Linux-Hardware Digest #309, Volume #9            Sat, 30 Jan 99 13:13:40 EST

Contents:
  Re: SB PCI 64 (Nicolas Kisselhoff)
  Re: Dell Inspiron 7000 (RonaldKriemann)
  Re: Linux becomes slow with more memory (Allen Crider)
  Monopoly Proprietarianism: Divided They Fall (Bob)
  PCMCIA socket for desktop pc (Pietro Marchionni)
  Re: Any New Brand Name PC Hardware Preconfigured with Linux ("mrushton")
  STB velocity 128 card ???? ("mrushton")
  Re: Linux becomes slow with more memory (Erik Rossen)
  Re: Notebook with 3D Rage LT Pro ("Dave Borgelt")
  Re: Clock weirdness..RH 5.2 and Dell Workstation 610 (James Youngman)
  Re: HOW TO TRANSFER XENIX DISK TO LINUX? (James Youngman)
  Re: Same Disk RAID and Mirroring ("Glenn C. Everhart")
  how do i enable serial support in the kernel? (Andrew Lee)
  Re: Linux keyboard? (For emacs use) (Erik Naggum)
  Re: XWindows on a 16MB TNT  :(  no go. ("Per Olsson")

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

From: Nicolas Kisselhoff <[EMAIL PROTECTED]>
Subject: Re: SB PCI 64
Date: Fri, 29 Jan 1999 18:39:45 +0100

Olivier GUILLEMOT wrote:
> 
> Hi,
> 
> I tried to install a Sound Blaster PCI64 but it doesn't seem to work.
> Does any body knows if there something special to do for it ??
> 
> thanks for it.
> 
> Olivier
> 
> --
> <>- olivier guillemot (mailto:[EMAIL PROTECTED]) -<>

Salut Olivier,

LE mieux est que tu récupère les sources du dernier Kernel 2.2.1 qui
viennent juste de sortir (les derniers drivers y sont et il est
globalement plus rapide) mais ce n'est pas indispensable.

Il faut de toute façon recompiler son noyau.

Choisis toutes les options que tu veux (lire kernel-howto en français ou
le guide du routard si tu ne l'as jamais fait).
Pour le son :

Sound card support : module
OSS sound modules : module
100 % sound blaster compatible SB16/32/64, ess jazz16 support : module
Generic OPL2/OPL3 FM synthétiser support : module
FM synthetiser (YM/OPL-) support : module
6850 UART support : module

Additionnal low level sound drivers : yes
AWE32 synth : module


Tu compiles : make dep ; make clean ; make zImage ; make modules ; make
modules_install.

Tu installes le nouveau noyau, tu reboot et tu lances sndconfig
normalement la détection est automatique sinon on te demande les
paramètres manuellement.

Voilà, amuse-toi bien ! (contacte-moi directement si cela ne marche pas)

Souhaiterais-tu participer à une mailing lit en français ?

Amicalement, Nicolas.

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

From: RonaldKriemann<[EMAIL PROTECTED]>
Subject: Re: Dell Inspiron 7000
Date: 29 Jan 1999 18:19:15 +0100

Spiros Boucouris <[EMAIL PROTECTED]> writes:

> Has anybody had any experiences (good or bad) installing redhat linux on
> this machine?  Any feedback / pointer will be greatly appreciated.

Take a look at

  http://www.eecs.umich.edu/~steveh/inspiron/

> Spiros

RGDS
  Ronald

-- 
       Indifference will certainly be the downfall of mankind,
                           but who cares ?
Ronald Kriemann                            Dept. of Computer Science,
[EMAIL PROTECTED]                 University of Kiel, Germany

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

From: Allen Crider <[EMAIL PROTECTED]>
Subject: Re: Linux becomes slow with more memory
Date: Fri, 29 Jan 1999 10:43:47 -0800



Gert Wollny wrote:
> 
> Hi all,
> 
> I have Redhat 5.1 running on following system:
> 
> MB:       HOT-433 486 PCI/ISA 256k Cache
> CPU:      Am5x86-P75
> RAM:      4*16 MB FPM 60ns
> GRAPHICS: MATROX MYSTIC 170
> NETWORK:  PCI NE2000 Clone
> 
> After I have upgraded from 48M RAM to 64M RAM linux becomes very slow,
> after switching to runlevel 3.
> Cachable Area is 64 MB with "Write Through" and 32 MB with "Write Back"
> but changing this gives no difference.
> 
> Currently I start the kernel with the MEM=48M option, and it workes like
> it did, befor I added the last 16MB SIMM.
> I have swapt the modules on the Board, but no changes - so the SIMMs
> seem to be okay.
> 
> The machine is dual boot with Win95, and the latter seems to be as fast
> as befor upgrading.
> 
> Any ideas?

