Linux-Hardware Digest #452, Volume #13 Sun, 20 Aug 00 16:13:10 EDT
Contents:
Miro PCTV Pro (Niek Smits)
Re: Netgear FA311 ethernet card Not working (Donald Becker)
Suse 64. + FritzX PC serial ("Holger")
Re: Improve Rotational Delay of HDD with Software Only (Thomas Tonino)
Re: SuSE 7.0 + Fritz!X USB? ("James Chew")
Re: Installing 2nd hard disk ("Peter T. Breuer")
Re: Installing 2nd hard disk ("Peter T. Breuer")
PCI MODEM .... (Fung)
Re: IOMEGA ppa troubles... (mst)
Re: Installing 2nd hard disk (Stewart Honsberger)
Re: Installing 2nd hard disk (Stewart Honsberger)
Re: MTRR error in Dual CPU system. (Ricardo Ferreira)
Re: Improve Rotational Delay of HDD with Software Only ("Ron Reaugh")
Re: Installing 2nd hard disk ("Peter T. Breuer")
Re: Does cdrecord really work with IDE CD-R?? (Douglas E. Mitton)
nVidia gforce2 & Xfree86-4.0.1 using 126M ram! (Tal Lancaster)
Accessing a remote Windows Printger from Linux ("B. Joshua Rosen")
----------------------------------------------------------------------------
From: Niek Smits <[EMAIL PROTECTED]>
Subject: Miro PCTV Pro
Date: Sun, 20 Aug 2000 18:00:41 +0200
This is a multi-part message in MIME format.
==============C39AB8D821575F5AF08578C9
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
How do I get my PCTV-Pro running on rh 6.2??, or is it not possible?
Greatz, Niek
==============C39AB8D821575F5AF08578C9
Content-Type: text/x-vcard; charset=us-ascii;
name="niek.smits.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Niek Smits
Content-Disposition: attachment;
filename="niek.smits.vcf"
begin:vcard
n:Smits;Niek
x-mozilla-html:FALSE
adr:;;;;;;
version:2.1
email;internet:[EMAIL PROTECTED]
fn:Niek Smits
end:vcard
==============C39AB8D821575F5AF08578C9==
------------------------------
From: [EMAIL PROTECTED] (Donald Becker)
Crossposted-To: comp.os.linux.networking
Subject: Re: Netgear FA311 ethernet card Not working
Date: Sun, 20 Aug 2000 16:19:33 GMT
In article <[EMAIL PROTECTED]>,
Ramin Sina <[EMAIL PROTECTED]> wrote:
>Hi all,
>Despite what is advertised on its box, netgear FA311 cards don't work
>with Linux (Atleast not with kernel 2.2.*) Tulip driver doesn't work
>with it. It comes with it's own linux driver that was written for 2.0.36
>and does not compile under Red Hat 6.2. Could someone please confirm
>this assessment;and are there any solutions to fix this?
A driver for released kernels (1.1.73 through 2.2.*) is at
http://www.scyld.com/network/ethercard.html
ftp://www.scyld.com/pub/network/natsemi.c
Don't forget that you will also need 'pci-scan' support.
A pre-2.4 version should be in the latest development kernels.
The Netgear bogusness Isn't My Fault.
--
Donald Becker [EMAIL PROTECTED]
Scyld Computing Corporation http://www.scyld.com
410 Severn Ave. Suite 210 Beowulf Clusters / Linux Installations
Annapolis MD 21403
------------------------------
From: "Holger" <[EMAIL PROTECTED]>
Subject: Suse 64. + FritzX PC serial
Date: Sun, 20 Aug 2000 18:57:20 +0200
Is there anybody with FritzX PC (serial) working with Suse 6.4 ???
------------------------------
From: Thomas Tonino <[EMAIL PROTECTED]>
Crossposted-To:
comp.arch.storage,comp.periphs,comp.sys.ibm.pc.hardware.misc,comp.sys.ibm.pc.hardware.storage
Subject: Re: Improve Rotational Delay of HDD with Software Only
Date: Sun, 20 Aug 2000 19:20:23 +0200
NAKAZATO Hajime wrote:
> I invented a software technique that improves rotational delay of HDD (only
> at reading).
