Alsa lib pcm.c: 7339:( and pcm_recover ) underrun occured

[rei...@ix.netcom.com]

Hello Leif,


Started a new subject heading in the hope to find a way
around this.

These underruns seem to be the major hangup now at this point with linrad here.

They seem to come randomly, previously I thought they had something to do
with my actions ( keyboard etc )

I have ALSA running and besides hoping to use Linrad I use
Iceweasal, the linux version of Firefox on this board.

Have not found any problems with audio using this browser, music
Youtubes etc.

Took the liberty to use Google again, searching for

"Alsa lib pcm.c: 7339:( and pcm_recover ) underrun occured "

It delivers many, many as usual, questions and no answers for
these overruns, happening in other programs using audio and perhaps
the audio sound hardware in Raspberry Pi.

I have included the lost attachment here were the underruns messages
are shown ( lower part of message )

I think they come in series of them and after a while it will stop
linrad. In the beginning, linrad keeps running at least it looks
like it is.

Yesterday I went outside the house, left Linrad running and 3 hours
later it was still running

I have an another USB sound card here somewhere in the shack of which the
rasberry users provide good reports in their reviews.

something like this

http://plugable.com/2014/11/06/how-to-switch-to-usb-audio-on-raspberry-pi


Do you think sampling rate of the audio output has anything to do with the underruniing
problem.

do you think linrad would find such a USB unit? and could it be selected?
Could the sample rate be set?

Just questions. I think to be stuck with out getting a solution for this!



gcc -o xlinrad -Wl,\
xmain.ol xvar.ol ui.ol lsetad.ol fft1.ol buf.ol lvar.ol radar.ol \
mouse.ol wide_graph.ol fft0.ol adtest.ol hires_graph.ol elad.ol \
baseb_graph.ol mix2.ol afc_graph.ol hwaredriver.ol affinity.ol \
sellim.ol palette.ol gifsave.ol timing.ol help.ol calibrate.ol \
calsub.ol calsub2.ol llsqvar.ol llsq.ol caliq.ol timf2.ol tx.ol \
powtim.ol selvar.ol fft2var.ol sigvar.ol screenvar.ol uivar.ol \
fft1var.ol fft3var.ol calvar.ol blnkvar.ol fft1_re.ol lxsys.ol \
powtvar.ol blank1.ol fft2.ol sdrip.ol loadusb.ol airspy.ol \
pol_graph.ol modesub.ol spur.ol cohsub.ol coherent.ol afedri.ol \
eme.ol spursub.ol freq_control.ol perseus.ol httpd.ol html_server.ol \
csplit.ol \
rxin.ol wcw.ol mix1.ol afcsub.ol morse.ol menu.ol fonts.ol rxout.ol \
tune.ol txtest.ol network.ol coh_osc.ol fm.ol soundcard.ol \
cwspeed.ol sdrvar.ol sdr14.ol keyboard_var.ol thrvar.ol pa.ol \
xsys.ol screen.ol cwdetect.ol tx_graph.ol txssb.ol pavar.ol lsdr.ol \
txvar.ol screensub.ol fft3.ol wxsys.ol meter_graph.ol wse.ol si570.ol \
extio.ol soft66.ol excalibur.ol usb2lpt.ol rtl2832.ol elektor.ol \
fcdpp.ol hid.ol pcie9842.ol bladerf.ol mirics.ol openhpsdr.ol \
oclprogs.ol \
\
/usr/lib/arm-linux-gnueabihf/libXext.so \
/usr/lib/arm-linux-gnueabihf/libX11.so \
/usr/lib/arm-linux-gnueabihf/libasound.so \
\
-lpthread -lm -ldl

You may want to disable pulseaudio in Ubuntu with this command:
sudo pasuspender -- ./xlinrad

root@raspberrypi:/home/linrad# ./xlinrad
Found Rafael Micro R820T tuner
Exact sample rate is: 230000.000533 Hz
Found Rafael Micro R820T tuner
Exact sample rate is: 230000.000533 Hz
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred
ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred

