Linux-Development-Sys Digest #449, Volume #6 Wed, 3 Mar 99 14:14:15 EST
Contents:
Re: procpc-1.2.9 'w' command??? (Juergen Heinzl)
bigphysarea patch on 2.2.2 ([EMAIL PROTECTED])
Re: PCI Fast Ethernet DEC 21143 Based Adapter ("B. James Phillippe")
Re: Some notes on glibc-2.1 and egcs-1.1.1 (Juergen Heinzl)
Re: scsiinfo: defective list not more than 512 entries (Andreas Schwab)
Re: waiting for milliseconds? (Andreas Schwab)
Re: Driver Programming ("Zefram Cochrane")
Re: Overclocking (was: Re: K6-2 and Linux, Are there any Bug?) (David)
Re: X Windows Graphics programing (ellis)
Re: Patch for 2.2.* and some Digital Celebris machines (Peter Samuelson)
Re: console and groups (Mike Cole)
gdb problems (Screech)
AudioPCI97 drivers (Michael Gilfix)
Re: booting NT & linux (Robert Schiele)
PCI Fast Ethernet DEC 21143 Based Adapter (Michael Pishos)
Re: How to understand Kernel Source code (Sami Tikka)
lockf() does it work? (Mohammed Abouzour)
UNIX Kernel GURU/MYSTIC Wanted@ - Job Opening!!! ([EMAIL PROTECTED])
Re: waiting for milliseconds? (Mark Hahn)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Juergen Heinzl)
Subject: Re: procpc-1.2.9 'w' command???
Date: Tue, 02 Mar 1999 20:52:36 GMT
In article <[EMAIL PROTECTED]>, Karl Keyte wrote:
>I've just built the package procpc-1.2.9 on a 2.2.1 kernel system with
>glibc-2.0.7 libraries. All seemed to compile fine and most of it runs
>without a problem.
>
>But...the 'w' command returns no users. uptime also gives the usual
[...]
Mail sent ... more in the glibc-FAQ.
Cheers,
Juergen
--
\ Real name : J�rgen Heinzl \ no flames /
\ EMail Private : [EMAIL PROTECTED] \ send money instead /
\ Phone Private : +44 181-332 0750 \ /
------------------------------
From: [EMAIL PROTECTED]
Subject: bigphysarea patch on 2.2.2
Date: Wed, 03 Mar 1999 09:48:35 GMT
I am using bigphysarea patch in my driver source and right now
I am using kernel 2.0.36. How do I port my driver on kernel 2.2.2 ?
Neeraj
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "B. James Phillippe" <[EMAIL PROTECTED]>
Subject: Re: PCI Fast Ethernet DEC 21143 Based Adapter
Date: Tue, 2 Mar 1999 23:14:36 -0800
On Tue, 2 Mar 1999, Michael Pishos wrote:
> does anyone know where i can get the driver for PCI Fast Ethernet DEC
> 21143 Based Adapter. and some sort of direction on how to do it.
Two drivers are included in the stock Linux kernel sources, in
drivers/net/tulip.c and drivers/net/de4x5.c. If you want more information,
you can start at Donald Becker's web page:
http://cesdis.gsfc.nasa.gov/people/becker/whoami.html
cheers,
-bp
--
B. James Phillippe . [EMAIL PROTECTED]
Software Engineer, WGT Inc. . http://www.terran.org/~bryan
------------------------------
From: [EMAIL PROTECTED] (Juergen Heinzl)
Subject: Re: Some notes on glibc-2.1 and egcs-1.1.1
Date: Tue, 02 Mar 1999 20:52:33 GMT
In article <7bh21o$q4v$[EMAIL PROTECTED]>, Jonathan Stott wrote:
>In article <[EMAIL PROTECTED]>,
>Ronald Cole <[EMAIL PROTECTED]> wrote:
>>[EMAIL PROTECTED] (Jonathan Stott) writes:
>>> gcc-2.8.1 (the current release) complies glibc-2.1 just fine, despite
>>> what the INSTALL notes claim [which, I suspect, is the real reason the
>>> release was pulled].
