Re: [Csnd-dev] possible issue with ALSA midi?

I see in the code that although it seems to accept -+rtmidi=alsa to select the alsaraw midi backend, the names of the midi modules are only alsaraw, alsaseq and devfile. Not sure what the intention was but I guess "alsa" should be added too (as an alias of alsaraw).

Prof. Victor Lazzarini
Maynooth University

Ireland

On 11 Nov 2024, at 20:34, Dave Seidel <[email protected]> wrote:


aha, interesting! Yes, that works. We should probably document this.

dave@phonon:~ $ csound -+rtmidi=alsaseq --midi-devices
5 MIDI input devices
0: hw:0,0 (Timer)
1: hw:0,1 (Announce)
2: hw:14,0 (Midi Through Port-0)
3: hw:24,0 (pisound MIDI PS-28625BA)
4: hw:128,0 (pisound-ctl)
3 MIDI output devices
0: hw:14,0 (Midi Through Port-0)
1: hw:24,0 (pisound MIDI PS-28625BA)
2: hw:128,0 (pisound-ctl)
overall amps: 0.0
overall samples out of range: 0
0 errors in performance
Elapsed time at end of performance: real: 0.003s, CPU: 0.003s

dave@phonon:~ $ csound -+rtmidi=alsaraw --midi-devices
1 MIDI input devices
0: hw:2,0 (pisound MIDI PS-28625BA)
1 MIDI output devices
0: hw:2,0 (pisound MIDI PS-28625BA)
overall amps: 0.0
overall samples out of range: 0
0 errors in performance
Elapsed time at end of performance: real: 0.004s, CPU: 0.004s

dave@phonon:~ $ csound -+rtmidi=devfile --midi-devices
0 MIDI input devices
0 MIDI output devices
overall amps: 0.0
overall samples out of range: 0
0 errors in performance
Elapsed time at end of performance: real: 0.001s, CPU: 0.001s

On Mon, Nov 11, 2024 at 3:12 PM vlz <[email protected] > wrote:

I checked and this error message comes from the MIDI side of things in the alsa module

static int32_t listDevicesM(CSOUND *csound, CS_MIDIDEVICE *list, int32_t isOutput){

int32_t count = 0;

char *s;

s = (char*) csound->QueryGlobalVariable(csound, "_RTMIDI");

if (strncmp(s, "alsaraw", 8) == 0) { /* ALSA Raw MIDI */

count = listRawMidi(csound, list, isOutput);

} else if (strncmp(s, "alsaseq", 8) == 0) {/* ALSA Sequencer */

count = listAlsaSeq(csound, list, isOutput);

} else if (strncmp(s, "devfile", 8) == 0) {

} else {

csound->ErrorMsg(csound, Str("rtalsa: Wrong callback."));

}

return count;

}

So maybe you can try one of the options: alsaraw, alsaseq or devfile (from the code, it’s raw midi, alsa sequencer midi or

midi file). These are the three options for listing but I guess “alsaraw” should have also been the same as “alsa” but

currently is not there.

> On 11 Nov 2024, at 20:04, Dave Seidel <[email protected] > wrote:

>

> Yes, -odac is working fine. I went back and checked, and I've been specifying "-+rtmidi=portmidi" for several years now on my Pis, going back to Cs6. So maybe this is an old/existing issue, which might make it low priority.

>

> On Mon, Nov 11, 2024 at 2:52 PM vlz <[email protected] > wrote:

> It could well be. Can you run -odac to it, does it make any sound?

>

>

> Prof. Victor Lazzarini

> Maynooth University

> Ireland

>

>> On 11 Nov 2024, at 19:48, Dave Seidel <[email protected] > wrote:

>>

>> I recently acquired a Pi 5, which I've equipped with a Pisound board (like my older Pis). I can build Cs7 without any issues, but I notice that when I query MIDI devices using ALSA as the interface, I get an error:

>>

>> dave@phonon:~/src/csound/build $ csound -+rtmidi=alsa --midi-devices

>> rtalsa: Wrong callback.0 MIDI input devices

>> rtalsa: Wrong callback.rtalsa: Wrong callback.0 MIDI output devices

>> rtalsa: Wrong callback. overall amps: 0.0

>> overall samples out of range: 0

>> 0 errors in performance

>> Elapsed time at end of performance: real: 0.001s, CPU: 0.001s

>>

>> If I specify portmidi, no errors:

>>

>> dave@phonon:~/src/csound/build $ csound -+rtmidi=portmidi --midi-devices

>> rtmidi: PortMIDI module enabled

>> 3 MIDI input devices

>> 0: 0 (Midi Through Port-0)

>> 1: 1 (pisound MIDI PS-28625BA)

>> 2: 2 (pisound-ctl)

>> 3 MIDI output devices

>> 0: 0 (Midi Through Port-0)

>> 1: 1 (pisound MIDI PS-28625BA)

>> 2: 2 (pisound-ctl)

>> overall amps: 0.0

>> overall samples out of range: 0

>> 0 errors in performance

>> Elapsed time at end of performance: real: 0.003s, CPU: 0.003s

>>

>> Same results on my Pi 4 running Cs7. I don't have a Pi running Cs6 for comparison. The Pi 4 is running bullseye; the Pi 5 is running bookworm.

>>

>> FWIW, I don't really care which interface I use, and I'm fine with using portmidi, but I thought this might indicate an issue.

>>

>> - Dave