Linux-Hardware Digest #421, Volume #14 Thu, 1 Mar 01 12:13:05 EST
Contents:
Re: pc/104 modem problems (glen)
Bad RAM ("Thom Lawrence")
Tandberg slr 100 and linux (Przemek Bak)
Re: Difference in disk performance between 2.2 & 2.4 (hac)
NEED Nvidia Geforce2 Chip Manual & Driver ("Panzer Lee")
Re: Can't access scsi drives (Mike Veenhuizen)
Harddisk performance ("Jerry Wong")
Re: NEED Nvidia Geforce2 Chip Manual & Driver ("Vigil")
Re: Help with LILO and ATA 100 on dual (Bj�rn Tore Sund)
Re: Harddisk performance ("Snarf")
Re: NEED Nvidia Geforce2 Chip Manual & Driver (Kenneth R�rvik)
Re: Big Drive, Reluctant BIOS, how to work around? ("D F")
Re: New SCSI drive partition information different for fdisk and (Pete Willemsen)
Re: Should I abandon SCSI? ("NewsReader2")
MPC8210, MVME2100 Linux Port (Wolfgang)
Re: MPC8210, MVME2100 Linux Port (Wolfgang)
Re: MPC8210, MVME2100 Linux Port (Wolfgang)
USB Modem ("Beast")
Re: Harddisk performance ("Hellraizer")
----------------------------------------------------------------------------
From: glen <[EMAIL PROTECTED]>
Subject: Re: pc/104 modem problems
Date: Thu, 01 Mar 2001 13:49:50 +0000
Well no-one offered me any suggestions, however the jumper settings that
came with the modem were actually
incorrect, so once I found the correct ones, then YES you can run a pc/104
internal modem under linux.
Glen
------------------------------
From: "Thom Lawrence" <[EMAIL PROTECTED]>
Subject: Bad RAM
Date: Thu, 1 Mar 2001 14:04:56 -0000
(sorry if outlook express makes this horrible for you all somehow :).
i have what i assume is some bad ram. i have a toshiba tecra 500CDT
laptop running linux, which comes with 16mb 'out of the box' and which
has an extra 48mb card to give a total of 64mb. at boot, if i press
escape or whatever and get into my bios settings, it has a little box
saying i have 65536k of ram. if however, i let the thing run as far as
its own self test, it will say 65280k is okay. it doesn't say 256k is
_bad_, it just doesn't count all the way. either way, all the signs are
that the ram is bad: i am getting weird gcc errors (that signal 11
thing), seg faults and all the right gubbins. it doesn't make anything
unusable as it is, just really awkward and takes lots of rebooting to
keep happy. i understand this is not the linux experience i deserve. :)
so i found out about rick van rein's bad ram patch and was really
impressed and excited, and made a memtest86 floppy. much to my
disappointment it didn't find any errors. i ran it overnight to do all
the tests and lots of passes. it still found nothing. so i'm a bit
stuck. i also took my 48mb of ram out to leave just the original 16mb,
the bios counted up to 16128 (again, 256k off) and memtest86 still
didn't find anything. i'm running another utility under linux
('memtester' or something) right now and it has found nothing on its
first pass. it _did_ find something when i ran it under midnight
commander, however. dunno if that means anything.
does anyone know what is happening and how i find the location of the
bad stuff? i was actually quite enthusiastic about having broken
hardware at first, knowing i could get linux to work around it. but i
just need to know how to find it. would/could my bios be hiding the bad
stuff? do i merely have 65280k of ram? i tried giving mem=65280k to the
kernel and i still got the same gcc errors etc. any help, experience or
pointers to utilities i can run would be greatly appreciated. cheers.
