linuxsampler-devel — This list is for discussing LinuxSampler development issues and ideas

Re: [Linuxsampler-devel] Just a report

[Mark K. <mar...@gm...>]

On Sat, 1 Jan 2005 21:03:04 +0100, Christian Schoenebeck
<sch...@so...> wrote:
> Es geschah am Samstag 01 Januar 2005 19:14 als Mark Knecht schrieb:

> 
> > Can gigdump extract certain samples from the gig file? Maybe I could
> > send a single sample and you could figure it out from there. Or tell
> > me how I might investigate this more here and I'll give it a try.
> 
> Not with gigdump, but with gigextract you can extract samples as .wav files
> (man gigextract). If you just want to extract certain sample(s), then look at
> the output of gigdump, and use the repective sample number(s) as shown there.

I have extracted one of the samples. It playing incorrectly in
alsaplayer and sounds jsut like it does in LS so this should tell you
what I'm hearing. I'm sending you the zipped sample off list. It's
about 200K bytes.

Can gigextract extract the sample without uncompressing it? If so I
can send you the actual sample and you can attempt to find the
problem.

gigextract is very cool. Good work!

- Mark

Re: [Linuxsampler-devel] Re: bugs

[Vladimir S. <ha...@gm...>]

Garett,

> Vlad, I have never tried to gdb over irc. I would be happy to learn how
> this is done if you think it would help. I'm pretty sharp with gdb and
> might just be able to get you the info you need myself (although I
> haven't familiarized myself with the ls code at all). Also, I could just
> make you a temporary account on my box and you could gdb through ssh.

sounds good. i'll be on irc.freenode.org #linuxsampler we can discuss
this in detail.
if you have time and desire, i'd be glad to hook you up with
everything especially since you are already sharp with gdb. then in
addition to your great testing and bugreports you'll be fixing bugs,
adding features in no time!
 
> I tried the lastest cvs ls and have not seen a disk stream error yet
> with > 2 channels. Right on!

good.
 
> Also, in regards to running ls as root... I just realized this morning
> that the voice dropping I had mentioned to the list last week was caused
> because I had not chmod +s linuxsampler .
> When I do not run ls as root on my system I get lots of audio artifacts
> and lots of voice dropping.

it might be an indication that your system doesn't have much free ram.
One of the major differences between LS root vs non-root is memory
can't be locked if it's not root. so it's probably getting swapped out
and then there will definitely be issues. you can test that theory by
doing swapoff -a and running non-root.
I do that sometimes when i'm debugging a hang and i don't want LS to
hang the whole system hard :)

Regards,
Vladimir.

[Linuxsampler-devel] Re: segfault report - No Region defined?

[Mark K. <mar...@gm...>]

On Sat, 1 Jan 2005 11:18:27 -0800, Mark Knecht <mar...@gm...> wrote:
> On Sat, 1 Jan 2005 11:10:14 -0800, Mark Knecht <mar...@gm...> wrote:
> >
> > I wonder if this could be something like the 'malformed MIDI' issue
> > that was reported by Garett yesterday? I'll see if amidirecord can
> > capture the incoming MIDI that causes the crash.
> >
> > - Mark
> >
> 
> And the winner is....... (drum roll) Control Change or Program Change?

OK, so I've spent some more time looking at the MIDI files in Pro
Tools. The big difference between the bass channel that causes the
segfault and the other two channels is that the bass channel uses
sustain pedal events for almost every note.

     222069 Control change          1   7  96
     222070 Program change          1  32
     222237 Note on                 1  50  76
     222501 Control change          1  64 127
     222628 Note on                 1  50   0
     222677 Control change          1  64   0
     222683 Note on                 1  57  70
     222821 Control change          1  64 127
     223038 Control change          1  64   0
     223054 Note on                 1  64  86
     223056 Note on                 1  57   0
     223351 Control change          1  64 127
     223461 Note on                 1  64   0
     223597 Note on                 1  57  63
     223612 Control change          1  64   0
     223776 Control change          1  64 127
     224017 Note on                 1  57   0
     224039 Note on                 1  50  73
     224050 Control change          1  64   0
     224292 Control change          1  64 127

Possibly LS is having trouble keeping up with that many controller
changes happening in real-time?

Jsut a guess.

- Mark

Re: [Linuxsampler-devel] Just a report

[Christian S. <sch...@so...>]

Es geschah am Samstag 01 Januar 2005 19:14 als Mark Knecht schrieb:
> If the info was going to be part of the app then I agree. I'm
> wondering if it is possible to write some small support app, just
> console based, that could query LS (or Linux itself) for LS memory
> usage?

