Linux-Misc Digest #464, Volume #27 Tue, 27 Mar 01 23:13:02 EST
Contents:
HELP ME!: Aiptek WEBCAM controls dont't work under Linux (OV511+/OV7620 (CYBERYOGI
=CO= Windler)
Re: Partitions and Sizes (Floyd Davidson)
Re: bash: what shell variable to set to search for shared libraries? (E J)
Re: man files for Java (E J)
Re: after Pan newsreader installed....... (Jan Schaumann)
Re: Getting my HP DeskJet 820Cse to work (Biogod)
Slackware runs on Yopy, the first Linux PDA in the world. ("jina")
Re: ALSA and Crystal Semiconductor CODECS: CS4237B and Suse 7.0 (Dances With Crows)
Re: On-demand dialing (Frank Hahn)
Re: Making .bat-file in linux (Dowe Keller)
Yahoo reports a Windows/Linux virus (Robert Lynch)
----------------------------------------------------------------------------
From: CYBERYOGI =CO= Windler <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,comp.os.linux.setup,linux.debian.user,redhat.config
Subject: HELP ME!: Aiptek WEBCAM controls dont't work under Linux (OV511+/OV7620
Date: Wed, 28 Mar 2001 04:20:40 +0200
I need to run 2 Aiptek "HyperVcam Mobile" USB webcams under Debian Linux 2.2
with kernel 2.4.1 on a PC with a DFI P5BV3+ Mainboard(rev.B2),an AMD K6 300MHz
CPU, 128MB RAM and a Venus Virge 2MB graphics card.
I am student of software-techniques and for my thesis I plan to develop a
sing- and gesture controlled music synthesizer with a 3D/VR user interface,
therefore I urgently need to get the cameras working.My problem is that the
webcam picture controls (brightness,contrast,hue etc.) in programs like
"xawtv" and "gqcam" seem to be completely ignored by the webcam driver and I
only get a framerate of about 2..4 fps. Sometimes the USB driver seems to
crash with results in pixel rubish spread all over the picture.
At the beginning also under Windoze98 I got tons of colourful pixel rubbish;
after some experiments with my USB port I discovered that connecting a Zobel
network (capacitor in series to resistor) as a terminator from pin "data+" to
GND removed the rubbish. Capacitors much >2nF(?) prevented the correct detec-
tion of the camera (only an "unknown USB device" was found); ones <100pf were
too small to get rid of all the rubbish. I don't remember well whether the
control sliders ever worked in Windoze before, but actually I have soldered
150pf in series to 560Ohm as well from "data+" to GND as from "data-" to GND
- the latter because I hoped it could fix the controls problem under linux,
but it didn't. The Windows ones work now,alsthough the automatic brightness
and aperture settings seem to re-activate themself sometimes spontaneously.
(After powerup the camera often needs to be re-plugged to be recognized cor-
rectly - I guess this is just a power supply problem which prevents the camera
from resetting well while all drive motors start spinning during power on.
The power supply is a 300W AT one of the brand "Fortron".)
When I start "xawtv" or "gqcam" under Linux, I just get a dim picture which is
slowly fading brighter and more colourful to a quite pale webcam picture,but
the picture control sliders don't work. In xawtv sometimes blue/green snow
appears during motion of its sliders, and afterwards the picture fades from
dim to bright again (because the webcam is resetted?), but the slider values
have no effect. After some time (or some program restarts?) the picture stopps
responding on slider movement at all. In gqcam when the automatic brightness
setting is on I can see the brightness slider move left and right,but the
framerate drops to 2 fps or so and the output picture is chopped by many pure-
ly white frames flickering through it. When I turn the automatic brightness
setting off, the white flickering is not present and manual slider operation
either restarts that dim-to-bright fading or does nothing.