> This technique makes HDD's rotational delay time half (only at reading). It
> needs no additional hardware. Load for CPU and memory is little.
It could be nice to use in _some_ areas. The idea is - as I read it - to
write everything to disk twice. This makes writing slower and halves
your storage capacity, but makes truly random accesses faster.
I can imagine using it for a read-mostly file system with files that are
mostly read in random order, but not for general use.
It could of course be expanded to triple or quadruple copies of data.
Could be a funny idea to shave something of the boot time of an OS or
something similar, but not for main storage.
Thomas
------------------------------
From: "James Chew" <[EMAIL PROTECTED]>
Subject: Re: SuSE 7.0 + Fritz!X USB?
Date: Mon, 21 Aug 2000 01:22:03 +0800
Where do you get SuSe 7.0?
Regards,
James.
<[EMAIL PROTECTED]> wrote in message news:8noep2$7fj$[EMAIL PROTECTED]...
> Is there anybody with a Fritz!X USB external ISDN card running under
> SuSE (7.0)?
>
> Martin
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,comp.os.linux.misc
Subject: Re: Installing 2nd hard disk
Date: 20 Aug 2000 17:17:57 GMT
In comp.os.linux.help [EMAIL PROTECTED] wrote:
: Peter T. Breuer <[EMAIL PROTECTED]> did eloquently scribble:
:> In comp.os.linux.help Dances With Crows <[EMAIL PROTECTED]> wrote:
:> : BTW, Peter, I've done "fdisk /dev/hdb", "mke2fs /dev/hdb{1,5,6,7,9}",
:> : "mount /dev/hdb1 /mnt/somewhere", and "cp -a /usr/local /mnt/somewhere"
:> : without rebooting or anything. Everything worked. This was with kernel
:> : 2.2.10....
:> What I'm getting at is that bit in the boot sequence where it lists
:> the partition table for each ide device it finds. I don't think that
:> code sequence gets repeated.
: It doesn't need to get repeated.
The kernel needs to reread the partition table.
: You add the new partition to /etc/fstab.
Eh? If the kernel was't aware of it before, it does you no good.
: You mount the new partition manually, and *ding* it's there.
Uh uh. No way, jose.
: All that sequence at the start does is check the current partitions in
: /etc/fstab and mount them.
No. You're thinking of the wrong part. That's not the boot sequence,
that's the init sequence.
:> The only chance you'd get to run it would
:> be when every single partition on that disk was dismounted, and fdisk
:> or someone issued a pleading ioctl to the kernel.
: Eh? I don't quite know WHAT you're thinking!
What I said. If you change the partition table on the DISK, you have to
adjust the corresponding map that the kernel holds of it, or they'll
differ and you'll get "interesting" results. I.e. you have to persuade
the kernel to reread the partition table from the disk.
: The whole point of the mount command is to allow for an extendable
: filesystem without the need to reboot!
I think you're confusing two parts of the software universe.
I haven't rebooted in weeks, so I can't show you the log. But if you
get a chance, look at the kernel boot sequence. You'll see the kernel
reads the partition tables of all devices on the ide controllers.
Nothing to do with mount.
Peter
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,comp.os.linux.misc
Subject: Re: Installing 2nd hard disk
Date: 20 Aug 2000 17:26:32 GMT
In comp.os.linux.help [EMAIL PROTECTED] wrote:
: Peter T. Breuer <[EMAIL PROTECTED]> did eloquently scribble:
:> In comp.os.linux.help Hiawatha Bray <[EMAIL PROTECTED]> wrote:
:> : The drive on my Linux box is full. I have an old HD I can put in, but I
:> : don't know how to configure Linux to recognize it and partition it
:> : correctly. How is this done? Thanks.
