Csound Csound-dev Csound-tekno Search About

[Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"

Date2015-12-05 13:08
FromPeter Burgess
Subject[Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?

Date2015-12-05 14:19
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?


Date2015-12-05 14:21
FromRory Walsh
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
It did.

On 5 December 2015 at 14:19, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?



Date2015-12-05 14:35
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
ah ok, sorry. maybe it was the old list that sends a copy straight to me

On Sat, Dec 5, 2015 at 2:21 PM, Rory Walsh <rorywalsh@ear.ie> wrote:
It did.

On 5 December 2015 at 14:19, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?




Date2015-12-05 15:45
Fromjpff
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
The DAG renit message should probably go as it says very little

Two immediatethoughts

1: did you combine stdout aand stderr?  I fnd (in tcsh)
     csound .... |& tee foo
ofte removes some confusions

2: Running under gdb in this case a ^c wen it hangs gives access to a 
backtrace and clues

Of ciyrse this is all linux/unix specific so may be irreklevent.

Also valgrind sometimes indicates where to look

and just thought....
3: does it hang the same with -j1

Date2015-12-05 16:16
FromMichael Gogins
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?


Date2015-12-05 16:17
FromMichael Gogins
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"

Yes, try valgrind, it can sometimes identify data races.

Regards,
Mike

On Dec 5, 2015 11:16 AM, "Michael Gogins" <michael.gogins@gmail.com> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?


Date2015-12-05 16:34
FromVictor Lazzarini
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?


Date2015-12-05 18:05
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?



Date2015-12-05 18:44
FromVictor Lazzarini
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?



Date2015-12-07 10:25
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?




Date2015-12-07 11:01
FromRory Walsh
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?





Date2015-12-07 11:40
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?






Date2015-12-07 11:52
FromRory Walsh
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
If you plan to to use the debugger from the command line, then learn to use it from the command line. If you plan to use the debugger in an IDE(which you already use for development), then learn to use it in your IDE. The underlying commands are the exact same, it's just a different interface. 

On 7 December 2015 at 11:40, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?







Date2015-12-07 11:57
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Hmmm, maybe you're right, I just thought I might benefit from learning what goes on under the hood.

On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
If you plan to to use the debugger from the command line, then learn to use it from the command line. If you plan to use the debugger in an IDE(which you already use for development), then learn to use it in your IDE. The underlying commands are the exact same, it's just a different interface. 

On 7 December 2015 at 11:40, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?








Date2015-12-07 12:12
FromRory Walsh
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Thankfully it's hard to learn to use a debugger in any context without learning about what's going on under the hood ;) 

On 7 December 2015 at 11:57, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hmmm, maybe you're right, I just thought I might benefit from learning what goes on under the hood.

On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
If you plan to to use the debugger from the command line, then learn to use it from the command line. If you plan to use the debugger in an IDE(which you already use for development), then learn to use it in your IDE. The underlying commands are the exact same, it's just a different interface. 

On 7 December 2015 at 11:40, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?









Date2015-12-07 12:19
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Lol, alright then, maybe I'll stick with my IDE

On Mon, Dec 7, 2015 at 12:12 PM, Rory Walsh <rorywalsh@ear.ie> wrote:
Thankfully it's hard to learn to use a debugger in any context without learning about what's going on under the hood ;) 

On 7 December 2015 at 11:57, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hmmm, maybe you're right, I just thought I might benefit from learning what goes on under the hood.

On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
If you plan to to use the debugger from the command line, then learn to use it from the command line. If you plan to use the debugger in an IDE(which you already use for development), then learn to use it in your IDE. The underlying commands are the exact same, it's just a different interface. 

On 7 December 2015 at 11:40, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?










Date2015-12-07 12:41
FromMichael Gogins
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"

Which IDE are you using? Most of them contain an embedded debugger. Did you say CodeBlocks? That one has a pretty good debugger but you might have to configure it.

Regards,
Mike

On Dec 7, 2015 7:19 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Lol, alright then, maybe I'll stick with my IDE

On Mon, Dec 7, 2015 at 12:12 PM, Rory Walsh <rorywalsh@ear.ie> wrote:
Thankfully it's hard to learn to use a debugger in any context without learning about what's going on under the hood ;) 

On 7 December 2015 at 11:57, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hmmm, maybe you're right, I just thought I might benefit from learning what goes on under the hood.

On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
If you plan to to use the debugger from the command line, then learn to use it from the command line. If you plan to use the debugger in an IDE(which you already use for development), then learn to use it in your IDE. The underlying commands are the exact same, it's just a different interface. 

On 7 December 2015 at 11:40, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?










Date2015-12-07 12:49
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Yeah, CodeBlocks. I've just yesterday found the 'watches' and 'running threads' windows, and found out how to get alot more output from the debugger. All this time I've been adding a pile of 'cout' to my code to work out what's going on. What a chump! :D

Pete

On Mon, Dec 7, 2015 at 12:41 PM, Michael Gogins <michael.gogins@gmail.com> wrote:

Which IDE are you using? Most of them contain an embedded debugger. Did you say CodeBlocks? That one has a pretty good debugger but you might have to configure it.

Regards,
Mike

On Dec 7, 2015 7:19 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Lol, alright then, maybe I'll stick with my IDE

On Mon, Dec 7, 2015 at 12:12 PM, Rory Walsh <rorywalsh@ear.ie> wrote:
Thankfully it's hard to learn to use a debugger in any context without learning about what's going on under the hood ;) 

On 7 December 2015 at 11:57, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hmmm, maybe you're right, I just thought I might benefit from learning what goes on under the hood.