Generally the outputted values for contrast,brightness etc. in the "camera
info" window stay the same and don't match the values set by the sliders,which
lets me conclude that the ov511 driver either never gets or ignores the values
sent from the sliders. When automatic brightness in gqcam is on, the webcam
picture seems to stay at a fixed brightness(without fading),but possibly this
rather happens because this option continuesly resets the webcam by the suc-
cessless attempt of sending brightness values to it than because it reponds to
the brighness commands properly. When the automatic brightness setting of the
ov511 driver module is turned off manually from the insmod command line, the
fading dim-to-bright effect doesn't occure,but the slider values remain ig-
nored.
When I set the picture size in gqcam to "quater", I get instead of a downsized
picture in 160x120(?) resolution just the upper left quater of the 320x240(?)
picture.Astonishingly this mode is displayed at a reasonable framerate, but
otherwise it is now in b/w while the blue colour bitplane(?) is scrolling/fli-
ckering through it like a poor TV reception interference of 2 mixing TV sta-
tion images.In the source code of the "ov511.c" driver in my kernel sources
2.4.1 I found for this mode the comment "FIXME",though this is likely a known
bug.
The Aiptek "HyperVcam Mobile" is reported by "gqcam" to be a OV511+ with a
OV7620 CCD chip. In the source code of the kernel's ov511 driver the controls
settings section of the OV7620 chip looks quite strange;apparently the value
of a multiplexed register is read, then a portion of its old value is added to
the new one and the result is written back to it or something like that. The
code from the OV7610(?) looks much more straight-forward. May the OV7620 code
be buggy or incomplete? Does anybody know why the camera sometimes makes pixel
rubbish under linux while it other times runs well for hours? Unloading and
re-inserting the ov511 kernel module doesn't help - only rebooting. (The rea-
son for the lousy framerate is likely my Virge graphics card,which higher
functions seem to be not supported by the Xfree86 driver.I will anyway re-
place that slow card soon because also its RAM seems to be a little faulty and
causes other kinds of pixel rubbish in certain situations.)
PLEASE tell me if the non-working camera controls in Linux are a bug in the
ov511 driver and if there is a fix for it. I urgently need to get these ca-
meras to work in a stabile way for the programming project of my thesis.
MAY THE SOFTWARE BE WITH YOU!
*============================================================================*
I CYBERYOGI Christian Oliver(=CO=) Windler I
I (teachmaster of LOGOLOGIE - the first cyberage-religion!) I
I ! I
*=============================ABANDON=THE=BRUTALITY==========================*
{http://www.informatik.fh-hamburg.de/~windle_c/e_index.html}
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: Partitions and Sizes
Date: 27 Mar 2001 16:33:46 -0900
Jason Green <[EMAIL PROTECTED]> wrote:
>Floyd Davidson <[EMAIL PROTECTED]> wrote:
>
>> if the intent to begin with is to use symlinks,
>> a conservative in sizing partitions can be taken (with some
>> exceptions), and of course several more partitions than appears
>> necessary can be used to advantage.
>
>> Remaining disk space, or new disk space if
>> another is added, are divided into /u1, /u2, /u3, /u4, etc.
>>
>> I typically make /home/floyd a symlink to something like
>> /u4/floyd (and do the same for any other user that has
>> significant disk requirements), and allow that entire partition
>> to be used by a single user.
>
>I don't quite follow the logic in this. That is, while I understand the
>logic in splitting the disk into partitions, and even giving over a whole
>partition to one user, I don't see how using symlinks makes any practical
>difference.
>
>For the example you gave, why wouldn't you simply keep /home/floyd as a
>mount point, and mount the partition there rather than at /u4?
In that particular case, it probably wouldn't make any
difference at all! But then, the person who owns that userid
might also have other userids, and they could all be on that one
partition. (Or whatever happens to work!)
>You still need to decide on a partition size, and you still need to
>transfer the contents onto the new partition. And on top if that, you
>fill the root directory with mount points.