:> Put drive in box. Boot system. Type fdisk /dev/hdb (or whatever) to
:> partition it. Run mke2fs on the partitions after one reboot.
:> You have to reboot unless someone has worked out a way to get the
:> kernel to rescan the table ...
: Reboot? Why? All you need to do is mount it.
You can't. As far as the kernel is concerned it won't be there. It
needs to reread the partition table from the disk. The image it
has of the table is out of date. It will only reread the table
under particular circumstances, and you have to ask it to. Fdisk
does ask it, afair, but if any of the disk is mounted at the time,
(for example), the ioctl will error.
: (You'll have to shut down to put the drive in of course, but once it's
: there, there should be no need for a reboot).
Think again. Why is this point so hard to grasp! You canot change the
ondisk sizes of partitions or create new ones without requiring the
kernel to reread the table. If it didn't reread and you changed the
size of /dev/hda1, say, then a mke2fs /dev/hda1 would use the old size
instead of the new one.
: No idea why a reboot is neccesary...
It guarantees that the kernel rereads the partition table. I told him
to do that because I couldn't guarrantee it for him in any other easily
describable way.
Peter
------------------------------
From: Fung <[EMAIL PROTECTED]>
Subject: PCI MODEM ....
Date: Mon, 21 Aug 2000 01:32:05 +0800
What is the step to install the PCI Modem in Linux ?
Some one say that using the commad setserial .. isapnp etc.
can any one help me ?
I already reading the modem How-to
but i'm still don't understand how to install
Any one have idea ?
< My Modem Model is Practical Intl 56K V90 VFM in PCI Mode >
The lable in the box say that it is a hardware modem..
------------------------------
From: mst <[EMAIL PROTECTED]>
Subject: Re: IOMEGA ppa troubles...
Date: Sun, 20 Aug 2000 13:47:49 -0400
January Weiner wrote:
>
> Hello,
> I am trying to install Debian 2.2 on a computer with an Iomega
> parallel port drive.
>
> I am desperate.
>
> While installing scsi modules, this is what I get if I try to
> install the apriopriate driver:
>
> /lib/modules/2.2.17/scsi/ppa.o: init_module: Device or resource busy
> Hint: this error can be caused by incorrect module parameters,
> including invalid IO or IRQ parameters
> /lib/modules/2.2.17/scsi/ppa.o: insmod /lib/modules/2.2.17/scsi/ppa.o failed
> /lib/modules/2.2.17/scsi/ppa.o: insmod ppa failed
>
To be able to use ppa as a module, you *must* have parallel port
support, which also must be loaded as a module, with the appropriate
parameters (i/o address, usually 0x3e8, and irq, usually 7). Check with
lsmod if you have the parport module loaded.
MST
------------------------------
From: [EMAIL PROTECTED] (Stewart Honsberger)
Crossposted-To: comp.os.linux.help,comp.os.linux.misc
Subject: Re: Installing 2nd hard disk
Reply-To: [EMAIL PROTECTED]
Date: Sun, 20 Aug 2000 18:47:59 GMT
On 19 Aug 2000 19:25:35 GMT, Peter T. Breuer wrote:
>: The drive on my Linux box is full. I have an old HD I can put in, but I
>: don't know how to configure Linux to recognize it and partition it
>: correctly. How is this done? Thanks.
>
>Put drive in box. Boot system. Type fdisk /dev/hdb (or whatever) to
>partition it. Run mke2fs on the partitions after one reboot.
>You have to reboot unless someone has worked out a way to get the
>kernel to rescan the table ...
Correct until the end. I installed a second drive to my system, added
partitions, and FDisk said something about calling IOCTL to recan table.
The only required re-boot was to physically install the drive (I didn't
have any hot-swappable equipment).
When I booted, I FDisk'ed the drive, mke2fs'ed the drive, modified fstab,
moved the contents of the /var partition (one of the ones I was moving
to the new drive) to a temp home, mounted the new /var partition on hdb,
moved the var-backup to its new home, and all was well.
