Linux-Hardware Digest #440, Volume #13 Thu, 17 Aug 00 22:13:08 EDT
Contents:
Re: Max Number of Serial Ports on one Machine ("John Graves")
Re: overclocking the nightshade serverboard ([EMAIL PROTECTED])
Re: yet another scsi problem (YASP) ([EMAIL PROTECTED])
Re: yet another scsi problem (YASP) (John Mamer)
Re: Max Number of Serial Ports on one Machine (Dances With Crows)
Re: Are there any that will work with the Intel i810? (D G)
Re: VIA SOUND with kernel 2.4.0-test7-pre4 (Glitch)
Re: Yamaha DS-XG and ALSA (Glitch)
Re: linksys question... ("David ..")
Does cdrecord really work with IDE CD-R?? (Arnold Selby)
----------------------------------------------------------------------------
From: "John Graves" <[EMAIL PROTECTED]>
Subject: Re: Max Number of Serial Ports on one Machine
Date: Thu, 17 Aug 2000 23:20:51 GMT
Thanks for the response. I was aware that there are significant hardware
limitations when it comes to putting multiple serial ports/boards into a
Linux-PC. What I am trying to figure out is how many serial ports can be
configured and supported by the Linux Kernel. (NB: we are using a MOXA
N-Port server with 16 serial ports which connects via the ethernet but to an
application, it looks just as if the serial ports were local - the app
accesses /dev/ttypX)
So, then, the question isn't so much "how many serial ports/boards can I
stuff into a x86 machine", but rather how many serial ports (aka /dev/ttyXX)
can the Linux Kernel support? Is there a kernel parameter or something that
I can look at?
Thanks again for the help.
John
"James Richard Tyrer" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> John Graves wrote:
>
> > Hello -
> >
> > I am trying to figure out the maximum number of serial ports that I can
> > configure on a Linux box. Is there any utility, kernel parameter,
kernel
> > source file, etc. that would tell me this information? Any assistance,
or
> > pointing in the right direction would be greatly appreciated.
> >
> > Thanks
> > John
>
> This is going to depend on the hardware you are using for serial ports.
ISA
> is limited to 4 or 6 and you cant't use them all at once unless you can
> assign seperate interupts. PCI will use more, but you will run out of PCI
> slots. Therefore, you need to lookinto one of the multiport boards that
are
> supported and determine if it is possible to use more than one of these
in
> your system.
>
> JRT
>
------------------------------
Subject: Re: overclocking the nightshade serverboard
From: [EMAIL PROTECTED]
Date: Thu, 17 Aug 2000 23:28:34 GMT
"Rob Love" <[EMAIL PROTECTED]> writes:
> I have a nightshade serverboard that is running 2 PII 400s. I would like
> to overclock them, but I can't seem to find any support or info on the web.
Why?
What are you doing that requires the bleeding-edge performance that 50
or 60MHz extra (apiece) will get you? Is it worth damaging or
destroying processors which are no longer manufactured and must be
bought used, or via a place like eBay?
I overclocked my P3-450 for a day to see what the fuss was all about.
What I discovered was that Windows NT would BSOD at random intervals,
and when it was actually working I noticed zero performance
improvements.
Considering that you have a cheap (about $300 for two processors)
upgrade path to 100MHz FSB P3's on that board, I really don't
understand why you want to overclock.
> I assume that it's because most users don't have this board and any
> company that buys the server isn't going to overclock it. Anyone
> have any suggestions?
Yes. Don't overclock. And for the love of God, don't listen to those
idiots running dual Celeron 333's at 1000MHz apiece. There are
reasons beyond a desire to get some more of your money that Intel
doesn't suggest overclocking CPUs, and why no serious machines are
_ever_ overclocked.