On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
If you plan to to use the debugger from the command line, then learn to use it from the command line. If you plan to use the debugger in an IDE(which you already use for development), then learn to use it in your IDE. The underlying commands are the exact same, it's just a different interface. 

On 7 December 2015 at 11:40, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
I know, but I don't know anything about the debugger commands, so I'd probably learn more about it if I used it from the command line. Up until this conversation I'd never thought to change any settings to get more info out of the debugger.

On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
No one said anything about having to use a debugger from the command line! CodeBlocks, along with most IDEs provide access to gdb through their graphical interface. 


Just don't forget to build and link against the debug version of Csound. Otherwise you won't be able to step through its code. 

On 7 December 2015 at 10:25, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Ah yes, I hadn't rebuilt with the new settings before debugging again.

I think I am starting to track down the bug now, it's not a csound or libsnd error at all. I'm sorry to hassle you guys about it. I've just realised that I reset csound just after each performance cycle, which is when it displays:

Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25

Nothing else comes up in the terminal until the next csound run at the moment, so that means, based on the terminal output alone, the freeze up could occur anywhere in my code between the two csound performances.

Victor: You're right ofcourse about learning the debugger, I've taken the easy way out by using an IDE to program in, which means I have yet to learn some of the basics, like how to build and debug from the command line and things like that. I am generally reliant on settings in the IDE. I should spend a bit of time learning to program from the command line, if nothing else, it will probably help me use the IDE better.

Cheers for all your continuing help and advice and guys!

Pete

On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
You need to build with the Debug configuration otherwise you will not be able to look up your symbols.

I suggest you take some time out learning how to use gdb, if you have not already done
so. Learning about valgrind is also useful,
but gdb is essential for the work you want to
do.

regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 18:05, Peter Burgess <pete.soundtechnician@GMAIL.COM> wrote:

Thanks for all your responses!

I forgot to clarify that I am on Linux (ubuntu 14.04) and using the Codeblocks IDE. It's also occured to me that I haven't changed over to 6.06 yet, should I do that? I still see lots of messages regarding bug fixes for the new version.

I've just looked up stderr. I don't think I've used that, but then I can't remember what I've done in the GUI code, nor do I know what the GUI library itself might do.

I don't know anything about debugger commands, and didn't know what to do with ^c (pressing shift c in terminal closes it) but I just looked up how to get a backtrace, and I've put bt 100 in the debugger initialisation commands in Codeblocks.This makes the debugger say "no stack" lol. I only get to see the callstack when an error is detected. It would be helpful if I could find a way to see the callstack when I want... that side, I've also changed some settings for my debugger that means it shows me lods more info now. I got this error just now:

[debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No such file or directory.

It looked like it was just stuck at the same point when it happened, but it also threw a segmentation fault. It hasn't thrown a segmentation fault or produced this error on any other occasion, so this might be a different error again... I'm researching the memcpy error now anyway, see if that leads me anywhere.

I have downloaded valgrind, but my app never quite gets going when I try running it under valgrind. Maybe I need to be more patient, just read it slows the run-speed down by around 30 times.

I will also test with -j1, and I'll get back to you guys with my results.

On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
There is an issue open on a related matter. It appears that there is either some memory corruption or there is something that is not quite reentrant (in the parsing stage). It seems to happen on OSX, not Linux.

I could not get to the bottom of it yet.

Regards

Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland

On 5 Dec 2015, at 16:16, Michael Gogins <michael.gogins@GMAIL.COM> wrote:

I got it. I too find sometimes  csound fails after many restarts.

Regards,
Mike

On Dec 5, 2015 9:20 AM, "Peter Burgess" <pete.soundtechnician@gmail.com> wrote:
Did the previous message get through? Usually I get a copy in my inbox straight away, but nothing has come through...

On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess <pete.soundtechnician@gmail.com> wrote:
Hi there. I am trying to work out why csound occasionally freezes after the output "libsndfile-1.0.25". I am running csound through the c++ API. It runs and resets csound many times over, and just occasionally it trips up as mentioned above.

The debugger won't pick it up, presumably because it's not throwing up any errors. I have tried looking through the output using verbose mode, but I'm not sure if that reveals anything either. Here is a portion of the output (I'm not sure how much is relevant, so I've included a reasonable chunk)

...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 1
removed instance of instr 1002
removed instance of instr 2010
deactivated all notes to time  20.212
frstoff = 0x3980e60
B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
dag_num_active = 1