--
Stewart Honsberger (AKA Blackdeath) @ http://tinys.cx/blackdeath
[EMAIL PROTECTED] (Remove 'thirteen' to reply privately)
Humming along under SuSE 6.4, Linux 2.4.0-test5
------------------------------
From: [EMAIL PROTECTED] (Stewart Honsberger)
Crossposted-To: comp.os.linux.help,comp.os.linux.misc
Subject: Re: Installing 2nd hard disk
Reply-To: [EMAIL PROTECTED]
Date: Sun, 20 Aug 2000 18:54:25 GMT
On 20 Aug 2000 07:42:00 GMT, Hiawatha Bray wrote:
>I tried this, but the copied /usr on my new larger disk seems to be missing
>files. When I try to run programs like Midnight Commander, they don't work
>any more. I went back to /etc/fstab and restored the mount for my original
>/usr partition, and everything went back to normal. Could the cp -a command
>be the wrong one for copying everything? Thanks.
You'll have to read the man page for 'cp' (man cp). One caveat you'll run
into when copying file is file permissions. Not all applications can deal
with their files suddenly being owned by 'root'. (Squid, for one, will bomb
on you if its cache is suddenly owned by root).
'cp -a' should do the trick for you, but I prefer to specify all options
verbosely rather than relying on a "Same as ... " approach. Also, I like
the -v (be verbose) option present so that I know what my system is doing,
also so that I know if there are any problems.
Me, I'd use;
cp -d -p -R -v /mnt/old /mnt/new
That worked like a champ for me when moving my /var partition around.
One could also possibly use the abilities of 'mc' to copy everything,
but I'd still stick with cp if I had the choice.
The other thing you could do, and it's something I've seen suggested
many times, is use 'tar' to archive all the files on the source partition
and expand them to the destination. This could be done in one or two steps
(either archive then extract, or use pipe to extract as it archives), but
again, cp worked for me so my reccomendation remains with it.
--
Stewart Honsberger (AKA Blackdeath) @ http://tinys.cx/blackdeath
[EMAIL PROTECTED] (Remove 'thirteen' to reply privately)
Humming along under SuSE 6.4, Linux 2.4.0-test6
------------------------------
From: Ricardo Ferreira <[EMAIL PROTECTED]>
Subject: Re: MTRR error in Dual CPU system.
Date: Sun, 20 Aug 2000 19:44:46 +0100
Hans wrote:
>
> Hello,
>
> My Linux system is
> Motherboard: Abit BP6 Socket 370
> CPU: Intel Celeron 466Mhz x2
> RAM: 64M SDRAM x1
> HDD: Western Digital Caviar 10.2GB x2
> CD-ROM: ATAP 40X
>
> dmesg shows the below messages.
>
> mtrr: your CPUs had inconsistent fixed MTRR settings
> mtrr: probably your BIOS does not setup all CPUs
This is perfectly normal. My Dual P3-550 oc 616 system with a
DFI-P2XBL/D board says the same and it works as a charm.
>
> What's mean?
It means the mtrrs are not the same in the two CPUs. MTRRs are registers
in the P3/Celeron that permit you to change how data is written to RAM.
It can be used to speed up video.
>
> I got really trouble this system. Especially, when I shutdown it,
> the system stop before shutting down eth0 (NE2000 PCI Network card.) and
> I just managed to shutdown it externally. And then, severe error report
> while booting fsck error "unmounting clearnly..." The fsck almost
> couldnot fix the partitions' illegal IO blocks. So, I used to spend
> horrible time only run fsck manually.
Thats because the system didnt unmount the filesystems when you shut it
down with the power button.
>
> And the worse stuff was the system stopped! whenever I did basic file
> handling:
> tar zxvpf linux-2.2.16.tar.gz # data crc error
>
> cp redhat-6.2-i386.iso /usr/software # in this case 'timeout waiting
> for DMA
> System stopped complaining 'DMA stuff'
Hmm, the last time this happened to me, it meant one of my HDDs was
about to die so i bought a new Fujitsu 20GB and now its working great.
