Linux-Hardware Digest #473, Volume #10 Sat, 12 Jun 99 13:13:37 EDT
Contents:
Re: Celeron or PII? (Lee Sau Dan ~{@nJX6X~})
Re: help: read ok, write fails, on scsi-atapi cd-rw (David Borowski)
geo_comp_addr: Cylinder number is too big ([EMAIL PROTECTED])
SUSE 6.1 - Adaptec1505 - Snapscan310 ("Dirk Demuynck")
Re: HP 720C doesn't print (Moritz Moeller-Herrmann)
Re: Which UPS ([EMAIL PROTECTED])
Re: Linux on Palm-PCs (David C)
Re: FAT32 ("Steven Buehler")
Re: eMachines as linux boxes (Gary Momarison)
Help with X Configuration - Unable to configure Display Adapter ("Vikram")
Re: Courier v.34 Modem (killbill)
Re: mt ... fsf not working (Paul Smicker)
Re: Linux on a Mitac 5033 (Mohd H Misnan)
Re: Graphic Card : ATI Expert 98 (Howard Mann)
Sonic Vibes sound card: problems! (Burra)
Re: Rockwell V.90 K56Flex Data, Fax, Speakerphone PCI Modem
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Lee Sau Dan ~{@nJX6X~})
Subject: Re: Celeron or PII?
Date: 11 Jun 1999 23:41:34 +0800
>>>>> "Johan" == Johan Kullstam <[EMAIL PROTECTED]> writes:
Johan> compiling code for the alpha is slower than compiling for
Johan> an i386.
Have you confirmed this saying by experiment?
Johan> the ia-32 arch has few registers and since they
Johan> all have special purposes, there is only one way they can
Johan> go. the alpha has many general purpose regs and hence many
Johan> ways to juggle them.
I think this theory is wrong. It is quite well known that code for a
CISC instruction set are much more difficult to optimize than code for
a RISC instruction set. (Beware: I'm talking about the *instruction
set* when I use the terms CISC vs. RISC. I know that P-II implements
the x86 instruction set using a RISC-like architecture behind the
scene. However, to the compiler, the instruction is still the x86
CISC set.)
Taking x86 and alpha (I don't know the alpha architecture, but I
assume that it is RISC) as examples. Since the alpha has lots of
registers and the registers are functionally identical to one another
(typical RISC characteristic), optimization is straight-forward. You
don't even have to choose, because all choices are identical. The
abundance of registers also means that the compile can be very
generous in allocating registers to code fragements, without
considering which variables are accessed more frequently than others.
For the x86, there are only a few registers. The registers are all
special-purpose. (e.g. ecx MUST be the counter for LOOP and REP[NE];
eax must be one of the multiplicands of a MUL, etc.) This introduces
constraints on how the registers can be used. The code optimizer must
take care of these constraints, making the optimization more
complicated and hence slower. Since there are too few registers,
they're a precious resource. So, the code optimizer must be very very
careful in allocating registers to variables. The optimizer must
carefully examine the code to find out which variable is used the most
frequently. It has also to figure out the life-time of the variables
so as to reuse registers smartly for different code fragments. In
summary, these make code optimization much more difficult than RISC.
So, if the "many general purpose regs" is a determing factor for the
kernel compilation time, I'd expect the alpha be faster.
Johan> the question really is, how fast can an alpha cross-compile
Johan> a kernel for an ia-32?
I'd like to see emprirical measurements, too.
--
Lee Sau Dan $(0,X)wAV(B(Big5) ~{@nJX6X~}(HZ)
.----------------------------------------------------------------------------.
| http://www.cs.hku.hk/~sdlee e-mail: [EMAIL PROTECTED] |
`----------------------------------------------------------------------------'
------------------------------
From: [EMAIL PROTECTED] (David Borowski)
Crossposted-To: comp.os.linux.help,comp.os.linux.questions,comp.os.linux.misc
Subject: Re: help: read ok, write fails, on scsi-atapi cd-rw
Date: 11 Jun 1999 14:34:34 GMT
Daniel M. Drucker ([EMAIL PROTECTED]) wrote:
: I just got a Smart & Friendly Speedwriter Plus. It is an ATAPI drive. I
: set up the SCSI-ATAPI emulation in the kernel. I am able to use the
: drive as a normal CD-ROM on /dev/scd0 without problems, so presumably
: the SCSI-ATAPI stuff is working. (?)