Who depends on 0 (instr 2000)?
*** 1 tasks
0(2000): status=AVAILABLE (watchList )
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
...
[ It repeats this line "DAG REINIT*************************" so many times it took me around a minute to scroll up to the top of it in terminal ]
...
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
DAG REINIT************************
removed instance of instr 2000
deactivated all notes to time  25.212
frstoff = (nil)
B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
Score finished in csoundPerform().
resetting Csound instance
removed instance of instr 0
auxlist for instr 2 (0x3970a30):
    auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
    auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
auxlist for instr 2 (0x3970a30):
auxlist for instr 11011 (0x39b1e00):
    auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
    auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
    auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
    auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
    auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
auxlist for instr 11011 (0x39b1e00):
auxlist for instr 2000 (0x3980e60):
    auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp 0x2daee38
    auxch at 0x3982830: size 159792, auxp 0x3986940, endp 0x39ad970
    auxch at 0x3982810: size 800, auxp 0x23dff70, endp 0x23e0290
    auxch at 0x3982890: size 176, auxp 0x2512ae0, endp 0x2512b90
    auxch at 0x3982870: size 304, auxp 0x2605670, endp 0x26057a0
auxlist for instr 2000 (0x3980e60):
inactive allocs returned to freespace
end of score.           overall amps:   2.5645   2.5633
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
32768 65536 sample blks of shorts written to csound/wav/1.2.wav (WAV)
signalflowgraph: csoundModuleDestroy(0x65f130)
virtual_keyboard real time MIDI plugin for Csound
WARNING: STK opodes not available: define environment variable RAWWAVE_PATH
(points to rawwaves directory) to use STK opcodes.
0dBFS level = 32768.0
Csound version 6.05 (double samples) Aug  3 2015
libsndfile-1.0.25


After this, it does nothing until you close the app or terminal window. Any thoughts?











Date2015-12-07 14:21
FromMichael Gogins
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
You are not a chump, you are learning.

You may be interested to read this:
http://michaelgogins.tumblr.com/How_to_Program

Regards,
Mike

-----------------------------------------------------
Michael Gogins
Irreducible Productions
http://michaelgogins.tumblr.com
Michael dot Gogins at gmail dot com


