Linux-Development-Sys Digest #96, Volume #7 Tue, 24 Aug 99 02:14:17 EDT
Contents:
Re: Can't compile network drivers (Peter Samuelson)
Re: Can't compile network drivers (Peter Samuelson)
Re: gcc internal compiler error, why ? (Peter Samuelson)
Re: Shared Libraries: what is the linux equivalent of "dllimport" and "dllexport"
(Peter Samuelson)
Re: loading an OS using Linux (H. Peter Anvin)
Re: TAO: the ultimate OS (Christopher Browne)
Re: need a _good_ kernel ftp site (Paul Kimoto)
Re: update_vm_cache (newsseeker)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: Can't compile network drivers
Date: 23 Aug 1999 02:22:19 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Stephen Torri <[EMAIL PROTECTED]>]
> I know that there is a driver from Realtek.
No, actually there's a driver from Don Becker, which at least one
vendor (Addtron) distributes on a floppy with their cards, patched
slightly for things like their own card model name. At least Addtron
didn't remove the original author/copyright info like some sites and/or
vendors have been known to do with Don's drivers in the past....
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: Can't compile network drivers
Date: 23 Aug 1999 02:15:36 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Stephen Torri <[EMAIL PROTECTED]>]
> I know that there is a driver from Realtek. What I want to know is
> why there are so many drivers listed in the menu when I use xconfig
> but are grayed out? I noticed that there were a large number of
> normal drivers missing from menuconfig as well. Just curious.
Because use of these entries depends on some other option you didn't
select earlier.
In the case of the 8139, according to the source (see the file
drivers/net/Config.in) it depends on CONFIG_EXPERIMENTAL, i.e. the very
first config option "Prompt for development and/or incomplete
code/drivers". Don't let that scare you off -- the driver seems to
work OK for me.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: gcc internal compiler error, why ?
Date: 23 Aug 1999 02:32:22 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Cute Panda <[EMAIL PROTECTED]>]
> I have no idea about the internal compiler errro of gcc, anybody knows ?
> please help, thanks a lot!
> ../../gcc/local-alloc.c:1974: Internal compiler error in function
> find_free_reg
Is it reproducible, i.e. can you *always* get the same error including
filename and line number? If so, you have a buggy gcc. If not, turn
down your overclocking a little or start wondering about that bargain
memory you just picked up. See http://www.bitwizard.nl/sig11/ for
various problems (most of them with gcc) caused by hardware.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Shared Libraries: what is the linux equivalent of "dllimport" and
"dllexport"
Date: 23 Aug 1999 04:13:21 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Jonas Utterstrom <[EMAIL PROTECTED]>]
> I am a C++ programmer so I am not trying to fuel the C vs C++
> flamewar. I just compare with the size of Windows dlls. And the fact
> is that i.e. Mozilla is twice as big in Linux.
Is that "Mozilla" or "Communicator"? If you're talking about
Communicator, recall that the Linux version has a copy of Motif
statically linked in, which could account for a great deal. Mozilla,
of course, won't have this problem....
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (H. Peter Anvin)
Subject: Re: loading an OS using Linux
Date: 24 Aug 1999 05:05:21 GMT
Reply-To: [EMAIL PROTECTED] (H. Peter Anvin)
Followup to: <[EMAIL PROTECTED]>
By author: Sandeep Kumar <[EMAIL PROTECTED]>
In newsgroup: comp.os.linux.development.system
>
> Hi everyone, I've grappled with this a while, so I'm seeking help. Briefly,
> I'd like to load xinu (for Pentium) using Linux. I think that version just
> runs in real mode.
>
> The xinu distribution from ftp://ftp.cs.purdue.edu/pub/Xinu/ compiles on a
> Linux box which I then copy to /boot, add an entry into /etc/lilo.conf, run
> lilo, and reboot. The system hangs with the message "Loading xinu".
>
> So, my question is: has anyone else done this yet? Are there any
> suggestions? What I would like to know, without serious debugging of the
> Linux Makefile is:
>
> o Is there a way to compile a program such as xinu so that it is linked to
> exactly the address to which it will be loaded by Lilo?
> o How do I indicate to Lilo to jump to xinu's start()?
> o Whatever else I need to know...
>
> I'm also thinking that I could conceivably compile xinu as a module (PIC)
> and, on load, just go into real mode or some such and call xinu's start(),
> perhaps after relocating it to where it needs to be? Any suggestions?
>
If it runs in real mode, you should be able to run it under DOSEMU.
-hpa
--
<[EMAIL PROTECTED]> at work, <[EMAIL PROTECTED]> in private!
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Reply-To: [EMAIL PROTECTED]
Date: Tue, 24 Aug 1999 04:41:25 GMT
On 23 Aug 1999 18:13:18 GMT, Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
> * what is an OS?
>what is an OS?
