Linux-Hardware Digest #371, Volume #10 Mon, 31 May 99 01:13:31 EDT
Contents:
Re: Dual Celeron's and SMP Performance Problems (CB)
RedHat 6, Vibra 16 ("Dave Zille")
FTape question. (AG)
Re: 286-to run I/O board? (Frank Hahn)
Re: Switch or Relay Output Cards under Linux? (bryan)
Re: Compact Flash with PCMCIA Adapter... (Jon Winters)
2.2.9 problem on Compaq 1850R (Chris Mauritz)
Re: IRQ 61, TLB IPI problem ("Rick Lindsay")
Re: /dev/sndstat file (Mohd H Misnan)
Re: First time Linux user with a simple question ("Danny Johnson")
Re: <$500US Linux Box (Eric Lawson)
FDD Tape Drive--The Saga Continues (Adam J)
Re: AMD 5x86/133Mhz (allen kurt savegnago)
YOURE NOT GOING TO BELIEVE THIS! 5722 ([EMAIL PROTECTED])
Re: File system for NT and Linux ("SuperPete")
Re: Compact Flash with PCMCIA Adapter... (a bolger)
Re: printing problem in Redhat (Tim Moore)
Re: kernel update (Tim Moore)
----------------------------------------------------------------------------
Date: Sun, 30 May 1999 19:32:56 -0700
From: CB <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.hardware.overclocking,csu.unix.linux
Subject: Re: Dual Celeron's and SMP Performance Problems
Totally Lost wrote:
>
> The work done by Tomohiro Kawada and others has been interesting.
<some verbiage and cache scaling diagram eliminated>
> SunOS svr20 5.5.1 Generic sun4m sparc SUNW,SPARCstation-20
>
> This particular machine didn't have a lot of
> memory and started paging above 32M, with 32K L1 and 1024K L2 cache
> sizes. Increasing the working set by a single cache line
> often reduces the performance to the next larger (and slower)
> memory object.
How closely does this system - it's hardware, software, and OS resemble the x86
SMP hardware, and Linux or NT? Is it truly a good analogy? For which OS? both?
> In a real system, there are many sources that scale into the working
> set size(s). The size of the active interrupt service routines and
> OS scheduler are often near the size of L1/L2 caches on many X86
> systems,
Which systems are we talking about here? 486? PII? Celeron? Xeon? PPro? Under
NT? Linux? How much variability is there between these? Aren't the PPro, PII,
Celeron, and Xeon rather different in their cache to CPU relationships? and thus
the observed situation arises that Celerons perform nearly identically to their
cousins - despite their small cache, as unlike all the other chips, the cache is
actually an integrated part of the processor silicon and runs at full speed -
even the Xeon, with full-speed cache, the L2 is a separate piece of silicon
connected via the PCB. The closest is the PPro perhaps... An attribute that
would appear to be a rather severe limitation turns out to be less so than one
might predict or expect.
> thus each interrupt flushes the cache and effectively runs
> at memory or L2 cache speeds. In an SMP system, both CPU's have to
> share the memory performance. If the L1/L2 caches are too small, then
> two processors are worse than one, because they will flush each other
> with distributed interrupts. The cost of flushing the cache by context
> switches and interrupts is two 2X ... memory cycles required to bring
> in the new context/interrupt plus the memory cycles required to replace
> the active process memory on exit. One reason that X86 SMP servers
> may scale, is that network interrupt loads that flush a uniprocessor,
> only impact one processor in a multiprocessor system ... especially
> if the interrupts are bound to a single processor. Ditto for the disk
> subsystems.
So this is an advantage of the x86 architecture as compared to your Sun
example?
<...>
> Understanding the locality and working set sizes, some X86 vendors
> have post processed their kernel builds to improve locality and
> placement. The work done by the SCO Unixware team on the object code
> segments has produced remarkable results. These vendors often have
> large kernels and feature sets, with same/better overall performance.
So it should be possible to optimize Linux for Celeron SMP - correct?
