Linux-Development-Sys Digest #787, Volume #6 Sat, 5 Jun 99 13:14:09 EDT
Contents:
Re: DLL's in Linux ? ("G. Sumner Hayes")
Re: NT driver writer new to Linux kernel/drivers (Daniel Robert Franklin)
Re: Framebuffer device? (y chen)
HELP, VFS can not boot (03,01) (y chen)
Re: TAO: the ultimate OS (Alexander Viro)
Re: TAO: the ultimate OS (Christopher B. Browne)
Re: TAO: the ultimate OS (Anthony Ord)
Re: TAO: the ultimate OS (Christopher B. Browne)
Re: HP JetDirects (using DLC) on Linux??? (Carl Anderson)
Re: Might Linux SMP write memory out of order? (David Wragg)
Re: DLL's in Linux ? ("Max Reason")
Re: TAO: the ultimate OS (Larry Blanchard)
Re: DLL's in Linux ? ("Max Reason")
Re: Might Linux SMP write memory out of order? ("Stefan Monnier "
<[EMAIL PROTECTED]>)
----------------------------------------------------------------------------
From: "G. Sumner Hayes" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: DLL's in Linux ?
Date: Mon, 31 May 1999 03:16:20 -0400
Kaz Kylheku wrote:
> It means that your Linux system is *severely* out of date.
> Jump-table based shared libraries are a relic from the a.out period
> of Linux.
[SNIP]
> On a modern Linux system, the output of ldd looks like this:
[SNIP]
> Next time you get bored, consider upgrading. :)
There are still a.out Linux distributions, even ones that keep up
to date on system utilities &tc. Talk to David Parsons if you
don't believe it. And there are some old a.out commercial
applications that haven't seen ELF versions released -- witness the
uproar when ZMAGIC binaries were (temporarily) broken two weeks
ago.
Even if you're using an old distribution, if you aren't on a public
network (where security is an issue) then there's no reason to upgrade
if things are working okay -- unless of course you want to.
a.out has a (fairly trivial) speed advantage over ELF. Not enough
to outweigh ELF's many other advantages IMO.
--Sumner
------------------------------
From: [EMAIL PROTECTED] (Daniel Robert Franklin)
Subject: Re: NT driver writer new to Linux kernel/drivers
Date: 5 Jun 99 08:28:49 GMT
[EMAIL PROTECTED] (Timothy J. Lee) writes:
>[EMAIL PROTECTED] (Holden McGroin) writes:
>|
>|Hi all. I'm an experienced NT kernel developer. I'm looking to get
>|back to my hacking roots and have some fun in the Linux world. Can
>|someone point me at some good documentation (besides the source) on
>|the linux kernel and development for it? You know, block diagrams,
>|sample driver, API suite yad yada.
>Rubini's Linux Device Drivers book should be helpful for writing
>drivers for 2.0.* and early 2.1.* kernels. Publisher is O'Reilly.
I would second that. This book rocks.
- Daniel
--
******************************************************************************
* Daniel Franklin - Postgraduate student in Electrical Engineering
* [EMAIL PROTECTED]
******************************************************************************
------------------------------
From: y chen <[EMAIL PROTECTED]>
Subject: Re: Framebuffer device?
Date: Fri, 04 Jun 1999 22:46:10 -0400
A very good url
http://www.uno.edu/~adamico/banshee
I found this by enter "savage3d linux" in yahoo.
Craig Graham wrote:
> How do I configure a 2.2.x kernal to enable
> the /dev/fb? framebuffer device?
> I've tried config/rebuild but it still refuses
> to appear...
>
> Craig Graham.
------------------------------
From: y chen <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: HELP, VFS can not boot (03,01)
Date: Fri, 04 Jun 1999 22:40:36 -0400
After building kernel 2.2.9 from 2.1.132, I can not
boot off floopy.
I compiled and made floppy disk by
"make bzdisk"
When i boot, I saw kernel loading...
and some info about CPU and PCI ...
Then i got:
Kernel panic, can not boot (3,1)..
I have no idea what this means.
Help will be highly appreciated!
Thanks ahead.
------------------------------
From: [EMAIL PROTECTED] (Alexander Viro)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: 5 Jun 1999 06:49:27 -0400
In article <[EMAIL PROTECTED]>,
Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
>Alexander Viro ([EMAIL PROTECTED]) wrote:
>: In article <[EMAIL PROTECTED]>,
>: Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
>: >I've written small amts of c++, what more
>: >familiarity with objects do you want?
>
>: ROFLMAO
>: Man, that's it. Exactly. Thanks for .sig, BTW.
