Linux-Development-Sys Digest #385, Volume #6 Mon, 8 Feb 99 06:14:07 EST
Contents:
Re: addition of system calls (Peter Samuelson)
Re: Adaptec 2930 SCSI adapter (Wayne Dubois)
Re: Nan and floating point related problems (Peter Samuelson)
Re: Linux apps in win2000 port news! (Matthew Hannigan)
Re: ssa driver for linux 2.2 (Peter Samuelson)
Re: Debugging the Linux Kernel (Peter Samuelson)
Re: Linux apps in win2000 port news! (Matthew Hannigan)
Calls for new security bits in 2.2x (leslie barstow)
Re: use theramin as input device (Michael Hirsch)
Re: Adding Stats to inetd? (Mark Hahn)
re: Notif-0.1 (Chris Willing)
Re: NFS - client cache (Dave Weis)
2.2.1 IPIP tunnel works differently? Documentation available? (Miguel Cruz)
Re: ssa driver for linux 2.2 (Oktay Istanbullu)
Ignorant Socalists (was disheartened gnome developer) ("Bob Taylor")
Re: include directories (Karl Heyes)
Re: Why I'm dumping Linux, going back to Windblows (Leslie Mikesell)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: addition of system calls
Date: 7 Feb 1999 21:24:17 -0600
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Marx Rajangam <[EMAIL PROTECTED]>]
> 1) I wrote a function
>
> asmlinkage int assoc(void)
> {
> return 1;
> }
You mean
asmlinkage int sys_assoc(void) ...
> 2) Then I added an entry for the system call in asm-i386/unistd.h file
> that looks like,
>
> #define __NR_assoc 164
Make sure this is the same header file seen from /usr/include/asm/ (via
symlink or whatever). Debian, for example, uses a fixed set of header
files rather than the symlinks.
I don't know about the rest of the procedure, never having written
system calls....
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Wayne Dubois)
Subject: Re: Adaptec 2930 SCSI adapter
Date: 8 Feb 1999 03:39:57 GMT
Reply-To: [EMAIL PROTECTED]
On 4 Feb 1999 19:46:26 GMT, Glen Morrell <[EMAIL PROTECTED]> wrote:
>Hello,
>
> Does anyone know about the Adaptec 2930 SCSI adapter,
>specifically, how it differs from the 2940 and whether there is a
>driver that works for linux?
>
> Thanks in advance,
>
> Glen Morrell
> [EMAIL PROTECTED]
>
I've been using the 2930 with 2 Seagate 4.2 GB barracuda drives with
the AIC7xxx driver quite sucessfully for some time now. The exact
driver (from my /proc) is:
Adaptec AIC7xxx driver version: 5.0.19/3.2.4
As far as I know, the main difference between it and the 2940 card is
that the 2930 is narrow SCSI vs Ultrawide for the 2940. For me, the
difference was about $200 (or so) ;-)
BTW... Since I bought the 2930 (about 9 mo ago) I haven't seen it in
stores or even mentioned in the newsgroups. Yours is the 1st mention
I've ever seen. I'd begun to wonder if I'd managed to buy the only
one!
- w
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Nan and floating point related problems
Date: 7 Feb 1999 22:06:48 -0600
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Maurizio de Cecco <[EMAIL PROTECTED]>]
> We are experimenting some strange behaviour related to fp under
> RedHat 5.1, gcc 2.8.1
Sounds like a compiler bug. If practical, try installing a recent egcs
and see if the bug is still there. egcs releases are generally
considered considerably stabler than gcc 2.8.x. Also, try compiling
your code with different optimization levels.
> We are working in reproducing the bug in a small example; meanwhile,
> anybody have ideas, or had similar experiences ?
If you can do that, you might also try running `gcc -S foo.c'; then
check the foo.S to see whether the assembly code looks right or wrong.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Matthew Hannigan)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.advocacy
Subject: Re: Linux apps in win2000 port news!
Date: 8 Feb 1999 04:20:31 GMT
In article <799h4s$2tp$[EMAIL PROTECTED]>,
M Sweger <[EMAIL PROTECTED]> wrote:
>
>Although X-Windows is specific to Unix,
Nope. Wasn't even started on Unix AFAIK.
There also implementations on Amiga and OS/2.
All are _implementations_ of X, not "emulations"
of Unix or anything else.