> Celron's with their 128K L2 cache can easily produce worse performance
> under modest to heavy load than a single processor in an SMP system.
Is this conjecture on your part or are you referring to some actual situation
that you are familiar with?
> In many server systems, even dual PII's with 512K L2 caches fail to
> scale or produce negative results under modest to heavy work loads.
What systems running what OS with what apps? A specifics might clarify your
example...
> In some applications, even Xenon's large caches fail to scale or
> produce negative results in comparision to uniprocessors of the same
> architecture. Simply put, it's not uncommon to find applications where
> two processors trashing the caches with SMP interrupts or shared memory
> produce much worse performance than the same system with only a single
> processor enabled.
Or: mis-matching of hardware and software will lead to less than optimal results
- yes this is a reasonable generalization. One shouldn't try to run Word 2K on a
286 for instance...
> Linux is not well multi-threaded, and has a pretty large foot print
> with very little locality in either the OS or it's primary applications.
Is this the most current information?
What about NT?
> I would expect that for many users, dual Celeron's may have a negative
> performance impact much of the time.
Well, this would be rather easy to check, would it not? Why rely on conjecture?
> To observe that Dual Celron's produce an observed improvement under
> certain lightly loading applications may be true. To assert that
> they scale under modest to heavy work loads is probably a huge mistake.
> To expect that they will scale well under Linux with heavy loads is
> almost certainly folly.
To make broad assertions with no examples is a relatively easy nag to ride -
right? Much more productive for someone to do some tests, post the results and
determine the actual limitations and/or advantages of particular configurations.
I am perfectly willing to entertain the possibility that some of your assertions
may be on the money - but you really didn't offer more than some general
theoretical arguments and cautions. Well, I don't have a Linux server to run
through the paces, so I'm not much help either...
CB
>
> John Bass
> UNIX Systems Consultant
>
> --== Sent via Deja.com http://www.deja.com/ ==--
> ---Share what you know. Learn what you don't.---
-----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
http://www.newsfeeds.com The Largest Usenet Servers in the World!
======== Over 73,000 Newsgroups = Including Dedicated Binaries Servers =======
------------------------------
From: "Dave Zille" <[EMAIL PROTECTED]>
Subject: RedHat 6, Vibra 16
Date: Sun, 30 May 1999 19:01:03 -0700
I am trying to get my ISA Vibra 16 card to work properly with RH 6.0.
I have given myself a crash course on the PNP tools included in RH 6.0, and
have made appropriate modifications to isapnp.conf and conf.modules, and the
sound services and card both initilalize OK during bootup (and there are no
other conflict error messages of any sort). However, once I start X, the
following behavior occurs:
- if i execute an MP3 file, for example, it starts to play with its' default
player (some sort of command line utility, I take it, not totally sure), and
it plays OK
- if I try to invoke any sort of sound from within X (whether it's a game,
system sound, or playing an MP3 with X11AMP), it does not work. Just does
nothing -- no errors that I can tell of.
Any help would be greatly appreciated. I am new to Linux, so if I sound a
little naive, please be sympathetic!
Regards,
Dave
------------------------------
From: AG <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: FTape question.
Date: Sun, 30 May 1999 19:39:01 -0400
Hello,
I'm having some difficulties writing to a DC 2000 QIC 40 cartridge.
I use a HP/Colorado T1000 tape back-up.
The tar commands seem to work fine, however, I get
errors
in /var/log/messages which state that the tape isn't
a
"Linux Raw Data Type".. I'm not sure how to format
the
tape so it knows that it isn't a DOS tape..
Each time a perform a back-up, it tells me that it
isn't a
"tar archive" so it skips over the data..
Any clues.. Thanks..
--
Alfred
________________________________________________
FAMU-FSU COE ASME Webmaster
http://www.eng.fsu.edu/~green
[EMAIL PROTECTED]
[EMAIL PROTECTED]
_________________________________________________
------------------------------
From: [EMAIL PROTECTED] (Frank Hahn)
Subject: Re: 286-to run I/O board?