‹ Previous | Next › | « Back to Sent

73 Rein W6SZ






-----Original Message-----
>From: Leif Asbrink <le...@sm5bsz.com>
>Sent: Jun 1, 2015 12:26 AM
>To: lin...@googlegroups.com
>Subject: Re: [Linrad] Raspi linrad audia
>
>Hello Rein,
>
>> Linrad is running now but is shutting down due to underruns it seems
>>
>> I did a new ./configure with help results in attachment 1
>>
>> and make xlinrad with messages in attachment 2
>No attachments, but does not matter. There is no info of
>interest since you were able to compile and start xlinrad.
>
>> ssb mode.
>> 230 kHz sampling
>> sound selections probably failing to do properly.
>????
>> 11% CPU and 24 % the other one, perhaps you can indicate
>> where I can read about the second.
>The second is the load due to the heaviest thread. Well under 100%
>so perfectly fine:-)
>Total load is 11% only.
>
>> noise somewhere selected for base band, see signal there. Red bar no effect
>> No audio produced.
>> Alsa lib pcm.c: 7339:( and pcm_recover ) underrun occured
>> At times I have seen here underruns and overruns messaging info
>> Source please, so I can read what is means and how to avoid it.
>> sure I have made errors in the sound card setup somewhere.
>> At one time a ran into error 1200
>You tried to select the OSS sound system which is not installed
>on your system. There is 4Front OSS on x86 platforms. Sometimes
>they provide better drivers than alsa. Maybe that is history now...
>
>> I have problem here what to, why and what is happening doing it.
>> Once I get a setup defined and saved to go return to, I should
>> be able to bootstrap up I think.
>OK.
>
>The first problem to solve is overrun error. Must not happen.
>The problem is that data arrives too fast for Linrad to attend
>to it in time. The buffer gets full and data is lost. Overrun.
>
>There can be many reasons.
>
>I suggest you start by installing gnome-system-monitor. Use it
>to monitor the total cpu load. You may well have a 100% CPU
>load problem due to the overhead caused by the desktop.
>If load is high, CPU sometimes above 95%, my guess is that this
>would be caused by the Gnome desktop environment. Then, install
>xfce4 which is a lightweight environment. Gnome may use up the CPU
>by emulating the visual effects that normally would have been done
>in a GPU.
>
>If you have an adequate margin on CPU load, peaks below 95%
>and average below 90% total cpu load, then something else is
>the problem. It could be due to latency. Bigger buffers might help.
>Then, in the U sub-menu, set max and min DMA rate equal and
>try different values. You can see the currently used DMA rate
>by pressing 'T' on the processing screen.
>
>You can set the priority higher. Up to real time if you tell you
>are an expert.
>
>If nothing of the above helps, I would conclude that something is
>wrong with your USB sub-system. Maybe something is wrong with
>your libusb-1.0 installation. You might get the latest from
>here: http://sourceforge.net/projects/libusb/files/libusb-1.0/libusb-1.0.19/
>if you do not already have libusb-1.0.19
>
>Maybe it is a USB2/USB3 problem. Try other USB ports. Maybe there
>is a BIOS where you can configure the USB system.
>
>Others have been sucessful (I think) Hopefully someone else can
>give better ideas.
>
>Anyway, as long as you have underrun errors, do not expect anything
>to work...
>
>73
>
>Leif
>
>
>--
>You received this message because you are subscribed to the Google Groups "Linrad" group.
>To unsubscribe from this group and stop receiving emails from it, send an email to linrad+un...@googlegroups.com.
>To post to this group, send email to lin...@googlegroups.com.
>For more options, visit https://groups.google.com/d/optout.

[Leif Asbrink]

Hello Rein,

> These underruns seem to be the major hangup now at this point with linrad here.

I am curious to know how you eliminated the overrun errors.

In case you still have some, you must be aware that loosing data at
the input means that there will be a shortage of data at the output
which is the phenomen called underrun error. (There is no data in the
output buffer.)

> They seem to come randomly, previously I thought they had something to do
> with my actions ( keyboard etc )
>
> I have ALSA running and besides hoping to use Linrad I use
> Iceweasal, the linux version of Firefox on this board.
>
> Have not found any problems with audio using this browser, music
> Youtubes etc.
>
> Took the liberty to use Google again, searching for
>
> "Alsa lib pcm.c: 7339:( and pcm_recover ) underrun occured "
>
> It delivers many, many as usual, questions and no answers for
> these overruns, happening in other programs using audio and perhaps
> the audio sound hardware in Raspberry Pi.

Sound systems may be extremely complicated. Linrad tries to force you
to select a sampling rate that can actually be used in the hardware,
but I am not so sure it is always sucessful.

If you have no overrun errors and a reasonable total CPU load
you might get overrun errors for several different reasons.

Maybe you have set a large size on the third fft? That would mean that
data comes out in very large blocks. On a not so fast computer
you might have to increase the output delay margin.