>
> My last trial was to reinstall Linux system formatting the Hard Disk.:(
>
> I just finished new kernel installation (2.2.16). It worked fine.
>
> Is there special BIOS setup for Dual CPU system? I cannot find MTRR
> related stuff in the MB manual I have.
>
> My first SMP machine is going through a lot of trouble. :(
> I wanna here SMP related info including kernel configuration and BIOS
> setup beyond the SMP-HOWTO document. <- I read it several times and
> followed the guide line.
>
If your overclocking your CPU, that could be the problem, as some
components dont react very well to overclocking.
Good luck
--
"One World, One Web, One Program" - Microsoft Ad
"Ein Volk, Ein Reich, Ein Fuhrer" - Adolf Hitler
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To:
comp.arch.storage,comp.periphs,comp.sys.ibm.pc.hardware.misc,comp.sys.ibm.pc.hardware.storage
Subject: Re: Improve Rotational Delay of HDD with Software Only
Date: Sun, 20 Aug 2000 19:06:12 GMT
Thomas Tonino wrote in message <[EMAIL PROTECTED]>...
>NAKAZATO Hajime wrote:
>
>> I invented a software technique that improves rotational delay of HDD
(only
>> at reading).
>> This technique makes HDD's rotational delay time half (only at reading).
It
>> needs no additional hardware. Load for CPU and memory is little.
It would require a near OS rewrite. The software would have to know about
the geometry of every different HD >MODEL & SIZE & FIRMWARE LEVEL<
encountered. Further to make it work the driver would have to keep track in
real time of the rotational position of the disks surface otherwise it's
chose the slower version to read half the time.
This is of course frequently done already using spindle counter-synced
hardware RAID 1(duplexed set) which requires little new software in the
host.
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,comp.os.linux.misc
Subject: Re: Installing 2nd hard disk
Date: 20 Aug 2000 19:06:53 GMT
In comp.os.linux.help Stewart Honsberger <[EMAIL PROTECTED]> wrote:
: On 19 Aug 2000 19:25:35 GMT, Peter T. Breuer wrote:
:>: The drive on my Linux box is full. I have an old HD I can put in, but I
:>: don't know how to configure Linux to recognize it and partition it
:>: correctly. How is this done? Thanks.
:>
:>Put drive in box. Boot system. Type fdisk /dev/hdb (or whatever) to
:>partition it. Run mke2fs on the partitions after one reboot.
:>You have to reboot unless someone has worked out a way to get the
:>kernel to rescan the table ...
: Correct until the end. I installed a second drive to my system, added
: partitions, and FDisk said something about calling IOCTL to recan table.
That's the one. The ioctl'll error out unless the drive is completely
dismounted at the time. I couldn't guarrantee it for him at his level
of expertise, so I asked him to reboot, which will ensure that every
partition is dismounted.
: The only required re-boot was to physically install the drive (I didn't
: have any hot-swappable equipment).
You'd have had to reboot if your root was on the drive and you were
repartitioning it, for example. That kind of situation is all I
intended to avoid by issuing the instruction as above.
Peter
------------------------------
From: [EMAIL PROTECTED] (Douglas E. Mitton)
Crossposted-To: comp.os.linux.misc,mailing.comp.cdwrite
Subject: Re: Does cdrecord really work with IDE CD-R??
Reply-To: [EMAIL PROTECTED]
Date: Sun, 20 Aug 2000 20:20:10 GMT
What distribution and kernel version are you using?
I get a very similar error with my Creative 6424 under kernel versions
2.2.14, 15 and 16. Kernel version V2.2.13 (and prior) work perfectly.
There have been some major edits to the scsi source code in the last 3
kernel releases.
At this point my work around is to use 2.2.13. This is not the best
solution BUT is all I have right now. If you do some searches on
dejanews with write_g1 you'll see you are not the only one BUT no one
seems to have come up with a fix.
Good luck!