Date: Mon, 31 May 1999 02:04:48 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 30 May 1999 03:35:54 GMT, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>Can I get a linux kernel small enough to use a ComputerBoards
>I/O board in a 286 AST Premium.I have added a 100M hard drive.I
>would like to use Basic to program it.??
>
Linux will only run on a 386 or higher. There is a project
to write a version that will run on a 286 but I don't know
how far along it is. I think the project is called ELKS.
--
Frank Hahn
How much does it cost to entice a dope-smoking UNIX system guru to
Dayton?
-- Brian Boyle, UNIX/WORLD's First Annual Salary Survey
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: Switch or Relay Output Cards under Linux?
Date: Sun, 30 May 1999 23:55:41 GMT
can you get by with X10 remote control units? you can use appliance
relays (16 of them) per house code. if you need more, use another
house code.
Art Botterell <[EMAIL PROTECTED]> wrote:
: Anyone have any leads or experience on switch/relay output boards under
: Linux? (I know they're used a lot in process-control aps, but that's about
: it...)
: I've got to make a recommendation to a client on whether we can do his
: project in Linux... I'd prefer to, but must solve the switch output problem
: first (the switch-closure outputs are needed to trip external alarms when
: certain things happen.)
: - Art Botterell
: Datacast Consultant
: [EMAIL PROTECTED]
--
Bryan
------------------------------
From: Jon Winters <[EMAIL PROTECTED]>
Crossposted-To: rec.photo.digital,alt.periphs.pcmcia
Subject: Re: Compact Flash with PCMCIA Adapter...
Date: Sun, 30 May 1999 21:57:44 -0500
Paul Rubin wrote:
>
> In article <[EMAIL PROTECTED]>,
> Jon Winters <[EMAIL PROTECTED]> wrote:
> >Hello,
> >
> >I've added a PCMCIA bay to my desktop machine running RedHat Linux 6.0.
> >
> >I'm trying to read a Compact Flash card from my digital camera using a
> >PCMCIA adapter.
> >
> >When I insert the card into the bay it beeps twice but I do not know the
> >command to mount the card.
>
> Try booting the machine with the card in the slot. Then look at the
> boot messages (/var/log/messages, I think), where PCMCIA card services
> gets started. You should see a device probe message saying something
> like "/dev/hde, 45 megabytes". Then you should be able to su to root
> and say something like
> mount -t msdos /dev/hde /mnt
> to put the flash card on /mnt (assuming /dev/hde is the right device).
I would like to thank you Paul.
You've solved my problem, got it mounted!
> I don't know how well Card Services handles the case when you put the
> card in while the system is running instead of before you boot. I'm
> afraid this is one area where Linux isn't so strong.
Hmm... still looking into that. Things got funky after ejecting the card
and re-inserting it.
Thanks again for your help!
--
Jon Winters // NBORG
http://www.newbug.org/
------------------------------
From: Chris Mauritz <[EMAIL PROTECTED]>
Subject: 2.2.9 problem on Compaq 1850R
Date: Mon, 31 May 1999 02:52:59 GMT
When I try to compile 2.2.9 on a Redhat 5.2 system on a vanilla
Compaq 1850 R (dual PII-450 with 192mb RAM and 3 4.3gig hot plug
disks), I get:
init/main.o: In function `get_options':
init/main.o(.text+0xa0): undefined reference to `__canary_death_handler'
init/main.o: In function `no_halt':
init/main.o(.text.init+0x34): undefined reference to `__canary_death_handler'
init/main.o: In function `mca_pentium':
init/main.o(.text.init+0x78): undefined reference to `__canary_death_handler'
init/main.o: In function `no_387':
init/main.o(.text.init+0xbc): undefined reference to `__canary_death_handler'
init/main.o: In function `copro_timeout':
init/main.o(.text.init+0x14c): undefined reference to `__canary_death_handler'
init/main.o(.text.init+0x200): more undefined references to `__canary_death_handler'
follow
make: *** [vmlinux] Error 1
179.630u 15.130s 1:43.62 187.9% 0+0k 0+0io 221691pf+0w
I've never seen this error before. Anyone have any insight? The linux
kernel has gotten *really* squirrelly lately. Things that used to be
rock solid are now starting to fall apart (like DAT tape support on
the aic7xxx controllers).