>>
>>Did you just test the intel port? I was under the impression that
>
>Yes, sorry for not mentioning that.
>
>>egcs-1.1.1 was recommended because it built glibc-2.1 reliably on
>>*all* the platforms that glibc-2.1 was ported to.
>
>But that ISN'T what the Install notes say (the files are on computer at
>home or I would try and find the exact wording - from memory, ISTR it
>says "gcc-2.8.1 does not work, you must use egcs-1.1.1" which is
>demonstrably false in at least one [very common] case).
[---]
You should always try to use the latest official release. Older versions
may not have all the features GNU libc requires. The current releases of
egcs (1.0.3 and 1.1.1) should work with the GNU C library (for powerpc see
question 1.5; for ARM see question 1.6).
{ZW} Due to problems with C++ exception handling, you must use EGCS (any
version) to compile version 2.1 of GNU libc. See question 2.8 for details.
[---]
- Some compilers produce buggy code. No compiler gets single precision
complex numbers correct on Alpha. Otherwise, the egcs 1.1 release should be
ok; gcc 2.8.1 might cause some failures; gcc 2.7.2.x is so buggy that
explicit checks have been used so that you can't build with it.
[---]
Fine so far, for C++ gcc-2.8.x is short of a no go, that is known and the
second part says *might* cause some failures, not "is going to", "will" or
"you can bet your compi on that and your soul is doomed to go to hell if
you use gcc". The snippet above is from the FAQ that comes with the glibc
source. Fair enough I dare say.
Sorry to correct that, it's just for there is some rambling going on
about that every now and then already.
>icbm://41.30.14N/81.36.36W/ [EMAIL PROTECTED] [EMAIL PROTECTED]
> Physicist for hire - http://poly.phys.cwru.edu/~jstott/resume.html
icbm; you've got your own Intercontinental Ballistic Missile ? Wow 8)
Cheers,
Juergen
--
\ Real name : J�rgen Heinzl \ no flames /
\ EMail Private : [EMAIL PROTECTED] \ send money instead /
\ Phone Private : +44 181-332 0750 \ /
------------------------------
From: Andreas Schwab <[EMAIL PROTECTED]>
Subject: Re: scsiinfo: defective list not more than 512 entries
Date: 03 Mar 1999 12:10:35 +0100
[EMAIL PROTECTED] (Dieter Rohlfing) writes:
[NEVER send a copy by mail without marking it as such!!!]
|> On 24 Feb 1999 12:29:04 +0100, Andreas Schwab
|> <[EMAIL PROTECTED]> wrote:
|>
|> >[EMAIL PROTECTED] (Dieter Rohlfing) writes:
|> >
|> >|> Allthough the program shows the number of entries, it lists only the
|> >|> first 512 entries of either list.
|> >
|> >There is a limit of 4096 bytes that one can send or receive via the
|> >SCSI_IOCTL_SEND_COMMAND interface.
|>
|> I already found out that fact. The program just reads the first 512
|> entries. Is there any SCSI_IOCTL_SEND_COMMAND like 'read next 512
|> entries' ?
The program doesn't read the first 512 entries, it uses what it gets from
the ioctl, which has a limit on 4096 bytes. There is no command that
tells a SCSI disk to return an arbitrary section from the defect list.
|> >|> Can anybody help?
|>
|> >Rewrite scsiinfo to use the SCSI generic (/dev/sg?) interface.
|>
|> I would like to modify scsiinfo, but I need info how to do it (specially
|> about the SCSI_IOCTL_COMMANDs, do you know any source to study).
You don't want to use SCSI_IOCTL_COMMAND, but rather read/write on
/dev/sg*. See drivers/scsi/scsi_ioctl.c and drivers/scsi/sg.c.
--
Andreas Schwab "And now for something
[EMAIL PROTECTED] completely different"
[EMAIL PROTECTED]
------------------------------
From: Andreas Schwab <[EMAIL PROTECTED]>
Subject: Re: waiting for milliseconds?
