Linux-Hardware Digest #75, Volume #10            Tue, 20 Apr 99 20:14:07 EDT

Contents:
  Re: Checking Dual Processor Loads? (David Konerding)
  Re: 3dfx VooDoo3 (Greg Fruth)
  Re: HDD Spindown - For a Year! (Stuart Lynne)
  ISA modem problem (root)
  Re: HDD Spindown - For a Year! (James Youngman)
  Epson Stylus Color 640 Resolution (Wolfgang Adrion)
  Promise Fasttrak IDE Raid ("Ehren Jarosek")
  Re: All the current OSes are idiotic (was Re: Is Windows for idiots?) ("Charles R. 
Lyttle")
  Re: is there a PCI modem that does work? (Mickey Stein)
  Re: Programmers are gods ("Don Dunlap")
  dma_intr: error ([EMAIL PROTECTED])
  Re: Programmers are gods (Dan Winker)
  Help! Running out of time (Garret Lee)
  Re: is there a PCI modem that does work? (Mickey Stein)
  Re: Linux installation on Generic box (Alexander Viro)

----------------------------------------------------------------------------

From: [EMAIL PROTECTED] (David Konerding)
Subject: Re: Checking Dual Processor Loads?
Date: 20 Apr 1999 15:45:56 GMT
Reply-To: [EMAIL PROTECTED]

On Sun, 18 Apr 1999 22:52:34 -0700, Jeff Johnson <[EMAIL PROTECTED]> wrote:
>Dan,
>
>    ...and.. yes the 686 (aka PPro in menuconfig/xconfig) is the right setting
>for Xeon CPUs. You may also want to watch during boot to see if both processors
>are recognized. Make sure that the kernel was compiled after the SMP=1
>statement was uncommented in /usr/src/linux/Makefile.
>
>-Jeff
>
>Dan Poynor wrote:
>
>> I have two Xeon 400Mhz processors in my Linux RH5.2 server and would like
>> to know they are both installed and working properly without shutting down
>> and opening the box.

Kernels in the 2.0 series do not report per-processor timings in top
or xosview.  You need a patch, or preferably, upgrade to the 2.2 series.

I would do:
grep "^processor" /proc/cpuinfo

Dave
================================================================================
Email: [EMAIL PROTECTED]    David Konerding     WWW: http://picasso.ucsf.edu/~dek
================================================================================
Snail: Graduate Group in Biophysics
Medical Sciences 926, Box 0446
University of California
San Francisco, CA 94143

------------------------------

From: [EMAIL PROTECTED] (Greg Fruth)
Subject: Re: 3dfx VooDoo3
Date: 19 Apr 1999 19:51:02 GMT

In article <7f71iu$[EMAIL PROTECTED]>, "G de Keating-Hart" 
<[EMAIL PROTECTED]> writes:
> Does any program actually use the 3Dfx card (even just voodoo2) in linux? 
> 
> Eric Bohn wrote in message <7f6kp9$2u6$[EMAIL PROTECTED]>...
> >In article <[EMAIL PROTECTED]>,
> >  sxltrex <[EMAIL PROTECTED]> wrote:
> >> I just bought a 3dfx VooDoo3 2000 PCI.  Does anyone have any idea if
> >> anyone is working on Linux drivers for this card?  (By the way, it kicks
> >> ass on Half-Life).

Any OpenGL application that links to a Voodoo hardware-enabled OpenGL (i.e.
Xi Graphics OpenGL or Metrolink OpenGL) or OpenGL-like library (i.e. Mesa
with the 3Dfx support turned on) will use the 3D acceleration.  (I don't know
if Xi Graphics or Metrolink actually *do* use the 3D acceleration, but they
at least have officially-licensed OpenGL products for Linux.)

I recently got my Voodoo2 working under Linux, and the frame rate for my
home-brew OpenGL program went up by a factor of 4 vs. software rendering.
I was actually expecting a greater speedup, but I still need to tweak and
optimize a bit.

