Audio (video) - Kindle. Getting it up and running, Code level.

[twobob]

Originally Posted by geekmaster

One thing about tiny little code with no error messages is that you get to rely on the general all-purpose error message called "segfault".

And then there are those darned built-in libraries that spoil the party by injecting their own error messages before aborting.

Of course, in "real" code (especially all the motion-control / robotics / factory automation stuff I did over the years), you need to recover gracefully from just about anything that can happen, logging it silently while not annoying the machine operator, and self-calibrating WHILE in operation. You cannot shut down operations just because the machine is failing, if your software can compensate for the progressive failure. But you also cannot just hit the reset button to restart after a large part of your machine goes through a brick wall and out into the parking lot because of a software glitch.

Now, the reason I brought this up, is that I am getting segfaults with your latest code. PEBKAC errors no doubt.

unsure which version I sent.
I call the current incarnation like ./Balsamic plughw 8k8bmono.wav 8000

but I think I added the argv[3] for alternate speed playback - extra testing options.

I removed all error handling more or less - aside from the obvious alsa bits that were already hanging around.

It may segfault when it gets to the end of the file. My current one does because I don't check for %WaveSize correctly (I'm monitoring the wave stream as video info)

I've abandoned all attempts to sync vid pending your next revelation.

My code is not hardy nor well programmed. But I've got the seggies down to expected ones my end. let me know approx where it segges and ill have a think - the strace of it should be utterly straightforward.

: ) It's possibly the polling failing but I dont think so.


as an aside I got a huge performance increase by simply

void gmplay4(void) {

	while (1)
	{
		if (frameCount %2==0)
		{
BLAH
                 }
	else
		{
			usleep(5);  // or 10 
		}

as opposed to the sync waiting. the minuscule difference in frame rate (seemingly like 5 or 6 solid ones rather than 6 or 7 sketchy ones) was not that worse. perhaps a tiny sleep could be worked in if we are using threads - they seem to like it

[geekmaster]

@twobob: didn't you just "celebrate your centenary post" like a day (or three) ago? I just noticed that your bicentenary just went past unheeded (until now). Congrats!

And my trimillenary post is coming up pretty darned quickly too.

[twobob]

Originally Posted by knc1

Strange, I haven't heard of that happening before.
Hmm...
What Java is it being run with?
Many distro's install the gnu versions of Java by default, you may have to manually install the for-real Oracle Tech. version of Java.
http://www.oracle.com/technetwork/java/index.html

- - -
PS: If you find a solution, please post what you did to fix it.

bash-4.1$ ec
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007f1b25ba6afb, pid=12551, tid=139754494949136
#
# JRE version: 6.0_24-b07
# Java VM: Java HotSpot(TM) 64-Bit Server VM (19.1-b02 mixed mode linux-amd64 compressed oops)
# Problematic frame:
# V [libjvm.so+0x691afb]
#
# An error report file with more information is saved as:
# /home/simon/hs_err_pid12551.log
#
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp
#
bash-4.1$

Just like that

[knc1]

I seem to be losing track, I thought you made an additional install of 32bit Slackware.
No matter ...

Since it is the Java interpreter that is segfaulting, I would try a different Java - - -
Maybe try the 32bit build or maybe try the IBM build.
(You can have them all installed at the same time.)

I can't recall if I ever did run Eclipse with a 64bit Java myself.

My notes from the last time I used Eclipse: http://minimodding.com/eclipse_tour

[twobob]

Originally Posted by knc1

I seem to be losing track, I thought you made an additional install of 32bit Slackware.
No matter ...

Since it is the Java interpreter that is segfaulting, I would try a different Java - - -
Maybe try the 32bit build or maybe try the IBM build.
(You can have them all installed at the same time.)

I can't recall if I ever did run Eclipse with a 64bit Java myself.

My notes from the last time I used Eclipse: http://minimodding.com/eclipse_tour

I nixxed the 32 bit in favour of leaving the space ready for a complete end to end rework of my current junk. It's the 64 bit I'm certain of it. Ah well. all the more reason to do a good job next time. Thanks very much

[knc1]

The 32bit java will run on your 64bit machine just fine.
= = =
My local time, plus xxx time zones == time to take a break.

[twobob]

Originally Posted by knc1

The 32bit java will run on your 64bit machine just fine.
= = =
My local time, plus xxx time zones == time to take a break.

Yup Yup, much appreciated. Wish you a happy break

[geekmaster]

I was able to adjust sound output on my K5 with amixer, and with alsamixer (text GUI) commands (built-in). It is a bit louder, but still rather quiet when doing:

cat 8k8bmono.wav|aplay

It is much too quiet with headphones, even with all volume controls set to max using those tools.

On the K5, alsamixer is full color on SSH, but on the K3 it is only monochrome.

Hmm... after playing with K3 alsamixer and increasing several headphone and speaker volumes, now I do not get any sound output.

UPDATE: Rebooting the K3 made the sound work again.

[twobob]

alsamixer is evil. avoid him

There are direct call IIRC Ill have a look when i get back

[knc1]

Originally Posted by geekmaster

I was able to adjust sound output on my K5 with amixer, and with alsamixer (text GUI) commands (built-in). It is a bit louder, but still rather quiet when doing:

cat 8k8bmono.wav|aplay

It is much too quiet with headphones, even with all volume controls set to max using those tools.

On the K5, alsamixer is full color on SSH, but on the K3 it is only monochrome.

Hmm... after playing with K3 alsamixer and increasing several headphone and speaker volumes, now I do not get any sound output.

UPDATE: Rebooting the K3 made the sound work again.

Must be a left-over from WinCE - if you move the mouse, you have to re-boot.

[geekmaster]

Originally Posted by knc1

Must be a left-over from WinCE - if you move the mouse, you have to re-boot.