>
>excuse me. how much of linux is written in c++?
non-fscking-sequitur. Has *no* relevance to your drivel on OOP
as a technics for writing anything. The level of familiarity stated above
("written small amounts of c++") is way insufficient for any meaningful
work on design.
>c++ is not my favorite language, so I don't use it.
If all you know about OO is C++ model, AND you don't use C++ - your claims
regarding OO design of anything are not based on experience, right?
Your similarity with one particular kind of vermin I know well enough
(Soviet philosophers, aka fecies from Marxism-Leninism department, aka
"Partija byla, est' i budet est'") had pissed me off enough to run a
little DN search. Impressive. Especially your drivel on sci.physics.
You mostly feed on the very natural reaction - people tend to assume
that one they are talking with knows what he's talking about. Fine,
but to imitate a clue that way one shouldn't overdo on this scum.
Main rule: *never* *ever* *try* *to* *pull* *the* *trick* *in* *a*
*place* *where* *you* *already* *did* *it*.
Let me summarize your, erm, achievements here: you demonstrated that you
don't know what you are talking about; you tried to play populism in
extremely clumsy way (come on, all it takes is a couple of hours on DejaNews
to get some backgound on the newsgroup); you demonstrated an utter lack of
clue on the culture ("all programmers are talking in buzzwords", yeah,
right...); you never backed a single claim you've made - buzzwords do *NOT*
count. The bottom line: you've failed. Miserably. I could argue your case
better than you did. You failed even to get a hint of background on the
words you've used - unpardonable sin for a troll. Free advice: if you want
to look how to do those things right - look for alt.syntax.tactical FAQ.
Overall: F-. Any fool can get replies. BFD. Sustaining the thread takes more
than that and it looks like it takes more than your abilities. Piss-poor.
Trolling for discussion is a pretty subtle art. If you want to do it -
at least *learn* how to do it.
Folks, I really recommend you to look at the postings of our wank^Wfriend
and estimate the general level. You know where to find DN.
As far as I'm concerned the fun is over. All too obvious and boring.
Pathetic. Sheesh... Trolls those days...
--
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid. Get yourself a better computer" - Dilbert.
------------------------------
From: [EMAIL PROTECTED] (Christopher B. Browne)
Crossposted-To: alt.os.linux,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 Jun 1999 13:45:07 GMT
On Fri, 4 Jun 1999 01:22:06 GMT, Vladimir Z. Nuri <[EMAIL PROTECTED]>
posted:
>C++ introduced the idea of objects.
I think that from this one line we can establish that you "lack clue"
about objects.
Bjarne Stroustrup (who, to those unaware, created C++) answers the
question "C++ got its Object-Oriented concepts from Smalltalk?" thus;
"No. C++ got the key notions of classes, derived classes, virtual
functions from Simula just like Smalltalk did."
Furthermore, those that wish to be considered literate about C++
and object orientedness should be aware that when asked if C++ is an
Object-Oriented language, its designer does not agree:
"C++ is a multi-paradigm language that supports Object-Oriented and
other useful styles of programming."
This answer also elaborates on the nature of object orientedness; it is
a *style,* as opposed to something more fundamental. That does not
imply that OO is unuseful; only that it is not the "silver bullet"
that many wish it were.
--
Those who do not understand Unix are condemned to reinvent it, poorly.
-- Henry Spencer <http://www.hex.net/~cbbrowne/lsf.html>
[EMAIL PROTECTED] - "What have you contributed to free software today?..."
------------------------------
From: [EMAIL PROTECTED] (Anthony Ord)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: Sat, 05 Jun 1999 14:35:10 GMT
On Fri, 04 Jun 1999 22:29:29 GMT, [EMAIL PROTECTED] (Jimen
Ching) wrote:
>Vladimir Z. Nuri ([EMAIL PROTECTED]) wrote:
>>but there is another way to do it. I've read the infamous
>>"cathedral vs. bazaar" model.. a rambling document if I ever
>>read one.. I was a bit shocked/annoyed with how meandering that thing
>>is and the fame it has received.
>
>Finally, someone said it. ;-) The fame that this document received is
>an indication of the type of audience you need to expect. Any well trained
>software engineer would see through that document after the first few
>paragraphs. I recently attended a talk by ESR. I was amazed by how
>willing people were in accepting the ideas in that paper. And these people
>are not just students, they are graduate students and professors in a
>university. They took everything ESR said to heart, without question.
>And when I raised a question, the audiance was the first to defend the
>ideas. This is not the type of environment I would expect from academia.
Of course it is. Academia is riddled with politics and
religion. For example look at the idea of Cold Dark Matter.