Sure

> Sorry. I do hear noise with these gig files. Underneath the noise I
> can hear the note, but badly distorted. It's a bass guitar, but I hear
> mostly high frequencies only and covered up with noise.
>
> > Mark, you mentioned there was still a problem with your Scarbee
> > J-Fingered Bass gig. What exactly was the problem?
>
> As stated above. I can record the gig from both apps and send you
> mp3's if it will help.

That would be fine.

> Can gigdump extract certain samples from the gig file? Maybe I could
> send a single sample and you could figure it out from there. Or tell
> me how I might investigate this more here and I'll give it a try.

Not with gigdump, but with gigextract you can extract samples as .wav files 
(man gigextract). If you just want to extract certain sample(s), then look at 
the output of gigdump, and use the repective sample number(s) as shown there.

> 3) More structural, but LS needs to support more than a stereo output.
> I record LS output into Ardour or Pro Tools and I need piano, drums,
> bass, all on separate stereo channels. Folks doing orchestral are used
> to recording maybe 8-10 separate groups of sounds for mixing later. LS
> doesn't seem to support anything like this today. We need some of the
> features in the DSP station portion of GSt, where we route LS channels
> to certain mixer elements and then attach certain mixer  elements to
> certain hardware outputs. (Or alsa_pcm channels.)

LS aready has that. But audio and MIDI device management (including routing) 
is not yet completely implemented in qsampler. But something like the 
following LSCP script would do it:

# load audio and MIDI driver
CREATE AUDIO_OUTPUT_DEVICE JACK
CREATE MIDI_INPUT_DEVICE ALSA

# increase the amount of output channels of the LS's Jack client e.g. to 4
SET AUDIO_OUTPUT_DEVICE_PARAMETER 0 CHANNELS=4

# connect to MIDI keyboard
SET MIDI_INPUT_PORT_PARAMETER 0 0 ALSA_SEQ_BINDINGS='72:0'

# set up 1st sampler channel (using default output channels, that is 0 and 1)
ADD CHANNEL
LOAD ENGINE gig 0
SET CHANNEL AUDIO_OUTPUT_DEVICE 0 0
SET CHANNEL MIDI_INPUT_DEVICE 0 0
LOAD INSTRUMENT '/home/me/ns_kit7free.gig' 0 0

# set up 2nd sampler channel (using output channels 2 and 3)
ADD CHANNEL
LOAD ENGINE gig 1
SET CHANNEL AUDIO_OUTPUT_DEVICE 1 0
SET CHANNEL MIDI_INPUT_DEVICE 1 0
LOAD INSTRUMENT '/home/me/freepiano.gig' 0 1
# this will alter the audio channel routing
# connect the engines output channel 0 to JACK clients output 2
SET CHANNEL AUDIO_OUTPUT_CHANNEL 1 0 2
# connect the engines output channel 1 to JACK clients output 3
SET CHANNEL AUDIO_OUTPUT_CHANNEL 1 1 3

# not necessary, just to see our setup
GET CHANNEL INFO 0
GET CHANNEL INFO 1

Sorry for the inconvenience. Once the release of LS is out, I will work on 
qsampler, so it covers all the possibilites given by the current version of 
LSCP.

CU
Christian

Re: [Linuxsampler-devel] Starting Probs

[rensinghoff <lui...@fr...>]

> Hope this helps,
> Mark


Nope !!

I keep on getting " Could not connect to server as a client"

[Linuxsampler-devel] Re: bugs

[Garett S. <shu...@co...>]

I do get the No Region defined message quite often when I hit a key 
that... has not region defined. :) However, I haven't seen a segfault 
coralated with this.

The malformed midi message was created by code similar to the following
ev1->data.note.channel=ev2->data.control.value
where ev1 is a noteon event and ev2 is a controller (sustain pedal) 
event. So I think that the channel value might be out of bounds. This 
code is pretty nonsensical and was written in error. However it did 
cause ls to segfault.

Vlad, I have never tried to gdb over irc. I would be happy to learn how 
this is done if you think it would help. I'm pretty sharp with gdb and 
might just be able to get you the info you need myself (although I 
haven't familiarized myself with the ls code at all). Also, I could just 
make you a temporary account on my box and you could gdb through ssh.

I tried the lastest cvs ls and have not seen a disk stream error yet 
with > 2 channels. Right on!

Also, in regards to running ls as root... I just realized this morning 
that the voice dropping I had mentioned to the list last week was caused 
because I had not chmod +s linuxsampler .
When I do not run ls as root on my system I get lots of audio artifacts 
and lots of voice dropping.