Considering that "wince" is the English word for a facial expression that is often used to express pain or great displeasure, Perhaps WinCE is aptly named.

[twobob]

Originally Posted by geekmaster

Considering that "wince" is the English word for a facial expression that is often used to express pain or great displeasure, Perhaps WinCE is aptly named.

Alsamixer is notoriously poor. the 13.1 Slackware kernel includes a version that is so bad if you use it you have to reboot the machine to get the INPUT working again on some devices.... Sound familiar? Don't even start me on the buggy monitoring and forgotten settings...

I would advise strongly against ever using (it in it's current state, which seems fairly stable at "poor"). There are better ways, I did include Alsamixer in my tips - but I also prefaced with - "If it's anything like my one it will be rubbish".

YMMV but it won't be high.

Spoiler:

void SetAlsaMasterVolume(long volume)
{
    long min, max;
    snd_mixer_t *handle;
    snd_mixer_selem_id_t *sid;
    const char *card = "default";
    const char *selem_name = "Master";

    snd_mixer_open(&handle, 0);
    snd_mixer_attach(handle, card);
    snd_mixer_selem_register(handle, NULL, NULL);
    snd_mixer_load(handle);

    snd_mixer_selem_id_alloca(&sid);
    snd_mixer_selem_id_set_index(sid, 0);
    snd_mixer_selem_id_set_name(sid, selem_name);
    snd_mixer_elem_t* elem = snd_mixer_find_selem(handle, sid);

    snd_mixer_selem_get_playback_volume_range(elem, &min, &max);
    snd_mixer_selem_set_playback_volume_all(elem, volume * max / 100);

    snd_mixer_close(handle);
}

http://www.alsa-project.org/alsa-doc...ple_mixer.html


Fails becase there is no "mixer" component declared by the alsa

Spoiler:

int SetAlsaMasterVolume(audio_volume_action action, long* outvol)
{

	 int ret = 0;
	    int fd, devs;


snd_mixer_t* handle;
snd_mixer_elem_t* elem;
snd_mixer_selem_id_t* sid;

static const char* mix_name = "Master";
static const char* card = "default";
static int mix_index = 0;

long pmin, pmax;
long get_vol, set_vol;
float f_multi;

snd_mixer_selem_id_alloca(&sid);

//sets simple-mixer index and name
snd_mixer_selem_id_set_index(sid, mix_index);
snd_mixer_selem_id_set_name(sid, mix_name);

    if ((snd_mixer_open(&handle, 0)) < 0)     return -1;
if ((snd_mixer_attach(handle, card)) < 0) {
    snd_mixer_close(handle);
    return -2;
}
if ((snd_mixer_selem_register(handle, NULL, NULL)) < 0) {
    snd_mixer_close(handle);
    return -3;
}

ret = snd_mixer_load(handle);

if (ret < 0) {
    snd_mixer_close(handle);
    return -4;
}
elem = snd_mixer_find_selem(handle, sid);
if (!elem) {
    snd_mixer_close(handle);
    return -5;
}

long minv, maxv;

snd_mixer_selem_get_playback_volume_range (elem, &minv, &maxv);
fprintf(stderr, "Volume range <%i,%i>\n", minv, maxv);

if(action == AUDIO_VOLUME_GET) {
    if(snd_mixer_selem_get_playback_volume(elem, 0, outvol) < 0) {
        snd_mixer_close(handle);
        return -6;
    }

    fprintf(stderr, "Get volume %i with status %i\n", *outvol, ret);
    /* make the value bound to 100 */
    *outvol -= minv;
    maxv -= minv;
    minv = 0;
    *outvol = 100 * (*outvol) / maxv; // make the value bound from 0 to 100
}
else if(action == AUDIO_VOLUME_SET) {
    if(*outvol < 0 || *outvol > VOLUME_BOUND) // out of bounds
        return -7;
    *outvol = (*outvol * (maxv - minv) / (100-1)) + minv;

    if(snd_mixer_selem_set_playback_volume(elem, 0, *outvol) < 0) {
        snd_mixer_close(handle);
        return -8;
    }
    if(snd_mixer_selem_set_playback_volume(elem, 1, *outvol) < 0) {
        snd_mixer_close(handle);
        return -9;
    }
    fprintf(stderr, "Set volume %i with status %i\n", *outvol, ret);
}

snd_mixer_close(handle);
return 0;

}

Fails for the same reason.

I suspect we need to do the alsactl setup satisfactorily before the programmatic side will work. One for my pile perhaps once I have this fft thingy happening.

Apologies that it is crappy, I can confirm it tried to mute the sound on mine too, fortunately I knew to go thru the settings one by one to reset them to play - while another shell played music, until I found the culprit, which on this occasion looked to be the actual master volume sigh.

definitely a flaky app, that will be best bypassed in the medium term via the code I have posted above.

[knc1]

I just posted static builds of xz (and lzma) utils -
slower than gzip, but will squeeze the life out of a .wav file if you need that.

[geekmaster]

I built xz for x86 and for arm. On the K3 it complains about missing glibc 2.6 (even when I static link it). I compressed 8kb8mono.wav on x86 using it. For that wave file, it compressed about 10% more than gzip but about 10% LESS than bzip2.

It's claim to fame seems to be for compressing linux log files.

[knc1]

Originally Posted by geekmaster

I built xz for x86 and for arm. On the K3 it complains about missing glibc 2.6 (even when I static link it). I compressed 8kb8mono.wav on x86 using it. For that wave file, it compressed about 10% more than gzip but about 10% LESS than bzip2.

It's claim to fame seems to be for compressing linux log files.

You should not get any complaints from the one I posted.
Also, I include the build notes so that other people can do the same now, or with a different version in the future.

Click the attachments icon on the DIY-KeK title line.