--
thom lawrence
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Przemek Bak)
Subject: Tandberg slr 100 and linux
Date: 1 Mar 2001 14:28:44 GMT
Does anybody know if Tandberg slr 100 streamer works with linux ?
przemol
------------------------------
From: hac <[EMAIL PROTECTED]>
Subject: Re: Difference in disk performance between 2.2 & 2.4
Date: Thu, 01 Mar 2001 14:37:42 GMT
Simon Turvey wrote:
>
> Hi,
> I've been running some speed tests between kernels using identical
> parameters to hdparm and have found 2.4.0 to be significantly slower
> than 2.2.18. Can anyone confirm this.
>
> For me the difference is from about 22MB/S to about 14MB/S
>
What drive? 2.4.0 does not enable UDMA66 by default; you need to use
the kernel parameter "ide0=ata66" to enable it.
--
Howard Christeller Irvine, CA [EMAIL PROTECTED]
------------------------------
From: "Panzer Lee" <[EMAIL PROTECTED]>
Subject: NEED Nvidia Geforce2 Chip Manual & Driver
Date: Thu, 1 Mar 2001 23:41:22 +0900
I study programming device driver .
Nowadays being interested in graphic chip driver, I can't get
graphic chip manual & driver source code. ( Chip vendors don't provide them
to individuals)
I want geforce2 chip manual & driver, but any other chips manual & driver is
ok.
Please, let me know how to get them.
...and have a nice day.. :-)
------------------------------
From: Mike Veenhuizen <[EMAIL PROTECTED]>
Subject: Re: Can't access scsi drives
Date: Thu, 01 Mar 2001 14:49:33 GMT
>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<
On 2/28/01, 4:10:39 PM, "D. Stimits" <[EMAIL PROTECTED]> wrote=20
regarding Re: Can't access scsi drives:
> "D. Stimits" wrote:
> >
> > > Mike Veenhuizen wrote:
> > >
> > > I have a system with two scsi 9M hard drives. The second one was a=
dded
> > > just two weeks ago. I installed Slackware on this drive. I edited
> > > lilo.conf and ran lilo, so that I could boot either linux on sda3,=
> > > linux on sdb1, or windows on sda1. It all worked perfectly. On the=
> > > weekend I recompiled the kernel on sdb1, and ran zlilo. After I
> > > rebooted I had a kernel panic, with the system unable to access th=
e
> > > hard drive. I booted up my floppy linux rescue disk, and when I ra=
n
> > > fdisk I received the message "Cannot access /dev/sda, or cannot ac=
cess
> > > /dev/sdb. If I tried to mount any of the drive partitions, I got t=
he
> > > message " device /dev/sd* is not a block device". I think the prob=
lem
> > > is with the scsi bus, but windows still works fine, and can access=
,
> > > format, partition etc. any of the drives. I have been using linux =
for
> > > a number of years, and this is a first. Any help?
> >
> > I've seen something similar. It seems that the failed boot device=20
causes
> > the scsi controller to not register the existence of the drive. Boot=
> > floppies based on regular boot allow the drives to be visible and ru=
n
> > fine, but rescue cd's and rescue floppies seem to not see the drives=
.=20
If
> > you run under a rescue though, even without the device being found, =
I
> > have had lilo -v -u succeed in uninstalling the failed boot sector
> > (reinstating the original).
> One note I forgot. The lilo -v -u is preceeded by a chroot to the=20
mount
> point of the drive onto the rescue media. Usually /mnt/sysimage/ or
> something similar.
> >
> > As to why it panicked, perhaps you didn't compile in the scsi suppor=
t
> > needed. Modules only work if the initial ramdisk is used, with=20
mkinitrd
> > (modules on the scsi drive can't be found without the ability to=20
access
> > scsi in the first place, but that ability requires scsi access for
> > modules to tell it how to...catch 22). I always compile scsi directl=
y
> > into a kernel, versus modules.
Ok! I got it working. After failing with lilo \v \u (probably because=20
I didn't have a saved boot record for lilo to use) I remembered I had=20
a lilo floppy in my collection. I was able to boot up /dev/sda3 which=20
is my / linux partition, ran lilo from there, and voila! All is well=20
again. I guess when I recompiled the kernel with zlilo, I didn't=20
provide support for scsi in the boot sequence.
