Linux-Development-Sys Digest #454, Volume #6 Fri, 5 Mar 99 10:13:54 EST
Contents:
Re: scsi tape driver I/O error on mag tape (Carlos Vidal)
APM error (yanyue)
Re: Overclocking (was: Re: K6-2 and Linux, Are there any Bug?) (Mike Hom)
Re: NFS Data Corruption between Linux 2.2.x and SunOS 5.5.1 (Michal Szymanski)
Re: Problems compiling linux-2.2.2 (Giandomenico De Tullio)
traffic shaper as module - why only one device ? (Antti Koskimaki)
Re: gprof : I would like to learn how to use it. (Andreas Schwab)
Port redirection in C question ("David Sisk")
Re: Scheduler policies (Andi Kleen)
links.. (Henrik Malmgren)
Re: linux threads and polling (Roope Anttinen)
Re: waiting for milliseconds? (Sami Tikka)
Re: PCI Bus master -- help please (Juan Bautista Romance)
Re: PROOF: Jesus *is* Lord of the Sabbath! (matevz bradac)
Re: X Windows Graphics programing ("Jean-Francois Rompre")
pthreads & resume/suspend (Alexander Sirotkin)
----------------------------------------------------------------------------
From: Carlos Vidal <[EMAIL PROTECTED]>
Subject: Re: scsi tape driver I/O error on mag tape
Date: Fri, 05 Mar 1999 09:01:20 +0000
[EMAIL PROTECTED] wrote:
> Hi there,
>
> I have a Digital TLZ07-DA tape drive attached to a Linux system (RH5.2). The
> scsi tape driver is /lib/modules/2.0.36-0.7/scsi/st.o. Running the following
> test program:
> ...
>
> I got the error:
>
> unable to forward space: -1
> Input/output error
Did you tried with the device /dev/nst0 (the
'non-rewind' device)?
--
Carlos Vidal
[EMAIL PROTECTED]
------------------------------
From: yanyue <[EMAIL PROTECTED]>
Subject: APM error
Date: 4 Mar 1999 14:31:32 GMT
Why I can shutdown all my computer in kernel2.2.2 but it is good in kernel
2.0.36. Please help me ‰
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
Crossposted-To:
alt.os.linux.slackware,comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.setup
From: Mike Hom <[EMAIL PROTECTED]>
Subject: Re: Overclocking (was: Re: K6-2 and Linux, Are there any Bug?)
Reply-To: [EMAIL PROTECTED]
Date: Thu, 4 Mar 1999 19:43:12 GMT
check out a real DOA at www.psychosis.com/doa to see
how to make a dual Celeron go to 500MHz each
for a total of 1GHz.
GBP wrote:
> >
> > the only chips i hear of that can be is celeron but besides that.... the
> > chip is not made to be over clocked..... if it was supposed to be over
> > clocked it would RUN at THAT SPEED not 50-100 MHz slower
>
> Its called multi tier pricing, and this isnt the only industry that does
> it. I means packaging essentially the same product two ways (or more)
> one for poorer people, one for richer. Some people will always want the
> better product and a are willing to pay more, since money is no object.
> So company X can jack the price on widgets up 25% and call them super
> widjets. But other people just can't afford the super widgets, and
> company X wants to sell to them too, even though they cant pay as much.
> So they sell "budget widgets". Both are just widgets made in the same
> plant. They dont want people to buy super widgets to know this, or they
> wont pay extra, so they create the fiction that bugjet widgets are
> somehow inferior. Lots of people will still buy the "inferior" budget
> widgets, beacuse they want widgets and thats what they can afford. They
> wish they could get super widgets but that doesnt stop them... budget
> widgets are better than non at all right???