For info on using the Voodoo[1,2,3,etc.] cards under Linux see:

        http://www.mesa3d.org/about.html        
        http://www-hmw.caribel.pisa.it/fxmesa/index.shtml
        http://glide.xxedgexx.com

Maybe if someone gets Half Life working under Linux I'd be able to play it
without it crashing my machine every 30 minutes!  :-)  Half Life rocks, but
it's extremely unstable.

FYI, see:

        http://www.winehq.com/Apps/query.cgi

Apparently several people have tried running Half Life under WINE, with
discouraging results.  It doesn't surprise me -- the thing barely runs
under DOS/Windows, much less under Linux/WINE!

--
Gregory Fruth ([EMAIL PROTECTED])

------------------------------

Crossposted-To: comp.os.linux.networking,comp.os.linux.portable
Subject: Re: HDD Spindown - For a Year!
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] (Stuart Lynne)
Date: Tue, 20 Apr 1999 22:55:07 GMT

In article <7fg0pb$[EMAIL PROTECTED]>,
David Peavey <[EMAIL PROTECTED]> wrote:
>I have an application whereby I would like to use a Linux machine as a
>network gateway.  This particular function requires a very high Mean time
>between failures (MTBF) - I.E. 10 years without failure.  I would like to
>set it up to powered up, with the necessary things loaded, and then left ..
>possibly forever.  I would like to be able to run the thing for upwards of a
>YEAR or so without needing the HDD.  Basically, the only time that the HDD
>would be required is for boot up when power returns after a power failure.
>I would turn off all CRON functions that access the HDD.  Any error or event
>logging could be buffered locally (for example in a ram drive) and then sent
>to a remote Monitor and Control system periodically (Say once per hour or
>once per day).  As far as I can tell, I don't believe the software apps need
>the HDD once they're loaded.
>
>My questions are:
>1)  Can Linux run without the primary HDD spinning?  (I would imagine so
>since laptops can run Linux - but for how long?)
>2)  If so, what are the implications of the swap file (could I replace it
>with a ram disk?), and
>3)  how could the the proc file system be handled (another ram disk?)
>4)  Is there a s/w application (or LINUX configuration) that spins down the
>HDD on command or after a timeout?

See http://edge.fireplug.net and http://www.linuxrouter.org for two similiar
router setups. Both will boot from various DOS formatted boot media. They setup
a RAM disk for running from. You will need a 486 or faster and 16MB of RAM.

For high MTBF I would suggest getting an 8 or 16MB FLASH IDE card. It looks
like an IDE drive to the system. But has zero moving parts.

-- 
Stuart Lynne <[EMAIL PROTECTED]>      604-461-7532      <http://edge.fireplug.net>
PGP Fingerprint: 28 E2 A0 15 99 62 9A 00  88 EC A3 EE 2D 1C 15 68

------------------------------

From: root <[EMAIL PROTECTED]>
Subject: ISA modem problem
Date: Wed, 21 Apr 1999 00:00:50 +0000

I bought a new IBM Aptiva m/c which inevitably came with a PCI WinModem.

So I bought another one (an ISA MicroComputer Research 'Mr' Modemw)
which has taken a little while to set up. When I set the jumpers to PnP,
Linux identified it using /sbin/isapnpdump and when set up in
/etc/isapnp.conf, the modem was recognised on booting although it was
not addressable as a serial device in any of the 'com' ports
(/dev/ttySX) I had configured it for.

In the end I set the hardware jumpers to force it to /dev/ttyS0 and
although I get an error message on boot, it now works fine. My problem
is now, I cannot share it with Windows (98) as Windows only seems to
want to recognise a PnP modem. So I now have two modems on my machine,
one for Linux and one the supplied PCI modem for Windows. I've only got
three expansion slots available and I want to put a NIC card in which
will leave me with none.