------------------------------
From: "Jerry Wong" <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.hardware.overclocking.amd,alt.windows98,alt.windows-me,hk.comp.pc,microsoft.public.win98.setup
Subject: Harddisk performance
Date: Thu, 1 Mar 2001 22:56:33 +0800
This is a multi-part message in MIME format.
=======_NextPart_000_000D_01C0A2A2.DCDF5780
Content-Type: text/plain;
charset="big5"
Content-Transfer-Encoding: quoted-printable
I want to install two harddisks in one IDE which supports ATA100. If one =
of the harddisk is ATA100 and the other is ATA33, will the former one =
work slower than that it should be due to the latter one?
--=20
http://members.hknet.com/~wong63124
(In Chinese Big 5)
http://members.hknet.com/~wong63124/linux.htm
(In English)
=======_NextPart_000_000D_01C0A2A2.DCDF5780
Content-Type: text/html;
charset="big5"
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=3Dbig5">
<META content=3D"MSHTML 5.50.4522.1800" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#c0c0c0>
<DIV><FONT size=3D2>I want to install two harddisks in one IDE which =
supports=20
ATA100. If one of the harddisk is ATA100 and the other is ATA33, will =
the former=20
one work slower than that it should be due to the latter =
one?</FONT></DIV>
<DIV><BR><FONT size=3D2>-- <BR></FONT><A=20
href=3D"http://members.hknet.com/~wong63124"><FONT=20
size=3D2>http://members.hknet.com/~wong63124</FONT></A><BR><FONT =
size=3D2>(In=20
Chinese Big 5)</FONT></DIV>
<DIV><FONT size=3D2></FONT> </DIV>
<DIV><A href=3D"http://members.hknet.com/~wong63124/linux.htm"><FONT=20
size=3D2>http://members.hknet.com/~wong63124/linux.htm</FONT></A><BR><FON=
T=20
size=3D2>(In English)</FONT></DIV></BODY></HTML>
=======_NextPart_000_000D_01C0A2A2.DCDF5780==
------------------------------
From: "Vigil" <[EMAIL PROTECTED]>
Subject: Re: NEED Nvidia Geforce2 Chip Manual & Driver
Date: Thu, 01 Mar 2001 15:15:16 +0000
If you could get Nvidia's 'closed source' driver source code, bless you.
"Panzer Lee" <[EMAIL PROTECTED]> had the audacity to claim:
> I study programming device driver . Nowadays being interested in graphic chip
>driver, I can't get graphic chip manual & driver source code. ( Chip
> vendors don't provide them to individuals) I want geforce2 chip manual & driver, but
>any other chips manual & driver is ok. Please, let me know how to
> get them.
--
.
------------------------------
From: [EMAIL PROTECTED] (Bj�rn Tore Sund)
Crossposted-To: comp.os.linux.setup
Subject: Re: Help with LILO and ATA 100 on dual
Date: 1 Mar 2001 15:16:09 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 28 Feb 2001 11:41:25 +0100, Luigi Cavallo wrote:
>
> Hi,
>
> I just installed RH 7.0 on a Dual PIII with ATA 100 HDs on the ATA100
> ide2.
> However, when I try to reboot from the HD the box hangs before arriving
> at the lilo prompt. Actually, it hangs with the first two uppercase
> charcaters of lilo, as
>
> LI
>
> If I boot from the floppy made during installation, and by feeding linux
> ide2=...
> it boots properly.
>
> I tried to install lilo both in the MBR as well as on /dev/hde, but the
> box behaves the same.
Sounds like a kernel without ATA 100 support. This is typically
not compiled into the packaged SMP kernel - it wasn't on my SuSE
7.0. I had exactly the same problem as you until I swapped to
an EIDE kernel (which I had to download from the SuSE ftp site) -
after which I had disk support, but no SMP support.
Compiling a new kernel was the solution to get both.