--
-Matt
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: ssa driver for linux 2.2
Date: 7 Feb 1999 22:29:31 -0600
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Andy Key <[EMAIL PROTECTED]>]
> Would anyone out there like to comment on how desirable they find
> putting SSA storage on Linux systems.
I think it would be really cool. Ever since we got our RS/6000 server
(a 5025-F50) I've wished I could convince people to install Linux on it
rather than AIX 4.3.1 which, for all its nice features, can be a PITA.
I am very unlikely ever to win this one, but I know there is *no*
chance as long as Linux doesn't support 45GB of our disks.
> SSAs wiring is considerably smaller and more manageable than SCSI,
> and multi-initiator (ie: HA or cluster) configurations are simpler.
> As SSA disks are wired in loops, there is even protection against
> single link failure here.
I love SSA wiring. What is it, 85 feet between connections? And a
nice small coax wire rather than 25, 50 or 68 pins. But when I tried
to put two hosts (our F50 and an old Micro Channel server) in the same
loop, neither one worked. Guess I needed to read more docs first about
how this is done and what the restrictions are.
> There are other vendors with SSA offerings too.
This is nice to hear. All I've ever heard about SSA was either from
IBM or from a friend who works for Symbios (aka LSI) -- so I thought it
might be some private venture between those two. Is SSA an open
standard? How about the specs for whatever available adapters?
Also, since SSA uses SCSI drives, I assume they communicate with the
machine via some sort of encapsulation protocol. Perhaps that could be
handled like the parallel-port ZIP drive -- as a module that perhaps
lives on top of the SSA hardware driver proper. (Can SSA handle
anything besides SCSI drives?)
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: Debugging the Linux Kernel
Date: 7 Feb 1999 22:36:13 -0600
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[<[EMAIL PROTECTED]>]
> How does one go about debugging the kernel?
Not easy. printk() is your friend. One other widespread technique is
to encapsulate your developing code in a dummy module (even if it will
eventually be core kernel code) so you can insmod and rmmod things at
will, thus drastically cutting down the compile-run cycle time.
Beyond that, they say that if you feed /usr/src/vmlinux and /proc/kcore
into gdb, it can at least examine, if not change, data structures.
I speak from absolutely no experience, however. One of these days I
plan to get involved with actual kernel development but so far I have
not found the time to pour into the learning curve involved.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Matthew Hannigan)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.advocacy
Subject: Re: Linux apps in win2000 port news!
Date: 8 Feb 1999 04:24:33 GMT
In article <[EMAIL PROTECTED]>,
D. J. Birchall <[EMAIL PROTECTED]> wrote:
>On 03 Feb 1999 15:57:53 -0500, Adam P. Jenkins <[EMAIL PROTECTED]> wrote:
>
>>to start a non-gui Unix app you'd still need to actually log into the
>>Unix machine, whereas if you had a Unix emulator you could just run
>>them on your NT machine too.
>
>I have to wonder what sort of performance an NT system emulating
>UNIX would deliver. Has anyone played with this?
>
Well, since it is not an emulator, probably pretty good.
You can ask Interix for a demo copy and try it yourself.
It is an native implementation of the Unix/Posix API and utilities.
Emulator is usualy the wrong word for these sort of things,
and should reserved for hardware emulation.
Wine Is Not an Emulator either.
--
-Matt
------------------------------
From: leslie barstow <[EMAIL PROTECTED]>
Subject: Calls for new security bits in 2.2x
Date: 8 Feb 1999 05:13:48 GMT
In my brief foray into the 2.2.x kernel looking for the SECURELEVEL
replacement, I did not notice any provision in the headers for user-level
calls to set/clear the new permissions bits. Are there any?
--
Les Barstow | Apple ][ Forever!
[EMAIL PROTECTED] |
| "How may I be honest with you today?"
Disclaimer: I didn't do it! | -- Tuvoc
------------------------------
From: Michael Hirsch <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: use theramin as input device
Date: 07 Feb 1999 22:39:02 -0500
Nix <$}xin{[email protected]> writes:
> Remember the days of typing on a ZX81's keyboard? This would be *worse*.
Okay, how about using sign language as the input? I mean, say, signed
exact English, not true American sign language. Sort of the
equivalent of "Dragon Dictate" for the hearing impaired.
--
Michael D. Hirsch Work: (404) 727-7940
Emory University, Atlanta, GA 30322 FAX: (404) 727-5611
email: [EMAIL PROTECTED] http://www.mathcs.emory.edu/~hirsch/
Public key for encrypted mail available upon request (or finger
[EMAIL PROTECTED]).