--
Eric McCoy ([EMAIL PROTECTED])
"hand, n. A singular instrument worn at the end of a human arm and
commonly thrust into somebody's pocket." - Ambrose Bierce
------------------------------
Subject: Re: yet another scsi problem (YASP)
From: [EMAIL PROTECTED]
Date: Thu, 17 Aug 2000 23:34:56 GMT
John Mamer <[EMAIL PROTECTED]> writes:
> I have a new external seagate ST318436LW (18GB) connected to my
> Adaptec AHA2940U/UW controller. The bios sees the disk drive, but linux
> can't get to it during the boot. I keep getting the message
> "Host 0 Abort (PID 4) timed out- resetting" over and over. I've used
> the controller successfully with my Jaz drive. I've removed everything
> from the SCSI bus, manually set the card termination to "on" (both low
> half and high half) and put a hardware terminator on the external
> drive. I'm suspecting either the cable or the terminator, but would
> greatly appreciate advice.
Do you have anything else on the bus? It may be that you should
actually have termination on the card off, depending on your
configuration. If you have wide and narrow SCSI devices on the same
chain, I think you have to put them in order and terminate high at the
narrow device. If you have a funky situation like two sockets for one
channel (say, one wide and one narrow), you may have to deal with
perverse termination rituals, like active high termination.
Someone or other (fairly) recently posted the voltages you should be
getting from a properly-terminated chain. If you have an accurate
voltmeter handy, you can check that out.
--
Eric McCoy ([EMAIL PROTECTED])
"hand, n. A singular instrument worn at the end of a human arm and
commonly thrust into somebody's pocket." - Ambrose Bierce
------------------------------
From: John Mamer <[EMAIL PROTECTED]>
Subject: Re: yet another scsi problem (YASP)
Date: Fri, 18 Aug 2000 00:06:48 GMT
At this point, I've taken everything off the bus. There are no internal scsi
devices, the only external device is the hard disc. I've manually told the
controller to terminate internally (I also fiddled around with the hi/low
on/off settings to no avail) and a physical terminator on the external drive.
Just for laughs, I tested the disc drive on my Windoze NT machine. It seems
to work fine using my 8945 card (2940 drivers). To test it, I did have to use
a different cable.
thanks
john
[EMAIL PROTECTED] wrote:
> John Mamer <[EMAIL PROTECTED]> writes:
>
> > I have a new external seagate ST318436LW (18GB) connected to my
> > Adaptec AHA2940U/UW controller. The bios sees the disk drive, but linux
> > can't get to it during the boot. I keep getting the message
> > "Host 0 Abort (PID 4) timed out- resetting" over and over. I've used
> > the controller successfully with my Jaz drive. I've removed everything
> > from the SCSI bus, manually set the card termination to "on" (both low
> > half and high half) and put a hardware terminator on the external
> > drive. I'm suspecting either the cable or the terminator, but would
> > greatly appreciate advice.
>
> Do you have anything else on the bus? It may be that you should
> actually have termination on the card off, depending on your
> configuration. If you have wide and narrow SCSI devices on the same
> chain, I think you have to put them in order and terminate high at the
> narrow device. If you have a funky situation like two sockets for one
> channel (say, one wide and one narrow), you may have to deal with
> perverse termination rituals, like active high termination.
>
> Someone or other (fairly) recently posted the voltages you should be
> getting from a properly-terminated chain. If you have an accurate
> voltmeter handy, you can check that out.
>
> --
> Eric McCoy ([EMAIL PROTECTED])
>
> "hand, n. A singular instrument worn at the end of a human arm and
> commonly thrust into somebody's pocket." - Ambrose Bierce
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: Max Number of Serial Ports on one Machine
Date: 18 Aug 2000 00:20:29 GMT
Reply-To: [EMAIL PROTECTED]
On Thu, 17 Aug 2000 23:20:51 GMT, John Graves wrote:
>Thanks for the response. I was aware that there are significant hardware
>limitations when it comes to putting multiple serial ports/boards into a
>Linux-PC. What I am trying to figure out is how many serial ports can be
>configured and supported by the Linux Kernel. (NB: we are using a MOXA
>N-Port server with 16 serial ports which connects via the ethernet but to an
>application, it looks just as if the serial ports were local - the app
>accesses /dev/ttypX)
>
>So, then, the question isn't so much "how many serial ports/boards can I
>stuff into a x86 machine", but rather how many serial ports (aka /dev/ttyXX)
>can the Linux Kernel support? Is there a kernel parameter or something that
>I can look at?