Bj�rn
--
Bj�rn Tore Sund "When in fear, and when in doubt;
Run in circles, scream and shout!"
[EMAIL PROTECTED] - Robert Heinlein
------------------------------
From: "Snarf" <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.hardware.overclocking.amd,alt.windows98,alt.windows-me,hk.comp.pc,microsoft.public.win98.setup
Subject: Re: Harddisk performance
Date: Thu, 01 Mar 2001 15:26:16 GMT
Man, it don't get any better. Spam, HTML format, and a purposesly vague
troll message.
Dave
Jerry Wong <[EMAIL PROTECTED]> wrote in message
news:97lnos$ljp$[EMAIL PROTECTED]...
I want to install two harddisks in one IDE which supports ATA100. If one of
the harddisk is ATA100 and the other is ATA33, will the former one work
slower than that it should be due to the latter one?
--
http://members.hknet.com/~wong63124
(In Chinese Big 5)
http://members.hknet.com/~wong63124/linux.htm
(In English)
------------------------------
Subject: Re: NEED Nvidia Geforce2 Chip Manual & Driver
From: [EMAIL PROTECTED] (Kenneth R�rvik)
Date: Thu, 01 Mar 2001 15:56:04 GMT
[EMAIL PROTECTED] (Panzer Lee) wrote in <97lmis$b17$[EMAIL PROTECTED]>:
>I study programming device driver .
>Nowadays being interested in graphic chip driver, I can't get
>graphic chip manual & driver source code. ( Chip vendors don't provide
>them to individuals)
>I want geforce2 chip manual & driver, but any other chips manual &
>driver is ok.
>Please, let me know how to get them.
The drivers from nVidia are proprietary, so you can't get the sources for
them, unless you have some really good friends at nVidia... ;)
You can always take a look at the opensourced drivers from the Xfree86
source, available from xfree86.org. These do not support hardware
accelerated openGL, though.
--
Kenneth R�rvik 91841353/22950312
Nordbergv. 60 A [EMAIL PROTECTED]
0875 OSLO home.no.net/stasis
------------------------------
From: "D F" <[EMAIL PROTECTED]>
Subject: Re: Big Drive, Reluctant BIOS, how to work around?
Date: Thu, 1 Mar 2001 10:56:56 -0500
David Christensen wrote in message ...
>Gene:
>
>> > Okay, I've been down this road once, and as I
>> > recall, I conquered the monster, but cannot
>> > locate the documentation as to how. Perhaps
>> > someone on this list has the answer...
>
>Get the Large Disk HOWTO:
>
> ftp://metalab.unc.edu/pub/Linux/docs/HOWTO/
>
>
>I just did a 30 GB Maxtor. 33.8 GB is a boundary. See the
HOWTO.
>
>
>> > Cylinders: 16383, Heads: 16, Sectors: 63
>
>That is a magic code meaning the disk is too big for
old-style CHS
>addressing. See the HOWTO.
>
>After trying all the various permutations in CMOS and going
back to
>"auto", my BIOS sees it as ~60,000 cyliners, 16 heads, and
63 sectors.
>The numbers multiply out to ~30 GB using a calculator.
>
>
>> > I did an install from scratch on a 40 GB maxtor
>> > with a 1 Gig / partition, a couple 128 Meg
>> > swap partitions, a 4 Gig /usr partition, and the
>> > rest in one large /home partition.
>
>I did a 10 MB boot partition (~200 cylinders, less than the
critical
>#1024), a 128 MB swap partition, a 1 GB root partition, and
saved the
>rest for later.
>
>A key piece of information is that checking for bad blocks
during
>formatting requires ~2 seconds per MB. So, the drive light
comes on,
>the drive makes very little noise (not the typical
click-click-click of
>formatting), and you think the installer has died! Be
patient -- allow
>~15 minutes per GB. The way I got this information was by
building a
>small system on a 240 MB disk, plugging in the big disk as
a second
>drive, and then partitioning/ formatting the big disk from
the command
>line ('mkfs -c -v' ?).