>
> In this case intel used 18 micron technology to make Celerons .. so they
> can run hotter than the older inferior 25 micron P2's. They are
> essentially dumping the celerons on the market at no profit to hurt AMD
> their only real threat. They intend to make all their money on the
> "high end" PII's even though the P2's are almost exactly the same they
> cost 2-4 times more!! The P2 is the "super widget". They developed the
> 18 micron manufacturing plants to make the upcoming P3 and beyond, but
> since the design wasnt ready yet they knocked off some Celerons while
> they were waiting.
>
> Personally i dont feel comfortable overclocking. But i can say that i
> should be almost impossible to melt a celeron due to the 18 micron
> design, and low offical clock speeds. The fast that Intel has been
> trying to devolope mechanisms to stop overclocking almost proves that
> its possible... For example due to design you cant run a celeron 300A
> at say 366, you have to go all the way up to 1.5x at 450mhz.. which
> actually works for a lot of people i hear! The P2-450 is like $500 a
> celeron 300A at 450 is $95 :) So lets say you burn one, your still $300
> ahead if the second one works, the odds are on the hacker's side here.
------------------------------
From: [EMAIL PROTECTED] (Michal Szymanski)
Subject: Re: NFS Data Corruption between Linux 2.2.x and SunOS 5.5.1
Date: 5 Mar 1999 10:30:29 GMT
On Thu, Feb 18, 1999 at 07:27:35PM +0000, Linus Torvalds wrote:
> In article <[EMAIL PROTECTED]>,
> Oliver Stahlhut <[EMAIL PROTECTED]> wrote:
> >
> >Data written to a SUN Solaris NFS server by a Linux 2.2.1 NFS client can
> >be truncated.
>
> It's a Solaris bug. See the attached information about official Sun bug
> ID numbers etc, and just contact your Sun representative.
Two questions:
1. Why older Linux kernels (2.0.XX) worked fine, if it is just a
Solaris bug? I used to compile big packages on a Linux machine
RH 4.2/2.0.35 on NFS-mounted directory on a Solaris server
and it worked rock-solid. Now with 2.2.1 I can just forget
about doing such things.
2. Any luck in finding the patch for Solaris 2.5.1 SPARC?
One comment:
For years, Linux NFS server was unable to serve Solaris clients properly
(a topic discussed many times on Usenet - any attempt to write
a hundred-MB file ended up in "linux server not responding")
Finally, with 2.2.1 and knfsd it started to work as expected.
And now, it is broken the other way. ;-(((((((((((((((((((((
regards,
Michal.
--
Michal Szymanski ([EMAIL PROTECTED])
Warsaw University Observatory, Warszawa, POLAND
------------------------------
From: Giandomenico De Tullio <[EMAIL PROTECTED]>
Subject: Re: Problems compiling linux-2.2.2
Date: Fri, 05 Mar 1999 10:23:11 +0100
Friedhelm Hinrichs wrote:
>
> Compiling the new kernel 2.2.2 ended with the following messages (I hope
> this does make sense to somebody):
>
> gcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2
> -fomit-frame-pointer -pipe -fno-strength-reduce -m486 -malign-loops=2
> -malign-jumps=2 -malign-functions=2 -DCPU=586 -c -o loopback.o
> loopback.c
> /usr/src/linux/include/net/sock.h: In function `sk_filter':
> /usr/src/linux/include/net/sock.h: In function `sk_filter_release':
> /usr/src/linux/include/net/sock.h:807: warning: implicit declaration
> /usr/src/linux/include/net/sock.h: In function `sk_filter_charge':
> /usr/src/linux/include/net/sock.h:817: dereferencing pointer to
Check
"
Lots of bugfixes. If you use the Socket filtering option
(CONFIG_FILTER), you'll
need this patch.
"
>From www.linuxhq.com
:)
Or... apply patch-2.2.2-ac7 :)
--
Windows98: Plug (lo scanner USB) and Pray (che non crashi il sistema)
Microsoft Outlook : dove vuoi vuoi vuoi vuoi vuoi postare oggi?
------------------------------
From: Antti Koskimaki <[EMAIL PROTECTED]>
Subject: traffic shaper as module - why only one device ?