Basically it is a one-trick pony. CDM has mass and nothing
else, just in the way that Ether (remember that?) would
transmit photons and yet have no mass or anything else. It
is needed because otherwise we wouldn't exist (remind anyone
of The Creator?) and despite looking for it, no one can find
it (remind anyone of The Creator?).
Is that not religion? Or blind advocacy, whatever you want
to call it.
BTW, I've got news - CDM doesn't exist.
>>a design document IS INDEED A START TO PROJECT X. and perhaps a far more
>>sensible one than writing code.
>
>Absolutely. Unfortunately, the audience you are speaking to will not
>likely to understand this. Because they are mostly uneducated juveniles,
>i.e. teenage, software developer wanabee. I'll admit there are a few
>gifted teenagers who can write better software then some of the 40 year
>olds I know. But these teenagers are rear.
Or rare even... ;-) Or are you saying they are anal and I've
misinterpreted it?
>The average usenet'er have
>no professional training in software development.
Of course not, but the average usenetter isn't on the
computer user groups. The average computer user groups
(excluding the computer games groups) has a very large
proportion of people who have professional training in
software development.
>My suggestion to you is to grow thicker skin. There will always be immature
>people who say "show me the code". A good software engineer will see the
>flaw in this line of thinking and ignore it. Anyone who has tried writing
>complex software know that a concept must be developed first, before the
>code can be developed. You can't write something if you don't have the
>ideas in your head. And if you don't write it down, you'll forget it.
But if I go back to my favourite saying - "A demonstration
is worth a thousand Usenet messages.". That always should be
bourn in mind. There is no need to write the whole thing,
just a few demos embedded in another OS (Linux is ideal
because you have access to the source) helps prove concepts.
>Enough with this rambling. Do what I do and ignore these people. To show
>you that there are people who are willing to discuss ideas, here are my
>thoughts on your essay.
>
>1. The first mistake you made was to umbrella the OS concept. An operating
>system is more than just what the user sees. The idea that an OS should be
>designed from the ground up to cater to the end-user is looking at only half
>the picture.
Agreed. Making the hardware dance will be the priority in
the early days. This is why it is always best to build on
another OS (as Linus did with Minix).
>Everything you suggested _can_ be done with Linux, the kernel.
>All OS's need a kernel. Most of the ideas you suggested are components or
>modules of the OS that sit on top of the kernel. Take the Hurd for instance.
>There is nothing to stop you from implementing _every_ feature you listed
>as a Hurd server. That is the power of the Hurd design.
>
>2. As someone else already pointed out, your essay may be ahead of its
>time. I.e. the hardware/software tools and techniques may not be available
>to implement every single feature of your OS. I.e. what language are you
>going to use? Some languages already have the needed tools you need. While
>others do not. If the wrong language is chosen, your project will fail
>because of the complexity that results from the choice of language. An
>example is the template library, STL. Stepanov initially used another
>language to implement the STL before he switched to C++. In his paper,
>he specifically said C++ allowed him to do the things he wanted to do with
>the STL containers. The original language simply was not powerful enough.
>Thinking about theory is fine, don't stop. But in that process, you must
>also keep an eye on the existing technology. Unless it is part of the
>design to implement needed tools, you don't want to over engineer.
>
>3. The third problem I saw can be summed up with, "theory without data".
>This is the same problem ESR had with all his papers. There are no data
>to support any of the ideas and concepts being discussed. Examples are
>not data.
They are if there are enough of them. I can't prove people
will die, but I have enough examples to put forward a
statistically significant case...
>For instance, do you have proof that the object file system
>design is superior to everything else? What are the alternatives? Did
>you do a study of file system and concluded that an object design is
>better? All I learned from your essay is that you believe that you can
>apply the technique of objects to a file system design. Well, duh. You
>can apply the technique of objects to anything. During the talk ESR gave,
>I asked him if "_open source_ should be applied to embedded software?"
>His answer was, "sure, why not." This is like saying:
>
>Avg Joe: "Fool, can I put a bullet through your head to see what happens?"
>Fool: "Sure, why not."
You can't expect the guy to do a full analysis of it while
on stage. You are always going to get quick answers in such
a situation. Can you think of a reason why it shouldn't be?
>Just because something can be done, doesn't mean it _should_ be done.
>Note, I am not suggesting you _can't_ do it. I'm only pointing out that,
>without data, there is no reason to do it.
>
>You have some good ideas. But the world is full of good ideas...
>
>Good luck.