Has anybody come across a similar problem, am I just being stupid...

Any thoughts appreiated. 

-- 
From: [EMAIL PROTECTED]
http://www.ross75.freeserve.co.uk
Using SuSE Linux 6.0 / NS Mozilla

------------------------------

From: James Youngman <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.portable
Subject: Re: HDD Spindown - For a Year!
Date: 20 Apr 1999 20:14:03 +0100

"David Peavey" <[EMAIL PROTECTED]> writes:

> 1)  Can Linux run without the primary HDD spinning?  (I would imagine so
> since laptops can run Linux - but for how long?)

man hdpam

> 2)  If so, what are the implications of the swap file (could I replace it
> with a ram disk?), and

You don't need a swap file if you have enough RAM.

> 3)  how could the the proc file system be handled (another ram
> disk?)

It's virtual anyway.   It's not attached to any block device.  See
/etc/fstab.


> 4)  Is there a s/w application (or LINUX configuration) that spins down the
> HDD on command or after a timeout?

See (1).

-- 
ACTUALLY reachable as @free-lunch.demon.(whitehouse)co.uk:james+usenet

------------------------------

From: Wolfgang Adrion <[EMAIL PROTECTED]>
Subject: Epson Stylus Color 640 Resolution
Date: Mon, 19 Apr 1999 23:26:44 +0200

Hi!

When configuring my printer with Yast only a resolution
of 360x360 is possible. Instead 720x720 should be possible
in theory. But then the printer is producing garbage ...:-((
Who can help me?


----------------
 Wolfgang Adrion 
mailto:[EMAIL PROTECTED]



------------------------------

From: "Ehren Jarosek" <[EMAIL PROTECTED]>
Subject: Promise Fasttrak IDE Raid
Date: Tue, 20 Apr 1999 14:05:57 -0500

Anyone have any information on how to get one of these to run under Linux?

Ehren



------------------------------

From: "Charles R. Lyttle" <[EMAIL PROTECTED]>
Crossposted-To: 
comp.lang.java.advocacy,comp.os.linux.advocacy,comp.os.ms-windows.advocacy
Subject: Re: All the current OSes are idiotic (was Re: Is Windows for idiots?)
Date: Tue, 20 Apr 1999 23:31:04 GMT

Martin Ozolins wrote:
> 
> Charles R. Lyttle wrote in message <[EMAIL PROTECTED]>...
> >Olaf Appelt wrote:
> >>
> >> >I have the Dogs/dogs problem on one of my jobs. One folder has about 40
> >> >files that begin with the letters "pro" and several that begin
> >> >"project". One of these I use much more than any of the others, so I
> >> >would like to name it "PROJECT" so it will stand out in windows explorer
> >> >and be easy to see. I don't care whether Windows uses the UC or not, I
> >> >want it. But Bill Gates insists that it can only be displayed as
> >> >"Project" along with all the other "Pro*" files and I have to read down
> >> >the list till I find it.
> >>
> >> That is configurable. Explorer->View->Folder Options->View->Allow all
> >> uppercase names
> >>
> >> Olaf
> >Found out why no one knew about this option : it isn't there. There is
> >no Folder Options. What OS were you running when you found that
> >configuraation?
> >
> That is an active desktop extension.  IE4 or above.
> >--
> >Russ Lyttle, PE
> ><http://www.flash.net/~lyttlec>
> >Thank you Melissa!
> >Not Powered by ActiveX
IE4 has been banned from work (for many reasons) so it isn't there. I
just installed IE4 on an older machine as part of VB6, and it didn't
show up. Possibly because Active Desktop didn't either. People at work
who have been gettin 98 and IE4 have been ending up with lots of
different results. My install came out to be only a little over 300MB,
but more typical installs seem to run 600-800MB.  So far the only thing
that seems to work everywhere is to enclose the folder or file name in
quotes. That works in all the unixen too, but is considered something to
be avoided.
-- 
Russ Lyttle, PE
<http://www.flash.net/~lyttlec>
Thank you Melissa! 
Not Powered by ActiveX

