Computer Shop <http://www.free-comp-shop.com>
--- Food, Shelter, Source code. ---
------------------------------
From: David Magda <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: 6 Jun 1999 20:23:50 GMT
In comp.os.linux.advocacy Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
[deleted the whole thing]
You will have to name it something else. Please see http://www.tao.co.uk.
--
David Magda <dmagda at ee.ryerson.ca>, 2nd Year Electrical Eng.
"Well," said Pooh, "what I like best--" and then he had to stop and think.
Because although Eating Honey was a very good thing to do, there was a
moment just before you began to eat it which was better than when you were,
but he didn't know what it was called. - A. A. Milne, The House at Pooh Corner
------------------------------
From: "FM" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: the ultimate OS
Date: Sun, 6 Jun 1999 13:45:10 -0700
G. Sumner Hayes wrote:
>Aside: If you want to use design documents to convince people, you'll
>need to demonstrate a firm understanding of the field.
Actually it merely needs to be a design document, but design of
implementation, not goals. The Tao document was a wish-list, rather
that how such a system should be put together.
Dan.
------------------------------
From: "FM" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: the ultimate OS
Date: Sun, 6 Jun 1999 13:36:35 -0700
Vladimir Z. Nuri wrote:
<snipped>
Overall, I found this very similar to an operating system I
was envisioning a few years ago, which was supposed to be a
key element in a house automation/network system I was
modeling. Of course my brain went nuts the moment I
considered implementation. My vague cricism follows.
> The entire OS is completely object oriented.
This is rather vague. What about Objected Orientation makes
it appropriate for low-level programming? Even if it were
possible to write an OS in OOP, it should be done only when
it provides better implementation. Of course whether the
above means the system should be written in OOP or merely
provide OOP interface isn't too clear.
>Objects have certain basic properties:
>
> * An object can be contained in other objects, and an object can
> "contain" multiple objects, or copies of them. These objects can
> be located "elsewhere," say distributed over a network.
I've always thought that the distinction of a directory as
a special file that contains subfiles very limiting. Indeed
I like the idea of a file being able to contain other
files, especially in view of HTML or any other document
structure that inherently compose of multiple objects.
index.html isn't always really an elegant way to do things.
> * An object has several basic elements: a *name*, *type*, *version*
> *interface*, *code*, and *data*. Some data is read-only while some
> data is readable and writeable.
I don't see what you mean by code, distinguishable from
interface. What is data? A raw binary stream? Data should not
be accessible directly at all should it?
> * Objects can be "moved around" to different "places" either on a
> local computer or remote computers.
How are interface/links maintained when an object is
transferred from one computer to another? For example if a
picture file, an object is transferred, are the methods for
display and manipulation automatically transferred as well?
> * Objects "hook up" to other objects and pass data back and forth.
> These links are all carefully recorded and entire object networks
> and trees can be traversed.
What is that supposed to mean and how is it supposed to be
implemented?
> There is another key way the existing OSes (both Linux and Win95) do
> not use objects in file systems, leading to endless difficulty. In
> Win95 and Linux both, a rather weird and troublesome convention of
> having extensions represent file types is used. In contrast, the type
> is an intrinsic property of the object and there is never any
> ambiguity.
So what happens if you download a file from a Linux machine?
How is the type determined? Does it carry "unspecified" tag
or are there some magical methods that automatically assign
the correct types to those files?
> The use of objects makes Tao the most modular system possible.
It could lead to such elegance but that depends much more on
implementation, which is more than hard. If I was thinking
about it, and you are thinking about it, many OS designers
have probably once stumbled onto it as well. When we don't
see a good implementation (if there is please let me know),
there is a good reason.
> The
> common artificial distinctions in current OSes between disparate
> aspects such as files, directories, etc. are all dissolved and
> subsumed into the object concept. Nothing other than objects exists on
> the system, and all objects are accessable to the user and software
> via impervious safeguards.
This is a minor issue but I'd like to see a file system that
lets a file have subfiles (or rational behind not allowing it)
>the object browser
> In Tao, a universal object browser is available and contains the
> complete logic and interface for dealing with the entire system. The
> object browser is a cross between a web browser, an OS application
> desktop, process manager, a file viewer, and a programming
> environment. It is very cleverly constructed such that only the amount
> of information relative to the sophistication of the user is revealed.
I think this is very close to what MS is trying to do with
explorer.exe. It's not really much more than a file manager/
user interface that can make use of plug-ins.
>optimization
>
> Tao is extremely intelligent and can handle multiple levels of
> caching. Local RAM and disk are both seen as intermediate caches to
> the network.
