Linux-Development-Sys Digest #661, Volume #8 Sat, 21 Apr 01 20:13:16 EDT
Contents:
Re: Can Linux kernel ported on supercomputer (using 16 processor) (Ulrich Weigand)
Re: Need pthread-0.9 for Linux on ARM architecture ("Arthur H. Gold")
Re: A Linux emulator for Linux, does this exist? (Grant Edwards)
Re: A Linux emulator for Linux, does this exist? (SammyTheSnake)
WWW: Linux Developer Network ("Misha")
Re: 82559 Intel Network Chip Driver (Greg Lee)
Re: IO system throughput (Paul Repacholi)
Re: Need pthread-0.9 for Linux on ARM architecture (Nix)
Re: howto properly access serial devices in Perl or C (Nix)
How to get all running process names (Hong Hsu)
How to get a number of processors (Hong Hsu)
Re: ide vs. scsi why so much slower ("Gene Heskett")
Re: How to get a number of processors ("Neil Butterworth")
Re: A Linux emulator for Linux, does this exist? (Jerry Kreps)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Ulrich Weigand)
Subject: Re: Can Linux kernel ported on supercomputer (using 16 processor)
Date: 21 Apr 2001 17:16:26 +0200
Greg Copeland <[EMAIL PROTECTED]> writes:
>I'm really not sure how related the S/390 is to his question. Keep in mind,
>as far as I can remember, Linux running on the 390 looks like a uniprocessor
>machine, rather, it can support say, 50,000+ uniprocessor machines. This,
>I think, is a completely different book than a single machine with 16+
>processors because of the overhead in scheduling, etc.
The S/390 is most certainly a SMP machine, both natively and also the virtual
machines under LPAR, VM, or VIF. The latest hardware models support up to
16 physical processors, and under VM you can assign up to 64 virtual processors
to any single guest. Linux for S/390 of course supports SMP, up to the usual
(Linux) limit of 32 processors on 32-bit archs (64 processors on 64-bit archs).
>Long story short, last I read, it was thought that Linux would scale well
>to 12 or so processors with a steady fall off out to about 16 processors.
The question is, what are you doing? User mode processes should scale
bascially linearly. In kernel mode, it depends on what areas of the kernel
you are exercising. There are still some things where lock contention gets
a problem, but for say the networking layer, the 2.4 kernel should scale
pretty well ...
However, this can really be decided only by running proper benchmarks; and
I unfortunately haven't seen any beyond 8-way. A 16-way Linux running
natively on a S/390 would really make an interesting scalability benchmark
possible ;-)
--
Dr. Ulrich Weigand
[EMAIL PROTECTED]
------------------------------
Date: Fri, 20 Apr 2001 19:00:17 -0500
From: "Arthur H. Gold" <[EMAIL PROTECTED]>
Subject: Re: Need pthread-0.9 for Linux on ARM architecture
Todd Anderson wrote:
>
> Anybody know where I can get a precompiled library or the source
> for libpthread-0.9.so?
>
> thanks,
>
> Todd
Versions of linuxthreads are `sync-ed' to glibc versions.
Go to http://www.gnu.org and follow the links; get the
version that corresponds to the glibc you're running.
HTH,
--ag
--
Artie Gold, Austin, TX (finger the cs.utexas.edu account
for more info)
mailto:[EMAIL PROTECTED] or mailto:[EMAIL PROTECTED]
--
Clone Bernie!
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: A Linux emulator for Linux, does this exist?
Date: Sat, 21 Apr 2001 16:50:07 GMT
In article <[EMAIL PROTECTED]>, Jonadab the Unsightly One wrote:
>> What about hot-swappable programs?
>
>That leads me to a feature I've been thinking about that I'd
>like to have...
>
>At some kind of signal from the user, the kernel writes
>the entire contents of RAM, plus the CPU registers and
>any other relevant data, to a big file on disk, saves
>a pointer to this so that it'll see it at boot time,
>and then just powers down.
>
>Next boot time, it sees that flag file, loads the big
>file off the disk to RAM, restores everything, and
>picks up right where it left off, in the middle of
>whatever it was doing.
How's this different than the existing "suspend to disk" feature?
--
Grant Edwards grante Yow! Of course, you
at UNDERSTAND about the PLAIDS
visi.com in the SPIN CYCLE --
------------------------------
From: [EMAIL PROTECTED] (SammyTheSnake)
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: A Linux emulator for Linux, does this exist?
Date: Sat, 21 Apr 2001 16:15:37 +0100
In article <9bkcra$5u0$[EMAIL PROTECTED]>, Philip Armstrong wrote:
>In article <[EMAIL PROTECTED]>,
>Jonadab the Unsightly One <[EMAIL PROTECTED]> wrote:
>>How much does a 390 cost?
>
>How long is a piece of string ?