: cdrecord reports the following:
: Cdrecord release 1.6 Copyright (C) 1995-1998 J�rg Schilling
: scsidev: '0,0,0'
: scsibus: 0 target: 0 lun: 0
: Device type : Removable CD-ROM
: Version : 0
: Response Format: 1
: Vendor_info : 'SAF '
: Identifikation : 'CD-RW4224A '
: Revision : '1.20'
: Device seems to be: Generic mmc CD-RW.
: When I go to actually write a CD-R using cdrecord, however, it hangs for
: 40 seconds and then spews an infinite series of these messages:
: scsi : aborting command due to timeout : pid 252, scsi0, channel 0, id
: 0, lun 0 Write (10) 00 00 00 03 41 00 00 10 00
: hdc: irq timeout: status=0xd0 { Busy }
: SCSI bus is being reset for host 0 channel 0.
: ide1: reset timed-out, status=0x90
: hdc: status timeout: status=0x90 { Busy }
: hdc: drive not ready for command
: SCSI host 0 abort (pid 252) timed out - resetting
: The drive's light remains on, and the drive cannot be read from until
: after a _physical_ power-cycle.
: Can anyone tell me what is going wrong, or how I can provide more
: information to make it clearer what is going wrong?
: Daniel Drucker
: [EMAIL PROTECTED]
Do you have anything on /dev/hdd the secondary slave? if so try disconnecting
it. I had a mitsumi that caused strange failures like that. Check jumpers
and try changing the ribbon cable.
check the log on bootup and see if there are any errors for the
scsi emulation. Does a -dummy test pass? Have you been able to write any cds?
On linux or any other operating system? If not It is remotely possible
there is a problem with the cdr.
------------------------------
From: [EMAIL PROTECTED]
Subject: geo_comp_addr: Cylinder number is too big
Date: Fri, 11 Jun 1999 21:57:31 GMT
I've rebuilt the kernel to accomodate some new hardware, but
when I run lilo to tell it where the new kernel image is, I
get the message:
geo_comp_addr: Cylinder number is too big (1135 > 1023)
This message appears even if I run lilo on the old lilo.conf
file as well. In any case, the new image is not pointed to and
when I reboot only the old image is accessible.
How can I get around this problem? I'm desperate.
thanks,
edmond
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: "Dirk Demuynck" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.misc,linux.dev.scsi,linux.scsi,nl.comp.os.linux,nl.comp.os.linux.installatie
Subject: SUSE 6.1 - Adaptec1505 - Snapscan310
Date: Sat, 12 Jun 1999 01:19:27 +0200
Hi,
Since a while I'm trying to setup my flatbed scanner Snapscan310 with no
luck so far. This is what I do
During the installation of SUSE, the Kernel SCSI module is loaded from the
'modules' disk ==>adaptec1505 (aha152x=0x140,11,7)
The systems tells me that it has recognized it witout any problems
Once the system is installed and rebooted these are the commands that are
used to continue
- SANE1_0_33.rpm is installed
- pnpdump>/etc/isapnp.conf
- uncomment into this file the IO adress/IRQ /ACT(Y) lines
- test the loading of the module with modprobe, which is OK
- edit /etc/conf.modules and add
alias scsi aha152x
options scsi aha152x=0x140,11,7
- reboot
During the rebooting , I get the several errors that the IOports setting are
wrong ?
after typing insmod aha152x=0x140,11,7,the card is again recognized
(sgchecks shows the Snapscan scanner) but when I launch Xsacnimage at that
time I get again an error saying that it is 'invalid parameters' and
scanning is not possible.
Can somehelp me into this issue or do I use a wrong procedure to set up
------------------------------
From: [EMAIL PROTECTED] (Moritz Moeller-Herrmann)
Crossposted-To:
alt.os.linux,alt.uu.comp.os.linux.questions,at.linux,comp.os.linux.setup,nl.comp.os.linux.installatie
Subject: Re: HP 720C doesn't print
Date: Sat, 12 Jun 1999 01:08:48 +0200
Reply-To: [EMAIL PROTECTED]
On Fri, 11 Jun 1999 20:13:15 +0200, Marc Mutz <[EMAIL PROTECTED]> wrote:
>root wrote:
^^^^^^
NO, NO, NO!! Create a new user! Read the manual. This is worse than using
Windows.
>> I've a HP 720C colour deskjet and I have SuSE 6.1 runing on my computer.
>> During instalation of the OS I had to choose a printer. Because HP 720C
>> isn't on the list, I've tried other printers (like cdjet550) on /dev/lp0
>> but neither will print.