Press 'T' on the processing screen and look at the D/A margin
and its minimum value.

Gnome with pulseaudio might add various mixers and sound effects just
like Microsoft might do. Such things may cause problems that may
also be solved by use of a larger output delay margin.

> I have included the lost attachment here were the underruns messages
> are shown ( lower part of message )
>
> I think they come in series of them and after a while it will stop
> linrad. In the beginning, linrad keeps running at least it looks
> like it is.
>
> Yesterday I went outside the house, left Linrad running and 3 hours
> later it was still running

I suggest you disable the output (press Z in the U menu.) Then run
Linrad for an hour. If you see a single underrun error, something is still
not correct on the input.

> I have an another USB sound card here somewhere in the shack of which the
> rasberry users provide good reports in their reviews.
>
> something like this
>
> http://plugable.com/2014/11/06/how-to-switch-to-usb-audio-on-raspberry-pi

No, leave the system sound as it is. Just plug in the USB soundcard and select it
in the U menu of Linrad. Then Linrad would not suffer from other programs
doing things to the Linrad output channel.

> Do you think sampling rate of the audio output has anything to do with the underruniing
> problem.

Maybe. It should not, but the sound system we use under Linux is not
anywhere near bug-free.

> do you think linrad would find such a USB unit? and could it be selected?
> Could the sample rate be set?

Yes on all.

> Just questions. I think to be stuck with out getting a solution for this!

You have not yet provided a single bit of useful information after
compilation was sucessful;-)

Underrun errors and overrun errors should not cause Linrad to shut down.
It should just mis-behave. Do you get an error message when Linrad shuts down?

One common reason for underrun errors is that you have too many programs open
and/or have selected large data arrays for Linrad. Then your "clever" operating
system would place Linrads working data on your swap file which with 100%
probability would cause an underrun error. Or several.

You might set storage times to minimum to make Linrad re-use memory
locations more often.

There is a system call mlock / munlock to prevent data from being
moved into the swap space. I have not (yet?) implemented that in Linrad
so if the OS thinks another application needs physical ram better,
Linrad would fail.

Something is obviously very wrong, but I have no info from you about
what Linrad is doing - except that the CPU load caused by Linrad is
not any problem at all.

73

Leif

[rei...@ix.netcom.com]

Hello Leif,

Got a print screen ( from using T ) just after it stopped,'

while running, the fft1 was reading 250 or so , on the picture "0" after stopping

http://www.nitehawk.com/rasmit/alin/linrad_T_stopped_256.jpg

I do not think I have the underrunning under control unfortunately ( it seems to be longer before it is happening

You say it is a matter of a buffer filling up. would that not getting better if the sound D/A would be running lower than 48 KHz

Tt is fixed here. I that a fuction of the Raspberry Pi chip?

It is really random I think whatever it is.

If I sample a real signal it appears to go much faster ( the stopping_ ) than sampling noise.

Hope you get all the info from the picture. Have to learn to make quick print screens with scrot?

Another question. I have very high base noise indicated, is that not too much gain in dongle gain setting? Is there a way to decrease that gain

( just mot Knowing how to adjust here, I guess )

73 Rein W6SZ







http"//www.nitehawk.com/rasmit/alin/linrad_T_stopped

-----Original Message-----
>From: Leif Asbrink <le...@sm5bsz.com>
>Sent: Jun 1, 2015 10:58 PM
>To: lin...@googlegroups.com

[Leif Asbrink]

Hello Rein,

> Got a print screen ( from using T ) just after it stopped,'
>
> while running, the fft1 was reading 250 or so , on the picture "0" after stopping
>
> http://www.nitehawk.com/rasmit/alin/linrad_T_stopped_256.jpg
>
> I do not think I have the underrunning under control unfortunately
> ( it seems to be longer before it is happening

Something is very wrong on the input side. What has happened
on the picture is that the input has stopped. The message "No input"
has appeared 4697 times.

The problem is in your system somehow. The rtlsdr is not sending data
any more. Maybe there is an error message in the terminal window.

> You say it is a matter of a buffer filling up. would that not
> getting better if the sound D/A would be running lower than 48 KHz
>
> Tt is fixed here. I that a fuction of the Raspberry Pi chip?
>
> It is really random I think whatever it is.
>
> If I sample a real signal it appears to go much faster ( the stopping_ ) than sampling noise.
>
> Hope you get all the info from the picture. Have to learn to make quick print screens with scrot?

Install gnome-system monitor and run it while Linrad is running.