Date: 03 Mar 1999 12:13:20 +0100
"Zefram Cochrane" <[EMAIL PROTECTED]> writes:
|> I thought about this sort of thing a few weeks ago: why would anyone want
|> to "wait" for something of the order of "milliseconds". *micro*seconds
|> I could understand, and *seconds* I could understand.
??? 1 milisecond = 1000 microseconds = 0.001 seconds
--
Andreas Schwab "And now for something
[EMAIL PROTECTED] completely different"
[EMAIL PROTECTED]
------------------------------
From: "Zefram Cochrane" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development,comp.os.linux.questions
Subject: Re: Driver Programming
Date: Tue, 2 Mar 1999 18:54:06 -0000
Olof Wolgast wrote in message <[EMAIL PROTECTED]>...
>Where?
Page 11 of the book says:
ftp://ftp.ora.com/pub/examples/linux/drivers
I haven't been there to look, so can't comment on that URL.
Richard [in PE12]
------------------------------
From: David <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux.slackware,comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Overclocking (was: Re: K6-2 and Linux, Are there any Bug?)
Date: Wed, 3 Mar 1999 11:43:08 +0000
i've had a dual pIII 500 machine that was actualy a pIII 500 and a PIII 450
clocked up that ran fine. exactly the same results as a real dual 500.
didn't notice the 450 getting any hotter either.
David
In article <[EMAIL PROTECTED]>, Christian Ravera <[EMAIL PROTECTED]> writes
>
>I'm running my K6-350 at 400 mhz, but I had to raise the voltage to 2.3V
>
>
>Christian Ravera
>[EMAIL PROTECTED]
>
>
>
>Gregory Leblanc wrote:
>
>> Not that this is a Linux question, but did you have to play with the
>> voltage to get it go that high? And what do you have for cooling on
>> that chip?
>> Greg
>>
>> On Mon, 22 Feb 1999 11:47:23 -0500, Andrius Kaikaris
>> <[EMAIL PROTECTED]> wrote:
>>
>> >I'm running K-6 2 300MHZ, overclocked to 375MHZ on a FIC VA-503 Mb, no probs
>at
>> >all, what so ever........... I use 72MB of EDO SIMM RAM
>> >
>> >Andy
>> >
------------------------------
From: [EMAIL PROTECTED] (ellis)
Crossposted-To: comp.os.linux.x
Subject: Re: X Windows Graphics programing
Date: 3 Mar 1999 00:46:38 GMT
In article <7ac0g2$s6e$[EMAIL PROTECTED]>,
Enki <[EMAIL PROTECTED]> wrote:
>is there a book or a group or a site where I can learn about programing
>graphix code in X windows
The X Toolkit Cookbook by Paul E. Kimball:
http://www.amazon.com/exec/obidos/ASIN/0139731326/ricksphotograpag
X Toolkit Intrinsics Reference Manual for Version 11 of the Window
System (Definitive Guides to the X Window System) by Tim O Reilly, Mark
Langley, David Flanagan (Editor):
http://www.amazon.com/exec/obidos/ASIN/1565920074/ricksphotograpag
--
http://www.fnet.net/~ellis/photo/
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: Patch for 2.2.* and some Digital Celebris machines
Date: 2 Mar 1999 19:46:37 -0600
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Christopher Neufeld <[EMAIL PROTECTED]>]
> The following patch fixes the problem by [...]
While I can't comment on the technical issues involved, I *can* say
that when you share a patch with people in Linux development, they will
usually thank you to use unified diffs (`diff -u'), since the format is
(according to me and everyone I've talked to) more readable than the
older context diff (`diff -c') format. GNU diff and GNU patch have had
support for Unified Diffs for quite awhile now so that's no longer an
excuse....
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Mike Cole)
Subject: Re: console and groups
Date: 3 Mar 1999 15:41:35 GMT
In article <[EMAIL PROTECTED]>, Slyglif Cain wrote:
>OK, this has probably been asked before, I can't can't seem to find anything
>on the subject when searching through DejaNews. Is there a way to setup Linux
>so that when a user is logged in via ttyX that they are a member of a
>particular group Y? I would this this would be useful for allowing access to
>certain devices depending on what terminal you are logged into (for example,
>if you are logged into the console, you should have access to /dev/dsp*, if
>you are logged into a particular serial line you might have access to some of
>the /dev/hd* devices).