>Try /dev/lp1 instead of /dev/lp0. It works for me...
SusE6.1 is a Linux 2.2 distribution. v2.2 kernels assign lp0 first.
--
Moritz Moeller-Herrmann [EMAIL PROTECTED] ICQ# 3585990 # Not only
Get my public pgp / gpg key from # Open Source(TM)
http://webrum.uni-mannheim.de/jura/moritz/pubkeymoritz # but also
KDE forever! Use Linux to impress your friends! # Open Minded!
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Which UPS
Date: Fri, 11 Jun 1999 14:35:29 GMT
I use an APC Back-UPS Pro 280. This is enough to run the computer AND
monitor for about 10 minutes. Most Unicies are currently supported,
but Linux is not. However, there are two daemons available that
support most APC products ... I use apcupsd which is available only in
binary form but is very configurable and has network support (shut down
remote machines if power goes out, etc.) and good monitoring (logging,
etc.).
Disclaimer: Yes, I'm an authorised APC reseller, so I'm biased.
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> Does anyone have a recommendation for a relatively inexpensive UPS
> that can be connected to a Linux box? The UPS is not really intended
> to support the box for any appreciable length of time - possibly 5-10
> minutes, but I do require it to connect to the serial port and perform
> a shutdown should a power-outage occur.
>
> That brings me to the second point, which is the software itself. Do
> any of the current UPS manufacturers actually supply monitoring
> software for Linux or is it something I will need to find from other
> sources such as Smart UPS Tools?
>
> Hopefully someone out there has already been through this mill and can
> point to their own successful implementation of a UPS.
>
> Hardware Packard Bell P350
> O/S : Slackware v3.5 : Kernel 2.0.34
>
> Peter Whelerton
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: David C <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.system,comp.os.linux.misc,comp.os.linux.development.apps
Subject: Re: Linux on Palm-PCs
Date: 12 Jun 1999 11:49:06 -0400
[EMAIL PROTECTED] (Chris Ho) writes:
> In article <[EMAIL PROTECTED]>, Christian Kleitsch wrote:
>>
>> As far as I know, a Linux version for the Palm Pilot and even one for
>> the Atari Portofolio has been already developed. Did anybody start to
>> develop a 'Pocket Linux' for Palm-size PCs? (I mean this devices
>> running now WinCE)
>>
>> If not, I would like to know if there is a way to synchronize
>> (appointments and to-do list would be enough ;) such a device with a
>> Unix-box.
>
> From my understanding, the compiled the kernel and got it to run on a
> Pilot. Unfortunately, there really is no file system on a Pilot so the
> kernel is kind of useless there, as only a memory manager. Unless they
> somehow came up with a RAM disk for the Pilots while I was looking the
> other way...
The only reference to this project I know about is
http://ryeham.ee.ryerson.ca/uClinux/
It looks quite interesting, but it appears that the page hasn't been
updated recently.
If anyone knows of a more recent site, please post it.
Thanks.
-- David
------------------------------
From: "Steven Buehler" <[EMAIL PROTECTED]>
Subject: Re: FAT32
Date: Fri, 11 Jun 1999 09:04:21 -0700
This does not work with Windows 98 FAT32 drives, at least not in OpenLinux.
Been there tried that.
It WILL work in DosLinux. That's the main reason I dumped OpenLinux.
SB
Brett Neely <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Yes. An example command:
>
> mount -t vfat /dev/hda1 /mnt/windows
>
> See `man mount` for additional information on the mount command.
>
> Cyberant wrote:
> >
> > Can I see FAT32 hard-drives in Linux? If so how? I have tried just about
> > everything I know of, but nothing.
> >
> > Thanx in advance
>
> --
> Brett Neely, Technical Support Engineer, Linuxcare, Inc.
> 415.354.4878 x269 tel, 415.701.7457 fax
> [EMAIL PROTECTED], http://www.linuxcare.com
> Linuxcare. At the center of Linux.