>
>:)
>
>I'm told IBM mainframe pricing is of the "turn them upside and shake
>them until all the spare money falls out" variety, but having never
>been in a position to actually want or need one of the beasts, I can't
^^^^
don't trust this man! he's lying! ;)
Cheers & God bless
SammyTheSnake
--
Sam.Penny @ Ntlworld.com | Looking for a computer related
Linux, Hardware & Juggling specialist :-) | job, if you can help, e-mail me :)
Wheels: bike, 'ickle bike, and unicycle. | /o \/
Boxen: K6-266@300, dual Celery500 & Nx486 | \__/\
------------------------------
From: "Misha" <[EMAIL PROTECTED]>
Subject: WWW: Linux Developer Network
Date: Sat, 21 Apr 2001 13:14:33 -0700
This is a multi-part message in MIME format.
=======_NextPart_000_0037_01C0CA65.019DCB80
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi, I'm the editor and founder of the Linux Developer Network web page.
I'm here to announce the web site in the website you find
* Libraries
* Tutorials
* Guides
* Manuals
Of Linux platform libraries like:
* GNOME
* C/C++
* PANGO
* Corba
* ...
The fun part of this web site is that you access the different kind of
libraries like a File Manager for example:
GNOME File Manager
This web site is under GPL and an entire web site is available for =
download
so you can browse it off-line.
You can help the development of the web site, e-mail MiSHA for more
information.
Thanks for the interest.
Hector Rivera Falu (MiSHA)
Htttp://www.lidn.org
=======_NextPart_000_0037_01C0CA65.019DCB80
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 5.50.4207.2601" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>Hi, I'm the editor and founder of the Linux =
Developer=20
Network web page.<BR><BR>I'm here to announce the web site in the =
website you=20
find<BR> * Libraries<BR> *=20
Tutorials<BR> * Guides<BR> *=20
Manuals<BR><BR>Of Linux platform libraries like:<BR> * =
GNOME<BR> * C/C++<BR> *=20
PANGO<BR> * Corba<BR> * =
...<BR><BR>The fun=20
part of this web site is that you access the different kind =
of<BR>libraries like=20
a File Manager for example:<BR> GNOME File =
Manager<BR><BR>This=20
web site is under GPL and an entire web site is available for =
download<BR>so you=20
can browse it off-line.<BR>You can help the development of the web site, =
e-mail=20
MiSHA for more<BR>information.<BR><BR>Thanks for the =
interest.<BR><BR>Hector=20
Rivera Falu (MiSHA)<BR>Htttp://www.lidn.org<BR><BR>
<DIV><FONT size=3D2></FONT> </DIV></BODY></HTML>
=======_NextPart_000_0037_01C0CA65.019DCB80==
------------------------------
From: Greg Lee <[EMAIL PROTECTED]>
Subject: Re: 82559 Intel Network Chip Driver
Date: 21 Apr 2001 17:56:04 GMT
[EMAIL PROTECTED] wrote:
> I'll add a little spice to the broth. I have a situation where eepro100.c
> (the driver for the 82559) prints out a KERN_ALERT of "eepro100:
> wait_for_cmd_done timeout!" and I go thru the source and find that
> wait_for_cmd_timeout is waiting on the port it was called with, which is
> always called with the argument ioaddr + SCBCmd where SMCCmd is 2 which
> maps to the command register in the 82559. There is a status register in the
> 82559 (SBCStatus at 0), but it is *never* checked to see if the command unit
> status is idle (meaning the command is executed). I further see that this
> driver works *most* of the time and only fails when transferring a large
> file across a 100baseT local LAN. This means, I think that the
> non-functional wait_for_cmd_done is timing out (with its hard coded count
> down from 1000) is loosing it with a fast processor and a fast network.
> So..Gentlemen, my question is "Am I diagnosing the problem correctly with
> the eepro100.c driver?".
I have this problem, also. I find that with late-model Linux
and X versions that the problem does not appear so long as I don't start
up X Windows. Since the video and network drivers are assigned the
same interrupt on my system, according to /proc/pci, I think interference
between the drivers might be involved.
--
Greg Lee <[EMAIL PROTECTED]>
------------------------------
Subject: Re: IO system throughput
From: Paul Repacholi <[EMAIL PROTECTED]>
Date: 22 Apr 2001 02:53:48 +0800
[EMAIL PROTECTED] (Alexander Viro) writes:
> In article <[EMAIL PROTECTED]>,
> Paul Repacholi <[EMAIL PROTECTED]> wrote:
> >The problem Linux, and unix in general has is that async IO is at
> >zero level in the user code. Chicken and egg to a large degree...
> fork() is cheap. _If_ you want async IO let a separate process to do
> normal IO to/from shared segment.
The fork is cheap in unix, the rest is the pits...