Fill the root directory with mount points??? :-)
Actually, in that case there is a /home partition, so the
portions of /home that get farmed out to other partitions are
not mounted on the root partition.
>The only reason I can think of is the order of mounting partitions from
>/etc/fstab becomes important.
>
>(I just know you are going to say the answer is in what I snipped ;-)
Nope. I just edited the article down a little more than I should
have, and didn't notice it until you pointed it out.
>> The reason for using three swap partitions is that occasionally
>> a nice small clean partition with nothing on it comes in handy,
>> and one or two of the swap partitions can be temporarily
>> formatted and used, while still allowing the system to have at
>> least some swap space.
>
>This is a useful tip, and something I'll bear in mind next time
>I have to partition a disk - thanks.
--
Floyd L. Davidson <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska) [EMAIL PROTECTED]
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: bash: what shell variable to set to search for shared libraries?
Date: Wed, 28 Mar 2001 02:19:47 GMT
export LD_LIBRARY_PATH=<blah>
or
LD_LIBRARY_PATH=<blah>
export LD_LIBRARY_PATH=<blah>
"P. Eric Chi" wrote:
> Hi,
>
> In csh, we can "setenv LD_LIBRARY_PATH ..." to set the path to search for
> shared libraries. What shell variable should I set in bash? Thanks very
> much,
>
> Eric.
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: man files for Java
Date: Wed, 28 Mar 2001 02:31:58 GMT
In java1.3 the man pages are
appletviewer.1 jarsigner.1 javadoc.1 jdb.1 rmic.1
serialver.1
extcheck.1 java.1 javah.1 keytool.1 rmid.1
tnameserv.1
jar.1 javac.1 javap.1 native2ascii.1 rmiregistry.1
To set the man page
in $HOME/.bash_profile add the following line
export PATH=/usr/java/jdk1.3/bin:$PATH
Why do you need to convert HTML for man pages? I use the file manager or
netscape and bookmark the HTML
pages. Very easy for me.
Eric Tetz wrote:
> Greetings...
>
> I want to do some Java development on Linux. My editor will do a man page
> lookup on a highlighted word, which really comes in handy when I'm
> programming in C, but is totally worthless for Java, as the Java docs are
> in HTML format.
>
> Are there manpage version of the Java does available, or is there a tool
> that can convert JavaDoc generated HTML into manpages?
>
> Cheers,
> Eric
------------------------------
From: [EMAIL PROTECTED] (Jan Schaumann)
Subject: Re: after Pan newsreader installed.......
Date: Wed, 28 Mar 2001 02:42:07 GMT
* Glitch wrote:
> I downlaoded glib/gtk 1.2.8 , after already having glib/gtk 1.2.6
> installed, b/c I needed the newer versions for Pan 0.9.6.
>
> I program in GTK and I wanted to make sure my programs still worked ok
> with the new version of GTK installed. I compiled one of my programs
> again and it compiled just fine, but I tried to execute it and got
> warnings.
>
> This is the warning:
> Gtk-WARNING **: Unable to locate loadable module in module_path: "libpixmap.so"
>
> My program still loads fine but all the theme stuff is missing. Basically
> the window is plain colored b/c the library above coudln't be found.
> HOwever the library above does exist on the system. It's at:
>
> /usr/lib/gtk/themes/engines/libpixmap.so
>
> This is the location of the old GTK, 1.2.6. I checked in the directory
> that the new GTK is in but I don't have the libpixmap.so file. So what
> can i do to get the new GTK to see the old lib file? Or should I do
> something else to get my GTK programs to be able to be affected by my
> current theme?
Some wild guesses:
The new GTK has been made the default location. If the new GTK does not
bring a libpixmap.so, you could try linking your application with this
library by specificing it as an argument to the linker. Or you could
try to copy the library from there to the new library-diretory. Or you
could create a symlink from the new directory to the old directory.