>
>======================================================================
>// Chris Giard (a.k.a. Slyglif Cain) | Life is like a box of dynamite,
>// EMAIL: [EMAIL PROTECTED] | you never know when it will go.
I use Debian, but I am sure it is atleast similar in most distros, so this
should help. Go into you etc directory, and edit login.defs (as root of
course). Look through the file, and somewhere near the end you should see a
line similar to:
CONSOLE_GROUPS floppy:audio:cdrom
and it has a very good description above it. In case your distro doesn't
here it is:
# List of groups to add to the user's supplementary group set
# when logging in on the console (as determined by the CONSOLE
# setting). Default is none.
#
# Use with caution - it is possible for users to gain permanent
# access to these groups, even when not logged in on the console.
# How to do it is left as an exercise for the reader...
I think this would be about what you need.
------------------------------
From: [EMAIL PROTECTED] (Screech)
Crossposted-To: comp.os.linux.development.apps
Subject: gdb problems
Date: Wed, 03 Mar 1999 16:33:28 GMT
This may sound kind of goofy, but when debugging a C program with gdb,
the debugger tries to find the source code to strol.c, which I don't
have and really don't want. I am running gdb 4.17 with gcc 2.8.1
under kernel 2.2.1 (RedHat 5.2). Am I missing something that I need
to do to disable this behavior, or is that normal? The debuggers on
the Solaris and HP boxes I've used don't do that, and I'm pretty sure
this didn't until recently (maybe related to the kernel upgrade?).
Any takers? Thanks...
Jason Lee | 186,000 miles per second. It's not just a
[EMAIL PROTECTED] | good idea. It's the law.
(Please doctor the addy appropriately)
------------------------------
From: [EMAIL PROTECTED] (Michael Gilfix)
Subject: AudioPCI97 drivers
Date: 3 Mar 1999 04:47:39 GMT
I've been searching for audiopci drivers for what seems like forever. I
found the drivers for the development version of the kernel (v2.0) but I want
to run the 2.0.36 or 2.0.35 kernel. Unfortunately, I can't seem to find a
patch. This seems baffling. If the drivers are available for the v2.0,
shouldn't there be a patch somewhere? Can anybody help? Thanks..
-- Mike
------------------------------
From: Robert Schiele <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.admin.misc,comp.os.ms-windows.nt.misc
Subject: Re: booting NT & linux
Date: Wed, 03 Mar 1999 17:19:44 +0100
Andi Braun wrote:
>
> How do I configure NT's bootmanager to start either NT on disk C: (or
> IDE 0, master) or Linux on D: (scsi, id 0)?
>
> Is there any chance or do I need to use Lilo?
You can use NT's bootmanager, but you also have to create a boot image
with lilo, which is then booted by the NT bootmanager.
>
> Thanks
>
> Andi
>
> PS Please reply via mail
------------------------------
From: Michael Pishos <[EMAIL PROTECTED]>
Subject: PCI Fast Ethernet DEC 21143 Based Adapter
Date: Tue, 02 Mar 1999 22:52:49 -0600
does anyone know where i can get the driver for PCI Fast Ethernet DEC
21143 Based Adapter. and some sort of direction on how to do it.
[EMAIL PROTECTED]
------------------------------
From: Sami Tikka <[EMAIL PROTECTED]>
Subject: Re: How to understand Kernel Source code
Date: 02 Mar 1999 22:08:58 +0200
I think you should also read Alessandro Rubini's "Linux Device
Drivers" from O'Reilly. I think you cannot write a device driver
without gaining some understanding about the workings of the kernel.
--
Sami Tikka, [EMAIL PROTECTED], http://www.iki.fi/sti/
------------------------------
Crossposted-To: comp.os.linux.development.apps
From: [EMAIL PROTECTED] (Mohammed Abouzour)
Subject: lockf() does it work?
Date: Wed, 3 Mar 1999 16:10:28 GMT
I have a question about lockf().