On Mon, Dec 7, 2015 at 7:49 AM, Peter Burgess
 wrote:
> Yeah, CodeBlocks. I've just yesterday found the 'watches' and 'running
> threads' windows, and found out how to get alot more output from the
> debugger. All this time I've been adding a pile of 'cout' to my code to work
> out what's going on. What a chump! :D
>
> Pete
>
> On Mon, Dec 7, 2015 at 12:41 PM, Michael Gogins 
> wrote:
>>
>> Which IDE are you using? Most of them contain an embedded debugger. Did
>> you say CodeBlocks? That one has a pretty good debugger but you might have
>> to configure it.
>>
>> Regards,
>> Mike
>>
>> On Dec 7, 2015 7:19 AM, "Peter Burgess" 
>> wrote:
>>>
>>> Lol, alright then, maybe I'll stick with my IDE
>>>
>>> On Mon, Dec 7, 2015 at 12:12 PM, Rory Walsh  wrote:
>>>>
>>>> Thankfully it's hard to learn to use a debugger in any context without
>>>> learning about what's going on under the hood ;)
>>>>
>>>> On 7 December 2015 at 11:57, Peter Burgess
>>>>  wrote:
>>>>>
>>>>> Hmmm, maybe you're right, I just thought I might benefit from learning
>>>>> what goes on under the hood.
>>>>>
>>>>> On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh  wrote:
>>>>>>
>>>>>> If you plan to to use the debugger from the command line, then learn
>>>>>> to use it from the command line. If you plan to use the debugger in an
>>>>>> IDE(which you already use for development), then learn to use it in your
>>>>>> IDE. The underlying commands are the exact same, it's just a different
>>>>>> interface.
>>>>>>
>>>>>> On 7 December 2015 at 11:40, Peter Burgess
>>>>>>  wrote:
>>>>>>>
>>>>>>> I know, but I don't know anything about the debugger commands, so I'd
>>>>>>> probably learn more about it if I used it from the command line. Up until
>>>>>>> this conversation I'd never thought to change any settings to get more info
>>>>>>> out of the debugger.
>>>>>>>
>>>>>>> On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh  wrote:
>>>>>>>>
>>>>>>>> No one said anything about having to use a debugger from the command
>>>>>>>> line! CodeBlocks, along with most IDEs provide access to gdb through their
>>>>>>>> graphical interface.
>>>>>>>>
>>>>>>>>
>>>>>>>> https://www.youtube.com/results?search_query=code%3A%3Ablocks+debugging
>>>>>>>>
>>>>>>>> Just don't forget to build and link against the debug version of
>>>>>>>> Csound. Otherwise you won't be able to step through its code.
>>>>>>>>
>>>>>>>> On 7 December 2015 at 10:25, Peter Burgess
>>>>>>>>  wrote:
>>>>>>>>>
>>>>>>>>> Ah yes, I hadn't rebuilt with the new settings before debugging
>>>>>>>>> again.
>>>>>>>>>
>>>>>>>>> I think I am starting to track down the bug now, it's not a csound
>>>>>>>>> or libsnd error at all. I'm sorry to hassle you guys about it. I've just
>>>>>>>>> realised that I reset csound just after each performance cycle, which is
>>>>>>>>> when it displays:
>>>>>>>>>
>>>>>>>>>> Csound version 6.05 (double samples) Aug  3 2015
>>>>>>>>>> libsndfile-1.0.25
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Nothing else comes up in the terminal until the next csound run at
>>>>>>>>> the moment, so that means, based on the terminal output alone, the freeze up
>>>>>>>>> could occur anywhere in my code between the two csound performances.
>>>>>>>>>
>>>>>>>>> Victor: You're right ofcourse about learning the debugger, I've
>>>>>>>>> taken the easy way out by using an IDE to program in, which means I have yet
>>>>>>>>> to learn some of the basics, like how to build and debug from the command
>>>>>>>>> line and things like that. I am generally reliant on settings in the IDE. I
>>>>>>>>> should spend a bit of time learning to program from the command line, if
>>>>>>>>> nothing else, it will probably help me use the IDE better.
>>>>>>>>>
>>>>>>>>> Cheers for all your continuing help and advice and guys!
>>>>>>>>>
>>>>>>>>> Pete
>>>>>>>>>
>>>>>>>>> On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini
>>>>>>>>>  wrote:
>>>>>>>>>>
>>>>>>>>>> You need to build with the Debug configuration otherwise you will
>>>>>>>>>> not be able to look up your symbols.
>>>>>>>>>>
>>>>>>>>>> I suggest you take some time out learning how to use gdb, if you
>>>>>>>>>> have not already done
>>>>>>>>>> so. Learning about valgrind is also useful,
>>>>>>>>>> but gdb is essential for the work you want to
>>>>>>>>>> do.
>>>>>>>>>>
>>>>>>>>>> regards
>>>>>>>>>>
>>>>>>>>>> Victor Lazzarini
>>>>>>>>>> Dean of Arts, Celtic Studies, and Philosophy
>>>>>>>>>> Maynooth University
>>>>>>>>>> Ireland
>>>>>>>>>>
>>>>>>>>>> On 5 Dec 2015, at 18:05, Peter Burgess
>>>>>>>>>>  wrote:
>>>>>>>>>>
>>>>>>>>>> Thanks for all your responses!
>>>>>>>>>>
>>>>>>>>>> I forgot to clarify that I am on Linux (ubuntu 14.04) and using
>>>>>>>>>> the Codeblocks IDE. It's also occured to me that I haven't changed over to
>>>>>>>>>> 6.06 yet, should I do that? I still see lots of messages regarding bug fixes
>>>>>>>>>> for the new version.
>>>>>>>>>>
>>>>>>>>>> I've just looked up stderr. I don't think I've used that, but then
>>>>>>>>>> I can't remember what I've done in the GUI code, nor do I know what the GUI
>>>>>>>>>> library itself might do.
>>>>>>>>>>
>>>>>>>>>> I don't know anything about debugger commands, and didn't know
>>>>>>>>>> what to do with ^c (pressing shift c in terminal closes it) but I just
>>>>>>>>>> looked up how to get a backtrace, and I've put bt 100 in the debugger
>>>>>>>>>> initialisation commands in Codeblocks.This makes the debugger say "no stack"
>>>>>>>>>> lol. I only get to see the callstack when an error is detected. It would be
>>>>>>>>>> helpful if I could find a way to see the callstack when I want... that side,
>>>>>>>>>> I've also changed some settings for my debugger that means it shows me lods
>>>>>>>>>> more info now. I got this error just now:
>>>>>>>>>>
>>>>>>>>>> [debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No
>>>>>>>>>> such file or directory.
>>>>>>>>>>
>>>>>>>>>> It looked like it was just stuck at the same point when it
>>>>>>>>>> happened, but it also threw a segmentation fault. It hasn't thrown a