-Jan
--
Jan Schaumann <http://www.netmeister.org>
unless ($humorEnabled || $canDetectSarcasm)
{printf "Please add Smileys where appropriate.";}
------------------------------
From: Biogod <[EMAIL PROTECTED]>
Subject: Re: Getting my HP DeskJet 820Cse to work
Date: Tue, 27 Mar 2001 21:31:18 -0500
Daniel Bostwick wrote:
> Hi,
> My problem is that I cannot get my HP DeskJet 820Cse to work under Linux. I
> have Mandrake 7.2 running kernel version 2.2.15. I am able to get a networked
> printer to work but not this one. I have tried they included print tool Printer
> Drake and the only driver that gets any acivity out of my printer is a pnm2ppa
> driver, and that causes the lights to blink. I have tried to run pdq but have
> not been able to understand it. Any help would be appriciated.
> Thanks,
> 12-year old Linux Enthusiast
>
> ---------------------------------------------------------
> 1 Linux Box
> 1 Dual Boot
> 1 Winders Box
I have the same printer as you. My mistake for buying it. The 820Cse printer
uses a closed protocol that HP won't release so that we can have a nice driver for
it. In order to set it up, I ran printtool as superuser. When printtool runs
click on the choose filter button and select the 820Cse print filter. It should
work...at least it did for me. I'm running RedHat 7.0...I'm not sure if that
would make any difference, though
------------------------------
From: "jina" <[EMAIL PROTECTED]>
Subject: Slackware runs on Yopy, the first Linux PDA in the world.
Date: Wed, 28 Mar 2001 12:00:18 +0900
Dear Sir or Ma'am
We, at G.Mate Inc., have developed and manufactured PDA based on Embedded
Linux OS since its
foundation in November of 1998.
The people from all of world paid great attention to our world-first
multimedia PDA by the name of YOPY
when it was exhibited in the CeBIT Show in Hanover of Germany and in the
Comdex/Fall in Las Vegas of
USA in February and November of 2000 respectively. Ever since we have been
devoting all our energy to
developing PDA integrating either Cellular module or Bluetooth module and
to the mobile communication
IMT-2000 PDA business for the next generation
In order to make Linux supporters and developers enthusiastically
participate in enlarging and developing
the environment of Linux and applied programs, we have started to sell the
development kit of multimedia
PDA based on Linux from January 21, 2001 through our website
(www.gmate.co.kr) and many people of the
world have shown their interests and encouragement generously.
The YDK product includes YOPY hardware, software environment, Linux Source
Package Solution, etc. By
sharing with you solutions and knowledge related to Embedded Linux PDA we
have developed so far, we
expect that there would be absolutely outstanding realization of desirable
Linux environment and
development of invaluable Linux PDA in the near future.
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Crossposted-To: comp.os.linux.setup
Subject: Re: ALSA and Crystal Semiconductor CODECS: CS4237B and Suse 7.0
Reply-To: [EMAIL PROTECTED]
Date: 28 Mar 2001 03:00:35 GMT
On Wed, 28 Mar 2001 01:44:48 -0000, Christopher W. Aiken staggered into
the Black Sun and said:
>On Tue, 27 Mar 2001 21:48:27 GMT, Keith Marjerison
><[EMAIL PROTECTED]> wrote:
>>1st, I'm a Linux newbe, so go eazy. How do I go about getting Suse to
>>recognize my TidalWave128 ISA sound card. It is PnP and the 'isadump'
>>utility :)sees the card but when I try and use 'YaST2' to install
>>'Generic' support for :)the 'CS42*' and the program stops saying the
>>'kernel' does not support it. :)What am I doing wrong? What is a good
>>reference for Linux ? i.e. Linux :)Unleashed? Thanks for any help in
>>advance.
>
>I got my CS4232 up and running by adding the following lines
>to my /etc/rc.d/boot.local file. The run the boot.local file
>or reboot.