How did you install librtlsdr and libusb-1.0 ? One (or both)
of these packages does not work properly. Maybe the system monitor
would show why. Do you have other things on the USB bus?
It looks like you are connected via Windows from another computer.

> Another question. I have very high base noise indicated,
> is that not too much gain in dongle gain setting?
> Is there a way to decrease that gain

The dongle is 8 bit and Linrad aligns data for the highest bit
to be used so the noise floor is 256 times higher than
normal. The parameter to use is First FFT amplitude that should
be something like 256 times below normal, 7 perhaps instead of 2000.

73

Leif

[rei...@ix.netcom.com]

Hello Lief,

Thanks. I like to go back to the beginning to provide you with
as much info as possible from the start.

>How did you install librtlsdr and libusb-1.0 ? One (or both)
>of these packages does not work properly. Maybe the system monitor
>would show why. Do you have other things on the USB bus?
>It looks like you are connected via Windows from another computer.

I understand now the use of configure AND the information is provides!

As far as I remember I never installed those packages by me as a
separate action!

They were present via the SVN procedure. But from the new run I can check
librtlsdr and libusb status ( or you can )


I have a WIFI dongle on the USB bus for internet, a keyboard
and x mouse and on and off a second audio chip.

This chip shops up under "U" and I select it if connected.



I used your video as a start

https://www.youtube.com/watch?v=bviyLefH8E0

( low end computer and debian


- Raspbeery pi comes with With debian

- It a low end computer.

- At this point I am mainly interested in waterfall work ( deep Space 8.4 GHz )

- I had gotten the impression that others here in the group were using the same
hardware as I have here.

- There is a lot of raspberry PI / debian info around ( reliable or not )



RTL-SDR domgle - Rsspberrry p i2 satnd alone.

I will remove the linrad directory, and reload the package.

And send you the results from the first .configure.

I hope you can see form that what is missing and so on

The problems I have with it stopping occur with the on board audio chip
and another audio card I have bee using * should not have done this I now
understand from you.

So I imagine with the new install, you might be able to tell me how
to clean up the installation as far as this second sound chip is concerned with
lsmod?

I hope if this approach is a problem, you let me know.

73 Rein W6SZ

-----Original Message-----
>From: Leif Asbrink <le...@sm5bsz.com>

[Leif Asbrink]

Hello Rein,

> Thanks. I like to go back to the beginning to provide you with
> as much info as possible from the start.

That is not possible unless you reinstall Debian from scratch.
Reinstalling Linrad will not give any information.
The configure script will just say that the things you need are
already installed.

> >How did you install librtlsdr and libusb-1.0 ? One (or both)
> >of these packages does not work properly. Maybe the system monitor
> >would show why. Do you have other things on the USB bus?
> >It looks like you are connected via Windows from another computer.
>
> I understand now the use of configure AND the information is provides!
>
> As far as I remember I never installed those packages by me as a
> separate action!
>
> They were present via the SVN procedure. But from the new run I can check
> librtlsdr and libusb status ( or you can )

Please install gnome-system-monitor and run in parallel with your
current Linrad installation.

I am suspicious about your environment. Modern programming typically
assumes that CPU load is no problem. That would not be true with
your system.

Do not waste time on a reinstall. It will most probably not give
any useful info.

> I have a WIFI dongle on the USB bus for internet, a keyboard
> and x mouse and on and off a second audio chip.
>
> This chip shops up under "U" and I select it if connected.

I have no experience with WIFI dongles, but I suspect it can
grab the USB bus and use it heavily, thus blocking the rtlsdr
operation. Can you disconnect it or do you need it to get access to
the system? Is there not an ordinary Ethernet connector you could use?

Keyboard, mouse and soundcards are harmless. They would not cause
any problem.

> I used your video as a start
> https://www.youtube.com/watch?v=bviyLefH8E0
> ( low end computer and debian
> - Raspbeery pi comes with With debian
> - It a low end computer.
>
> - At this point I am mainly interested in waterfall work ( deep Space 8.4 GHz )
> - I had gotten the impression that others here in the group were using the same
> hardware as I have here.

Yes, I think so.

> - There is a lot of raspberry PI / debian info around ( reliable or not )
> RTL-SDR domgle - Rsspberrry p i2 satnd alone.
> I will remove the linrad directory, and reload the package.

That will NEVER help or provide any useful information.

> And send you the results from the first .configure.
> I hope you can see form that what is missing and so on

That will not be possible.