>>>>>>>>>> segmentation fault or produced this error on any other occasion, so this
>>>>>>>>>> might be a different error again... I'm researching the memcpy error now
>>>>>>>>>> anyway, see if that leads me anywhere.
>>>>>>>>>>
>>>>>>>>>> I have downloaded valgrind, but my app never quite gets going when
>>>>>>>>>> I try running it under valgrind. Maybe I need to be more patient, just read
>>>>>>>>>> it slows the run-speed down by around 30 times.
>>>>>>>>>>
>>>>>>>>>> I will also test with -j1, and I'll get back to you guys with my
>>>>>>>>>> results.
>>>>>>>>>>
>>>>>>>>>> On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini
>>>>>>>>>>  wrote:
>>>>>>>>>>>
>>>>>>>>>>> There is an issue open on a related matter. It appears that there
>>>>>>>>>>> is either some memory corruption or there is something that is not quite
>>>>>>>>>>> reentrant (in the parsing stage). It seems to happen on OSX, not Linux.
>>>>>>>>>>>
>>>>>>>>>>> I could not get to the bottom of it yet.
>>>>>>>>>>>
>>>>>>>>>>> Regards
>>>>>>>>>>>
>>>>>>>>>>> Victor Lazzarini
>>>>>>>>>>> Dean of Arts, Celtic Studies, and Philosophy
>>>>>>>>>>> Maynooth University
>>>>>>>>>>> Ireland
>>>>>>>>>>>
>>>>>>>>>>> On 5 Dec 2015, at 16:16, Michael Gogins
>>>>>>>>>>>  wrote:
>>>>>>>>>>>
>>>>>>>>>>> I got it. I too find sometimes  csound fails after many restarts.
>>>>>>>>>>>
>>>>>>>>>>> Regards,
>>>>>>>>>>> Mike
>>>>>>>>>>>
>>>>>>>>>>> On Dec 5, 2015 9:20 AM, "Peter Burgess"
>>>>>>>>>>>  wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> Did the previous message get through? Usually I get a copy in my
>>>>>>>>>>>> inbox straight away, but nothing has come through...
>>>>>>>>>>>>
>>>>>>>>>>>> On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess
>>>>>>>>>>>>  wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>> Hi there. I am trying to work out why csound occasionally
>>>>>>>>>>>>> freezes after the output "libsndfile-1.0.25". I am running csound through
>>>>>>>>>>>>> the c++ API. It runs and resets csound many times over, and just
>>>>>>>>>>>>> occasionally it trips up as mentioned above.
>>>>>>>>>>>>>
>>>>>>>>>>>>> The debugger won't pick it up, presumably because it's not
>>>>>>>>>>>>> throwing up any errors. I have tried looking through the output using
>>>>>>>>>>>>> verbose mode, but I'm not sure if that reveals anything either. Here is a
>>>>>>>>>>>>> portion of the output (I'm not sure how much is relevant, so I've included a
>>>>>>>>>>>>> reasonable chunk)
>>>>>>>>>>>>>
>>>>>>>>>>>>>> ...
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> removed instance of instr 1
>>>>>>>>>>>>>> removed instance of instr 1002
>>>>>>>>>>>>>> removed instance of instr 2010
>>>>>>>>>>>>>> deactivated all notes to time  20.212
>>>>>>>>>>>>>> frstoff = 0x3980e60
>>>>>>>>>>>>>> B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
>>>>>>>>>>>>>> dag_num_active = 1
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Who depends on 0 (instr 2000)?
>>>>>>>>>>>>>> *** 1 tasks
>>>>>>>>>>>>>> 0(2000): status=AVAILABLE (watchList )
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> ...
>>>>>>>>>>>>>
>>>>>>>>>>>>> [ It repeats this line "DAG REINIT*************************" so
>>>>>>>>>>>>> many times it took me around a minute to scroll up to the top of it in
>>>>>>>>>>>>> terminal ]
>>>>>>>>>>>>> ...
>>>>>>>>>>>>>
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> removed instance of instr 2000
>>>>>>>>>>>>> deactivated all notes to time  25.212
>>>>>>>>>>>>> frstoff = (nil)
>>>>>>>>>>>>> B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
>>>>>>>>>>>>> Score finished in csoundPerform().
>>>>>>>>>>>>> resetting Csound instance
>>>>>>>>>>>>> removed instance of instr 0
>>>>>>>>>>>>> auxlist for instr 2 (0x3970a30):
>>>>>>>>>>>>>     auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
>>>>>>>>>>>>>     auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
>>>>>>>>>>>>> auxlist for instr 2 (0x3970a30):
>>>>>>>>>>>>> auxlist for instr 11011 (0x39b1e00):
>>>>>>>>>>>>>     auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
>>>>>>>>>>>>>     auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
>>>>>>>>>>>>>     auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
>>>>>>>>>>>>>     auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
>>>>>>>>>>>>>     auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
>>>>>>>>>>>>> auxlist for instr 11011 (0x39b1e00):
>>>>>>>>>>>>> auxlist for instr 2000 (0x3980e60):
>>>>>>>>>>>>>     auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp
>>>>>>>>>>>>> 0x2daee38
>>>>>>>>>>>>>     auxch at 0x3982830: size 159792, auxp 0x3986940, endp
>>>>>>>>>>>>> 0x39ad970
>>>>>>>>>>>>>     auxch at 0x3982810: size 800, auxp 0x23dff70, endp
>>>>>>>>>>>>> 0x23e0290
>>>>>>>>>>>>>     auxch at 0x3982890: size 176, auxp 0x2512ae0, endp
>>>>>>>>>>>>> 0x2512b90
>>>>>>>>>>>>>     auxch at 0x3982870: size 304, auxp 0x2605670, endp
>>>>>>>>>>>>> 0x26057a0
>>>>>>>>>>>>> auxlist for instr 2000 (0x3980e60):
>>>>>>>>>>>>> inactive allocs returned to freespace
>>>>>>>>>>>>> end of score.           overall amps:   2.5645   2.5633
>>>>>>>>>>>>>        overall samples out of range:        0        0
>>>>>>>>>>>>> 0 errors in performance
>>>>>>>>>>>>> Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
>>>>>>>>>>>>> 32768 65536 sample blks of shorts written to csound/wav/1.2.wav
>>>>>>>>>>>>> (WAV)
>>>>>>>>>>>>> signalflowgraph: csoundModuleDestroy(0x65f130)
>>>>>>>>>>>>> virtual_keyboard real time MIDI plugin for Csound
>>>>>>>>>>>>> WARNING: STK opodes not available: define environment variable
>>>>>>>>>>>>> RAWWAVE_PATH
>>>>>>>>>>>>> (points to rawwaves directory) to use STK opcodes.
>>>>>>>>>>>>> 0dBFS level = 32768.0
>>>>>>>>>>>>> Csound version 6.05 (double samples) Aug  3 2015
>>>>>>>>>>>>> libsndfile-1.0.25
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> After this, it does nothing until you close the app or terminal
>>>>>>>>>>>>> window. Any thoughts?
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>