>
>--jc
Regards
Anthony
--
=========================================
| And when our worlds |
| They fall apart |
| When the walls come tumbling in |
| Though we may deserve it |
| It will be worth it - Depeche Mode |
=========================================
------------------------------
From: [EMAIL PROTECTED] (Christopher B. 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: Sat, 05 Jun 1999 13:55:07 GMT
On 5 Jun 1999 05:19:53 -0400, Alexander Viro <[EMAIL PROTECTED]> posted:
>In article <[EMAIL PROTECTED]>,
>Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
>>I've written small amts of c++, what more
>>familiarity with objects do you want?
>
>ROFLMAO
>Man, that's it. Exactly. Thanks for .sig, BTW.
Indeed. I too have "written small amounts of C++," and make no claim
that it has anything to do with familiarity with objects.
That will definitely get kept as a .signature
--
"I've written small amts of C++, what more familiarity with objects do
you want?" -- Vladimir Z Nuri <[EMAIL PROTECTED]>
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/corba.html>
------------------------------
From: Carl Anderson <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.admin,comp.os.linux.development,comp.os.linux.hardware,comp.os.linux.networking,linux.redhat.development,linux.redhat.misc
Subject: Re: HP JetDirects (using DLC) on Linux???
Date: Sat, 05 Jun 1999 04:44:36 -0400
> >
> > Does ANYONE know of a way to communicate with the old JetDirect boxes that
> > do not support TCP/IP? They support IPX/SPX and DLC (not LPD).
> >
look at the perl scripts in HOWTO-Printing for older jetdirect cards
Older cards (rev A02.XX) at least won't allow you to bind to port 9100
and use
an active IPX connection at the same time. So turn off IPX support.
Carl Anderson
[EMAIL PROTECTED]
------------------------------
From: David Wragg <[EMAIL PROTECTED]>
Subject: Re: Might Linux SMP write memory out of order?
Date: 04 Jun 1999 23:41:27 +0000
"Stefan Monnier <[EMAIL PROTECTED]>"
<[EMAIL PROTECTED]> writes:
> >>>>> "David" == David Wragg <[EMAIL PROTECTED]> writes:
> > x86 is just as (un)reasonable as Alpha, MIPS, PPC, Sparc, in this
> >regard: If you access multiple shared variables without the
> >protection of synchronisation primitives (mutexes etc.), you need
> >memory barriers to avoid getting bitten.
>
> But the guarantees are not all the same. MIPS has particularly strong
> guarantees, while Alpha has them unusually weak. For example
> (assuming a and b are both set to zero):
>
> store a, 1 read b;
> store b, 1 read a;
>
> if `read b' returns 1, then `read a' is IIRC guaranteed by x86 to
> return 1 as well.
The Intel manuals have always said that for P6 processors reads can be
carried out speculatively and in any order. So I don't think this
guarantee holds.
> > implied by machine code. The way to tell the compiler what you want
> >is using volatile (or function calls). The way to tell the processor
> >what
>
> Hmmm... does ANSI C really enforce such a constraint on function calls
> ? That would make it quite a bit harder for compilers to implement
> inlining correctly.
You are correct, ANSI C doesn't require it. I should have said
"external function calls". C compilers can (and do) rearrange memory
accesses in and around inline function calls. However, lots of
software (e.g. every pthreads implementation I've seen) relies on the
compiler not rearranging memory acesses around calls to functions
unless it can prove that such rearrangements are valid, which it an
only do if it has seen the souce code for the called function. So I
didn't want to imply that volatile is the only way to restrict the
compiler.
David Wragg
------------------------------
From: "Max Reason" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: DLL's in Linux ?
Date: Sat, 5 Jun 1999 06:26:44 -0000
Richard R Urena wrote in message <7isphp$[EMAIL PROTECTED]>...
> So I got bored yesterday and ran ldd on all the files in my system.
> Part of the output reads:
> /usr/bin/basic libm.so.4 (DLL Jump 4.5pl26)
> (BTW, I didn't know I had BASIC on my machine; this is
> really a dream come true for me.)
> But anyway, there were several other programs with
> mention of these "DLL Jump"s
> What exactly does this mean?
This is 80% off topic, but I noticed you like the idea of having
a BASIC for Linux. I thought you might like to know that you
can download a freeware package called XBasic for Linux.
XBasic is an complete development system with integrated
editor, compiler, debugger and interactive GuiDesigner that
lets you layout windows to your liking then convert them to
functions in your program (back & forth in fact). Assuming
you have an ELF Linux, XBasic will probably work for you.
I know somebody recently downloaded and installed XBasic
on their redhat 5.1 system and it worked right away. If you're
interested, you can read about XBasic and download at:
http://www.maxreason.com/software/xbasic/xbasic.html .
XBasic only runs when XWindows is running of course.