>
>echo "Setup CS4232 Sound"
>modprobe soundcore
>modprobe sound
>modprobe ad1848
>modprobe uart401
>modprobe cs4232 io=0x530 irq=7 dma=1 dma2=3 mpuio=0x330 mpuirq=9
SuSE 7.x uses ALSA, not the kernel OSS drivers, by default. Chris gave
the OSS solution, which should work provided you can find the right
values for all the parameters. Every cs42xx I've seen has used IRQ 5,
for instance, and "0" for dma2, but try various things... the worst you
can do is lock up the system. Keith, have you tried "alsaconf"?
As for a good Linux book, _Running Linux_ by Matt Walsh (published by
O'Reilly and Associates) is a pretty good low-to-mid-level book. If you
got the "Professional" SuSE distro, you should have a pretty
comprehensive set of manuals sitting around. (The manuals are much less
comprehensive in the "Personal" distro. Sigh.) The following sites can
be very useful for Linux users:
http://linuxdoc.org/HOWTO/
http://linuxnewbies.org/
file:/usr/share/doc/packages/
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (Frank Hahn)
Subject: Re: On-demand dialing
Date: Wed, 28 Mar 2001 03:10:02 -0000
Reply-To: [EMAIL PROTECTED]
On Tue, 27 Mar 2001 17:12:11 -0500, Chris Coyle <[EMAIL PROTECTED]> wrote:
>
> "Dennis" <[EMAIL PROTECTED]> wrote in message
> news:99pu0u$nf6$[EMAIL PROTECTED]...
> > I have set up IP masquerade on Red Hat Linux 7.
> >
> > From the server and other Windows workstations I can browse the web. No
> > problem.
> >
> > However I need to manually establish the ppp dial-up connection from the
> > server each time a Windows machine wants to connect.
> >
> > I want the workstations to be able to launch that ppp connection
> > automatically from their browser(& mail client).
>
> I did this on RH6.2 by downloading pppsetup from slak,
> but then I discovered you can set it up through some
> optional part of linuxconf. I don't know if its the
> same on RH7 but I would advise you to browse through
> EVERY branch of linuxconf to see if its there.
>
There may be other ways around your problem. The diald package at
one time included a script that could be used to control diald. I
use to export this so that it displayed on the Windows machine. You
could use this to stop and start diald instead of waiting for it
to time out.
Another option was a program called dialmon. It was a Windows client
that could also control diald. Take a look at issue 33 of the Linux
Gazette. The address is http://www.linuxgazette.com.
The address for dialmon is: http://www.quaking.demon.co.uk/dialmon.html
--
Frank Hahn
There is no such thing as fortune. Try again.
------------------------------
From: [EMAIL PROTECTED] (Dowe Keller)
Subject: Re: Making .bat-file in linux
Date: 27 Mar 2001 19:28:00 -0800
On Tue, 27 Mar 2001 12:56:30 GMT, Lew Pitcher <[EMAIL PROTECTED]> wrote:
>On 26 Mar 2001 22:33:01 GMT, Jeremiah DeWitt Weiner
><[EMAIL PROTECTED]> wrote:
>
>>Dowe Keller <[EMAIL PROTECTED]> wrote:
>>> Let me explain. Under Un*x type OSes, if the first line of a file
>>> contains:
>>> #!/usr/bin/foo
>>> Then the shell passes the file to the program "foo", living in the
>>> /usr/bin directory. The program "foo" then interprets the file.
>>
>> Dowe, I knew that. In fact, if you look at earlier articles, you'll
>>see I posted that exact fact. I was asking why /bin/bash is used for files
>>which DO NOT contain the #! line.
>
>I believe that it's not /bin/bash in particular, but is instead the
>default shell for your session that is used on scripts that do not
>contain the #! line.
>
>In other words, if you change your default shell to ksh, then it will
>be ksh that will interpret the script by default. Likewise, if you
>change to zsh, then zsh interprets the script.