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Subject: Re: Adding Stats to inetd?
Date: 8 Feb 1999 05:56:12 GMT
> Has anyone done any work at providing inetd with some statistics?
the normal solution is to write an interposer, not unlike tcpwrappers.
------------------------------
From: [EMAIL PROTECTED] (Chris Willing)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.x
Subject: re: Notif-0.1
Date: 2 Feb 1999 09:37:20 GMT
Reply-To: [EMAIL PROTECTED]
You could do an anonymous ftp to virek.vwis.com then
cd pub/jhallen and get the file.
I had the same browser problem.
chris
In article <[EMAIL PROTECTED]>,
Frank Hale <[EMAIL PROTECTED]> writes:
|> Joseph H Allen wrote:
|> >
|> > It verified fine. Several people have already downloaded it. The only
|> > problem I can think of is that the period on the end is not part of the
|> > link (sorry). It's just a period:
|> >
|> > ftp://virek.vwis.com/pub/jhallen/notif-0.1.tar.Z
|> >
|>
|> I tried it again and it still doesn't work for me, It will resolve to
|> the IP but it never connects, I guess it times out. Do you have any
|> mirror sites? I would like to take a look at the toolkit.
|>
|> --
|> From: Frank Hale
|> Email: [EMAIL PROTECTED]
|> ICQ: 7205161
|> Website: http://www.franksstuff.com/
|>
|> "I say line-ux you say lynn-ux,
|> whats the difference? Its still better than windows"
--
Chris Willing Ph: (61-2) 9351 3005
Vislab, A28 Fax: (61-2) 9351 7726
University of Sydney Email: [EMAIL PROTECTED]
NSW 2006 Australia http://www.vislab.usyd.edu.au/staff/chris/
------------------------------
From: Dave Weis <[EMAIL PROTECTED]>
Subject: Re: NFS - client cache
Date: Sun, 7 Feb 1999 21:39:21 -0600
> >>>>> "Leszek" == Leszek Gryz <[EMAIL PROTECTED]> writes:
> > Do you know if linux has implemented NFS client cache?
At last summer's linux expo, a group had done this and presented a talk
about it. search around with altavista and you should be able to find it.
djweis
--
David Weis | 10520 New York Ave, Des Moines, IA 50322
[EMAIL PROTECTED] | Voice 515-278-0133 Ext 231
http://www.plconline.com/ |
------------------------------
From: [EMAIL PROTECTED] (Miguel Cruz)
Crossposted-To: comp.os.linux.networking
Subject: 2.2.1 IPIP tunnel works differently? Documentation available?
Date: 8 Feb 1999 08:05:10 GMT
Is there any documentation of how the IPIP encapsulation (tunl0 device) is
supposed to work in 2.2.1? I can't get it to do anything to save my life. I
would prefer the most technical documentation possible short of having to
reacquaint myself with the source after years of blissful severance. Thanks
muchly.
miguel
------------------------------
From: [EMAIL PROTECTED] (Oktay Istanbullu)
Subject: Re: ssa driver for linux 2.2
Date: 8 Feb 1999 09:56:28 GMT
Reply-To: [EMAIL PROTECTED]
In comp.os.linux.development.system,
Peter Samuelson <[EMAIL PROTECTED]> wrote:
>I think it would be really cool. Ever since we got our RS/6000 server
>(a 5025-F50) I've wished I could convince people to install Linux on it
>rather than AIX 4.3.1 which, for all its nice features, can be a PITA.
>I am very unlikely ever to win this one, but I know there is *no*
>chance as long as Linux doesn't support 45GB of our disks.
You are right, a driver for ssa is lacking and I can't understand
why the ssa driver project a time ago has been canceled. ssa is
very well adapted to ftp, www and other servers which major part
run linux.
>(...) Is SSA an open
>standard? How about the specs for whatever available adapters?
>
I think that ssa is an open standard (as open as scsi) ,
the ANSI specs could easily
be found (cf my last message) and there was a www server about it
but it doesn't work any more (www.ssaia.org)
>Also, since SSA uses SCSI drives, I assume they communicate with the
>machine via some sort of encapsulation protocol. Perhaps that could be
>handled like the parallel-port ZIP drive -- as a module that perhaps
>lives on top of the SSA hardware driver proper. (Can SSA handle
>anything besides SCSI drives?)