Date2015-12-07 15:00
FromPeter Burgess
SubjectRe: [Csnd-dev] csound occasionally freezes after "libsndfile-1.0.25"
Haha, I know, and it's probably a good thing I didn't know about the extra debugger tools really. It forced me to think really hard about what my program was doing instead, and there's no harm in that.

I'll give that a look. I've been pooling my learning resources recently. Just started reading "The Linux Programming Interface" as well. I'm reading it from the start, and it's amazing how much basic stuff I didn't know! Every time I think I'm really starting to nail programming (high level programming at least), something happens that shows me that I'm still only scratching the surface.

On Mon, Dec 7, 2015 at 2:21 PM, Michael Gogins <michael.gogins@gmail.com> wrote:
You are not a chump, you are learning.

You may be interested to read this:
http://michaelgogins.tumblr.com/How_to_Program

Regards,
Mike

-----------------------------------------------------
Michael Gogins
Irreducible Productions
http://michaelgogins.tumblr.com
Michael dot Gogins at gmail dot com


On Mon, Dec 7, 2015 at 7:49 AM, Peter Burgess
<pete.soundtechnician@gmail.com> wrote:
> Yeah, CodeBlocks. I've just yesterday found the 'watches' and 'running
> threads' windows, and found out how to get alot more output from the
> debugger. All this time I've been adding a pile of 'cout' to my code to work
> out what's going on. What a chump! :D
>
> Pete
>
> On Mon, Dec 7, 2015 at 12:41 PM, Michael Gogins <michael.gogins@gmail.com>
> wrote:
>>
>> Which IDE are you using? Most of them contain an embedded debugger. Did
>> you say CodeBlocks? That one has a pretty good debugger but you might have
>> to configure it.
>>
>> Regards,
>> Mike
>>
>> On Dec 7, 2015 7:19 AM, "Peter Burgess" <pete.soundtechnician@gmail.com>
>> wrote:
>>>
>>> Lol, alright then, maybe I'll stick with my IDE
>>>
>>> On Mon, Dec 7, 2015 at 12:12 PM, Rory Walsh <rorywalsh@ear.ie> wrote:
>>>>
>>>> Thankfully it's hard to learn to use a debugger in any context without
>>>> learning about what's going on under the hood ;)
>>>>
>>>> On 7 December 2015 at 11:57, Peter Burgess
>>>> <pete.soundtechnician@gmail.com> wrote:
>>>>>
>>>>> Hmmm, maybe you're right, I just thought I might benefit from learning
>>>>> what goes on under the hood.
>>>>>
>>>>> On Mon, Dec 7, 2015 at 11:52 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
>>>>>>
>>>>>> If you plan to to use the debugger from the command line, then learn
>>>>>> to use it from the command line. If you plan to use the debugger in an
>>>>>> IDE(which you already use for development), then learn to use it in your
>>>>>> IDE. The underlying commands are the exact same, it's just a different
>>>>>> interface.
>>>>>>
>>>>>> On 7 December 2015 at 11:40, Peter Burgess
>>>>>> <pete.soundtechnician@gmail.com> wrote:
>>>>>>>
>>>>>>> I know, but I don't know anything about the debugger commands, so I'd
>>>>>>> probably learn more about it if I used it from the command line. Up until
>>>>>>> this conversation I'd never thought to change any settings to get more info
>>>>>>> out of the debugger.
>>>>>>>
>>>>>>> On Mon, Dec 7, 2015 at 11:01 AM, Rory Walsh <rorywalsh@ear.ie> wrote:
>>>>>>>>
>>>>>>>> No one said anything about having to use a debugger from the command
>>>>>>>> line! CodeBlocks, along with most IDEs provide access to gdb through their
>>>>>>>> graphical interface.
>>>>>>>>
>>>>>>>>
>>>>>>>> https://www.youtube.com/results?search_query=code%3A%3Ablocks+debugging
>>>>>>>>
>>>>>>>> Just don't forget to build and link against the debug version of
>>>>>>>> Csound. Otherwise you won't be able to step through its code.
>>>>>>>>
>>>>>>>> On 7 December 2015 at 10:25, Peter Burgess
>>>>>>>> <pete.soundtechnician@gmail.com> wrote:
>>>>>>>>>
>>>>>>>>> Ah yes, I hadn't rebuilt with the new settings before debugging
>>>>>>>>> again.
>>>>>>>>>
>>>>>>>>> I think I am starting to track down the bug now, it's not a csound
>>>>>>>>> or libsnd error at all. I'm sorry to hassle you guys about it. I've just
>>>>>>>>> realised that I reset csound just after each performance cycle, which is
>>>>>>>>> when it displays:
>>>>>>>>>
>>>>>>>>>> Csound version 6.05 (double samples) Aug  3 2015
>>>>>>>>>> libsndfile-1.0.25
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Nothing else comes up in the terminal until the next csound run at
>>>>>>>>> the moment, so that means, based on the terminal output alone, the freeze up
>>>>>>>>> could occur anywhere in my code between the two csound performances.