-Garett

Mark Knecht wrote:

>I have qsampler set up with three instruments - piano, bass and drums.
>Piano and drums are working well. Bass MIDI is muted in Pro Tools. The
>instant I turned on the bass MIDI channel the segfault occured:
>
>10:41:17.396 Save session: "/home/mark/Chick.lscp".
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 88
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 55
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 55
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 55
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 52
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 55
>Triggering new voice failed!
>gig::Voice: No Region defined for MIDI key 55
>Triggering new voice failed!
>  
>
>>>>FATAL ERROR: Segmentation fault (SIGSEGV) occured! <<<
>>>>        
>>>>
>Showing stack trace...
>No debugger found
>Killing LinuxSampler...
>lscp_client_recv: recv: Connection reset by peer
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>lscp_client_call: send: Broken pipe
>10:44:09.597 Client disconnecting...
>10:44:09.615 Client disconnected.
>10:44:09.617 Server was stopped with exit status 0.
>
>I do not know if the 'No region defined' messages were for the
>piano/drum channels or for the bass.
>
>How can I provide better info? This is a PlanetCCRMA FC2 box. gdb is
>not currently installed so I'll look into that. How do I compile
>LS/qsampler to give you better info?
>
>Thanks,
>Mark
>
>
>-------------------------------------------------------
>The SF.Net email is sponsored by: Beat the post-holiday blues
>Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
>It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
>_______________________________________________
>Linuxsampler-devel mailing list
>Lin...@li...
>https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
>  
>

[Linuxsampler-devel] Re: segfault report - No Region defined?

[Mark K. <mar...@gm...>]

On Sat, 1 Jan 2005 11:10:14 -0800, Mark Knecht <mar...@gm...> wrote:
> 
> I wonder if this could be something like the 'malformed MIDI' issue
> that was reported by Garett yesterday? I'll see if amidirecord can
> capture the incoming MIDI that causes the crash.
> 
> - Mark
> 

And the winner is....... (drum roll) Control Change or Program Change? 

[mark@Godzilla mark]$ arecordmidi --port 64:0 --dump
Chick-bass-midi.mid Waiting for data. Press Ctrl+C to end
_______Tick Event_________________ Ch _Data__
      53411 Control change          1   7  96
      53411 Program change          1  32
      53591 Note on                 1  50  76
      53855 Control change          1  64 127
      53981 Note on                 1  50   0
      54031 Control change          1  64   0
      54037 Note on                 1  57  70
      54174 Control change          1  64 127
      54392 Control change          1  64   0
      54408 Note on                 1  64  86
      54410 Note on                 1  57   0
      54704 Control change          1  64 127
      54816 Note on                 1  64   0
      54950 Note on                 1  57  63

Pro Tools does not seem to have a way to not send these values. It
filters on the input side by not on the transmit side. I don't know
how to filter this on the LS machine. Since Garett reported the
malformed MIDI messages issue WRT a MIDI filter maybe that's not the
issue at all, but rather he just received these same sort of messages
also?

This MIDI file is a commercially available file that works fine with GSt.

Cheers,
Mark

[Linuxsampler-devel] Re: segfault report - No Region defined?

[Mark K. <mar...@gm...>]

On Sat, 1 Jan 2005 10:48:59 -0800, Mark Knecht <mar...@gm...> wrote:
> I have qsampler set up with three instruments - piano, bass and drums.
> Piano and drums are working well. Bass MIDI is muted in Pro Tools. The
> instant I turned on the bass MIDI channel the segfault occured:
> 
<SNIP>
> >>> FATAL ERROR: Segmentation fault (SIGSEGV) occured! <<<
> Showing stack trace...
> No debugger found
> Killing LinuxSampler...
> lscp_client_recv: recv: Connection reset by peer
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> lscp_client_call: send: Broken pipe
> 10:44:09.597 Client disconnecting...
> 10:44:09.615 Client disconnected.
> 10:44:09.617 Server was stopped with exit status 0.
> 
> I do not know if the 'No region defined' messages were for the
> piano/drum channels or for the bass.

The 'no region defined' messages are not part of this problem. I have
redone the crash using only the bass MIDI channel. I get the crash
with none of those message. The appear to be for the drum channel. (I
think.I'll check later.)

I wonder if this could be something like the 'malformed MIDI' issue
that was reported by Garett yesterday? I'll see if amidirecord can
capture the incoming MIDI that causes the crash.

- Mark

Re: [Linuxsampler-devel] Starting Probs

[Mark K. <mar...@gm...>]

Luigi,
   Hi! Welcome! A New Year with New People! Great!

   I run LS by first starting Jack from qjackctl and then starting
qsampler. qsampler starts and runs LinuxSampler for me. I see all the
LS messages in the bottom window of qsampler.