> And "if" above is a pretty serious one - 99% programmers _can't_
> write safe asynchronous code. And I mean algorithm level - like "you
> should not let two threads of execution add elements into the same
> double-linked list unless you protect the critical area". If that
> requires considerable amount of thinking, poking around with
> debugger, etc. - sorry, but it means that one should stay the fsck
> out of any asynchronous code. And for absolute majority of
> programmers that's exactly the case. Worse than buffer overflows,
> worse than inability to live without garbage collector - most of
> multithreaded programs are choke-full of races that should have
> shown up on the first code review done by anybody even remotely
> competent. And yes, I'm talking about code that had been reviewed
> and intended for real use. So thank you very much, I'd rather see
> async stuff of any kind used only when it's absolutely necessary.
I grew up on it, so it is sort of the 'obvious' way to go! But you are
VERY right. They have no clue, because they have never had a chance to
do it! The course never touch that sort of thing now days. One from years
ago started with an assebly program to real, add and print 2 numbers.
In a few week (about 6 I think) you where running on the bare metal,
and having to do it with several terminals!
> It's bad enough on the kernel side where that stuff _is_ necessary (and
> people still manage to fuck up in the mind-boggling ways - destructors
> called before removing references from shared data structures, etc.).
> In the applications code... <shudder> Been there. seen that, nearly vomited
> my guts out while reviewing the crap.
Ah, met C++ I see :)
> Sigh... Sorry about the rant - just spent an hour dealing with a "Threads!
> Are! Wonderful!" duhveloper. Most of that time went on showing him why
^^^^^^^^^^
> this, this and that is _not_ going to work. And yes, the list included
> manipulation of the shared data structures without any exclusion, use
> of global variables by async code in a way that was completely bogus
> _and_ not needed, memory management stuff done in signal handlers and broken
> beyond belief. deadlocks "dealt with" by sending SIGKILL if potential
> victims made no progress for too long, etc., etc. Finally gave up and
> sent him to hell... If Dave Null feels like continuing attempts to educate
> this idiot - more power to him.
cringe...
A standard I used for RSX was to have the IO call parameter block, the
IO Status Block, and the buffer pointers all contiguous in a standard
layout. The 'AST return value' was set to be the address of the IOSB,
so you had a free poinert to the status when you entered your AST
code, the buffer info next, and if it went foobar, the stuff for a
retry was in a known place. I later extended that to 2 IOPBs so I
could bounce in and out with a very short code path.
You can do the same on VMS. Makes it easy to drive hardware to its
physical limit!
--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
Raw, Cooked or Well-done, it's all half baked.
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: Need pthread-0.9 for Linux on ARM architecture
Date: 21 Apr 2001 20:23:18 +0100
On Fri, 20 Apr 2001, Todd Anderson yowled:
> Anybody know where I can get a precompiled library or the source
> for libpthread-0.9.so?
It is in the GNU C Library's add-on package, at
ftp.gnu.org:/pub/gnu/glibc/glibc-linuxthreads-2.2.2.tar.gz; you need
glibc itself to compile it (you untar the add-on package from the glibc
directory and it slots right in to glibc).
--
`There are other possibilities for stopping people from snoring;
for instance, just killing them.' --- Rainer
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To:
comp.os.linux.hardware,comp.os.linux.development.apps,de.alt.comm.isdn4linux
Subject: Re: howto properly access serial devices in Perl or C
Date: 21 Apr 2001 20:25:26 +0100
On Sat, 21 Apr 2001, Jonadab the Unsightly One gibbered:
> A language easier than Perl?
>
> [Adds Expect to languages-to-investigate list.]
It's dependent upon TCL. Be warned.
( ;) )
--
`There are other possibilities for stopping people from snoring;
for instance, just killing them.' --- Rainer
------------------------------
From: Hong Hsu <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: How to get all running process names
Date: Sat, 21 Apr 2001 16:57:21 -0400
Hi all,
I am newer for Linux and wondering there is API which allows me to get
process related information, such as process names, process ID and
process' memory usage. How ps or top get these information? Is source
code of ps or top available somewhere? If I need to dig into kernel
symbol table via /dev/kmem, where the data structure of kernel symbol
table is defined?
Thank you for your time,
Hong
[EMAIL PROTECTED]
------------------------------
From: Hong Hsu <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: How to get a number of processors
Date: Sat, 21 Apr 2001 17:16:52 -0400
Hi all,
Here is my quick question. My application needs to know how many
processors running in the host machine, is there API which allows me to
get a number of processors?