>
> The OS of a computer has undergone many modifications over time. At
> first it was designed as a very rudimentary system that simply loaded
> programs and interfaced with the basic hardware of the computer, i.e.
> tape or disk drive, keyboard, monitor. The modern OS has grown into a
> system of astonishing breadth and complexity with much code dedicated
> to creating the GUI and supporting diverse hardware (including
> networks). Microsoft (MS) NT code numbers in the tens of millions of
> lines.
>
> Mainframe OSes often are even more complex, however the world does
> seem to be standardizing on Win95, NT, and Unix in general. The
> so-called battle between MS-based OSes and non-MS OSes is still in
> play with people taking sides whether MS will totally dominate the OS
> niche or whether an upstart such as Linux may threaten the MS
> hegemony.
Impressive spiel.
It's rather unfortunate that this doesn't actually provide any insight
into the (good question) of "what is an OS?"
It would probably have been a wiser move to at least indicate that
"OS" represents the two words "Operating System."
> I have mused on a name to call it for a long time, but recently came
> up with the idea of naming it "Tao". Tao is a very subtle Eastern
> philosophy espoused in simple poetry with short stanzas by the mystic
> Lao Tzu several centuries B.C. My goal is to formulate an operating
> system with the same combination of simplicity and power voiced in
> this philosophy and maybe spark the creativity, enthusiasm, and most
> ideally, the labor of others in achieving it.
Sadly, there is already an important software package that already has
that name, namely The Ace ORB.
> Here are the key design criteria of Tao, all of which will be expanded
> in the sections below:
>
> * The system favors end-user convenience and simplicity more than
> any other criteria. Many OSes claim to do this but actually make
> subtle choices that favor "developer convenience" over user
> convenience. For example users often have to perform complicated
> installation tasks that could all be eliminated but are not, due
> to programmer laziness or lack of development time. The system
> must be extremely easy to use. If Win95 is considered "user
> friendly", then Tao is even far superior to this, say
> "conscientious". There are many aspects of Win95 that one could
> argue are "user friendly" (compared to earlier very crude or
> rudimentary means such as command line interfaces), but are
> definitely not "conscientious".
> * The system has a strong element of uniformity and coherence, both
> internally in the code and externally in all interfaces, such as
> for devices and the user GUI, via a totally unified
> object-oriented system. Many OSes are designed and various
> elements are continually added over time, and the system becomes
> "legacy software" that is hard to maintain and loses cohesion. Tao
> embraces the idea that software is continuously evolved and
> upgraded, including the OS itself.
> * The system is based on completely open standards and hence has
> excellent interoperability. It is also based on a continual
> creation of new open standards within it. In contrast commercial
> OS companies are often hostile toward open standards.
> * The system has excellent security. Viruses are impossible to
> contract on the system based on its design. In contrast people
> must be somewhat paranoid and go through all kinds of contortions
> to avoid infection on existing systems. Also, end users changing
> system configuration parameters will *never* paralyze the OS.
> Accidental corruption of the system integrity is virtually
> impossible; the system is "foolproof" no matter how ingenious the
> fool!
> * The system has extreme integrity and reliability. Many common
> failures associated with existing OSes are impossible in Tao due
> to stability features. Huge amounts of unnecessary time are wasted
> on fixing or debugging unreliable OSes today. In Tao, increasing
> functionality and components (new hardware and software) does not
> noticeably increase complexity or unreliability of the system.
> Sophisticated debugging features make troubleshooting easy even
> for the novice.
> * The user is almost always insulated from low-level concepts such
> as hard drives, cards, devices, memory, etc. For example a user
> might add a new hard drive, and simply see available "space"
> increase, without any notion of some objects being located on one
> drive or another.
The killer question is always that of when push comes to shove, and it
is necessary to choose between these "extreme" requirements just which
will be favored for any given design element.
None of these are actually design elements; merely constraints...
>the object
>
> In Tao the center of the OS universe is the "object" which is vaguely
> analogous to a C++ object except that it is intrinsic to the entire
> OS, any language can be associated with its code, and it has a unique
> version, as well as other differences. The entire OS is completely
> object oriented. 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.
> * 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.
> * Objects can be "moved around" to different "places" either on a
> local computer or remote computers.
> * 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.
Hmmm. Sounds like someone heard about CORBA, and mistook an Object
Request Broker for an Operating System.
> Every mass-produced OS so far does not use objects for certain
> intrinsic functions. For example a disk directory tree can be thought
> of as a single object that contains subobjects-- subdirectories. But
> the systems do not treat file organization in this way.
>
> Tao would not have a file system in the conventional sense. Disk space
> is mapped onto an object. Subobjects can be created within this
> object.
Ah. A system based on persistent objects.
> 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.
>
> The use of objects makes Tao the most modular system possible. 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.
