hi all ,
i am trying to compile two externals of yves's on to windows . there is one issue with threads [i am using pthreads for maximum unix compatibility] that i seem to be having though . there is no usleep() for windows as far as i can tell . however i did fine a function called Sleep() . from my brief browsing of code on the internet . i thought the two for the more part were equivalent in their functionality : to put a thread/process to sleep for a specified amount of time . however , i seem to get *buffer overruns* when i play with the externs . everything compiles fine and is even loaded fine , it seems to be when the external tries to draw to the screen that i get the buffer overrun . the externals i am trying to compile over are sonogram~ 0.9 and cooled~ 0.6 .
thanks .
joge .
Hi,
usleep() takes time in microseconds, Window's sleep() needs milliseconds... this makes sleep() sleep 1000 times longer than wanted if used with the same value...
Olaf
"joge ." schrieb:
hi all ,
i am trying to compile two externals of yves's on to windows . there is one issue with threads [i am using pthreads for maximum unix compatibility] that i seem to be having though . there is no usleep() for windows as far as i can tell . however i did fine a function called Sleep() . from my brief browsing of code on the internet . i thought the two for the more part were equivalent in their functionality : to put a thread/process to sleep for a specified amount of time . however , i seem to get *buffer overruns* when i play with the externs . everything compiles fine and is even loaded fine , it seems to be when the external tries to draw to the screen that i get the buffer overrun . the externals i am trying to compile over are sonogram~ 0.9 and cooled~ 0.6 .
thanks .
joge .
hi,
Olaf's help is very useful here because "sleep" on UNIX takes a number of seconds as argument ( thus, different from Windows' sleep ).
If you ported the code "as is", there should be a delay of 100 seconds before drawing anything, i guess that's not what you expect.
So, divide THREAD_SLEEP_TIME by 1000.
Next, i don't really know what you mean by "buffer overrun", maybe because i never used PD on Windows ???
cheers,
yves
Olaf Matthes wrote:
Hi,
usleep() takes time in microseconds, Window's sleep() needs milliseconds... this makes sleep() sleep 1000 times longer than wanted if used with the same value...
Olaf
"joge ." schrieb:
hi all ,
i am trying to compile two externals of yves's on to windows . there is one issue with threads [i am using pthreads for maximum unix compatibility] that i seem to be having though . there is no usleep() for windows as far as i can tell . however i did fine a function called Sleep() . from my brief browsing of code on the internet . i thought the two for the more part were equivalent in their functionality : to put a thread/process to sleep for a specified amount of time . however , i seem to get *buffer overruns* when i play with the externs . everything compiles fine and is even loaded fine , it seems to be when the external tries to draw to the screen that i get the buffer overrun . the externals i am trying to compile over are sonogram~ 0.9 and cooled~ 0.6 .
thanks .
joge .
hi,
Olaf's help is very useful here because "sleep" on UNIX takes a number of seconds as argument ( thus, different from Windows' sleep ).
i took care of the microseconds to milliseconds issue . thanks .
If you ported the code "as is", there should be a delay of 100 seconds before drawing anything, i guess that's not what you expect.
So, divide THREAD_SLEEP_TIME by 1000.
Next, i don't really know what you mean by "buffer overrun", maybe because i never used PD on Windows ???
i am not sure what a buffer overrun is either . i am imagining its just means that the program is writing to an array past the end of the declared array space .
i have narrowed it down to crashing after the the record button is pressed that the sample is done playing .
cheers,
yves
Olaf Matthes wrote:
Hi,
usleep() takes time in microseconds, Window's sleep() needs
milliseconds... this
makes sleep() sleep 1000 times longer than wanted if used with the same
value...
Olaf
"joge ." schrieb:
hi all ,
i am trying to compile two externals of yves's on to windows . there
is
one issue with threads [i am using pthreads for maximum unix
compatibility]
that i seem to be having though . there is no usleep() for windows as
far
as i can tell . however i did fine a function called Sleep() . from my brief browsing of code on the internet . i thought the two for the more part were equivalent in their functionality : to put a thread/process
to
sleep for a specified amount of time . however , i seem to get *buffer overruns* when i play with the externs . everything compiles fine and
is
even loaded fine , it seems to be when the external tries to draw to the screen that i get the buffer overrun . the externals i am trying to
compile
over are sonogram~ 0.9 and cooled~ 0.6 .
thanks .
joge .
here is some info on the Buffer Overrun term :
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vclib/html/ vclrf_set_security_error_handler.asp
[you might have to copy both lines individually and glue them back together in your browser]
also here is the exact message that i get from the Visual C++ Runtime :
Buffer overrun detected!
Program: D:\downloads\pd\pd-0.35-test23\bin\pd.exe
A buffer overrun has been detected which has corrupted the program's internal state. The program cannot safely continue execution and must now be terminated.
.........
examining the code some more , the error seems to happen when the extern [i am test all of this on sonogram~ frist] tries to draw the sonogram after the record message is hit .
joge .
hi,
Olaf's help is very useful here because "sleep" on UNIX takes a number of seconds as argument ( thus, different from Windows' sleep ).
i took care of the microseconds to milliseconds issue . thanks .
If you ported the code "as is", there should be a delay of 100 seconds before drawing anything, i guess that's not what you expect.
So, divide THREAD_SLEEP_TIME by 1000.
Next, i don't really know what you mean by "buffer overrun", maybe because i never used PD on Windows ???
i am not sure what a buffer overrun is either . i am imagining its just means that the program is writing to an array past the end of the declared array space .
i have narrowed it down to crashing after the the record button is pressed that the sample is done playing .
cheers,
yves
Olaf Matthes wrote:
Hi,
usleep() takes time in microseconds, Window's sleep() needs
milliseconds... this
makes sleep() sleep 1000 times longer than wanted if used with the same
value...
Olaf
"joge ." schrieb:
hi all ,
i am trying to compile two externals of yves's on to windows . there
is
one issue with threads [i am using pthreads for maximum unix
compatibility]
that i seem to be having though . there is no usleep() for windows as
far
as i can tell . however i did fine a function called Sleep() . from
my
brief browsing of code on the internet . i thought the two for the
more
part were equivalent in their functionality : to put a thread/process
to
sleep for a specified amount of time . however , i seem to get
*buffer
overruns* when i play with the externs . everything compiles fine and
is
even loaded fine , it seems to be when the external tries to draw to
the
screen that i get the buffer overrun . the externals i am trying to
compile
over are sonogram~ 0.9 and cooled~ 0.6 .
thanks .
joge .
If you ported the code "as is", there should be a delay of 100 seconds before drawing anything, i guess that's not what you expect.
So, divide THREAD_SLEEP_TIME by 1000.
Next, i don't really know what you mean by "buffer overrun", maybe because i never used PD on Windows ???
i am not sure what a buffer overrun is either . i am imagining its just means that the program is writing to an array past the end of the declared array space .
i have narrowed it down to crashing after the the record button is pressed that the sample is done playing .
well, i can only guess here.
when recording is finished, a thread is created in order to send all graphic data to the GUI.
uncomment the following line :
// post( "sonogram~ : stopped recording" );
and tell me if this appears in the log.
regards,
yves.
for fun, i translated my page to french with google and got this : http://ydegoyon.free.fr/fr.html
that's funny for those who can read french, sorry.
-
joge . -
Olaf Matthes -
Yves Degoyon