------------------------------

From: Mickey Stein <[EMAIL PROTECTED]>
Subject: Re: is there a PCI modem that does work?
Date: Mon, 19 Apr 1999 15:10:02 -0700

Hi Wade,

     Multitech makes a REAL PCI modem that most likely works but you should talk
to them to verify it. I used to have the part # laying around but at the moment,
all I can remember is that it ends in ZPX (PCI) so just go to www.multitech.com
and do a search on it and maybe write them an email concerning linux. It's a
little pricy but I'm betting it'll work too.

         Good luck,

                Mick


[EMAIL PROTECTED] wrote:

> Ok, I've read all the posts about internal PCI modems being pieces of crap
> winmodems, etc...the problem I have is that I have a 2U rackmount ATX case
> that only allows me 2 pci cards as internal devices, the ISA slots are not
> usable. We're putting these in at our co-locator.
>
> We don't have room to use an external modem, not even by strapping it on the
> side or something, because we're using every inch of our rack space. An
> external modem is, unfortunately, just not an option.
>
> So, I've heard about (although not directly) people who have gotten an
> internal PCI modem to work with linux. If anyone has done this and can tell
> me the exact brand and model I would greatly appreciate it.
>
> Thanks,
>
> Wade Albright
> Online Partners
> [EMAIL PROTECTED]
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own


------------------------------

From: "Don Dunlap" <[EMAIL PROTECTED]>
Crossposted-To: 
comp.lang.java.advocacy,comp.os.linux.advocacy,comp.os.ms-windows.advocacy,comp.os.linux.setup
Subject: Re: Programmers are gods
Date: Tue, 20 Apr 1999 10:39:53 -0500

I agree wholeheartedly with your method of commenting.  A description above
the code should be enough in MOST cases.  There are exceptions.  Many of the
comments that I see are merely a repeat of the code itself and useless if
previous comments aren't used.  Always write your comments with the thought
in mind that the person reading them is rather stupid and needs very clear
instructions.

Don D.


John Burton <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> root wrote:
> >
> > Leslie Mikesell wrote:
> >
> > > In article <[EMAIL PROTECTED]>,
> > > Chris Costello <[EMAIL PROTECTED]> wrote:
> > > >In article <7ffjuv$k85$[EMAIL PROTECTED]>, Leslie Mikesell wrote:
> > > >> And by the way, comments really have nothing to do with making
> > > >> code understandable and when the code is broken there is no
> > > >> reason to trust the comments.
> > > >
> > > >char achar, bchar, cchar, dchar, echar, fchar, gchar, hchar,
> > > >     ichar, jchar, kchar;
> > > >
> > > >int dsiz = M_SIZE;
> > > >
> > > >int
> > > >func(int a, char b, char *c, char **d)
> > > >{
> > > >       if (a == b == *c == (d[0][0]+d[1][a])) {
> > > >               c = malloc(sizeof(a+b+c+(*d)));
> > > >
> > > >               memccpy(c, *d, kchar, dsiz);
> > > >
> > > >               return strlen(c) + d[0][a] % (a + (b*2));
> > > >       } else
> > > >               memcpy(c, *d, dsiz);
> > > >
> > > >       return *d[0][a % 3];
> > > >}
> > > >
> > > >   This is broken code, but it's hard to figure out the real
> > > >purpose of it.  Wouldn't comments help here?
> > >
> > > Only to the extent that if the code and the comments disagree
> > > you can assume that they are probably both wrong.  Otherwise
> > > if it is the comment that is wrong you will be mislead.  Given
> > > a limited amount of time, I'd rather see the effort going into
> > > making the code understandable.  I do find it helpful to put
> > > the data into structs and comment the struct declarations though.
> >
> > All data is valid in the solution of a problem -- if nothing more than
> > to tell you the original author was a idiot and not to trust anything
> > that they did.  No comments means half the information is missing.
> > Even if the code does something correctly, how do you know what
> > the author intended it to do if they didn't bother to write it down?
> >
>
> Ummm...different school of thought...write the code so the algorithm is
> understandable *without* comments. Use identifiers & function names that
> are concise, yet provide enough information to understand what is going
> on. The purpose and function of the code should be immediately apparent
> to another programmer. Comment only as much as necessary. Since it is
> difficult to grasp what a hunk of code does if it is longer than 1 page,
> try to keep functions short.
>
> Compare coding to writing a report... use whitespace to set off
> logically organized sections of code and use enough verbage to get your
> point across, but not so much that it puts the reader to sleep...
>
> Ideally, you sholdn't need *any* comments in the body of a
> function/procedure. You give a 1 sentence prologue which contains a
> description of the function (if you need more than one sentence, perhaps
> your function is not cohesive enough and needs to be redesigned), a
> description of the interface, and a definition of each of the
> parameters. Do all of this *before* you write the code. Then write the
> code as clearly as possible remembering that in *most* cases, coding
> "tricks" are not necessary because the compilers can figure out its own
> efficiency "tricks", but clear, understandable coding can save the next
> programmer hundreds of hours and thousands of grey or yanked out
> hair...;-)
>
> *ALWAYS* code for the *other* programmer, because someday *you* may be
> the *other* programmer...
>
> John