Also interesting: You can also download a fully compatible
XBasic for windows at the same web-site. Yes, you heard
right - XBasic programs developed on either implementation
run on the other without any changes - even if they have tons
of graphics and GUI in them. Of course if ya can't stand
Windows, who cares about that?
------------------------------
From: Larry Blanchard <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: 05 Jun 1999 09:38:24 PDT
Reply-To: [EMAIL PROTECTED]
Bill Vermillion wrote:
>
> ________________________________________________
> /BCD-ROCKED |
> |]]] ] ]] |
> | ]]] ] |
> | |
> |111111111111111111111111111111111111111111111111|
> |]222222]2222222222222222222222222222222222222222|
> |3]3333]33333333333333333333333333333333333333333|
> |44]444444]44444444444444444444444444444444444444|
> |55555555]555555555555555555555555555555555555555|
> |66666]666666666666666666666666666666666666666666|
> |777777777777777777777777777777777777777777777777|
> |888888888888888888888888888888888888888888888888|
> |9999]9999999999999999999999999999999999999999999|
> |________________________________________________|
>
> Bill
>
What's really bad is I can still read that :-).
--
Larry Blanchard - Old roses, old motorcycles, and old trains
Homo Sapiens is a goal, not a description.
------------------------------
From: "Max Reason" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: DLL's in Linux ?
Date: Sat, 5 Jun 1999 06:41:24 -0000
Max Reason wrote in message <37594f90@coconut-wireless>...
>Richard R Urena wrote in message <7isphp$[EMAIL PROTECTED]>...
>> So I got bored yesterday and ran ldd on all the files in my system.
>> Part of the output reads:
>> /usr/bin/basic libm.so.4 (DLL Jump 4.5pl26)
>> (BTW, I didn't know I had BASIC on my machine; this is
>> really a dream come true for me.)
>> But anyway, there were several other programs with
>> mention of these "DLL Jump"s
>> What exactly does this mean?
>
> This is 80% off topic, but I noticed you like the idea of having
> a BASIC for Linux. I thought you might like to know that you
> can download a freeware package called XBasic for Linux.
> XBasic is an complete development system with integrated
> editor, compiler, debugger and interactive GuiDesigner that
> lets you layout windows to your liking then convert them to
> functions in your program (back & forth in fact). Assuming
> you have an ELF Linux, XBasic will probably work for you.
> I know somebody recently downloaded and installed XBasic
> on their redhat 5.1 system and it worked right away. If you're
> interested, you can read about XBasic and download at:
> http://www.maxreason.com/software/xbasic/xbasic.html .
> XBasic only runs when XWindows is running of course.
> Also interesting: You can also download a fully compatible
> XBasic for windows at the same web-site. Yes, you heard
> right - XBasic programs developed on either implementation
> run on the other without any changes - even if they have tons
> of graphics and GUI in them. Of course if ya can't stand
> Windows, who cares about that?
I forgot to mention, XBasic is a variant of BASIC, but it's
also a capable language. The entire XBasic development
environment is written in XBasic, and it's function protocol
is C / OS compatible so you can call C libraries and Linux
system calls all you like. XBasic is implemented only with
calls to the OS and xlib, not C libraries.
------------------------------
From: "Stefan Monnier <[EMAIL PROTECTED]>"
<[EMAIL PROTECTED]>
Subject: Re: Might Linux SMP write memory out of order?
Date: 05 Jun 1999 12:41:08 -0400
>> store a, 1 read b;
>> store b, 1 read a;
>> if `read b' returns 1, then `read a' is IIRC guaranteed by x86 to
>> return 1 as well.
> The Intel manuals have always said that for P6 processors reads can be
> carried out speculatively and in any order. So I don't think this
> guarantee holds.
The x86 architecture is defined as guaranteeing `processor consistency' which
does guarantee the above. The P6 can indeed execute reads speculatively and in
any order, but the processor detects if a write happened to that same location
between the speculative load and the retirement, in which case the load is
re-executed. For some reason, they didn't go all the way to provide sequential
consistency, although it's not much more difficult (as SGI showed).
> only do if it has seen the souce code for the called function. So I
> didn't want to imply that volatile is the only way to restrict the
> compiler.
But using function calls is only a hack that relies on knowledge about how the
compiler+linker works. If you switch to another compiler+linker which puts
abstract-syntax-trees in .o and .so files and does the actual (while program)
compile at "link"-time, then you might get screwed.
MIPS compilers can do something like that if requested (except for .so files,
which still allows you to get what you want without necessarily incurring the
cost of a `volatile').
Stefan
------------------------------
** 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
******************************