Date: 05 Mar 1999 12:47:04 +0200
I tested shaper with kernel 2.2.1 (as module), and it seemed to even
work quite well. But. I have a need for more than just one
shaper. A look at source showed there is code to support four devices,
but that's only if I compile shaper in kernel (which I don't want to
just in case it happens to hang - like docs warn me - or something)
What is The Reason shaper is written to support only one device when
compiled as module ? Modifications seemed _too_ easy and straight
forward to me so I can't help thinking there is a good reason why it
hasn't been written so at the fist place.
If there's no any good reason, has anyone allready done the
modifications ? (made patch available ?)
And who is maintaining the (shaper) code nowadays ?
Thanks in advance,
Antti Koskim�ki
--
Antti Koskimaki <[EMAIL PROTECTED]>
replace everything since the word 'for' with 'fi' (for Finland)
------------------------------
From: Andreas Schwab <[EMAIL PROTECTED]>
Subject: Re: gprof : I would like to learn how to use it.
Date: 05 Mar 1999 12:05:27 +0100
Gilles Tabary <[EMAIL PROTECTED]> writes:
|> Hello.
|>
|> gprof is a profiler tool. It gives runtime information on programms : call
|> graph, time used in each functions and so on...
|>
|> I would like to learn how to use it.
|>
|> Is some boody knows an URL or anythig which can help me?
|> I do NOT want any URL pointing to linux/unix manual pages. I need somthing
|> like a guide.
Try the gprof manual (the real one, not the man page). It has a tutorial.
--
Andreas Schwab "And now for something
[EMAIL PROTECTED] completely different"
[EMAIL PROTECTED]
------------------------------
From: "David Sisk" <[EMAIL PROTECTED]>
Subject: Port redirection in C question
Date: Wed, 3 Mar 1999 20:59:56 -0500
Is there a/what is the system call in Linux (from C) to redirect an INET
socket request to a socket server listening on a different port?
For instance, say you have a socket server listening on port 2000 and a
different socket server listening on port 2001. A socket connection request
from a client shows up at port 2000. How can the socket server listening on
port 2000 send the request to the socket server listening on port 2001?
I can think of one way to do this: let the client connect to the server on
port 2000, have the server on port 2000 send it the port number to connect
to, then have the client disconnect from the server on port 2000 and
reconnect to the server on port 2001.
Is there a way to accomplish the same effect without having the network
round trip? (I think there is, I just don't know it!)
Please respond via email as well as posting if you would be so kind.
Thanks and regards,
--
David C. Sisk
The Unofficial ORACLE on NT site
http://www.ipass.net/~davesisk/oont.htm
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: Scheduler policies
Date: 03 Mar 1999 20:54:25 +0100
Jos� Mar�a Fern�ndez Gonz�lez <[EMAIL PROTECTED]> writes:
> Hi newsgroup!!!
>
> I know the Linux scheduler, and I would like to know if there is any project
^^^^^^^^^^^^^^^^^^^^^^^^^^
really?
> which the next target: to have the chance of choosing among some scheduling
> policies when you compile your Linux kernel (e.g. real time, time sharing,
> etc...).
man sched_setscheduler(2)
If you need an additional hart rt scheduler see http://www.rtlinux.org
-Andi
--
This is like TV. I don't like TV.
------------------------------
From: Henrik Malmgren <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.misc,comp.os.linux.networking
Subject: links..
Date: Thu, 04 Mar 1999 22:24:55 +0100
Linux and programming links.....
Go to : http://www.fly.to/thewizz/
NOTE: Don't forget to look under "tons of prgramming links"......it
really are tons of programming links, and it a good programmer who have
done that page.
--
/Henrik!
Unix, msdos and NT. Also known as the good, the bad and the ugly...
-------------------------------------------------------------------
Linux is a fact...Microsoft is a question...And the answer is NO...