------------------------------

From: [EMAIL PROTECTED]
Subject: dma_intr: error
Date: Mon, 19 Apr 1999 20:23:07 GMT

Hi, The other night while cron was running the normal house keeping
operations I got the following dma errors:

04:02:14 mypc kernel: hda: dma_intr: status=0x51 { DriveReady SeekComp
lete Error }
04:02:14 mypc kernel: hda: dma_intr: error=0x04 { DriveStatusError }

I was wondering if anyone could tell me how I could find out more info on
these error messages, and if it could be a sign of problems with the hard
disk.  A search of dejanews brought up 5 hits on the subject but they werent
of much use to me.

Thanks in advance

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

------------------------------

From: [EMAIL PROTECTED] (Dan Winker)
Subject: Re: Programmers are gods
Crossposted-To: 
comp.lang.java.advocacy,comp.os.linux.advocacy,comp.os.ms-windows.advocacy,comp.os.linux.setup
Date: Mon, 19 Apr 1999 22:11:05 GMT

This one is in our released code.  The programmer is gone.
Apparently he forgot to go back and fix it:

/*...
*   Algorithms:
*
*       reverse ?? 16 bit CRC algorithm with zz polynomial as defined in
*           ABOOK
...*/

------------------------------

From: Garret Lee <[EMAIL PROTECTED]>
Subject: Help! Running out of time
Date: Mon, 19 Apr 1999 21:30:37 GMT

I'm trying to get Linux into our systems and I have been given a limited
time to show off Linux's benefits. However, I just ran out of disk
space. I have been able to get another disk, now I need the procedure to
add a disk to Linux without disrupting the code developement that has
already taken place. 

The Facts:

I need the original disk and the new disk to complete the job. They are
SCSI disks. The new disk is 9 Gigs. Can I have 9 Gig partition or do I
need to break it up?

I need to preserve all the original data and partitions.

Thanks in advance for the assistance. 

Garret

------------------------------

From: Mickey Stein <[EMAIL PROTECTED]>
Subject: Re: is there a PCI modem that does work?
Date: Mon, 19 Apr 1999 15:14:10 -0700


http://www.multitech.com/products/FamilyHomePages/tn.asp

The above is a better link for the modem I was talking about but their server
seems to be down right now.