Can you upgrade your cache to 512k?

> 
> Thank you for your time
> 
> Gert
> 
> --
> Remove NOSPAM to reply or mailto:[EMAIL PROTECTED]
> Max-Planck-Institute of Cognitive NeuroSience      http://www.cns.mpg.de
> 
> http://gerti.home.pages.de

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

From: Bob <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Monopoly Proprietarianism: Divided They Fall
Date: Fri, 29 Jan 1999 14:13:40 -0500

Bob wrote:

> Jim Harper wrote:
>
> > I think the bigger problem is that places like GTE, and the like, do not
> > want educated users on their systems.

McNealy Says Don't Buy PCs, Buy
             Services

             "The most ridiculous way to get information
             is through a PC," declares Sun's CEO.

             by Mary Lisbeth D'Amico, IDG News Service
             January 28, 1999, 12:22 p.m. PT

              Information systems departments, he said,
             should hand over as many tasks as possible to
             service providers. He recommended, for
             example, that corporations shut down their e-mail
             services

             In fact, he said, companies should get everything
             they can free over the Internet from service
             providers--free news, free spreadsheets, free
             presentation graphics--negating the need for
             what he called a "four-way Pentium NT hairball"
             to do all those things.

             To meet the challenge, ISPs have to keep
             services and content device-neutral and base all
             services on Internet protocol, McNealy said.
             *****Users should not have to use Windows*****
             to view data or run services. The model for this vision
             is one in which simple, reliable appliances--devices
             such as a telephone--link users up to a network.

             "You know how to boot up a telephone? You pick it up,"
             he quipped.


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

From: Pietro Marchionni <[EMAIL PROTECTED]>
Subject: PCMCIA socket for desktop pc
Date: Fri, 29 Jan 1999 19:49:06 +0100

This is a multi-part message in MIME format.
==============1C067E8397CCAEB6FB903B7E
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi all,

I just buyed one SCM Microsystems SWAPBOX PC CARD ADAPTER (sbi-d2p card
reader) that is a PCMCIA multiple slot socket for desktop Pc, and I
would like to use it with my linux too.
Is there any driver for this device?
If not there is a device like this that I can use with linux?
Thanks in advance.




==============1C067E8397CCAEB6FB903B7E
Content-Type: text/x-vcard; charset=us-ascii;
 name="pietro.marchionni.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Pietro Marchionni
Content-Disposition: attachment;
 filename="pietro.marchionni.vcf"

begin:vcard 
n:Marchionni;Pietro
tel;cell:0347/3882824
tel;fax:+3906/50770365
tel;work:+3906/50770358
x-mozilla-html:FALSE
url:www.systeam.it
org:Systeam Spa;DAP
adr:;;V.le Eroi di Cefalonia, 37;Rome;;00128;Italy
version:2.1
email;internet:[EMAIL PROTECTED]
title:Ing.
fn:Pietro Marchionni
end:vcard

==============1C067E8397CCAEB6FB903B7E==


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

From: "mrushton" <[EMAIL PROTECTED]>
Subject: Re: Any New Brand Name PC Hardware Preconfigured with Linux
Date: 30 Jan 1999 16:29:48 GMT



Dave Wright <[EMAIL PROTECTED]> wrote in article
<[EMAIL PROTECTED]>...
> Hello,
> 
> Anyone know of good, brand new pre-configured w/ Linux PC hardware
> available from reputable PC vendors? What do you recommend?
> 
> Thanks in advance,
> Dave
> [EMAIL PROTECTED]
> 


I thought Dell is offering Red Hat Linux as an option on thier machines. 
Dell makes a real good product.  We bought them at work and I bought one
for home.  We had a few glitches like a bad onboard NIC on a few machines
but other than that they were good. The sent a tech to fix it for us and we
were back in the saddle again. 


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

From: "mrushton" <[EMAIL PROTECTED]>
Subject: STB velocity 128 card ????
Date: 30 Jan 1999 16:34:41 GMT

I have an STB Velocity128 AGP card 4MB.  Has anyone had any luck in getting
one to work with Linux ????



============================================================================


Mike Rushton  [EMAIL PROTECTED]  

Some favorite Fallen Flag roads :
    D&H     LV      CNJ     DL&W    PNER    The Laurel Line

Some favorite fallen Colleries : 
    Huber   Harry E (Bucket of Blood)   Sullivan Trail   Prospect

Railroad and Anthracite Homepage :
    http://www.geocities.com/SoHo/Lofts/6444  
    http://NEPA.railfan.net 

====================================================================

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

From: Erik Rossen <[EMAIL PROTECTED]>
Subject: Re: Linux becomes slow with more memory
Date: Sat, 30 Jan 1999 17:33:37 +0100

I ran into a similar problem this past November when I upgraded my
memory from 32Mb to 64Mb.  Look for the comp.os.linux.hardware thread
"Adding RAM caused Linux box to run slower. Why?" on Dejanews.