Arnold Selby <[EMAIL PROTECTED]> wrote:
>I have failed bout 40 timess in succession attempting to write the
>same 7 audio wav files
>to various (memorex and TDK) CDRs.
>After 4 coasters, I shifted to -dummy.
>/root/cdrecord/cdrecord: Input/output error. write_g1: scsi sendcmd:
>
>retryable error
>status: 0x2 (CHECK CONDITION)
>CDB: 2A 00 00 00 08 55 00 00 1B 00
>Sense Bytes: 70 00 05 00 00 00 00 06 00 00 00 00 24 00 00 00
>Sense Key: 0x5 Illegal Request, Segment 0
>Sense Code: 0x24 Qual 0x00 (invalid field in cdb) Fru 0x0
>Sense flags: Blk 0 (not valid)
>
>Fixating time: 0.004s
>/root/cdrecord/cdrecord: Input/output error. mode select g1: scsi
>sendcmd: retryable error
>status: 0x2 (CHECK CONDITION)
>CDB: 55 10 00 00 00 00 00 00 10 00
>Sense Bytes: 70 00 05 00 00 00 00 06 00 00 00 00 26 00 00 00
>Sense Key: 0x5 Illegal Request, Segment 0
>Sense Code: 0x26 Qual 0x00 (invalid field in parameter list) Fru 0x0
>Sense flags: Blk 0 (not valid)
>cmd finished after 0.001s timeout 40s
------------------------------------------------
Doug Mitton - Brockville, Ontario, Canada
'City of the Thousand Islands'
EMail: [EMAIL PROTECTED]
http://www.cybertap.com/dmitton
Other: mitton.dyndns.org
SPAM Reduction: Remove "x." from my domain.
------------------------------------------------
------------------------------
From: Tal Lancaster <[EMAIL PROTECTED]>
Subject: nVidia gforce2 & Xfree86-4.0.1 using 126M ram!
Date: Sun, 20 Aug 2000 19:30:37 GMT
I have just installed an Elsa Gladiac card (nVidia gforce2 GTS) on my
machine. Along with XFree86-4.0.1 (and the nVidia 0.94 tarballs).
Now X is up and running. However it is soaking up all of my memory
(from top):
Size RSS %CPU %MEM COMMAND
117M 112M 0.1 875 X
What is going on here? What can I do to fix it?
On one of my other machines running Linux with XFree86-4.0.1 for another
card I get more respectable values from top:
Size RSS %CPU %MEM COMMAND
26M 25M 0.1 10.2 X
Tal
--
Tal Lancaster
The RenderMan Repository (http://www.renderman.org/RMR/)
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "B. Joshua Rosen" <[EMAIL PROTECTED]>
Subject: Accessing a remote Windows Printger from Linux
Date: Sun, 20 Aug 2000 15:44:56 -0400
We need to be able to print to a printer on a Windows98 system from a
Redhat 6.2 Linux system over the Internet. Both machines have static IP
addresses and DSL connections. The Windows system has the Norton
firewall product on it.
I've done the following, using printtool I added an SMB printer with the
following parameters
Names: lpx
Spool Directory: /var/spool/lpd/lpx
Hostname of Printer Server: IP number of Windows system
IP number:
Printer Name: HPDeskJet
User
Password
Workgroup: workgroup
Input Filter: HP Deskjet 720
The Windows system has printer sharing turned on. The firewall did
detect an attempt to access the system when I attempted to print a test
page, but nothing happened at the printer. I've never been successful in
printing to any Windows printer, even on a LAN, so there is clearly
something that I don't understand. I'm strictly a Unix/Linux person so I
know very little about Windows so please provide step by step
instructions for what needs to be done on the Windows side.
1) Have I set up the printer correctly on the Linux side?
2) How do we set up the Norton Firewall software to permit access to the
printer from the Linux system? (preferably we would like to limit access
to the Linux system's IP address).
3) Is there something else that needs to be set up the Windows side?
Thanks,
Josh
------------------------------
** 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
******************************