> The problems I have with it stopping occur with the on board audio chip
> and another audio card I have bee using * should not have done this I now
> understand from you.

I expect you to see this problem even if you press 'Z' in the 'U' sub-menu
to disable the soundcard.

> So I imagine with the new install, you might be able to tell me how
> to clean up the installation as far as this second sound chip is concerned with
> lsmod?
>
> I hope if this approach is a problem, you let me know.

Yes, it is not going to help or improve in any way.

1) What is your total system load?

2) Can you eliminate USB WIFI?

You have a problem - and there is a reason. The installation
procedure would give no hint about what the problem might be
and what might be the reason for it.

73

Leif

[rei...@ix.netcom.com]

Hello Leif,

Thanks again. On the reloading business.

The raspberry pi is running from a SD card, every new start debian gets reloaded.

Are you saying that the debian part gets modified by Linrad activity? In other words
should I reload the SD card? New installed programs, such as Linrad, get also stored
on the SD card in a other section of the memory. That might also not be quire true?

Other subject.

Your suggestion of removing the WIFI adapter was right on target. So the random errors
do not occur any longer. Very well. I can run Linrad over night without it stopping
with the underruns.

The underruns still happen however , Mostly with some of the mouse activity related,
about that more later.

The mouse is only visible when I move it. In one of the video's you mentioned something
on how to fix this, I can not locate that now where that was. So please how.

Not sure on the Gnome-system-monitor. I loaded it, did a ./configure, during that procedure
my HDMI went dark, do not know the reason why. but linux was still running
I did not get the info from the configure and do not know how to repeat the installation in
the hope to recover the lost info during the black screen phase

Can you suggest how to repeat the installation of the system-monitor if needed and what is the
command to actually start the s-monitor. I guess, a ridiculous question, but afraid I do not know
the answer or where to find those answers. Hate to have to ask all ths stuff, I guess linux is the reason.

As a matter of fact I do nor really understand the installation of packages procedure if there is
such a thing. in Linrad it is clear "make" will inform what to do.
Do not have the idea this is a universal approach. Could be wrong there of course as well!


Then what is the function or should be the function of the bottom window ( gain ) below the LO
setting. It starts most of the time with 15 dB. If I try to change this it goes to 0 and can
not be edited and going form 15 to 0 dB is not making any visible change or having effect.

I did the changed for the 8 bit correction for 2 V maximum from Dongle ADC. The base line went
from somewhere of 85 to 90 dB to 35 dB. ( S 6-7 on S meter )

Is there gain control for the dongle? I guess the dongle gain is too high, just a guess at
this point though.

73 Rein W6SZ



-----Original Message-----
>From: Leif Asbrink <le...@sm5bsz.com>
>Sent: Jun 2, 2015 11:00 PM
>To: lin...@googlegroups.com
>Subject: [Linrad] Re: Alsa lib pcm.c: 7339:( and pcm_recover ) underrun occured
>

[Gunter Bohn]

Hallo Rein,

I am using raspberry-pi with fifi-sdr on usb port and I had first the
problem with
underrun for a while.
I think, the problem ist the usb-management by the raspberry-pi.
Any request via usb is a potential error for the usb-dongle.
By using a usb-splitter the underrun-problem was extremly.
I have reduced all on usb (only mouse/ keyboard - keyboard with touch-pad).
Now it runs all normal without underrun in CW-mode.


Gun / DL2RUG

[rei...@ix.netcom.com]

Hello Gun,

Thanks for the heads up on the 7339 underrun error.

What is a usb-splitter? External USB ports?

The random errors are gone now but I still experience mouse
driven 7339 errors i think.

It appears that any not audio related activity on USB is a no no with the raspberry pi.

They show up as 1199 linrad errors, "too short output delay margins"

I have had that on 1000 as suggested , 2000 and even 3000 when I get a too low memory error.

It is still happening. Also I can't keep the cursor alive with F10 ( per Leif's video )

Leif suggested to install gnome_system-monitor. And after talking to a linux person got it
installed, much easier than expected. Seeing the 4 core loads now. All around 30 - 35 %
with 15 and 27 % indicated by Linrad.

The audio output from the raspberry pi is very low. Are you using it for audio output?

Starting to feel good about this project. Lots more to learn though, but is going to be working
I think.

[Leif Asbrink]

Hello Rein,

> Thanks again. On the reloading business.
> The raspberry pi is running from a SD card, every new start debian gets reloaded.
>
> Are you saying that the debian part gets modified by Linrad activity? In other words
> should I reload the SD card? New installed programs, such as Linrad, get also stored
> on the SD card in a other section of the memory. That might also not be quire true?