>>>>>>>>>
>>>>>>>>> Victor: You're right ofcourse about learning the debugger, I've
>>>>>>>>> taken the easy way out by using an IDE to program in, which means I have yet
>>>>>>>>> to learn some of the basics, like how to build and debug from the command
>>>>>>>>> line and things like that. I am generally reliant on settings in the IDE. I
>>>>>>>>> should spend a bit of time learning to program from the command line, if
>>>>>>>>> nothing else, it will probably help me use the IDE better.
>>>>>>>>>
>>>>>>>>> Cheers for all your continuing help and advice and guys!
>>>>>>>>>
>>>>>>>>> Pete
>>>>>>>>>
>>>>>>>>> On Sat, Dec 5, 2015 at 6:44 PM, Victor Lazzarini
>>>>>>>>> <Victor.Lazzarini@nuim.ie> wrote:
>>>>>>>>>>
>>>>>>>>>> You need to build with the Debug configuration otherwise you will
>>>>>>>>>> not be able to look up your symbols.
>>>>>>>>>>
>>>>>>>>>> I suggest you take some time out learning how to use gdb, if you
>>>>>>>>>> have not already done
>>>>>>>>>> so. Learning about valgrind is also useful,
>>>>>>>>>> but gdb is essential for the work you want to
>>>>>>>>>> do.
>>>>>>>>>>
>>>>>>>>>> regards
>>>>>>>>>>
>>>>>>>>>> Victor Lazzarini
>>>>>>>>>> Dean of Arts, Celtic Studies, and Philosophy
>>>>>>>>>> Maynooth University
>>>>>>>>>> Ireland
>>>>>>>>>>
>>>>>>>>>> On 5 Dec 2015, at 18:05, Peter Burgess
>>>>>>>>>> <pete.soundtechnician@GMAIL.COM> wrote:
>>>>>>>>>>
>>>>>>>>>> Thanks for all your responses!
>>>>>>>>>>
>>>>>>>>>> I forgot to clarify that I am on Linux (ubuntu 14.04) and using
>>>>>>>>>> the Codeblocks IDE. It's also occured to me that I haven't changed over to
>>>>>>>>>> 6.06 yet, should I do that? I still see lots of messages regarding bug fixes
>>>>>>>>>> for the new version.
>>>>>>>>>>
>>>>>>>>>> I've just looked up stderr. I don't think I've used that, but then
>>>>>>>>>> I can't remember what I've done in the GUI code, nor do I know what the GUI
>>>>>>>>>> library itself might do.
>>>>>>>>>>
>>>>>>>>>> I don't know anything about debugger commands, and didn't know
>>>>>>>>>> what to do with ^c (pressing shift c in terminal closes it) but I just
>>>>>>>>>> looked up how to get a backtrace, and I've put bt 100 in the debugger
>>>>>>>>>> initialisation commands in Codeblocks.This makes the debugger say "no stack"
>>>>>>>>>> lol. I only get to see the callstack when an error is detected. It would be
>>>>>>>>>> helpful if I could find a way to see the callstack when I want... that side,
>>>>>>>>>> I've also changed some settings for my debugger that means it shows me lods
>>>>>>>>>> more info now. I got this error just now:
>>>>>>>>>>
>>>>>>>>>> [debug]1546    ../sysdeps/x86_64/multiarch/memcpy-ssse3-back.S: No
>>>>>>>>>> such file or directory.
>>>>>>>>>>
>>>>>>>>>> It looked like it was just stuck at the same point when it
>>>>>>>>>> happened, but it also threw a segmentation fault. It hasn't thrown a
>>>>>>>>>> segmentation fault or produced this error on any other occasion, so this
>>>>>>>>>> might be a different error again... I'm researching the memcpy error now
>>>>>>>>>> anyway, see if that leads me anywhere.
>>>>>>>>>>
>>>>>>>>>> I have downloaded valgrind, but my app never quite gets going when
>>>>>>>>>> I try running it under valgrind. Maybe I need to be more patient, just read
>>>>>>>>>> it slows the run-speed down by around 30 times.
>>>>>>>>>>
>>>>>>>>>> I will also test with -j1, and I'll get back to you guys with my
>>>>>>>>>> results.
>>>>>>>>>>
>>>>>>>>>> On Sat, Dec 5, 2015 at 4:34 PM, Victor Lazzarini
>>>>>>>>>> <Victor.Lazzarini@nuim.ie> wrote:
>>>>>>>>>>>
>>>>>>>>>>> There is an issue open on a related matter. It appears that there
>>>>>>>>>>> is either some memory corruption or there is something that is not quite
>>>>>>>>>>> reentrant (in the parsing stage). It seems to happen on OSX, not Linux.
>>>>>>>>>>>
>>>>>>>>>>> I could not get to the bottom of it yet.
>>>>>>>>>>>
>>>>>>>>>>> Regards
>>>>>>>>>>>
>>>>>>>>>>> Victor Lazzarini
>>>>>>>>>>> Dean of Arts, Celtic Studies, and Philosophy
>>>>>>>>>>> Maynooth University
>>>>>>>>>>> Ireland
>>>>>>>>>>>
>>>>>>>>>>> On 5 Dec 2015, at 16:16, Michael Gogins
>>>>>>>>>>> <michael.gogins@GMAIL.COM> wrote:
>>>>>>>>>>>
>>>>>>>>>>> I got it. I too find sometimes  csound fails after many restarts.
>>>>>>>>>>>
>>>>>>>>>>> Regards,
>>>>>>>>>>> Mike
>>>>>>>>>>>
>>>>>>>>>>> On Dec 5, 2015 9:20 AM, "Peter Burgess"
>>>>>>>>>>> <pete.soundtechnician@gmail.com> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> Did the previous message get through? Usually I get a copy in my