   I run as a normal user but I have all the LSM modules installed and
running here. I don't know if you would have problems running as a
user without them as I haven't tried.

Hope this helps,
Mark


On Sun, 27 Feb 2005 19:56:28 +0100, rensinghoff
<lui...@fr...> wrote:
> Hi List !
> 
> I am having trouble to start linuxsampler/qsampler
> 
> I always get "could not bind to server socket" error in qsampler and
> "Server exit status"
> 
> can i run linuxsampler as user ? or is it necessary to be root ?
> 
> what about jack-support ? Do i have to start linuxsampler with any
> option, unfortunately there is no --help ;-(
> 
> I tried to find any linuxsampler getting started or HOWTO with no
> success
> 
> Thanks for your help in advance
> 
> -------------------------------------------------------
> The SF.Net email is sponsored by: Beat the post-holiday blues
> Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
> It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
> _______________________________________________
> Linuxsampler-devel mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
>

[Linuxsampler-devel] Starting Probs

[rensinghoff <lui...@fr...>]

Hi List !


I am having trouble to start linuxsampler/qsampler

I always get "could not bind to server socket" error in qsampler and
"Server exit status"

can i run linuxsampler as user ? or is it necessary to be root ?

what about jack-support ? Do i have to start linuxsampler with any
option, unfortunately there is no --help ;-(

I tried to find any linuxsampler getting started or HOWTO with no
success

Thanks for your help in advance

[Linuxsampler-devel] segfault report - No Region defined?

[Mark K. <mar...@gm...>]

I have qsampler set up with three instruments - piano, bass and drums.
Piano and drums are working well. Bass MIDI is muted in Pro Tools. The
instant I turned on the bass MIDI channel the segfault occured:

10:41:17.396 Save session: "/home/mark/Chick.lscp".
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 88
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 55
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 55
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 55
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 52
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 55
Triggering new voice failed!
gig::Voice: No Region defined for MIDI key 55
Triggering new voice failed!
>>> FATAL ERROR: Segmentation fault (SIGSEGV) occured! <<<
Showing stack trace...
No debugger found
Killing LinuxSampler...
lscp_client_recv: recv: Connection reset by peer
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
lscp_client_call: send: Broken pipe
10:44:09.597 Client disconnecting...
10:44:09.615 Client disconnected.
10:44:09.617 Server was stopped with exit status 0.

I do not know if the 'No region defined' messages were for the
piano/drum channels or for the bass.

How can I provide better info? This is a PlanetCCRMA FC2 box. gdb is
not currently installed so I'll look into that. How do I compile
LS/qsampler to give you better info?

Thanks,
Mark

Re: [Linuxsampler-devel] Just a report

[Mark K. <mar...@gm...>]

On Sat, 1 Jan 2005 17:15:52 +0100, Christian Schoenebeck
<sch...@so...> wrote:
> Happy new year everybody! :)

And to you!

> 
> Wow, Vladimir was really assiduous and that even on new years eve!
> 
> Es geschah am Samstag 01 Januar 2005 03:12 als Mark Knecht schrieb:
> > > Memory stats could be looked at by tools like "top"
> >
> > Yes, but that doesn't show what each gig file or each sample, etc., is
> 
> We could easily collect the amount of cached sample data in the
> InstrumentResourceManager. But I think we'll better postpone that for the
> time after the first release.

If the info was going to be part of the app then I agree. I'm
wondering if it is possible to write some small support app, just
console based, that could query LS (or Linux itself) for LS memory
usage?

> 
> > When Christian and I discussed this problem, probably about a year
> > ago, it was my understanding he thought that the compression
> > algorythmn might be different than other files and hence this causes
> > the problem. I don['t think we ever tested that idea fully.
> 
> I thought you only had problems with LS's interpretation of articulation
> informations. If there's a problem wirh the decompression algorithm you would
> hear noise or something. Or did I get you wrong?

Sorry. I do hear noise with these gig files. Underneath the noise I
can hear the note, but badly distorted. It's a bass guitar, but I hear
mostly high frequencies only and covered up with noise.

> 
> Mark, you mentioned there was still a problem with your Scarbee J-Fingered
> Bass gig. What exactly was the problem?

As stated above. I can record the gig from both apps and send you
mp3's if it will help.

Can gigdump extract certain samples from the gig file? Maybe I could
send a single sample and you could figure it out from there. Or tell
me how I might investigate this more here and I'll give it a try.