Thank you for your time,
Hong
[EMAIL PROTECTED]
------------------------------
Date: 21 Apr 2001 17:8:33 -0500
From: "Gene Heskett" <[EMAIL PROTECTED]>
Subject: Re: ide vs. scsi why so much slower
Gene Heskett sends Greetings to bill davidsen;
bd> In article <[EMAIL PROTECTED]>,
bd> Eric Taylor <[EMAIL PROTECTED]> wrote:
bd>| So, I get this:
bd>|
bd>| scsi 33 meg/second
bd>| ide 4.5 meg/second - no dma
bd>| ide 12.7 meg/second - with dma
bd>| Can someone explain why such a difference. The
bd>| ide drives are ata 100 7200 rpm. Not sure about
bd>| the scsi device.
bd> DMA is nice, did you put the drive in ATA/100 mode? You should
bd> download the current hdparm (4.1 I believe) and set the options:
bd> -d1 -u1 -m8 -c3 -W1 -A1
bd> which I got from the IDE God, Mark Lord, and see if that make a
bd> small improvement of about 500%.
I did this on a new 46 gigger running on a PDC20267 card, and the -tT
test went up about half a megabyte, and offers the comment: "hmm,
suspicious results, possibly not enough free memory to test"
Cheers, Gene
--
Gene Heskett, CET, UHK |Amiga A2k Zeus040, Linux @ 500mhz
email gene underscore heskett at iolinc dot net
#Amiga based X10 home automation program EZHome, see at:#
<http://www.thirdwave.net/~jimlucia/amigahomeauto>
This messages reply content, but not any previously quoted material,
is � 2001 by Gene Heskett, all rights reserved. Due to recent
changes in M$ lusers TOS, mail from msn.com, msn.net, microsoft.com,
microsoft.net, hotmail.com, and hotmail.net is auto-deleted, unread.
--
------------------------------
From: "Neil Butterworth" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: How to get a number of processors
Date: Sat, 21 Apr 2001 23:34:40 +0100
"Hong Hsu" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
> Hi all,
>
> Here is my quick question. My application needs to know how many
> processors running in the host machine, is there API which allows me to
> get a number of processors?
The processes on a Linux system appear as directories in the /proc
filesystem. To get a count of the current number of processes I guess you
could count all entries in /proc which begin with a digit. This seems kind
of inefficient though...
NeilB
------------------------------
From: Jerry Kreps <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: A Linux emulator for Linux, does this exist?
Date: Sat, 21 Apr 2001 18:48:07 -0500
Jonadab the Unsightly One wrote:
> [EMAIL PROTECTED] (Mario Klebsch) wrote:
>
> > What about hot-swappable programs?
>
> That leads me to a feature I've been thinking about that I'd
> like to have...
>
> At some kind of signal from the user, the kernel writes
> the entire contents of RAM, plus the CPU registers and
> any other relevant data, to a big file on disk, saves
> a pointer to this so that it'll see it at boot time,
> and then just powers down.
>
> Next boot time, it sees that flag file, loads the big
> file off the disk to RAM, restores everything, and
> picks up right where it left off, in the middle of
> whatever it was doing.
>
> A few things wouldn't work properly when restored,
> most notably apps that rely on the outside world
> (like internet stuff), printing that's in progress,
> and anything that earns its living paying very close
> attention to the time of day. Some kinds of external
> devices might also have issues.
>
> But most things should be just fine. And it would
> be an *extremely* handy feature. You could get to
> the other OS for a while, or turn the computer off
> and unplug it for an electrical storm, or move
> across state lines, or whatever, without having
> to finish up all your open tasks.
It's already been done, probably more than once.
About 20 years ago I encountered a machine from
the Logical Corp. called "David". It was a state
machine with the properties you desired. Skeptical,
I reached down and yanked the power cord out of
the wall before the demonstrator had an opportunity
to block me. When powered back up it returned
*exactly* to where it was and what it was doing.
It also included a natural language processor.
I ended up buying a franchise for a 12 state area
to sell an Apple based peripheral card that
employed an OS called "SAVVY". It was written
in FORTH, a beautiful languge, and featured
an excellent natural english programming language
and database. It's inventor, James Dow III, the
great grandson of Bell Star, the outlaw, didn't file
patents and only told one other person how is
pattern recognition and parsing system worked.
The two of them could not develop a multi-user,
multi-tasking version of the system without help,
so it never progressed beyond a single user system.
When then IBM PC came out they released a
software only version, to get away from the
peripheral card mfg problems, but it's parsing system
wasn't truely AI. It was just a clever hash which
began with a preprocessor that took an english
phrase, say, "List all the customers in Neu York
who have perchased widjets since 12/29/83",
and alphabetized the letters, removed redunant
letters and searched a dictionary with that pattern.
If only one pattern was found it would excute the
matched command. If more than one pattern was
found it would show the commands and ask which
one the operator wanted. I didn't pay much
attention to the natural language processor because
the FORTH based programming language combined
with a dictionay based BTREE database was a very
powerful, bottom up development environment.
I miss it.
JLK
------------------------------
** 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 by posting to the
comp.os.linux.development.system newsgroup.
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
******************************