>
The protocol is scsi compatible, an ssa protocol maps scsi one: ssa-s2p
protocole maps scsi 2, all the fields are the same but their use
are different. It is the reason why It is easier to take scsi driver
sources and modify them than making an ssa part which will use scsi
part.
If someone is used to hacking and scsi, I need a little help, I can share
my work on the net.
Oktay.
[EMAIL PROTECTED]
------------------------------
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] ("Bob Taylor")
Subject: Ignorant Socalists (was disheartened gnome developer)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.x
Date: Fri, 29 Jan 1999 06:00:52 GMT
In article <[EMAIL PROTECTED]>,
Michael Powe <[EMAIL PROTECTED]> writes:
[snip]
>
> Marcin> Greed is human characteristics, not political
> Marcin> characteristics, stupid.
>
> Fortunately, your lack of manners is not a human characteristic, but
> one of persons not brought up in a proper home. You have my sympathy
> for your deprivation.
You are wrong on *both* counts.
> >> Even the corporations, billionaires, and politicians all
> >> understand that capitalism is destructive; they ceased to
> >> believe in it after the Great Depression.
>
> Marcin> Actually, Great Depression was caused by government, or
> Marcin> Federal Reserve ineptly replacing clearing system worked
> Marcin> out by banks, which successfully defended against runs on
> Marcin> banks on the beginning of 20th century for example.
>
> A shame your expertise does not match your presumption. The 19th
> Century and early 20th Century were just a long cycle of economic
> growth, recession and depression. I don't know to what "runs on
> banks" you are referring, or why, since the major economic problems of
> recessions are not associated with these.
If you don't know what "runs on banks" means then you are also ignorant
concerning the events of and shortly after 1929.
> Marcin> Actually, the body of evidence against socialism is
> Marcin> staggering. There were lots of various flavors of
> Marcin> socialism, all of them failed.
>
> There's no such body of evidence. All the evidence is that
> cooperative projects are far more effective in reaching goals than
> tooth-and-claw competition. Certainly, cooperation is more effective
> than the monopolistic capitalism which you unthinkingly worship.
My yes! You really *do* live in another universe. It is a *fact* that
Socialism has failed miserably. Furthermore Capitalism and a free
market created the most effective, powerful and largest economy the
world has ever seen while Socialism has created misery, poverty and
a dirty environment. This is *fact*. Please take your Socialist
Propaganda back to the sewer from which it came. Any attempt to show
you the facts that are before your very eyes does nothing more than
prove how BLIND you are. You insufferable, arrogant pseudo
intellectual.
Please note Followup-To.
Bob
--
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Subject: Re: include directories
Date: Mon, 08 Feb 1999 10:28:29 -0500
George Telford wrote:
> Sebastian Bo�ung wrote:
> >
> > Hi,
> >
> > I�m kind of new to Linux and to C++ as well. When I tried my first
> > little
> > program:
> >
> > #include <iostream.h>
> > main()
> > {
> > cout << "test";
> > }
> >
> > the compiler complained it could not find iostream.h. I figured that
> > this
> > is because of the file not being in a directory that is searched for
> > header files. I found iostream.h in /usr/include/g++/ and added the
> > directory
> > to the path (using the -I option of gcc). The compiler now complains
> ---snip--- ^
> |
> |
> This is your problem ------------------|
> You're trying to compile a C++ program (iostream.h) with the C compiler
> (stdio.h)
> Try changing the gcc to g++ (and reverting back to #include
> <iostream.h>)
>
Correct, also note that the standard says #include <iostream>, although you
may
find that iostream.h is acceptable on most implementations.
karl.
------------------------------
From: [EMAIL PROTECTED] (Leslie Mikesell)
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Why I'm dumping Linux, going back to Windblows
Date: 29 Jan 1999 00:01:06 -0600
In article <[EMAIL PROTECTED]>,
Nix <$}xin{[email protected]> wrote:
>Jan Andres <[EMAIL PROTECTED]> writes:
>
>> I think this is the way Unix is meant to be used: You don't have
>> multiple programs that partially do the same thing, as this would be
>> like reinventing the wheel.
>
>The exception that proves the rule: awk sed perl
Historically those didn't all exist at the same time. Awk does
things sed doesn't, perl came later and does even more. Sed
is nearly a straight superset of grep, though. If sed had been
done first, operating as grep could have been a command line
option.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
** 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
******************************