"reload" ?? I can not follow your thinking. "debian part"???

I guess your SD card is equivalent to a hard disk. As a first step
you install a Linux distribution.
Different Linux distributions include different sub-sets of all
the packages in the distribution file. The Debian installation on this
particular computer lists 36711 different packages as an example.

As a second step you place the Linrad distribution package in
a directory and run the configure script.

The configure script will tell you which packages you have to install
in case they are not already present. This will be different in different
distributions, but once you followed the instructions and installed for
example nasm, the package nasm will permanently be a part of your
Debian installation.

The same for all the other packages that configure helps you to install.
Nothing of that is a Linrad installation. It is just the necessary
additions Linrad needs in your Debian installation.

Once many enough packages are installed you can install Linrad
with the command make xlinrad (or similar)

If you want to de-install Linrad, just remove the Linrad directory.

Your Debian installation would of course not be changed by that.
If you subsequently want to install Linrad you would find that
the configure script does not tell you you have to install
anything.

> Other subject.
>
> Your suggestion of removing the WIFI adapter was right on target. So the random errors
> do not occur any longer. Very well. I can run Linrad over night without it stopping
> with the underruns.

OK. There must be a reason for this. Easiest is to just not use it,
but maybe better to try to find out what the problem is.

> The underruns still happen however , Mostly with some of the mouse activity related,
> about that more later.

OK.

> The mouse is only visible when I move it. In one of the video's you mentioned something
> on how to fix this, I can not locate that now where that was. So please how.

Sorry, I do not remember. Maybe someone else on the list remembers...

> Not sure on the Gnome-system-monitor. I loaded it, did a ./configure, during that procedure
> my HDMI went dark, do not know the reason why. but linux was still running
> I did not get the info from the configure and do not know how to repeat the installation in
> the hope to recover the lost info during the black screen phase

Just run configure again. It will provide the same output.
There is no reason to run system-monitor during setup. The fact something crashed
while you were running ./configure points to a problem in Debian itself.

> Can you suggest how to repeat the installation of the system-monitor if needed and what is the
> command to actually start the s-monitor. I guess, a ridiculous question, but afraid I do not know
> the answer or where to find those answers. Hate to have to ask all ths stuff, I guess linux is the
> reason.

"repeat installation" is a typical Windows thing. That is not going to help in
Linux. (Only very seldomly...)

Why would you want to "repeat the installation of the system-monitor"?

> As a matter of fact I do nor really understand the installation of packages procedure if there is
> such a thing. in Linrad it is clear "make" will inform what to do.
> Do not have the idea this is a universal approach. Could be wrong there of course as well!

I hope the text at the top of this mail has clarified a bit.

> Then what is the function or should be the function of the bottom window ( gain ) below the LO
> setting. It starts most of the time with 15 dB. If I try to change this it goes to 0 and can
> not be edited and going form 15 to 0 dB is not making any visible change or having effect.

That window controls gain or attenuation. You might try -30 or something.
In case you selected to use hardware AGC, then the gain setting does not work.

> I did the changed for the 8 bit correction for 2 V maximum from Dongle ADC. The base line went
> from somewhere of 85 to 90 dB to 35 dB. ( S 6-7 on S meter )
>
> Is there gain control for the dongle?

Yes.

> I guess the dongle gain is too high, just a guess at
> this point though.

Probably:-)

73

Leif

[Leif Asbrink]

Hello Rein,

> It appears that any not audio related activity on USB is a no no with the raspberry pi.
>
> They show up as 1199 linrad errors, "too short output delay margins"
>
> I have had that on 1000 as suggested , 2000 and even 3000 when I get a too low memory error.

You might try to set the Linrad priority higher and/or you could try
to set the DMA rate lower.

> It is still happening. Also I can't keep the cursor alive with F10 ( per Leif's video )

Which video? Was it really showing a Linux version?

> Leif suggested to install gnome_system-monitor. And after talking to a linux person got it
> installed, much easier than expected. Seeing the 4 core loads now. All around 30 - 35 %
> with 15 and 27 % indicated by Linrad.

OK. So you do not have an overload problem.

What about memory and swap? (only relevant if you still have underrun errors
while not inducing USB errors from mouse or keyboard)

> The audio output from the raspberry pi is very low.

Presumably you should run alsamixer or similar.

73

Leif