> 
> Vladimir, you are right regarding the disk thread. Of course it makes sense to
> share one single disk thread between all instances of the same sampler engine
> type. I also planned to change that long ago, but did not do it yet due to
> priorities. The engine concept should also be changed a bit, but I propose to
> do these things after the upcoming release.
> 

WRT to the upcoming release I think you guys are really very close. In
my mind there are just a couple of things that must happen either
before or soon after release:

1) More a qsampler issue possibly but based on yesterday's work I'm
having lots of problems saving and restoring setups. I store *.lscp
file and then it won't restore. LS reports that it cannot set up
channels or parameters. I would encourage the group to take a close
look at this and make sure it won't look flaky to new users. (I expect
almost all users to end up using qsampler.)

2) The Scarbee issue as above. It's a pretty popular set of libraries
used by a lot of folks. If you do release without it fixed then we
need to state it doesn't work up front and say when we'll fix it.
(Basic errata issue)

3) More structural, but LS needs to support more than a stereo output.
I record LS output into Ardour or Pro Tools and I need piano, drums,
bass, all on separate stereo channels. Folks doing orchestral are used
to recording maybe 8-10 separate groups of sounds for mixing later. LS
doesn't seem to support anything like this today. We need some of the
features in the DSP station portion of GSt, where we route LS channels
to certain mixer elements and then attach certain mixer  elements to
certain hardware outputs. (Or alsa_pcm channels.)

Those are my thoughts for now. I've just finished build Vladimir's
latest updates so I'll do some more testing today and see how it goes.
My main goal right now is to do some 30-100 voice recordings using Pro
Tools and LS only to see how well it holds up. It broke yesterday so I
hope Vladimir's work helps.

Cheers,
Mark

Re: [Linuxsampler-devel] Just a report

[Christian S. <sch...@so...>]

Happy new year everybody! :)

Wow, Vladimir was really assiduous and that even on new years eve!

Es geschah am Samstag 01 Januar 2005 03:12 als Mark Knecht schrieb:
> > Memory stats could be looked at by tools like "top"
>
> Yes, but that doesn't show what each gig file or each sample, etc., is

We could easily collect the amount of cached sample data in the 
InstrumentResourceManager. But I think we'll better postpone that for the 
time after the first release.

> When Christian and I discussed this problem, probably about a year
> ago, it was my understanding he thought that the compression
> algorythmn might be different than other files and hence this causes
> the problem. I don['t think we ever tested that idea fully.

I thought you only had problems with LS's interpretation of articulation 
informations. If there's a problem wirh the decompression algorithm you would 
hear noise or something. Or did I get you wrong?

Mark, you mentioned there was still a problem with your Scarbee J-Fingered 
Bass gig. What exactly was the problem?

Vladimir, you are right regarding the disk thread. Of course it makes sense to 
share one single disk thread between all instances of the same sampler engine 
type. I also planned to change that long ago, but did not do it yet due to 
priorities. The engine concept should also be changed a bit, but I propose to 
do these things after the upcoming release.

CU
Christian

Re: [Linuxsampler-devel] bugs

[Vladimir S. <ha...@gm...>]

Ok, so i've submitted a fix that should take care of the issue with
streams leaking out when multiple channels are used.
i know there are definitely some issues there still, but hopefully
those will not surface during "normal" use.
For example, one issue is this:
CREATE AUDIO_OUTPUT_DEVICE ALSA CARD='0,0' SAMPLERATE=48000
ADD CHANNEL
LOAD ENGINE GigEngine 0
SET CHANNEL AUDIO_OUTPUT_DEVICE 0 0
CREATE MIDI_INPUT_DEVICE ALSA
SET CHANNEL MIDI_INPUT 0 0 0 0
SET MIDI_INPUT_PORT_PARAMETER 0 0 ALSA_SEQ_BINDINGS='128:0'
LOAD INSTRUMENT 'test.gig' 0 0
SET CHANNEL VOLUME 0 0.15

everything works fine so far but if i just do this one command again:
SET CHANNEL AUDIO_OUTPUT_DEVICE 0 0
then something gets connected incorrectly, some state gets corrupted
somewhere and if i try to trigger any voices after that LS crashes.
I haven't had a chance to look at the crash yet . . . martini is
kicking in now and i'm going to go relax or something :)

hopefully we'll clean up most of the bugs during this year :)))

good night everybody and happy new year :)