-------------------------------------------------------------------
------------------------------
From: Roope Anttinen <[EMAIL PROTECTED]>
Subject: Re: linux threads and polling
Date: 5 Mar 1999 12:09:56 GMT
Reply-To: [EMAIL PROTECTED]
Sean MacLennan <[EMAIL PROTECTED]> wrote:
> I am having difficulty finding documentation on Linux threads.
> I need to guarentee that a thread wakes up every 10-20ms and accesses
> a board. I tried a usleep() in a while loop but once in a while I get
> a 100-150ms delay.
> Any ideas appreciated.
Take a look at "man setitimer". Set the polling thread to wait for SIGALARM
and ingnore the signal in other threads. Use the ITIMER_REAL to decrement in
realtime.
That may be much more precise approach that using usleep(). Too bad that
(according to the manpage) signal queues aren't supported so some timer
signals may get lost :(
Roope
--
MicroSoft? is that some kind of a toilet paper?
PS: Look for address here, not from headers. And remove NOSPAM's
___________________________________________________________________________
[EMAIL PROTECTED] / [EMAIL PROTECTED]
+358 9 812 7567 / +358 500 445 565 / +358 49 445 565
http://myy.helia.fi/~anttiner/index.html
===========================================================================
Helsinki Business Polytechnic - Institute of information technology
------------------------------
From: Sami Tikka <[EMAIL PROTECTED]>
Subject: Re: waiting for milliseconds?
Date: 04 Mar 1999 07:26:18 +0200
Take a look at the real time clock driver.
(kernel source)/Documentation/rtc.txt
--
Sami Tikka, [EMAIL PROTECTED], http://www.iki.fi/sti/
------------------------------
From: Juan Bautista Romance <[EMAIL PROTECTED]>
Subject: Re: PCI Bus master -- help please
Date: 04 Mar 1999 18:52:57 +0100
[EMAIL PROTECTED] (Albrecht Dre�) writes:
> Hi all,
>
> I am currently writing a kernel driver for ver. 2.0.36, but I have some serious
> problems with PCI bus master dma mode (btw, the PCI chip is an AMCC S5933).
>
> In some cases, data transfer via bus master dma simply hangs the complete
> kernel, that is: no reaction to keyboard or any other input, no kernel message
> at all (e. g. oops), ... This behaviour is highly sensitive to the setting of
> the PCI_LATENCY_TIMER register (although I could not find a setting which does
> not produce any crashes). Could anybody explain the meaning of this register's
> value? And what value should be used? Where to find some more info?
>
> Thanks in advance, Albrecht.
--
The PCI_LATENCY_TIMER on S5933 and other PCI chipsets says to PCI controller
how many PCI clocks the S5933 play as PCI Master frozing PCI bus. Therefore
This parameter defines the burst length on PCI. It's strongly depended of
your PCI controller. On typical PC motherboards, this number is optimized for 64.
The value on S5933 is decremented once every 8 PCI clocks after FRAME# is
active.
Our SLink card has a S5933 as well, and this parametter has been sweepped from
6 to 128. We found an strong step response in the Bandwidth performance from 20
ticks to end, from BW1 = 25Mbytes to 40Mbytes/sec ( nominal response ).
However the user/driver applications works perfectly in two cases.
http://ific.uv.es/tical/rod/slink.html
You can find a nice ( S5933 Hardware ) DMA Description at:
http://www.amcc.com/pdfs/BusMast.pdf
But One thing. Whatever value Hangs the Kernel?
If is it. I guess that you have a touble with the addresing registers values.
Remember that you need bus adresses:
tmp->pci_addr = virt_to_bus( ( volatile void* )tmp->k_ptr );
Is you memory a good memory? The allocated block must be removed from the
swapping play and must be contigous as well.