------------------------------
From: Gary Momarison <[EMAIL PROTECTED]>
Subject: Re: eMachines as linux boxes
Date: 11 Jun 1999 09:02:21 -0700
[EMAIL PROTECTED] writes:
> Is there a howto on building a system someehere, thx
Look for "Building a Computer" in
http://www.aa.net/~swear/pedia/hardware-misc.html
And "Raven" in
http://www.aa.net/~swear/pedia/hardware.html
--
Look for Linux info at http://www.dejanews.com/home_ps.shtml and
Gary's Encyclopedia at http://www.aa.net/~swear/pedia/index.html
------------------------------
From: "Vikram" <[EMAIL PROTECTED]>
Subject: Help with X Configuration - Unable to configure Display Adapter
Date: Sat, 12 Jun 1999 20:53:45 +0530
Hi,
I have a Cyrix Graphics Express Card which is built-in to my motherboard
with a Cyrix 233MMX processor and I am unable to configure my machine to
run X Windows on Red Hat Linux 5.2
Please Help.Is it possible or not ?
Vikram
------------------------------
From: killbill <[EMAIL PROTECTED]>
Subject: Re: Courier v.34 Modem
Date: Fri, 11 Jun 1999 15:35:43 GMT
In article <q_T73.81$[EMAIL PROTECTED]>,
"John Connell" <[EMAIL PROTECTED]> wrote:
> I am *very* new to Linux. I have a USR Courier v.34 external modem.
> I am trying to set it up with mucho problems. Please note that this
> is the first real OS I have used. I
Welcome! All operating systems have their warts, but Linux has
deterministic and logical warts, and once you beat a Linux wart, it
remains in submission. Windows warts seem more random, if not malicious
(I'm convinced Plug-n-Play has something personal against me), and the
operating system lacks the granularity of control to really resolve the
problems.
Congratulations, you have found your first Linux wart!
> I am using RH6.0 and KDE is my GUI of choice.
Excellent information to provide. This is very helpfull to solve the
problem.
> I have copied my ppp files from /usr/doc to /usr/bin and edited
> appropriately,
Thats a good start.
> also edited hosts and resolv.conf files. KPPP ID's my modem, will
> query it, dials my ISP tries to log on
This means that more then likely the modem end of things is working
great, your problems likely have nothing to do with the modem, but
rather with the scripting for how your ISP talks to you when you dial it
up.
> quits and redials again and again and again.....
It is likely not getting an expected response characters from your ISP,
as set up in the ppp scripts.
> I just really need a step by step
> procedure, have read the "How-tos and the man pages, but am doing
> nothing but getting more confused.
Actually, your post indicates you understand more of the problem then
you realize. PPP can be pretty sticky to set up sometimes. The good
news is once you figure it out for your ISP, you likely will never have
to mess with it again.
> If I use minicom I can get to my ISP's server type in my ID and PW
> then get a NETWORK/LOGIN as KEYWORDS,
Make sure that this is what your ppp configuration files expect (pun
intended). You may have Linux enter another command or commands at this
point, something like startppp or ppp or somesuch. Try typing help at
this prompt, and see if you can get any information from the system.
Whatever command you type should bring up funny squiggles and other
random characters. Most ISP's do not need this extra step, they
automatically detect a PPP connection, but some require it be started
manually on their end.
Maybe when it says "NETWORK/LOGIN" it is asking you a question... "do
you want a network connection (i.e. PPP) or a login session?". Try
typing in NETWORK, and see if funny characters start showing up.
I have not set up RH 6.0 for PPP yet, but I seem to recall 5.2 had a
pretty nice ppp setup option under linuxconf that set things up pretty
easily. Run linuxconf and see if you can get it working through there.
> If I login again
> I get a No CARRIER response. Beats me......
This may be something funny with your modem not getting reset, or
hanging up between your attempts. Make sure your modem is getting
reset. Within minicom, I think ATH is the hang up string, and ATZ will
reset the modem. I would try linuxconf first, however. I am using
windowmaker, and do not have experience with KDE, so I don't know how
the kpppd system is set up. I do remember linuxconf making things
pretty straightforward (albiet a little tedious).
Feel free to email me with more specific questions... I will help if I
can.
--
Bil Kilgallon ([EMAIL PROTECTED])
--"I believe, what I believe, has made me what I am. I did not make
it, It is making me, it is the very truth of God, not the invention
of any man". Rich Mullins, quoting G.K. Chesterton.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Paul Smicker <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.misc
Subject: Re: mt ... fsf not working
Date: Fri, 11 Jun 1999 08:44:51 -0700
Thanks for all your helpful responses and suggestions, I'll be giving them a try
first thing Monday...
Paul Smicker
Computer Specialist
WestEd
Michael Meissner wrote:
> [EMAIL PROTECTED] writes:
>
> > andreas,
> >
> > paul's specified the tape not to rewind /dev/nst0.