I noticed that if I create a file and lock it using lockf:
fd = open( filename, O_RDWR|O_CREATE, 0777);
rc = lockf( fd, F_TLOCK, 0);
// do not call close(fd) while program is still running
and if the program was killed abnormally (with SIGKILL for instance)
the lock file is still there. So if I try to restart my program again
lockf returns something like (Resource Temporarily unavailable).
Isn't the OS supposed to guarantee that the lock gets deleted when the
program terminates? Is there a work around for this problem?
I also noticed sometimes, if I wait (say 5 minutes) and try to run the
program again, lockf successeds! (other times it doesn't)
Any help would be appreciated.
Thanks
Moe
--
M. Abouzour
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux,comp.os.linux
Subject: UNIX Kernel GURU/MYSTIC Wanted@ - Job Opening!!!
Date: Tue, 02 Mar 1999 20:34:11 -0800
==============E2DC4E1A66880075A0A43ED4
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
MYSTIC: Remote from or beyond human comprehension; baffling human
understanding; unkowable; obscure; mysterious
GURU: A mentor in spiritual and philosophical topics; a spiritual
parent or teacher.
Does that describe you??? If it does, I am looking for you. I need to
hire a UNIX Kernel GURU/MYSTIC as a Chief Engineer. You will chart the
course of product development and work directly with the CTO and R & D.
This is an absolute INCREDIBLE job in Sunnyvale, CA in the heart of the
Silicon Valley... where it ALL happens. Or so we who live and work here
think!!!
Everyone has to live in their own dillusion.
Great start-up company and we will give you lots of money, lots of stock
and demand lots of hours. Need to know anything else??? Okay, so in
all seriousness, I am really looking for the right candidate to fill
this job and offer a fun environment to work in.
H
--
Herb Alston
MIT RECRUITING
[EMAIL PROTECTED]
415.285-8110
415.641-1515Fax
==============E2DC4E1A66880075A0A43ED4
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<b><font face="Arial Black"><font size=+2>MYSTIC</font></font></b>:
Remote from or beyond human comprehension; baffling human understanding;
unkowable; obscure; mysterious
<p><b><font face="Arial Black"><font size=+2>GURU</font></font></b>:
A mentor in spiritual and philosophical topics; a spiritual parent or teacher.
<p>Does that describe you??? If it does, I am looking for you.
I need to hire a UNIX Kernel GURU/MYSTIC as a Chief Engineer. You
will chart the course of product development and work directly with the
CTO and R & D.
<p>This is an absolute INCREDIBLE job in Sunnyvale, CA in the heart of
the Silicon Valley... where it ALL happens. Or so we who live and
work here think!!!
<p>Everyone has to live in their own dillusion.
<p>Great start-up company and we will give you lots of money, lots of stock
and demand lots of hours. Need to know anything else??? Okay,
so in all seriousness, I am really looking for the right candidate to fill
this job and offer a fun environment to work in.
<p>H
<p>--
<br>Herb Alston
<br>MIT RECRUITING
<br>[EMAIL PROTECTED]
<br>415.285-8110
<br>415.641-1515Fax
<br> </html>
==============E2DC4E1A66880075A0A43ED4==
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Subject: Re: waiting for milliseconds?
Date: 3 Mar 1999 18:10:14 GMT
>> I thought about this sort of thing a few weeks ago: why would anyone want
>> to "wait" for something of the order of "milliseconds". *micro*seconds
>> I could understand, and *seconds* I could understand.
that's incredibly unimaginative: I'm running right now, a psych
stimulus that generates 21 Hz stimuli, which means I want to wait
around 48 ms. I actually do it in a busy loop, waiting for the TSC
to reach a particular value, since that gives me around 3 ns resolution.
(actually, that's a lie, since RDTSC takes ~20 clocks on a PII!)
ultimately, it would be nice to see, in 2.3, perhaps, complete
removal of the 10ms-based kernel clock. there are already people
working on this, and it wouldn't effect user-level compatibility,
since select/times/etc could be fiddled to behave the same...
10 ms is way long by today's standards...
------------------------------
** 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
******************************