C
--
Christopher Mauritz
[EMAIL PROTECTED]
------------------------------
From: "Rick Lindsay" <lcs@"jump".net>
Crossposted-To: alt.comp.periphs.mainboard.asus,comp.os.linux.development.system
Subject: Re: IRQ 61, TLB IPI problem
Date: Sun, 30 May 1999 20:48:33 -0500 (CDT)
Reply-To: "Lindsay Computer Systems" <lcs@"jump".net>
On Sun, 30 May 1999 21:36:14 GMT, Andrew Daviel wrote:
>I've got a dual Celeron system, ASUS 100MHz m/b. running RedHat 6.0 Linux
>Dual 366MHz 370-pin PGA o/c to bus 85, CPU 450.
>After running hard for 10 hours or so (mprime + setiathome)
>I get :
>Unexpected IRQ vector 61 on CPU #1
>Stuck on TBL IPI wait (CPU #0)
>This with kernel 2.2.5, and also 2.3.2 SMP
>First time with 2.3.2, the system locked.
>With 2.2.5, it hung in there for a couple of minutes with
>lots of these errors, ethernet couldn't interrupt etc. then hung.
>Hung to black screen (was in console mode, running softdog)
>3finger salute no good, reset button no good (maybe didn't hold it
>>4 seconds), front panel power off no good. Pulled the power for
>1 minute - that worked, eventually.
>Ideas ??
Sure, reproduce the problem with a pair of Slot 1 specification processors!
Rick Lindsay, Lindsay Computer Systems, http://www.lc-sys.com
Austin, Texas. 512-719-5257. Asus based systems, Asus Products.
Advanced Systems.
This message is SHAREWARE, please register...
------------------------------
From: [EMAIL PROTECTED] (Mohd H Misnan)
Subject: Re: /dev/sndstat file
Date: 30 May 1999 14:22:38 GMT
On Sat, 29 May 1999 15:52:34 -0700, Grant Lowe wrote:
>Mohd H Misnan wrote:
>> Why do you think it was wrong? /dev/sndstat is used by sound driver and the
>> config that you can see by doing cat /dev/sndstat is produced by sound driver.
>> If you think that your /dev/sndstat is wrong, then you may have misconfigured
>> your sound driver.
>
>I though it was wrong becase when I did a cat /dev/sndstat, I get the message: No
>such file or directory.
>Yet if I look with ls in the directory, it's there.
Does your soundcard works? Anyway, reading the 2.2.X kernel source code,
/dev/sndstat will be gone for good in the next release of stable 2.4.X or 3.0.X.
Anyway, check you sound driver if it is configured correctly.
--
|Mohd Hamid Misnan|[EMAIL PROTECTED] / [EMAIL PROTECTED] |i|
|MacOS 8.6 + |http://www.geocities.com/SiliconValley/3319/ |M|
|AMD K6-2/300 + |We want to take over the world, but we don't have |a|
|Linux 2.2.9 i586 |to do it tomorrow. It's OK by next week - Linus T.|c|
------------------------------
From: "Danny Johnson" <[email protected]>
Crossposted-To: alt.uu.comp.os.linux.questions,comp.os.linux.help
Subject: Re: First time Linux user with a simple question
Date: Mon, 31 May 1999 03:00:34 GMT
Well, I've had a few problems with XWindows, but after a trying a lot of
different settings, I got it to work. What distro are you using?
David Lentz wrote in message <[EMAIL PROTECTED]>...
>Hello all,
> I am a first time Linux user and I must say I like it a lot! I have a
>problem though, my system has a Diamond Stealth GII 460 card and I can't
>quite seem to get the most out of it. The resolution I get is like
>300xsomething. Has anyone got a Diamond Stealth II 460 to work
>correctly?