/usr/src/linux/Documentation/devices.txt seems to say that you can put 255
serial ports in, if you have 4 64-port serial cards. Stallion makes
64-port serial cards, and the devices run from /dev/ttyE0 to
/dev/ttyE255. You will have to create most of those device nodes using
MAKEDEV or mknod, and you will probably have to compile a module to make
use of Stallion cards.
wrt the old-style ttyp* devices, they've been replaced for the most part
by the /dev/pts filesystem. There's an option for the kernel at compile
time to determine whether /dev/pts/ can be used as a filesystem, and
the maximum number of pseudo-terminals available. Default is 256; you
can bump it up to 2048 (that should be enough, right?)
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Those who do not understand Unix are
http://www.brainbench.com / condemned to reinvent it, poorly.
=============================/ ==Henry Spencer
------------------------------
From: D G <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Are there any that will work with the Intel i810?
Date: Thu, 17 Aug 2000 17:31:42 -0700
Hammer wrote:
>
> In article <[EMAIL PROTECTED]>,
> D G <[EMAIL PROTECTED]> wrote:
> > I just got back from the LinuxWorldExpo. I was surprised to see that
> > Intel had a booth there specifically for linux on the i810
> > motherboards. The guy there recommended XFree86 4.0.1 and the 2.4.0
> or
> > 2.3.99whatever_the_latest kernel. Otherwise, the 2.2.16 kernel should
> > be OK, although he recommended using the XFCom server from
> > support.intel.com with that kernel.
> >
> > There were two modules that the i810 relies on, one is the agpgart,
> and
> > I can't remember the other. But the other module will be in the 2.4
> > kernel and only partially exists in the 2.2.16 kernel. (I'm just
> > paraphrasing what I remember.)
> >
>
> Interesting, thanks.
>
> I have been told (not yet tried it) that you can disable the onboard
> sound in the bios, then put in your own PCI sound card and skip AC'97
> completely. Been meaning to try it. I might go the ALSA route too, but
> I'm not relishing the 4-6 hours I suspect it will take, after watch the
> [alsa-users] mailing list for a couple weeks :( Or, maybe I'm just a
> wimp :)
It took me about a half-hour to download, compile, and setup. Very
easy. The hardest part was automating it in /etc/conf.modules. I
didn't have any problems with the driver itself.
--
DG
e-mail is: [EMAIL PROTECTED]
(remove the Z's--they're what I do when I read SPAM!)
------------------------------
Date: Thu, 17 Aug 2000 21:14:18 -0400
From: Glitch <[EMAIL PROTECTED]>
Subject: Re: VIA SOUND with kernel 2.4.0-test7-pre4
are u using OSS or ALSA?
Michael Meding wrote:
>
> Hi everyone,
>
> I have a problem with my onboard sound on my MSI-6330 K7TPRO.
>
> The thing is xmms plays *wav's and mp3's fine, but Quake3 and DOOM
> LEGACY do not rpoduce anything. Also different OpenGL demos do not
> deliver sound, the sound files are played solely by xmms fine.
>
> Doom Legacy jumps at me with sequencer not configured. So I guess it is
> just some messed up link or configuration somewhere.
>
> Any clues anyone ?
>
> Thanks in advance.
>
> Greetings
>
> Michael
>
> PS Mailing me would be nice
------------------------------
Date: Thu, 17 Aug 2000 21:20:12 -0400
From: Glitch <[EMAIL PROTECTED]>
Subject: Re: Yamaha DS-XG and ALSA
As i'm not using linux right now i can't help you as much as i normaly
could. All I can say is I followed everything the instructions said for
my particular situation (sony vaio with ds-xg sound card) and had
everything working except programs which used OSS drivers. For those
programs to work someone said i still needed OSS emulation turned on
with ALSA. AFter I did that everything worked fine.