Exactly. This was one of the more central points to my
vision, a system where the only distinctions between
RAM/Disk/Network are distance and speed, where all system
resources are accessed in similar manner. I felt that
this is important for the house automation/network system
since the distinction between permanent storage and
temporary storage is insignificant in server-type systems.
But then I later found that these ideas have already been
implemented in certain systems (Single-Address Space
Operating Systems).
>conversion
>
> A very major aspect of Tao is a unified system whereby a set of
> objects are available to convert objects in one form to objects in
> another form. So for example a file "object" in one "format" (type)
> can be converted to one in another based on the inherent objects in
> the system. In Tao, there is total knowledge about the entire object
> network and the paths through which an object may be converted.
You mean there are ways (filter objects?) through which the
system can be taught how to convert different object types.
You can't expect an OS is to have total knowledge of how to
read and write every single object type in the world.
> Overall, much of the Tao OS can be seen as a streamlined conversion
> between data and code in a way that is not well understood or
> practiced today. For example in many programming languages there is a
> debate between "interpreted", "compiled", "just-in-time", etc.-- but
> all code is equivalent and these concepts are relevant only as
> optimization features within the OS, analogously to disk caching.
> Objects that are used frequently should be compiled, objects that are
> used less frequently can be just-in-time compiled, and infrequently
> used objects can be interpreted "on the fly". A well-constructed OS
> could automatically make the best decisions with the help of some
> occasional explicit tweaking by programmers.
Ouch this is way too close to my idealism :)
> In Tao, the user is given powerful control over the object system. The
> distinction between the application created by a company and what is
> created by a user is blurred. A user can access the internal
> connections of the objects in software and modify them to his needs.
This, if I read correctly, forces open source model on
the developers and may not be welcomed at all by developers.
I had thought of such idea and felt that there should be a
way to let the developers package and hide their object
structures. After all there's got to be an informal way to
escape this exposure anyways, so why not make it formal
through "the powerful security model provided by the OS"?
> All components
> grow or shrink in "size" (memory and processing time) very smoothly,
> dynamically, and invisibly as demand fluctuates. The system always
> "flows smoothly" even when almost anything breaks.
There always is a trade-off between efficiency and
flexibility. We can only hope that efficienty will be less
of a concern in future computing than it has been.
> The security is such that a virus or vicious application is simply
> impossible to create on the system, in the same way that a
> microprocessor can protect some areas of code as "read only" without
> any exception. Nothing that an application is permitted to do can have
> a pathological effect on the system.
How about an Operating System upgrade? I was thinking
of a system that actually puts absolute restrictions on
every process, where it cannot modify anything beyond
its scope (sandbox model). To do anything useful with
such a system however, UI must be given some special
priviledges, to make it possible for a user to
temporarily extend its scope (modifying a document not
within its object structure, for example, would be
considered out of scope, for any process).
> In Tao *everything* is *always* "plug-and-play". All hardware, all
> software, everything. Anything can be connected or disconnected at any
> time with the system handling it elegantly in every situation.
These are hardware restrictions that you cannot deal
with as an OS developer. Unless you consider the message
"Unable to use the device" as elegant handling, the
above is impossible.
> * Some may complain that a totally object-oriented OS may be slow,
> or that other ideas here cannot be efficiently implemented. But I
> generally disagree. I think good ingenuity and design decisions
> will actually make the speed of the Tao OS as fast or faster than
> existing systems, because of the overall coherence of the system.
> And again Moore's law is always applicable, such that a more
> complex OS embodying the demanding aspects of Tao will become more
> feasible in the near future. There will be a price to pay for
> every aspect of Tao but in my opinion the end result will be well
> worth it.
In other words, Tao is going to be slow than other less
elegant systems, but Moore's Law and its flexibility will
more than compensate for it. Nothing wrong with that,
but let's cut the BS ("cannot be efficiently implemented
... I disagree")
> * Open standards will be difficult to achieve with many companies
> thinking that they cannot make money without proprietary systems,
> and the cooperation of many companies that think they compete. But
> it may be that Linux is showing that now even an entire OS can be
> "open" and benefit the entire worldwide computing community while
> supporting companies can still be economically viable.
Before that, think about how TAO breaks all the existing
standards.
> Tao is not merely an open OS but in fact an open computation universe.
> I think its design goals are inherently desirable and practical and
> will eventually be achieved in some form one day not too far in the
> future.
I like the system a lot, just as I did my Operating System,
which seems nearly identical, but I'd wait before I see
some implementation or even specific design as to how.
Dan, ever as dreaming.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