>
>
>> > I also remember something about a jumper
>...
>> ONLY WORKS with the so-called MAX-BLAST
>
>I think you're right about the jumper -- it's pointless.
>
>
>> The BIOS is dated in mid-1999, which is supposedly
>> able to handle drives with more than 4096 cylinders.
>
>Keeping trying CMOS permutations until you see a CHS
combination that
>multiplies out to 40 GB on a calculator.
>
>
>> ... throw away the motherboard as soon as I can get to a
>> computer swap meet. Looks like I need to stick
>> to Intel MBs, or at least away from the off-brands
>> like Jetway.
>
>I am loosing faith in Intel MB's. I bought an inexpensive
Shuttle
>Spacewalker (#AV61) specifically for Linux (RH6.2) and have
been very
>pleased with it. It's currently running the 30 GB Maxtor.
>
>HTH,
>
>--
>David Christensen
>[EMAIL PROTECTED]
>
The only problem that I can see for you is if your BIOS
actually hangs during boot. If that's the case, you need a
BIOS upgrade.
If it boots fine but doesn't detect the drive, you're okay.
Linux kernels since 2.2.16/2.3.29, IIRC, will access a drive
bigger than 33 GB just fine. All of this is in the Large
Disk HOWTO, BTW. I installed a 40.9 GB Maxtor in my box last
week. I had to flash the BIOS, upgrade the kernel and set
the drive to None in the BIOS (recent Linux kernels don't
need the BIOS to access the drive, anyway.) After that, I
simply changed the logical geometry to 4865/255/63 and built
a partition table on the drive. (Divide your LBA sectors by
(255*63) to figure the number of cylinders.) The drive even
works fine in WIndows, now. I can't stress enough the
importance of the Large Disk HOWTO for what you're trying to
do...
Dave Fluri North Bay, Ontario Canada
(The opinions herein are mine. I do not speak for my
employer unless I expressly indicate otherwise.)
------------------------------
From: Pete Willemsen <[EMAIL PROTECTED]>
Subject: Re: New SCSI drive partition information different for fdisk and
Date: Thu, 01 Mar 2001 09:08:44 -0700
Rod,
Your suggestions worked great. I'm still not sure of why cfdisk
initially created my paritions incorrectly. I'll investigate that if I
have time.
In any event, I zeroed out the partition table, had PartitionMagic
create *any* Linux partitions I needed and then I installed Linux. The
CHS geometry reported by PartitionMagic is now in sync with what fdisk
reports: 2213 Cyl, 255 Heads, and 63 Sectors.
Thanks again for your help.
Pete Willemsen
Rod Smith wrote:
> 1) In Linux, use "dd if=/dev/zero of=/dev/sdb bs=512 count=1". This will
> completely wipe out the partition table, along with any flaky
> information it might contain that could be the root of the problem.
> 2) Since you mentioned you've got Partition Magic, I'd recommend
> booting it and using it to create new partitions. I believe that PM
> has some way to report the CHS geometry, so check that it doesn't use
> 1 for the number of heads or cylinders. If it does, start over or try
> another tool.
> 3) When you re-install or restore Linux, check with fdisk and cfdisk to
> see that the CHS geometry is being reported consistently.
------------------------------
From: "NewsReader2" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Thu, 01 Mar 2001 16:37:03 GMT
"Ron Reaugh" <[EMAIL PROTECTED]> wrote in message
news:Xqgn6.8642$[EMAIL PROTECTED]...
>
> NewsReader2 <[EMAIL PROTECTED]> wrote in message ...
> >
> >"Ron Reaugh" <[EMAIL PROTECTED]> wrote in message
> >news:7g5n6.6016$[EMAIL PROTECTED]...
> >>
> >> J.B. Nicholson-Owens wrote in message ...
> >> >Donovan Rebbechi wrote:
> >> >> RAID 0 is great for high performance desktops, but if you want a
> >> >> storage solution that's also *reliable*, it's useless.