I vaguely remember having a lot of unresolved symbol problems but I
think it ended up being caused by using insmod snd-ymfpci instead of
modprobe..as modprobe by definition(i assume) probes for all modules
that are needed to make sure everything that is needed is loaded the
first time.
HTH
Brandon
alien8 wrote:
>
> I recently purchased Corel Linux and I've got it running in a dual-boot with
> Win98. This computer's motherboard is Intel's MU440EX which has the Yamaha
> DS-XG onboard PCI audio.
> I've followed linuxnewbie.org's ALSA installation how-to, and everything
> went great up until the command 'modprobe snd-ymfpci' which brought up the
> following error:
>
> ----------------------------------------------------------------------------
> ----
>
> /lib/modules/2.2.12/misc/snd.o: unresolved symbol printk_Rsmp_1b7d4074
> soundcore: Device or resource busy
> snd-mixer: No such file or directory
> snd-mixer: No such file or directory
> snd: No such file or directory
>
> ----------------------------------------------------------------------------
> ----
>
> I tried rebooting anyways and I get to the log-in screen fine, but as soon
> as it starts loading KDE, everything freezes up and the only solution is to
> 3-finger salute, and then reboot to console and rename /etc/conf.modules to
> something else and rename the backup.
>
> Ive tried the following versions of the ALSA drivers:
> 0.5.8
> 0.5.9
> 0.5.9a
> and I always get the same results. I've also tried opening up the Task
> Manager and sending the Kill Signal to all open audio processes (kwmsound
> and kaudioserver) which were marked as Sleep and Zombie respectively, but to
> no avail.
>
> I've searched all the forums here with no luck, as well as read all of
> ALSA's documentation and I've come to the following decision: If I can't get
> it to run I'll have to go out and buy an old SB ISA card, but I'd really
> rather get this one working... Any suggestions?
>
> -------
> alien8
> www.planetunreal.com
> www.planetunreal.com/modcentral
------------------------------
From: "David .." <[EMAIL PROTECTED]>
Subject: Re: linksys question...
Date: Thu, 17 Aug 2000 20:19:04 -0500
Frederick Artiss wrote:
>
> Oh gosh, good luck! You chose the one network card that people have been
> having problems with in linux. The problem is that Linksys changed their
> card and the linux people haven't had a chance to write a good driver
> for it. (Probably it's a tulip). I've heard some people have had luck
> recompiling the kernel with lines supplied by Linksys, but I found it
> less trouble to just replace the card -- especially if you're new to the
> OS.
I use Linksys cards in 3 of my systems and they work great. I did have
to replace the old module and recompile the tulip module that came with
the card in order to get it to work.
If you replace the card I would replace it with a 3Com card instead of
some off the wall brand. My 3Com cards also work great!!
--
Confucius say: He who play in root, eventually kill tree.
Registered with the Linux Counter. http://counter.li.org
ID # 123538
------------------------------
Date: Thu, 17 Aug 2000 21:47:27 -0400
From: Arnold Selby <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,mailing.comp.cdwrite
Subject: Does cdrecord really work with IDE CD-R??
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.
I have tried cdrecord 1.8, 1.9 and 1.10a; and have also upgraded my
drive to the latest firmware on my mitsumi
the annoyance and frustration is that cdrecord will fail (almost)
every time, but it will fail writing different
tracks. Once (only once), it did dummy-write all tracks, but on a
retest, it failed.
If it would fail consistantly, i would have given up sooner
I can dual boot and have successfully written CD's with the same
hardware under Window's 98
( yeahhh, i know)
my testing script is:
uname -a
dmesg|egrep -i "amd|233|Memory:.*kern"
eject
eject -t
START=$(date)
date
nice -10 ~/cdrecord/cdrecord -v speed=1 dev=0,00,0 -fs=6m -dummy
-audio ~/tmp_cd/track0*.cdda.wav
echo START=$START
date
speed=2 guarantess a quicker failure
I run this in console mode (no X windows) or even init S, and nice it
to make sure cdrecord gets all the time it needs
the failures are similar:
Linux 2.2.14 #2 SMP Sun Jul 30 14:36:14 EDT 2000 i586 unknown
CPU0: AMD AMD-K6tm w/ multimedia extensions stepping 02
Detected 233867666 Hz processor.