Exactly, I mentioned bash because that was the shell that the OP
mentioned using, and I wanted to make it clear that this was shell
dependent, rather than a feature of the OS. AFAIK, However, all
UN*X shells exhibit this behavior.
--
[EMAIL PROTECTED] http://www.sierratel.com/dowe
WARNING - the content of this message may be erroneous, misspelled and
perhaps even flammable. It also contains small parts that could cause
asphyxiation. NOT RECOMMENDED FOR CHILDREN UNDER 3 YEARS OF AGE
------------------------------
From: Robert Lynch <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Yahoo reports a Windows/Linux virus
Date: Tue, 27 Mar 2001 19:55:12 -0800
"First Virus to Infect Both Windows, Linux Emerges"
http://dailynews.yahoo.com/h/nm/20010327/wr/virus_winux_dc_1.html
==
By Eric Lai
SAN FRANCISCO (Reuters) - A computer virus that can infect PCs
running either the ubiquitous
Windows operating system or the increasingly popular Linux (news
- web sites) operating system
emerged on Tuesday, which its discoverers say is a world first.
The virus, dubbed ``W32.Winux'' by the company that first
reported it, anti-virus firm Central
Command, is not destructive and does not appear to have infected
any computers yet.
Still, the virus sets a disturbing precedent.
``We didn't think this was possible,'' said Keith Peer,
President and Chief Executive of Medina,
Ohio-based Central Command. ``It's a real step forward for virus
writers.''
Another anti-virus maker, McAfee.com Corp. (NasdaqNM:MCAF -
news), said it had not seen the virus and could not confirm
reports of W32.Winux.
W32.Winux spreads by infecting executable programs that run
either on later versions of Windows from Microsoft Corp.
(NasdaqNM:MSFT - news) -- including 95, 98, Me, NT and 2000 --
or the various flavors of Linux, a free operating system that is
gaining ground among techies and businesses.
Users can set off the dormant virus by either double-clicking on
an infected program or e-mail attachment. After it is activated,
the
virus automatically searches for all nearby Windows or Linux
applications of at least 100 kilobytes in size, which it then
proceeds to
infect.
Central Command, which first received the virus via an anonymous
e-mail originating in the Czech Republic early Tuesday
afternoon, said a virus writer named Benny claiming affiliation
to a known group of virus writers called 29A, appears to be the
culprit.
Benny and 29A have been implicated as being behind a number of
other previous viruses that have been considered technically
innovative but not particularly destructive.
In late 1999, a virus that masqueraded as a fix for the
Millennium Bug made its way around the Internet. The 29A group
claimed
credit for that virus, as well as another one emerging last
September called ``Stream'' which experts said was particularly
clever at
disguising itself from anti-virus software.
The W32.Winux virus is written in a primitive computer language
called 'assembly language', which is what allows it to infect
either
Windows or Linux programs, Peer said.
With Linux's growing popularity, an increasing number of PC
users are installing both Windows and Linux on the same computer.
Despite its ability to jump between different operating systems,
W32.Winux is not a fast-spreading virus.
For one, Peer said that the virus appears to be limited to
spreading only on PCs running Intel Pentium processors -- meaning
it
could not spread to Sun Microsystems Inc. (NasdaqNM:SUNW - news)
servers running Linux.
And unlike more recent worm-type viruses like Melissa or Love
Letter, W32.Winux cannot automatically e-mail itself to other
Internet
users worldwide.
``It's rather old-fashioned in that way,'' Peer said.
While there are thousands of viruses swimming around on Windows
computers worldwide, there are relatively few for Linux -- an
estimated less than 50, Peer said, which he attributed to the
lack of virus writers targeting the Linux operating system.
Central Command says it has developed a cure for the virus at
its Web site (http://www.avx.com).
--
Robert Lynch Berkeley CA USA [EMAIL PROTECTED]
------------------------------
** 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.misc.
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-Misc Digest
******************************