>
>Thank you,
> David Lentz
> [EMAIL PROTECTED]
------------------------------
From: Eric Lawson <[EMAIL PROTECTED]>
Subject: Re: <$500US Linux Box
Date: Sun, 30 May 1999 21:06:01 -0700
[EMAIL PROTECTED] wrote:
> I'm a poor college student and a linux newbie interested in building a
> machine (w/o monitor) for *less than $500*. I am interested in any
> suggestions anyone would make regarding main boards, processors,
> sound/video/network cards, etc that they have had postive experiences
> with, running linux and XFree86. I am still undecided between a slot 1,
> socket 370, or socket 7 system, and I want to know how safe it is to
> overclock chips like the celeron 300a and the AMD K6-2 266, 300 or 350.
>
> Thanks,
> Omar
>
> --== Sent via Deja.com http://www.deja.com/ ==--
> ---Share what you know. Learn what you don't.---
For a *real cheap good time* check to see if your school has a surplus
property sales outlet. I stopped by the one at ASU in Tempe and walked
out, after paying $50, with a 486DX4100 with an 850Meg hard drive and 24Meg
of RAM. Also included was a 3C509 Ethernet card, keyboard, serial mouse
and 28.8Kbps modem. SVGA Monitors were $50, but I didn't need the monitor.
The computers were stacked on tables and there was a bench with a working
keybaord, monitor, mouse and CPU where stuff from the tables could be
tested before you put down the money.
The stuff is a few years old, and not as fast as the 450MHz PIIs, but you
don't have to worry about winmodems, win_power _cords :-) , and such
stuff.
It works perfectly--it is what I'm using right now (I added a 4.3GB hard
drive). It handles SAMBA and IP masquerading for my wife's computer,
automatic dialing to the ISP, X- windows, and getting email from the
ISP and putting it into two mailboxes (so we don't have to wonder if she/he
got the email on the other computer).
------------------------------
From: [EMAIL PROTECTED] (Adam J)
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: FDD Tape Drive--The Saga Continues
Date: Sun, 30 May 1999 22:31:45 -0500
Again, thanks for the help. For all who haven't followed, I have just
installed a colorado 250MB floppy controller tape drive on a P166 running
redhat 5.0. I have tried to back up to it using tar, and, having gotten
the right switches in the command line, I get this error:
>tar (grandchild): Cannot open archive /dev/ftape: Input/output error
>tar (grandchild): Error is not recoverable: exiting now
>Broken pipe
I thought that it might be bvecause the tape needs formatting, but using
mkfs yields the same error. Advice?
Thanks,
Adam J
------------------------------
Subject: Re: AMD 5x86/133Mhz
From: [EMAIL PROTECTED] (allen kurt savegnago)
Date: Mon, 31 May 1999 03:01:53 GMT
RU�John Hong wrote:
RU�>
RU�> Anyone running with one of these? Anyhow, I o/c this to 160Mhz
RU�> (4x40Mhz) and everything was going just fine until I tried recompiling
RU�> the kernel (ended up with Sig11). Going back to 133Mhz (4x33Mhz)
RU�I've been running this baby at 160MHz for several years now. No problems
RU�whatsoever! Compiling a kernel goes just fine too.
RU�--
RU�Groeten / Regards,
RU�-=-Reinout
RU�[ have a look at my homepage: http://www.cs.vu.nl/~rmvschou/ ]
Yo,
I have an old Shuttle 533-HOT mobo and could never overclock. Might have
been something to do with slow cache chips or the cpu might not have
been able to handle it. Runs O.K. at 133Mhz. I read where some lucky joe's
have been able to push it to 180Mhz with good results. That is a moot
point now though, since Petium prices have gone down.
Hell, just to get the cache chips to upgrade the cache to 1mb, would cost
more than a new FIC motherboard!