Memory: 63272k/65536k available (904k kernel code, 416k reserved, 888k
data, 56k init)
Sun Aug 13 14:33:11 EDT 2000
Cdrecord 1.9 (i586-pc-linux-gnu) Copyright (C) 1995-2000 J�rg Schilling
TOC Type: 0 = CD-DA
scsidev: '0,00,0'
scsibus: 0 target: 0 lun: 0
Linux sg driver version: 2.1.36
Using libscg version 'schily-0.1'
atapi: 1
Device type : Removable CD-ROM
Version : 0
Response Format: 1
Vendor_info : 'MITSUMI '
Identifikation : 'CR-2801TE '
Revision : '1.10'
Device seems to be: Philips CDD-522.
Using generic SCSI-3/mmc CD-R driver (mmc_cdr).
Driver flags : SWABAUDIO
FIFO size : 6291456 = 6144 KB
Track 01: audio 78 MB (07:44.53) no preemp
Track 02: audio 106 MB (10:33.29) no preemp
Track 03: audio 41 MB (04:09.44) no preemp
Track 04: audio 112 MB (11:10.20) no preemp
Track 05: audio 72 MB (07:11.40) no preemp
Track 06: audio 68 MB (06:45.06) no preemp
Track 07: audio 91 MB (09:04.36) no preemp
Total size: 573 MB (56:50.29) = 255772 sectors
Lout start: 574 MB (56:52/22) = 255772 sectors
Current Secsize: 2352
ATIP info from disk:
Indicated writing power: 5
Is not unrestricted
Is not erasable
ATIP start of lead in: -11325 (97:31/00)
ATIP start of lead out: 336225 (74:45/00)
Disk type: Long strategy type (Cyanine, AZO or similar)
Manuf. index: 22
Manufacturer: Ritek Co.
Blocks total: 336225 Blocks current: 336225 Blocks remaining: 80453
RBlocks total: 345460 RBlocks current: 345460 RBlocks remaining: 89688
Starting to write CD/DVD at speed 1 in dummy mode for single session.
Waiting for reader process to fill input buffer ... input buffer ready.
Starting new track at sector: 0
Track 01: Total bytes read/written: 81943680/81943680 (34840 sectors).
Starting new track at sector: 0
Track 02: Total bytes read/written: 111712944/111712944 (47497 sectors).
Starting new track at sector: 0
Track 03: Total bytes read/written: 44001216/44001216 (18708 sectors).
Starting new track at sector: 0
Track 04: Track 04: 112 of 112 MB written (fifo 98%).
Track 04: Total bytes read/written: 118223280/118223280 (50265 sectors).
Starting new track at sector: 0
Track 05: Track 05: 72 of 72 MB written (fifo 100%).
Track 05: Total bytes read/written: 76098960/76098960 (32355 sectors).
Starting new track at sector: 0
Track 06: 4 of 68 MB written (fifo 98%).
/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)
write track data: error after 5016816 bytes
Sense Bytes: 70 00 00 00 00 00 00 06 00 00 00 00 00 00 44 FF 5D FA
Writing time: 2521.141s
Fixating...
WARNING: Some drives don't like fixation in dummy mode.
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
/root/cdrecord/cdrecord: fifo had 6980 puts and 6885 gets.
/root/cdrecord/cdrecord: fifo was 0 times empty and 5292 times full, min
fill was 93%.
START=Sun Aug 13 14:33:11 EDT 2000
Sun Aug 13 15:15:42 EDT 2000
help will be greatly appreciated
arnold
------------------------------
** 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
******************************