The problem (at least on my machine) was that the motherboard was
incapable of caching more RAM than 32Mb, even though it could address up
to 64Mb.  Linux uses high memory before low memory, hence the cache was
almost never getting used and performance dropped roughly in half. 
Under Windows 3.1 the difference was unnoticable since W3.1 uses low
memory first and I didn't use any monster programs under W3.1.

The solution was to incorporate the slram patch into my kernel.  This
patch allows one to reserve a specified amount of upper RAM to use as a
RAM swap disk and forces Linux to park itself in the cachable region of
RAM.  Not as nice as having all of RAM covered by cache, but better than
not using the new 32Mb module I had just bought.  The patch can be found
at http://www.andrew.cmu.edu/~keryan/slram/ .

Good luck,

Erik Rossen
[EMAIL PROTECTED]

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

From: "Dave Borgelt" <[EMAIL PROTECTED]>
Subject: Re: Notebook with 3D Rage LT Pro
Date: Sat, 30 Jan 1999 10:30:43 -0600
Crossposted-To: comp.os.linux.setup,comp.os.linux.x

Thanks for your reply.  I think I have resolved my problem, and maybe yours,
too.

There is a description on setting up a Dell Inspiron 7000 on the web at:
www.eecs.umich.edu/~steveh/inspiron and under the heading of X Windows, Free
X Server, Method 1 my computer (WinBook XL2) works using the libc6 version
of the patches (I'm using RH 5.2).

I've only used the machine with X for about an hour after my successful
install but all seems to be perfect.  This could work for you, too.

My thanks to Steve Hsieh and the others on this site for their work.  You've
been a life saver.

Dave

Derek Mason wrote in message <[EMAIL PROTECTED]>...
>Dave,
>
>You're not alone in this.  I recently got a Fortune 850 Notebook (at work)
with
>a similar display.
>The Linux here is SuSe v5.3 which is issued with XFree86 3.3.2
>
>As far as I can tell the 3D Rage LT wasn't supported in even 3.3.3.1
however, a
>brief look on wednesday at the SuSe SDB seemed to indicate that the SuSe
>modified versions XCom would. I recently got 3.3.3.1-8 libc5 version c/w
xf86
>and a new xf86setup.
>
>It still didn't work to' :(((
>
>I'm working on it with a great deal of help from others in the Linux
community.
>
>DM
>
>Dave Borgelt wrote:
>
>> I have a notebook with 1024 x 768 display (Winbook XL2) and the 3D Rage
LT
>> Pro AGP.  I have installed the XFree86 3.3.3.1 from Redhat's site and am
>> running Redhat 5.2.
>



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

From: James Youngman <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.questions,comp.os.linux.setup
Subject: Re: Clock weirdness..RH 5.2 and Dell Workstation 610
Date: 28 Jan 1999 23:08:55 +0000

[EMAIL PROTECTED] (Rick Crelia) writes:

> I have a new Dell Precision Workstation 610, 400MHz Xeon CPU,
> 1MB cache, 256MB RAM, blah blah blah. I have installed RH 5.2
> on it and have found the clock to be unreliable. Overtime, I
> lose several ticks and end up way off the actual time. In fact,
> after leaving it set with the correct time at 23:30 last night,
> when I returned this a.m. at 08:30, the time according to the
> workstation was only 01:09.

Turn off APM in the BIOS, or reconfigure (rebuild) your kernel to
enable kernel APM support.

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

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

From: James Youngman <[EMAIL PROTECTED]>
Subject: Re: HOW TO TRANSFER XENIX DISK TO LINUX?
Date: 28 Jan 1999 23:22:57 +0000

"damatex" <[EMAIL PROTECTED]> writes:

> - I have installed the XENIX drive as a slave:  mount says:
> 
>    Unable to read Xenix/SystemV/Coherent superblock.

Recompile your kernel, including support for Xenix/SystemV/Coherent superblocks.

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

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

From: "Glenn C. Everhart" <[EMAIL PROTECTED]>
Crossposted-To: comp.arch,comp.arch.storage,alt.os.linux,comp.periphs
Subject: Re: Same Disk RAID and Mirroring
Date: Sat, 30 Jan 1999 12:56:47 -0500

I tried some single disk mirroring awhile back and had the fun
of watching the disk go into "wash-rinse-dry" shaking mode.

Single disk RAID would seem best if it avoided head seeks,
which are slow and cause mechanical vibration. Thus I'd
think this could be a useful idea if done so that head seeks
would not be needed...stay within the cylinder.

Complications are that modern disks don't have fixed size
cylinders, and usually hide their real geometry. They also
have very considerable ECC already.