Best regards
Kurt Savegnago
[EMAIL PROTECTED]
* 1st 2.00 #6533 * "I drank WHAT?!" -Socrates
--
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.networking
Subject: YOURE NOT GOING TO BELIEVE THIS! 5722
Date: 31 May 1999 02:18:09 GMT
ruwihxzwbjtmpbzfpscbgvdynxcjkdirrtdhjrtnhesiw
------------------------------
From: "SuperPete" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: File system for NT and Linux
Date: Sun, 30 May 1999 23:09:32 -0500
How about HPFS? NT 4 has "undocumented" support for HPFS and Linux supports
it well.
Linux NTFS read-only support works very well, read-write is scary, and it is
definitely there. To enable it you will have to enable "prompt for
experimental" whatever at the beginning and turn on NTFS read only to be
prompted for RW.
-Pete
Wolfgang Ganzert <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>Hello,
>
>I want to use WinNT 4.0 and Linux to share the same data partition on a
>disk. (size 9GB). The problem is that Linux does not support NTFS in the
>"normal" distribution of the Kernel 2.036. On the other side WinNT does
>not recognize ext2fs from Linux.
>For FAT32 a special driver for WinNT is needed. Also the file attributes
>rwx from Linux ar not supported on the FAT32. This is a problem when
>running shell scripts from FAT32 on Linux.
>
>What would be the best solution to share the large data?
>
>Any suggestions?
>
>Thank You!
>
>Sincerely
>
>Wolfgang
>
>
>P.S.:Since sometimes I have problems accessing the news-server I would
>be thankfull for a direct e-mail to me.
>
------------------------------
Date: Mon, 31 May 1999 11:15:07 +0000
From: a bolger <[EMAIL PROTECTED]>
Crossposted-To: rec.photo.digital,alt.periphs.pcmcia
Subject: Re: Compact Flash with PCMCIA Adapter...
it should see your pcmcia reader as a "drive" on your machine
go to file mgr and see if it exists there
andy
Jon Winters wrote:
> Hello,
>
> I've added a PCMCIA bay to my desktop machine running RedHat Linux 6.0.
>
> I'm trying to read a Compact Flash card from my digital camera using a
> PCMCIA adapter.
>
> When I insert the card into the bay it beeps twice but I do not know the
> command to mount the card.
>
> Any help is appriciated.
>
> Thanks in advance.
> --
> Jon Winters // NBORG
> http://www.newbug.org/
------------------------------
Date: Sun, 30 May 1999 21:29:24 -0700
From: Tim Moore <[EMAIL PROTECTED]>
Subject: Re: printing problem in Redhat
Try ghostscript 5.10 or higher.
--
direct replies substitute timothymoore for user name
"Everything is permitted. Nothing is forbidden."
WS Burroughs.
------------------------------
Date: Sun, 30 May 1999 21:32:22 -0700
From: Tim Moore <[EMAIL PROTECTED]>
Subject: Re: kernel update
/usr/src/linux is a link.
1. delete the linux link
2. make a directory 'linux-2.2.7' (or whatever)
3. dump the kernel bits into the directory
4. relink linux to linux-2.2.7.
[tim@asus src]$ pwd
/usr/src
[tim@asus src]$ ls -l
total 5
lrwxrwxrwx 1 root root 12 Jan 8 22:18 linux ->
linux-2.0.36
drwxr-xr-x 13 root root 1024 Nov 12 1998 linux-2.0.34
drwxr-xr-x 13 root root 1024 Nov 12 1998 linux-2.0.35
drwxr-xr-x 17 root root 2048 Apr 26 06:12 linux-2.0.36
drwxr-xr-x 7 root root 1024 Oct 14 1998 redhat
Dr Death wrote:
>
> The documents seem to agree that the lInux kernel is in /usr/src/linux. I
> dont seem to have this dir. I have /usr/src/redhat but i cant find the
> kernel files there. I am trying to update with 2.2. Ive unziped the files to
> a temp dir called linux.
>
> Can anyone tell me what to do next
>
> Thanks
> Ian
>
> [EMAIL PROTECTED]
--
direct replies substitute timothymoore for user name
"Everything is permitted. Nothing is forbidden."
WS Burroughs.
------------------------------
** 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.hardware) 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-Hardware Digest
******************************