==========================================================================
Juan Bautista Romance Gallego _/ _/_/_/_/ _/ _/_/_/_/
IFIC, Instituto de Fisica Corpuscular _/ _/ _/ _/
Universidad de Valencia _/ _/ _/ _/
Av. Doctor Moliner, 50 _/ _/_/_/ _/ _/
E-46100 Burjassot, VALENCIA _/ _/ _/ _/
SPAIN _/ _/ _/ _/_/_/_/
PHONE 34 6 398 31 92 - FAX 34 6 386 45 83
E-mail: [EMAIL PROTECTED], [EMAIL PROTECTED]
==========================================================================
------------------------------
From: matevz bradac <[EMAIL PROTECTED]>
Crossposted-To:
alt.society.underwear,comp.sys.mac.misc,comp.sys.amiga.hardware,fr.rec.voyages
Subject: Re: PROOF: Jesus *is* Lord of the Sabbath!
Date: Thu, 04 Mar 1999 21:05:17 +0100
Apropriate quote from bad religion's "i want to conquer the world":
"Hey brother christian with your fine and mighty errand,
your actions speak so loud I can't hear a word you're saying...."
rgds,
matevz
David Allen wrote:
> And you would think that this guy is only out to net techno nerds with
> the groups he posts to. Am I the only one that doesn't care that Jesus
> is lord?
>
> {Insert rant that would generally piss this guy off but that I'm too
> tired to type out}
>
> OBVIOUSLY you've mistaken me for someone who vaguely gives a shit about
> your pathetic little life, christian.
------------------------------
Crossposted-To: comp.os.linux.x
From: "Jean-Francois Rompre" <[EMAIL PROTECTED]>
Subject: Re: X Windows Graphics programing
Date: Thu, 4 Mar 1999 17:37:54 GMT
If you are interested in 3D graphics programming, you have at least two
possibilities:
PEX and OpenGL. PEX is an extension to X, and is free, but manuals are
scarce: the
best tutorial and reference books, PEXLib Programming Manual and its
accompanying
reference by Gaskin and Talbot respectively ( O'Reilly) are out-of-print.
OpenGL on the other hand, has a wealth of literature available, but is not
public domain:
the only free OpenGL implementation I know of is by Mesa, which works well
on my
system but is not officially OGL-compliant.
>From what I know so far from both systems, PEX offers a structure hierarchy
to store
components for a better management of complexity, but has several subsets
allowed in
its implementation(s) thereby reducing portability. OpenGL does not have
the high level
tools of PEX ( although there are libraries around to do that ), but
strictly enforces a
minimum set of capabilities and guarantees portability.
BTW, I found a great source to get the out-of-print PEX books: unlike
amazon, you
don't have to give a credit card number without even knowing whether they
have the
book or not:
http://www.abebooks.com/cgi/abe.exe/routera^_pr=oi^_ph=1^_bi=2183410
PEXlib Reference Manual. 3D Programming in X by Talbott,Steve O'Reilly &
Associates,1992 Softcover, VG, shows little wear. (Keywords: computer)
The price of the book is US$ 20.00
Please reference the seller's book # 1543 when ordering.
The seller is A Novel Idea
2805 San Mateo NE , Albuquerque, NM, U.S.A., 87110.
<mailto:[EMAIL PROTECTED]> Ph: 505-883-6217.
Hope this helps.
J.-F.
_______________________________________
ellis <[EMAIL PROTECTED]> wrote in article <7bi0pe$tgl$[EMAIL PROTECTED]>...
> 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: Alexander Sirotkin <[EMAIL PROTECTED]>
Subject: pthreads & resume/suspend
Date: 5 Mar 1999 13:32:20 GMT
Is there any way to implement resume/suspend with pthreads ?
Or is there any thread library (LWP or something) which supports
suspend/resume and can be compiled on Linux ?
I'm not concerned with effeciency or any other similar issues,
I just need to make a project (University) that should
be submitted on ancient SunOS and should use LWP's suspend/resume
and I want to make it on Linux and afterwards port to SunOS's LWP.
------------------------------
** 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
******************************