> > ^
> >
> > paul,
> >
> > i came across the same problem. here's one of many solutions:
> > dd if=/dev/null of=/dev/nst0
> > then you can fire up restore.
>
> Rev'ing mt up to 0.5b (or down to 0.4) will also fix the problem. FWIW, RedHat
> 5.2 ships 0.5 (ie, the buggy version), and RedHat 6.0 ships 0.5b.
>
> --
> Michael Meissner, Cygnus Solutions
> PMB 198, 174 Littleton Road #3, Westford, Massachusetts 01886
> email: [EMAIL PROTECTED] phone: 978-486-9304 fax: 978-692-4482
------------------------------
From: [EMAIL PROTECTED] (Mohd H Misnan)
Subject: Re: Linux on a Mitac 5033
Date: 12 Jun 1999 14:43:59 GMT
On Fri, 11 Jun 1999 22:30:30 +0100, Michael Treacy wrote:
>Hi,
>
>Has anyone installed Linux on a Mitac 5033??....I'm getting one in a few
>days and I am going to keep it for Linux.
>
>I am especially curious about getting the touch-pad mouse to work.
>
>Any help would be appreciated.
I'm using Mitac 5033 and has no problem running Linux (RH 5.1 upto 6.0) on it.
The touch-pad work flawlessly and you can swith it with PS/2 mouse w/o a problem
during a boot-up time (you can swith it by going to standby to disk and startup
too). If you want the XF86Config, just mail me, surely my reply address need to
be edited.
--
|Mohd Hamid Misnan |[EMAIL PROTECTED] / [EMAIL PROTECTED] |i|
|iMac/233 RevB+MacOS 8.6 |http://www.geocities.com/SiliconValley/3319/ |M|
|Mitac 5033/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: Howard Mann <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Graphic Card : ATI Expert 98
Date: 12 Jun 1999 16:30:03 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Charles E Taylor IV) writes:
> In article <[EMAIL PROTECTED]>,
> Chuck Snively <[EMAIL PROTECTED]> writes:
>> Hi,
>>
>> When you run Xconfigurator, you don't have to chose a curtain video card. You can
>> choose "Unsupported Card", go to the next screen and choose the SVGA server. I
>> have done this. This works, I have the same card.
>
> Why would you do this, when Xconfigurator works with this card as is?
>
> Well, as is in the latest release of X they made available on their
> errata pages for Redhat 5.2 ...
>
I believe some clarification is in order:
When a program like XConfigurator is used, one may be able
to specify one's videocard from the proffered list.
This will put the name of the videocard in the Graphics Device
section of XF86Config and specify the correct X -server to be linked
to X. Great.
However, having the "name" of the videocard in XF86Config
is of no consequence per se.
Specifying and linking the correct X-server to X is, obviously,
crucial.
Thus, choosing " unlisted" or "unsupported" card, but then specifying
the correct X-server is indeed O.K.
Cheers,
--
Howard Mann
http://www.newbielinux.com
(a LINUX website for newbies)
Smart Linuxers search at: http://www.deja.com/home_ps.shtml
------------------------------
From: Burra <[EMAIL PROTECTED]>
Subject: Sonic Vibes sound card: problems!
Date: 11 Jun 1999 17:31:22 GMT
OK here is the problem, Red Hat's sound card config recognizes the sound
card, but when it plays a sound sample I can't here anything. This is what
is in my dmesg log:
sv: unloading
sv: version v0.14 time 17:09:49 May 14 1999
sv: found adaptor at io 0xda00 irq 10 dmaa 0xac00 dmac 0xac10 revision 3
and does it one more time.
in my messages log:
dma timed out?
write: chip lookup? dmasz fragsz 256 count 65536 hwptr64 swptr64
modprobe: can't locate module sound-slot-1
modprobe: can't locate module sound-service-10
I have sound support compiled in the kernel and the sonic vibes driver from
the kernel.
What's wrong anyone know???????
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Re: Rockwell V.90 K56Flex Data, Fax, Speakerphone PCI Modem
Date: 11 Jun 1999 17:31:24 GMT
Richard Natal wrote:
>
> Just tow questions.....
> -How can I do my Rockwell V.90 K56Flex Data, Fax, Speakerphone PCI Modem
run
> under Linux ???
> -Where can I find drivers about my modem ???(for linux of course)
>
> Thanks !
>
I have a Rockwell modem working fine under Linux. Just make sure it's not a
winmodem and your fine. As long as it is a real hardware modem you do not
need drivers, just choose the right com port and you should be fine.
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
** 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
******************************