      Mick


[EMAIL PROTECTED] wrote:

> Ok, I've read all the posts about internal PCI modems being pieces of crap
> winmodems, etc...the problem I have is that I have a 2U rackmount ATX case
> that only allows me 2 pci cards as internal devices, the ISA slots are not
> usable. We're putting these in at our co-locator.
>
> We don't have room to use an external modem, not even by strapping it on the
> side or something, because we're using every inch of our rack space. An
> external modem is, unfortunately, just not an option.
>
> So, I've heard about (although not directly) people who have gotten an
> internal PCI modem to work with linux. If anyone has done this and can tell
> me the exact brand and model I would greatly appreciate it.
>
> Thanks,
>
> Wade Albright
> Online Partners
> [EMAIL PROTECTED]
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own


------------------------------

From: [EMAIL PROTECTED] (Alexander Viro)
Crossposted-To: 
comp.os.linux.setup,redhat.hardware.arch.intel,alt.os.linux,linux.redhat,alt.linux,aus.computers.linux,linux.redhat.install
Subject: Re: Linux installation on Generic box
Date: 20 Apr 1999 20:05:09 -0400

In article <[EMAIL PROTECTED]>,
Alex Balboa (HPTi|crimmins)  <balboa> crossposted to hell and back:

>I am currently trying to install Redhat 5.2 Linux on the following
>system configuration:
>System board brand: Tyan ATX S16820 Tahoe 2 ATX (Pentium II PCI ATX)
>CPU: Pentium II 266 MHZ (dual) with 512 K of cache)
>Memory: 512 MB EDO (GENERIC)
>Floppy 3.5" 1.44MB FDD
>Hard drive: Seagate Barracuda ST19171W (SCSI:1 drive at 4 GB and 6
>additional ones at 9.1 GB)
>CDROM: Toshiba XM-5701TA (12-Speed Fast SCSI2)
>Tape Drive: HP C5133A
>Network Card: Intel EtherExpress Pro (32Bit PCI 10/100 BT)
>Video Card: Matrox Millenium (8MB)
>Sound Card: Creative Lab SB AWE 64 Gold (ISA 20 Bit 4 MB)
>SCSI Adapter: Adaptec (2 of them) model AHA 2940 U/UW
>Zip drive: IOmega 100MB SCSI
>JAZ drive: IOmega 1GB SCSI
>PCMCIA card reader:  ANTEC DESCARTES 761345-64113
>
>I was wondering if some one has installed Redhat 5.2 Linux on such a
>system and, if so,
>please advise on any special procedures or drivers for any of the above
>components that I would need.
>Thank you in advance (please forward your replies to the newsgroup and
>my email)
>Sincerely,
>Alex Balboa

        As for your question - nothing special needed. Except one thing
(see below).

[snip butt-ugly sig lacking normal delimiter and exceeding McQ 2 times]
[snip HTML copy of the above tossed in for no good reason]

BTW, you've chosen an excellent set of groups to ask your question:
comp.os.linux.help
comp.os.linux.setup
comp.os.linux.hardware
comp.os.linux.development.apps
comp.os.linux.development.system
redhat.hardware.arch.intel
alt.os.linux
alt.uu.comp.os.linux.questions
linux.redhat
comp.os.linux.questions
comp.os.linux.networking
alt.linux
aus.computers.linux
comp.os.linux.advocacy
linux.redhat.install
comp.os.linux.x
comp.os.linux.prog
comp.os.linux.m68k

Impressive. Really. comp.os.linux.development.system is, no doubt, the
best possible place for such question. Well, second to best -
comp.os.linux.m68k is THE place for question about non-m68k hardware.
There are several additional hits (.advocacy, several non-existant groups,
.development.apps, .networking, yodda, yodda), but they are somewhat
less original.

And now for the problem with your setup. Actually I was slightly wrong -
it requires *two* things. One of them being an ability to learn how to
use something (Linux, USENET, whatever) and another... a brain.

[Newsgroups somewhat trimmed]

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

------------------------------


** 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
******************************

Reply via email to