On Sat, 1 Jan 2005 03:02:34 -0500, Vladimir Senkov <ha...@gm...> wrote:
> Everybody,
> 
> While looking at this problem i discovered that we have somewhat of a
> design problem or maybe i just don't understand the design.
> Today we have a diskthread for each channel.
> So while there are limits to numbers of streams and voices those
> limits seem to only apply per channel.
> And there is some code in the disk thread that assumes that there is
> only one diskthread.
> So . . . i'm not sure if i should fix the diskthread not to assume
> that it is the only one or if i should make diskthread unique per
> sampler rather than per channel?
> 
> I suppose if instruments "live" on different drives (for example) it
> could make more sense to have separate diskthread for each channel.
> but what about the limits? we should probably rethink limits anyways,
> so for for now i'll see if there is an easy way to fix diskthread not
> to assume that it's the only instance.
> 
> Regards,
> Vladimir.
> 
> > Also, I have been getting disk stream errors when I load 3 or more
> > channels at once. Whichever channel I am playing will eventually (after
> > a minute or two) begin to show disk stream errors. If I begin playing on
> > another channel it will get disk stream errors until there are only two
> > channels that do not have disk stream errors. These two channels will
> > work correctly until another channel is added.
> 


-- 
Regards,
Vladimir

Re: [Linuxsampler-devel] bugs

[Vladimir S. <ha...@gm...>]

Everybody,

While looking at this problem i discovered that we have somewhat of a
design problem or maybe i just don't understand the design.
Today we have a diskthread for each channel.
So while there are limits to numbers of streams and voices those
limits seem to only apply per channel.
And there is some code in the disk thread that assumes that there is
only one diskthread.
So . . . i'm not sure if i should fix the diskthread not to assume
that it is the only one or if i should make diskthread unique per
sampler rather than per channel?

I suppose if instruments "live" on different drives (for example) it
could make more sense to have separate diskthread for each channel.
but what about the limits? we should probably rethink limits anyways,
so for for now i'll see if there is an easy way to fix diskthread not
to assume that it's the only instance.

Regards,
Vladimir.

> Also, I have been getting disk stream errors when I load 3 or more
> channels at once. Whichever channel I am playing will eventually (after
> a minute or two) begin to show disk stream errors. If I begin playing on
> another channel it will get disk stream errors until there are only two
> channels that do not have disk stream errors. These two channels will
> work correctly until another channel is added.

Re: [Linuxsampler-devel] bugs

[Vladimir S. <ha...@gm...>]

Garett,

> I was preparing an email to the list but you just solve these problems
> too fast! :)

only simple ones :)

> >>Also, I guess this is not necessarily a bug, but I have gotten segfaults
> >>from malformed midi messages.
> >>
> >>
> >
> >definitely a bug, we should never segfault. could you send those
> >messages in a midi file?
> >
> >
>  the midi messages where malformed in an alsa sequencer filter program.
> I will try to figure out what the messages contained tomorrow.

one other option is online over the irc gdb session :)
this way you might not have to figure out the messages . . . might be
easier if the problem is simple enough.
Another option is build with increased logging level . .  might or
might not show anything.
The best way is online/irc gdb session imho.

Regards,
Vladimir.

Re: [Linuxsampler-devel] bugs

[Garett S. <shu...@co...>]

Vladimir Senkov wrote:

Happy new Year!

>Hello Garett,
>
>Happy new year!!!
>it's coming up here in 8 minutes :)
>
>  
>
>>Hello Vladimir, I began to notice the invalid synthesis mode error when
>>I would create a channel using one instrument from a gig file, then
>>remove the channel and add another channel with a different instrument
>>from the same gig. After building ls from cvs 15 minutes ago the problem
>>has not appeared.
>>    
>>
>
>cool. please let the mailing list know when you experience problems
>this way someone will fix it sooner :)
>i'm glad the fix seems to work. i'm surprised i haven't seen this
>problem myself, it was kind of a trivial bug, variable was not
>properly initialized.
> 
>  
>
I was preparing an email to the list but you just solve these problems 
too fast! :)

>>Also, I have been getting disk stream errors when I load 3 or more
>>channels at once. Whichever channel I am playing will eventually (after
>>a minute or two) begin to show disk stream errors. If I begin playing on
>>another channel it will get disk stream errors until there are only two
>>channels that do not have disk stream errors. These two channels will
>>work correctly until another channel is added.
>>    
>>
>
>Ok, so unfortunately disk streaming problems are not over yet. I suspected that.
>I'll try to reproduce the problem using the description you provided.
>If there is any more minimalistic step by step reproduction sequence
>please let me know.
>
>  
>
>>Also, I guess this is not necessarily a bug, but I have gotten segfaults
>>from malformed midi messages.
>>    
>>
>
>definitely a bug, we should never segfault. could you send those
>messages in a midi file?
>  
>
 the midi messages where malformed in an alsa sequencer filter program. 
I will try to figure out what the messages contained tomorrow.