> >> >
> >> >I wouldn't say it's useless--that depends on the data stored on the
> >volume
> >> >and how well the site has prepared for failure. But I agree--RAID 0
is
> a
> >> >high price to pay for failure.
> >>
> >>
> >> Nope, just keep good backups like is required with a single HD.
> >
> >Not good enough at 3:00pm the next day!
> >
> >Just keep to a single drive for greater reliability and stability.
> >
> >Choose 10K+ SCSI to improve workstation performance
>
>
> All nonsense. Nothing supplants a good backup scheme at 3:00 or at any
> other time. Fast inexpensive EIDE RAID 0 plus an appropriate
> backup/checkpoint scheme is just as reliable as a SCSI solution.
>
>
Yeah sure Ron!
All that extra crap that goes on with the Promise
controller/firmware/software/drivers and half the reliability of a single
drive is asking for trouble in a high value corporate environment. As well
those ATA drives being made as cheaply as possible and the warranty hassles
associated with them forget it!
Stick with non-RAID-0 for ATA if you go ATA.
Get a grip!
------------------------------
From: Wolfgang <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.powerpc
Subject: MPC8210, MVME2100 Linux Port
Date: Thu, 01 Mar 2001 17:43:35 +0100
Hi !
Is anybody working on a Linux Port for the MPC8240 or has some
information about such a port ?
Wolfgang
------------------------------
From: Wolfgang <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.powerpc
Subject: Re: MPC8210, MVME2100 Linux Port
Date: Thu, 01 Mar 2001 17:45:51 +0100
Wolfgang wrote:
> Hi !
>
> Is anybody working on a Linux Port for the MPC8240 or has some
> information about such a port ?
>
> Wolfgang
Sorry for the wrong subject !
It must be "MPC8240, MVME2100 Linux Port"
Wolfgang
------------------------------
From: Wolfgang <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.powerpc
Subject: Re: MPC8210, MVME2100 Linux Port
Date: Thu, 01 Mar 2001 17:46:12 +0100
Wolfgang wrote:
> Hi !
>
> Is anybody working on a Linux Port for the MPC8240 or has some
> information about such a port ?
>
> Wolfgang
Sorry for the wrong subject !
It must be "MPC8240, MVME2100 Linux Port"
Wolfgang
------------------------------
From: "Beast" <[EMAIL PROTECTED]>
Subject: USB Modem
Date: Fri, 2 Mar 2001 11:12:02 +0100
Hello!
I've got the question - does kernel 2.4.x correctly support USB modems?
I'm interested in Pentagram Fate (model with USB Hub) - is it working under
Linux?
Thanks..
------------------------------
From: "Hellraizer" <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.hardware.overclocking.amd,alt.windows98,alt.windows-me,hk.comp.pc,microsoft.public.win98.setup
Subject: Re: Harddisk performance
Date: Thu, 1 Mar 2001 16:52:28 -0000
My eyes fell out! HELP! I squinted to see the message and pop, out they
come. OOoppps theres one - squish - oops
--
Hellraizer
I realised I was god when I started to pray one day then I realised I was
talking to myself
"Snarf" <[EMAIL PROTECTED]> wrote in message
news:sStn6.295954$[EMAIL PROTECTED]...
> Man, it don't get any better. Spam, HTML format, and a purposesly vague
> troll message.
> Dave
>
> Jerry Wong <[EMAIL PROTECTED]> wrote in message
> news:97lnos$ljp$[EMAIL PROTECTED]...
> I want to install two harddisks in one IDE which supports ATA100. If one
of
> the harddisk is ATA100 and the other is ATA33, will the former one work
> slower than that it should be due to the latter one?
>
> --
> http://members.hknet.com/~wong63124
> (In Chinese Big 5)
>
> http://members.hknet.com/~wong63124/linux.htm
> (In English)
>
>
------------------------------
** 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 comp.os.linux.hardware.
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
******************************