Including safeguards that prevent a user from creating combinations of
objects that provide an isomorphism that *looks* like a file? :-)
>security
>
> 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. In Tao there are robust limits on
> all finite resources such as memory, disk space, screen space, and
> processing time such that no processing can take on any pathological
> aspects. Certain operations such as formatting a disk cannot even be
> called by non-OS software components.
Most excellent.
How does this map onto conventional security concepts such as security
domains, mandatory access controls, access control lists, and such?
>Some will object to any proposal for a new system, focusing solely on
>spurious elements such as the reputation or background of the creator
>or proponents as the gauge for its validity rather than the inherent
>ideas, invoking the "genetic fallacy".
I think it hardly likely that anyone will feel the need to focus on
your reputation or background. Focusing on the ideas, or lack
thereof, should be quite adequate.
>Others may object that merely coming up with such an idea does not
>involve any progress toward the goal espoused.
Coming up with the idea of having an operating system having a whole
pile of constraints is hardly progress.
Architectural definitions would be, if they were presented.
--
"Bother," said Pooh, as he deleted his root directory.
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/oses.html>
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: need a _good_ kernel ftp site
Date: 24 Aug 1999 01:17:46 -0500
Reply-To: [EMAIL PROTECTED]
In article <gvgw3.1815$[EMAIL PROTECTED]>, Phil Howard wrote:
> I would need to
> figure out which site is actually reliable and stable. Does anyone
> know which ones are?
Usually I use katelyn.optilinkcomm.net, kernel.varesearch.com,
yamato.terrabox.com, or som12308.som.cwru.edu (all incarnations
of ftp.us.kernel.org).
--
Paul Kimoto <[EMAIL PROTECTED]>
------------------------------
From: newsseeker <[EMAIL PROTECTED]>
Subject: Re: update_vm_cache
Date: Mon, 23 Aug 1999 23:48:03 -0500
Peter Samuelson wrote:
> [newsseeker <[EMAIL PROTECTED]>]
> > I put the same question to the linux-kernel mailing list and got
> > greeted with cries of anguish that this question was asked again for
> > the >1000th time. However they did give me an answer which is:
> [answer snipped]
>
> To you and to Hee-Chul Yun:
>
> Do not, repeat, DO NOT run development kernels without following the
> development process! This means reading linux-kernel or at least
> enough of it to have some idea where development is going and which
> patches to avoid.
You have to admit that if a person reads the linux-kernel mailing list
every other day, they can very easily miss a patch that should have been
avoided. And, perhaps it is apparent to you as to where to find all the
kernel change logs (much less putting together what works and what doesn't
out of the change logs for each and every change), but for someone like me
trying to learn where all the sources are and trying to put things
together so maybe one day I could contribute, it isn't intuitive as to
where all the information sources are. I've been monitoring the 'list'
for 2 months, and I've yet to see a location mentioned for the new
change-log, or collection of change-logs (freshmeat.net gives better info
as to where the latest changelog is). BTW, please don't try to tell me to
use a mirror....too many times I've seen where the mirrors are sometimes
weeks behind the real site.
> I don't mean to take the tone of a flame because obviously it's not
> just you guys: a lot of people try to run development kernels when they
> shouldn't.
I'm not flaming either, but it is frustrating to catch up to knowing where
or what everything is...and at times it feels like me, being an
'outsider', never will catch up.
> But a general question: What exactly did you hope 2.3.12
> would do for you that 2.2.5 wouldn't?
I run 2.2.10smp - it has some features in the networking that is more
stable than 2.2.5. In my case, I tried out 2.3.12 just to check it
out...curiosity about the USB support, and wanted to see if I could see
any differences in the 'improved' smp handling. Are you telling me I
shouldn't run 2.3.12 just to check it out? BTW: kernel-2.3.12 actually
is more stable for me than kernel-2.2.10smp (ie - everything including
sound works consistently where I have some inconsistencies with
kernel-2.2.10smp ... please don't bother telling me that I have 'hardware
problems' since hardware problems seldom disappear with the change of
software).
> For myself, I run 2.3 to keep up
> on the development process and maybe contribute the odd patch when I
> see something wrong. The fact that you didn't know vfat was broken
> from 2.3.7 until 2.3.13 tells me that you must have other reasons.
You deduce wrong. I didn't know vfat was broken because I didn't read the
changelog (much less trying to find the full thing) back to where it got
broken.....I just read it for 2.3.12 foolishly thinking that if there was
still a problem in 2.3.x, it would also show up for the changelog in the
most current version of 2.3.x. I despise having to read the @#$%
changelog from 2.3.1 to 2.3.x just to determine what still works, and what
doesn't in the current 2.3.x version.
> --
> Peter Samuelson
> <sampo.creighton.edu!psamuels>
Don't worry, I won't be posting to the linux-kernel mailing list since I
know 'outsiders' aren't welcome.
--nwskr
------------------------------
** 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
******************************