>>>>>>>>>>>> inbox straight away, but nothing has come through...
>>>>>>>>>>>>
>>>>>>>>>>>> On Sat, Dec 5, 2015 at 1:08 PM, Peter Burgess
>>>>>>>>>>>> <pete.soundtechnician@gmail.com> wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>> Hi there. I am trying to work out why csound occasionally
>>>>>>>>>>>>> freezes after the output "libsndfile-1.0.25". I am running csound through
>>>>>>>>>>>>> the c++ API. It runs and resets csound many times over, and just
>>>>>>>>>>>>> occasionally it trips up as mentioned above.
>>>>>>>>>>>>>
>>>>>>>>>>>>> The debugger won't pick it up, presumably because it's not
>>>>>>>>>>>>> throwing up any errors. I have tried looking through the output using
>>>>>>>>>>>>> verbose mode, but I'm not sure if that reveals anything either. Here is a
>>>>>>>>>>>>> portion of the output (I'm not sure how much is relevant, so I've included a
>>>>>>>>>>>>> reasonable chunk)
>>>>>>>>>>>>>
>>>>>>>>>>>>>> ...
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> removed instance of instr 1
>>>>>>>>>>>>>> removed instance of instr 1002
>>>>>>>>>>>>>> removed instance of instr 2010
>>>>>>>>>>>>>> deactivated all notes to time  20.212
>>>>>>>>>>>>>> frstoff = 0x3980e60
>>>>>>>>>>>>>> B 20.053 .. 20.211 T 20.211 TT 20.211 M:   1.3811   1.3810
>>>>>>>>>>>>>> dag_num_active = 1
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Who depends on 0 (instr 2000)?
>>>>>>>>>>>>>> *** 1 tasks
>>>>>>>>>>>>>> 0(2000): status=AVAILABLE (watchList )
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>>> ...
>>>>>>>>>>>>>
>>>>>>>>>>>>> [ It repeats this line "DAG REINIT*************************" so
>>>>>>>>>>>>> many times it took me around a minute to scroll up to the top of it in
>>>>>>>>>>>>> terminal ]
>>>>>>>>>>>>> ...
>>>>>>>>>>>>>
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> DAG REINIT************************
>>>>>>>>>>>>> removed instance of instr 2000
>>>>>>>>>>>>> deactivated all notes to time  25.212
>>>>>>>>>>>>> frstoff = (nil)
>>>>>>>>>>>>> B 20.211 .. 25.211 T 25.211 TT 25.211 M:   0.0625   0.0625
>>>>>>>>>>>>> Score finished in csoundPerform().
>>>>>>>>>>>>> resetting Csound instance
>>>>>>>>>>>>> removed instance of instr 0
>>>>>>>>>>>>> auxlist for instr 2 (0x3970a30):
>>>>>>>>>>>>>     auxch at 0x397a9e0: size 48, auxp 0x231a100, endp 0x231a130
>>>>>>>>>>>>>     auxch at 0x3976a90: size 32, auxp 0x235c7f0, endp 0x235c810
>>>>>>>>>>>>> auxlist for instr 2 (0x3970a30):
>>>>>>>>>>>>> auxlist for instr 11011 (0x39b1e00):
>>>>>>>>>>>>>     auxch at 0x39c9e70: size 48, auxp 0x254db90, endp 0x254dbc0
>>>>>>>>>>>>>     auxch at 0x39c4988: size 64, auxp 0x231a150, endp 0x231a190
>>>>>>>>>>>>>     auxch at 0x39c09f0: size 16, auxp 0x2374f90, endp 0x2374fa0
>>>>>>>>>>>>>     auxch at 0x39bcae8: size 32, auxp 0x235c830, endp 0x235c850
>>>>>>>>>>>>>     auxch at 0x39b8b98: size 16, auxp 0x2147350, endp 0x2147360
>>>>>>>>>>>>> auxlist for instr 11011 (0x39b1e00):
>>>>>>>>>>>>> auxlist for instr 2000 (0x3980e60):
>>>>>>>>>>>>>     auxch at 0x3982850: size 8296, auxp 0x2dacdd0, endp
>>>>>>>>>>>>> 0x2daee38
>>>>>>>>>>>>>     auxch at 0x3982830: size 159792, auxp 0x3986940, endp
>>>>>>>>>>>>> 0x39ad970
>>>>>>>>>>>>>     auxch at 0x3982810: size 800, auxp 0x23dff70, endp
>>>>>>>>>>>>> 0x23e0290
>>>>>>>>>>>>>     auxch at 0x3982890: size 176, auxp 0x2512ae0, endp
>>>>>>>>>>>>> 0x2512b90
>>>>>>>>>>>>>     auxch at 0x3982870: size 304, auxp 0x2605670, endp
>>>>>>>>>>>>> 0x26057a0
>>>>>>>>>>>>> auxlist for instr 2000 (0x3980e60):
>>>>>>>>>>>>> inactive allocs returned to freespace
>>>>>>>>>>>>> end of score.           overall amps:   2.5645   2.5633
>>>>>>>>>>>>>        overall samples out of range:        0        0
>>>>>>>>>>>>> 0 errors in performance
>>>>>>>>>>>>> Elapsed time at end of performance: real: 7.552s, CPU: 3.158s
>>>>>>>>>>>>> 32768 65536 sample blks of shorts written to csound/wav/1.2.wav
>>>>>>>>>>>>> (WAV)
>>>>>>>>>>>>> signalflowgraph: csoundModuleDestroy(0x65f130)
>>>>>>>>>>>>> virtual_keyboard real time MIDI plugin for Csound
>>>>>>>>>>>>> WARNING: STK opodes not available: define environment variable
>>>>>>>>>>>>> RAWWAVE_PATH
>>>>>>>>>>>>> (points to rawwaves directory) to use STK opcodes.
>>>>>>>>>>>>> 0dBFS level = 32768.0
>>>>>>>>>>>>> Csound version 6.05 (double samples) Aug  3 2015
>>>>>>>>>>>>> libsndfile-1.0.25
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> After this, it does nothing until you close the app or terminal
>>>>>>>>>>>>> window. Any thoughts?
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>