> 
>Regards,
>Vladimir.
>
>
>-------------------------------------------------------
>The SF.Net email is sponsored by: Beat the post-holiday blues
>Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
>It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
>_______________________________________________
>Linuxsampler-devel mailing list
>Lin...@li...
>https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
>  
>

Re: [Linuxsampler-devel] bugs

[Vladimir S. <ha...@gm...>]

Hello Garett,

Happy new year!!!
it's coming up here in 8 minutes :)

> Hello Vladimir, I began to notice the invalid synthesis mode error when
> I would create a channel using one instrument from a gig file, then
> remove the channel and add another channel with a different instrument
> from the same gig. After building ls from cvs 15 minutes ago the problem
> has not appeared.

cool. please let the mailing list know when you experience problems
this way someone will fix it sooner :)
i'm glad the fix seems to work. i'm surprised i haven't seen this
problem myself, it was kind of a trivial bug, variable was not
properly initialized.
 
> Also, I have been getting disk stream errors when I load 3 or more
> channels at once. Whichever channel I am playing will eventually (after
> a minute or two) begin to show disk stream errors. If I begin playing on
> another channel it will get disk stream errors until there are only two
> channels that do not have disk stream errors. These two channels will
> work correctly until another channel is added.

Ok, so unfortunately disk streaming problems are not over yet. I suspected that.
I'll try to reproduce the problem using the description you provided.
If there is any more minimalistic step by step reproduction sequence
please let me know.

> Also, I guess this is not necessarily a bug, but I have gotten segfaults
> from malformed midi messages.

definitely a bug, we should never segfault. could you send those
messages in a midi file?
 
Regards,
Vladimir.

[Linuxsampler-devel] bugs

[Garett S. <shu...@co...>]

Hello Vladimir, I began to notice the invalid synthesis mode error when 
I would create a channel using one instrument from a gig file, then 
remove the channel and add another channel with a different instrument 
from the same gig. After building ls from cvs 15 minutes ago the problem 
has not appeared.

Also, I have been getting disk stream errors when I load 3 or more 
channels at once. Whichever channel I am playing will eventually (after 
a minute or two) begin to show disk stream errors. If I begin playing on 
another channel it will get disk stream errors until there are only two 
channels that do not have disk stream errors. These two channels will 
work correctly until another channel is added.

Also, I guess this is not necessarily a bug, but I have gotten segfaults 
from malformed midi messages.

Vladimir Senkov wrote:

>Mark, i've submitted a fix. please take an update from CVS, rebuild
>and try in again, i think it should take care of this problem.
>
>
>On Fri, 31 Dec 2004 21:54:02 -0500, Vladimir Senkov <ha...@gm...> wrote:
>  
>
>>Mark, i think i just found a bug by looking at the code for a few minutes.
>>I'm going to fix it now and let you know, so don't worry about reproducing.
>>
>>On Fri, 31 Dec 2004 21:34:54 -0500, Vladimir Senkov <ha...@gm...> wrote:
>>    
>>
>>>Mark,
>>>
>>>This is definitely a bug:
>>>gig::Synthesizer: Invalid Synthesis Mode: 241
>>>and it causes the server to terminate.
>>>do you consistently get it every time?
>>>If so, i'd like to figure out the minimal set of variables to get this
>>>to reproduce it. i should be able to fix it easily then.
>>>
>>>Regards,
>>>Vladimir.
>>>
>>>      
>>>
>>--
>>Regards,
>>Vladimir
>>
>>    
>>
>
>
>  
>

Re: [Linuxsampler-devel] Problems with LS/qsampler loading gigs

[Mark K. <mar...@gm...>]

Building now. Won't have much more time to test tonight but will start
with it first thing tomorrow.

Thanks! And again, have a very fun and safe New Year's Eve!

- Mark


On Fri, 31 Dec 2004 22:07:20 -0500, Vladimir Senkov <ha...@gm...> wrote:
> Mark, i've submitted a fix. please take an update from CVS, rebuild
> and try in again, i think it should take care of this problem.
> 
> 
> On Fri, 31 Dec 2004 21:54:02 -0500, Vladimir Senkov <ha...@gm...> wrote:
> > Mark, i think i just found a bug by looking at the code for a few minutes.
> > I'm going to fix it now and let you know, so don't worry about reproducing.
> >
> > On Fri, 31 Dec 2004 21:34:54 -0500, Vladimir Senkov <ha...@gm...> wrote:
> > > Mark,
> > >
> > > This is definitely a bug:
> > > gig::Synthesizer: Invalid Synthesis Mode: 241
> > > and it causes the server to terminate.
> > > do you consistently get it every time?
> > > If so, i'd like to figure out the minimal set of variables to get this
> > > to reproduce it. i should be able to fix it easily then.
> > >
> > > Regards,
> > > Vladimir.
> > >
> >
> >
> > --
> > Regards,
> > Vladimir
> >
> 
> --
> Regards,
> Vladimir
> 
> -------------------------------------------------------
> The SF.Net email is sponsored by: Beat the post-holiday blues
> Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
> It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
> _______________________________________________
> Linuxsampler-devel mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
>

Re: [Linuxsampler-devel] Problems with LS/qsampler loading gigs

[Vladimir S. <ha...@gm...>]

Mark, i've submitted a fix. please take an update from CVS, rebuild
and try in again, i think it should take care of this problem.