I have no info on error distribution but suspect that firmware
bugs (and response to power glitches) and in some cases file
structure brain damage are at least comparable with hardware
failure for lost data. It has been so in my experience.

I wonder if something like journalling writes to a second
platter (perhaps every few minutes, so for multiple writes to
a sector one would need only one journal) could be useful. A
backup of the entire volume plus this journal could reconstruct
the volume at any given instant after the full backup. A side
benefit would be a virtual undelete. The cost would be the
journal, which would grow without bound; you'd need to
re-backup now and then.

Log structured filesystems already do this kind of thing, but
have other problems (slow read). However, journalling addons
may well exist in Linux, seeing theconsiderable file system
research going on for that platform.

Andy Glew wrote:
> 
> comp.arch readers may know that I am very interested
> in issues of how reliability for personal computers.
> 
> In this instance, how to provide the reliability of RAID
> and/or mirroring to computers that have only one disk
> spindle, since the vast majority of PCs have only one
> or two hard disks.
> 
> Which causes me to ask if anyone has support for
> single disk RAID or mirroring, whether as a Windows
> installable filesystem, in UNIX, even better in standard
> LINUX, or possibly at the level of the disk controller
> or disk drive?
> 
> And, similarly, whether anyone has any error distribution
> data that would suggest whether this would be effective.
> 
> I imagine single disk RAID as providing a 5th, parity, track,
> for every 4 data tracks on disk. Dibs as to whether the parity
> block rotates or not.
> 
> Such track oriented single disk RAID would allow an error that
> hit any single disk block or track to be recovered. The usual
> performance impact (nothing for reads, a significant impact 4X for
> random writes, etc.).  Hybridized with mirroring, read performance
> might improve.
> 
> Such single disk RAID would not help continued uptime for
> whole disk failures, but would help data integrity and recovery
> if the disk could be repaired. It would also help data integrity
> for storage errors - bad blocks, bad tracks, etc.
> 
> Which is why I ask about the distribution of disk errors:
> Are multi-block disk errors common, motivating a track oriented
> RAID rather than a RAID approach that kept parity blocks on the
> same track as their data blocks?  Are errors that take out
> multiple tracks common, motivating interleaving of data tracks
> protected by different parity tracks?
> 
> And, overall, I am interested in whether this is available on any
> system I am likely to want to use on my home computer - LINUX
> or NT.
> 
> ---
> 
> My question is motivated by the prospect of a weekend spent backing up
> my home systems to CD-R, and by my unhappiness at having lost data
> on existing storage media, both main disk and archives. (I hope CD-Rs
> last longer than my old magnetic QIC tapes and Jazz disks.)
> 
> I would happily lose 20% of my storage media in order to increase reliability.
> I would not be so happy about the loss of performance on writes, although
> a hybrid mirroring strategy would help, as would a strategy that did not
> protect recently written data, but which, overnight and in idle cycles,
> protected likely to be read only data.
> 
> On my active hard disks, 95% of my storage is read-only - installed software,
> etc. - so the loss would be small.
> 
> On my backup and archive storage, such protection is almost a no-brainer.
> 
> ---
> 
> So: is it out there?

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

From: [EMAIL PROTECTED] (Andrew Lee)
Subject: how do i enable serial support in the kernel?
Date: Sat, 30 Jan 1999 18:29:04 GMT

more specifically, where is the option for it in make menuconfig?

tia

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

From: Erik Naggum <[EMAIL PROTECTED]>
Crossposted-To: 
comp.os.linux.misc,comp.unix.questions,comp.unix.misc,comp.emacs,comp.editors
Subject: Re: Linux keyboard? (For emacs use)
Date: 30 Jan 1999 18:03:11 +0000

* [EMAIL PROTECTED] (Tim Smith)
| Since the usual rule is not to *randomly* capitalize, but rather to
| capitalize all words that begin sentences, no information is lost.  Think
| of it as an xor.

  which means that words that already should be capitalized are no longer
  distinguished from words that shouldn't.  as far as I can see, one does
  not downcase proper names, so XOR is wrong.  XOR would have preserved
  information.  this is an OR operation.

| What information did you think was being destroyed, anyway?

  whether a word is a proper name.  if you never use proper names, no
  information is lost.  most people do, however.

#:Erik
-- 
  SIGTHTBABW: a signal sent from Unix to its programmers at random
  intervals to make them remember that There Has To Be A Better Way.

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

From: "Per Olsson" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.x
Subject: Re: XWindows on a 16MB TNT  :(  no go.
Date: Fri, 29 Jan 1999 20:59:30 +0100

allacircle wrote:

>Does bpp stand for bits per pixel??

IMHO, Yes.

/Per Olsson

--
If everything else fails, read the instructions. ;)




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


** 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