On Fri, 31 Dec 2004 21:54:02 -0500, Vladimir Senkov <ha...@gm...> wrote:
> Mark, i think i just found a bug by looking at the code for a few minutes.
> I'm going to fix it now and let you know, so don't worry about reproducing.
> 
> On Fri, 31 Dec 2004 21:34:54 -0500, Vladimir Senkov <ha...@gm...> wrote:
> > Mark,
> >
> > This is definitely a bug:
> > gig::Synthesizer: Invalid Synthesis Mode: 241
> > and it causes the server to terminate.
> > do you consistently get it every time?
> > If so, i'd like to figure out the minimal set of variables to get this
> > to reproduce it. i should be able to fix it easily then.
> >
> > Regards,
> > Vladimir.
> >
> 
> 
> --
> Regards,
> Vladimir
> 


-- 
Regards,
Vladimir

Re: [Linuxsampler-devel] Problems with LS/qsampler loading gigs

[Vladimir S. <ha...@gm...>]

Mark, i think i just found a bug by looking at the code for a few minutes.
I'm going to fix it now and let you know, so don't worry about reproducing.

On Fri, 31 Dec 2004 21:34:54 -0500, Vladimir Senkov <ha...@gm...> wrote:
> Mark,
> 
> This is definitely a bug:
> gig::Synthesizer: Invalid Synthesis Mode: 241
> and it causes the server to terminate.
> do you consistently get it every time?
> If so, i'd like to figure out the minimal set of variables to get this
> to reproduce it. i should be able to fix it easily then.
> 
> Regards,
> Vladimir.
> 


-- 
Regards,
Vladimir

Re: [Linuxsampler-devel] Problems with LS/qsampler loading gigs

[Vladimir S. <ha...@gm...>]

Mark,

This is definitely a bug:
gig::Synthesizer: Invalid Synthesis Mode: 241
and it causes the server to terminate.
do you consistently get it every time?
If so, i'd like to figure out the minimal set of variables to get this
to reproduce it. i should be able to fix it easily then.

Regards,
Vladimir.

Re: [Linuxsampler-devel] Problems with LS/qsampler loading gigs

[Mark K. <mar...@gm...>]

On Fri, 31 Dec 2004 21:20:42 -0500, Vladimir Senkov <ha...@gm...> wrote:
> i've just remembered that the gui probably hasn't implemented the
> "background gig loading" feature. so you have to set the timeout to
> something really high for the time being. 30 seconds (or however long
> it takes to load the biggest gig) should do it.
> 
Right. I set it to 6 seconds and seem to get the gigs loaded. However,
when I start Pro Tools playing against 3 gig files - a jazz piece with
piano, bass and drums, it plays for 10 seconds and then has problems
and disappears from QJC's connections window:

18:21:30.595 Channel 0 reset.
18:21:55.123 Channel 1 created.
Starting disk thread...OK
Loading gig file '/mnt/audio1/Gigs/Pianos/Bardstown Audio/The
Bosendorfer Imperial Grand Version 2.2.gig'...OK
Loading gig instrument...OK
Caching initial samples...OK
18:22:45.352 Channel 1 reset.
18:22:59.075 Channel 2 created.
Starting disk thread...OK
Loading gig file '/mnt/audio1/Gigs/Bass/East West Bass/Fender/2PU
SJZHRD B.gig'...OK
Loading gig instrument...OK
Caching initial samples...OK
18:22:59.971 Channel 2 reset.
18:23:17.008 Channel 3 created.
Starting disk thread...OK
Loading gig file '/mnt/audio1/Gigs/Drums/Wizzo/Ambience Kit XXL.gig'...OK
Loading gig instrument...OK
Caching initial samples...OK
gig::Synthesizer: Invalid Synthesis Mode: 241
18:24:11.520 Client disconnecting...
18:24:11.539 Client disconnected.
18:24:11.540 Server was stopped with exit status -1.

Re: [Linuxsampler-devel] Problems with LS/qsampler loading gigs

[Vladimir S. <ha...@gm...>]

i've just remembered that the gui probably hasn't implemented the
"background gig loading" feature. so you have to set the timeout to
something really high for the time being. 30 seconds (or however long
it takes to load the biggest gig) should do it.