From geodancor at aol.com Sat Sep 1 05:28:11 2012 From: geodancor at aol.com (Dan Corneliu Georgescu) Date: Sat, 1 Sep 2012 08:28:11 -0400 (EDT) Subject: [CM] weighting not recognized Message-ID: <8CF56546268A476-FEC-69FD3@webmail-m054.sysops.aol.com> Hi, Following example doesnot work because ?weighting does not name aclase?. (define(chordy) (let ((pat (new cycle :of (list (new chord :of '(c4 e4 g4)) (new chord :of (new heap :of '(df4 ef4 gf4 af4 bf4 df5 ef5) :for 3)) (new chord :of '(e4 a4 cs5)) (new weighting :of '(bf3 (ef1 :weight .5 :max 2)) :for 1))))) (process repeat 32 do (doeach (k (next pat)) (output (new midi :keynum k:time (now)))) wait .5))) Could anyone please help me tosolve this problem? Thanks! Cor -------------- next part -------------- An HTML attachment was scrubbed... URL: From mail at peterodoherty.net Mon Sep 3 01:16:22 2012 From: mail at peterodoherty.net (Peter O'Doherty) Date: Mon, 03 Sep 2012 10:16:22 +0200 Subject: [CM] New user - a few questions Message-ID: <50446756.5030707@peterodoherty.net> Hi list, I'm eager to have a look at CM and have a number of questions: I would prefer to use emacs instead of Grace (I'm on Ubuntu). What's the best way to do this? Are there specific installation instructions? I would also like to use Common Music Notation. Is this still supported/ included? I notice that CM also outputs fomus files. Is this now the preferred way to deal with notation, instead of CMN? Many thanks, Peter -- //============================= -> Peter O'Doherty -> http://www.peterodoherty.net -> mail at peterodoherty.net -> https://joindiaspora.com/people/70716 //============================= -------------- next part -------------- An HTML attachment was scrubbed... URL: From bil at ccrma.Stanford.EDU Mon Sep 3 04:32:46 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Mon, 3 Sep 2012 04:32:46 -0700 Subject: [CM] New user - a few questions In-Reply-To: <50446756.5030707@peterodoherty.net> References: <50446756.5030707@peterodoherty.net> Message-ID: <20120903112520.M37264@ccrma.Stanford.EDU> cmn is still supported in the sense that I'll fix bugs that look easy to fix, and I make sure it works in sbcl as new versions of that lisp come along. It's not part of the SAL/s7 version of Common Music (I think Fomus is what you want). Someday. I might translate it to scheme, but I hesitate -- it makes heavy use of CLOS which I now consider a huge mistake, and I really only intended it to serve the composer's need for a quick view of an algorithm's output. As I got more and more requests for calligraphic enhancements, I got less and less interested. From mail at peterodoherty.net Mon Sep 3 04:49:12 2012 From: mail at peterodoherty.net (Peter O'Doherty) Date: Mon, 03 Sep 2012 13:49:12 +0200 Subject: [CM] New user - a few questions In-Reply-To: <20120903112520.M37264@ccrma.Stanford.EDU> References: <50446756.5030707@peterodoherty.net> <20120903112520.M37264@ccrma.Stanford.EDU> Message-ID: <50449938.1080904@peterodoherty.net> On 09/03/2012 01:32 PM, Bill Schottstaedt wrote: > cmn is still supported in the sense that I'll fix bugs that look > easy to fix, and I make sure it works in sbcl as new versions of > that lisp come along. It's not part of the SAL/s7 version > of Common Music (I think Fomus is what you want). Someday. > I might translate it to scheme, but I hesitate -- it makes > heavy use of CLOS which I now consider a huge mistake, > and I really only intended it to serve the composer's need for > a quick view of an algorithm's output. As I got more and > more requests for calligraphic enhancements, I got less > and less interested. > > Thanks for the information. I'll take a look at the fomus interface once I get it all installed (though my instinct is that CMN is must more interesting - as I'm not interested in sound synthesis or "calligraphic enhancements"). About the installation process... I followed these instructions from the help file. rm -rf sndlib curl -o sndlib.tar.gz ftp://ccrma-ftp.stanford.edu/pub/Lisp/sndlib.tar.gz tar -zxf sndlib.tar.gz rm sndlib.tar.gz cd sndlib ./configure CC=g++ make cd .. rm -rf cm svn co http://commonmusic.svn.sf.net/svnroot/commonmusic/trunk cm cd cm premake --verbose --target gnu --sndlib ../sndlib make At the make stage I get an error: ==== Building cm ==== mkdir -p obj/cm/Release g++ -MMD -D "JUCE_IOS=0" -D "JUCE_QUICKTIME=0" -D "JUCE_OPENGL=0" -D "JUCE_USE_FLAC=0" -D "JUCE_USE_OGGVORBIS=0" -D "JUCE_USE_CDBURNER=0" -D "JUCE_USE_CDREADER=0" -D "JUCE_WEBBROWSER=0" -D "JUCE_CHECK_MEMORY_LEAKS=0" -D "JUCE_JACK=1" -D "COMMONMUSIC=1" -D "WITH_SNDLIB=1" -D "SVNVERSION=\"`/usr/bin/svnversion`\"" -D "LINUX=1" -D "NDEBUG=1" -I "src" -I "../sndlib/" -O2 -o "obj/cm/Release/Metronome.o" -c "src/Metronome.cpp" mkdir -p obj/cm/Release g++ -MMD -D "JUCE_IOS=0" -D "JUCE_QUICKTIME=0" -D "JUCE_OPENGL=0" -D "JUCE_USE_FLAC=0" -D "JUCE_USE_OGGVORBIS=0" -D "JUCE_USE_CDBURNER=0" -D "JUCE_USE_CDREADER=0" -D "JUCE_WEBBROWSER=0" -D "JUCE_CHECK_MEMORY_LEAKS=0" -D "JUCE_JACK=1" -D "COMMONMUSIC=1" -D "WITH_SNDLIB=1" -D "SVNVERSION=\"`/usr/bin/svnversion`\"" -D "LINUX=1" -D "NDEBUG=1" -I "src" -I "../sndlib/" -O2 -o "obj/cm/Release/Syntax.o" -c "src/Syntax.cpp" mkdir -p obj/cm/Release g++ -MMD -D "JUCE_IOS=0" -D "JUCE_QUICKTIME=0" -D "JUCE_OPENGL=0" -D "JUCE_USE_FLAC=0" -D "JUCE_USE_OGGVORBIS=0" -D "JUCE_USE_CDBURNER=0" -D "JUCE_USE_CDREADER=0" -D "JUCE_WEBBROWSER=0" -D "JUCE_CHECK_MEMORY_LEAKS=0" -D "JUCE_JACK=1" -D "COMMONMUSIC=1" -D "WITH_SNDLIB=1" -D "SVNVERSION=\"`/usr/bin/svnversion`\"" -D "LINUX=1" -D "NDEBUG=1" -I "src" -I "../sndlib/" -O2 -o "obj/cm/Release/CmSupport.o" -c "src/CmSupport.cpp" src/CmSupport.cpp: In function ?void mp_send_note(s7_pointer, s7_pointer, s7_pointer, s7_pointer, s7_pointer)?: src/CmSupport.cpp:1457:32: error: cannot convert ?s7_pointer {aka s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double s7_number_to_real(s7_scheme*, s7_pointer)? src/CmSupport.cpp:1464:31: error: cannot convert ?s7_pointer {aka s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double s7_number_to_real(s7_scheme*, s7_pointer)? src/CmSupport.cpp:1471:31: error: cannot convert ?s7_pointer {aka s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double s7_number_to_real(s7_scheme*, s7_pointer)? src/CmSupport.cpp:1478:31: error: cannot convert ?s7_pointer {aka s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double s7_number_to_real(s7_scheme*, s7_pointer)? src/CmSupport.cpp:1485:32: error: cannot convert ?s7_pointer {aka s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double s7_number_to_real(s7_scheme*, s7_pointer)? src/CmSupport.cpp: In function ?void cm_shell(char*)?: src/CmSupport.cpp:186:14: warning: ignoring return value of ?int system(const char*)?, declared with attribute warn_unused_result [-Wunused-result] make[1]: *** [obj/cm/Release/CmSupport.o] Error 1 make: *** [cm] Error 2 [~/cm]$ Any ideas what the problem might be? Thanks, Peter -- //============================= -> Peter O'Doherty -> http://www.peterodoherty.net -> mail at peterodoherty.net -> https://joindiaspora.com/people/70716 //============================= From taube at illinois.edu Mon Sep 3 05:13:28 2012 From: taube at illinois.edu (Heinrich Taube) Date: Mon, 3 Sep 2012 07:13:28 -0500 Subject: [CM] New user - a few questions In-Reply-To: <20120903112520.M37264@ccrma.Stanford.EDU> References: <50446756.5030707@peterodoherty.net> <20120903112520.M37264@ccrma.Stanford.EDU> Message-ID: yes i need to build against the latest sndlib and see what the problems are, its on my list but i have almost no time lately to do this! ill try to have something by next friday, apologies for your wait -- its not for lack of interest on my part! On Sep 3, 2012, at 6:32 AM, Bill Schottstaedt wrote: > cmn is still supported in the sense that I'll fix bugs that look > easy to fix, and I make sure it works in sbcl as new versions of > that lisp come along. It's not part of the SAL/s7 version > of Common Music (I think Fomus is what you want). Someday. > I might translate it to scheme, but I hesitate -- it makes > heavy use of CLOS which I now consider a huge mistake, > and I really only intended it to serve the composer's need for > a quick view of an algorithm's output. As I got more and > more requests for calligraphic enhancements, I got less > and less interested. > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From cm at quidecco.de Mon Sep 3 05:31:11 2012 From: cm at quidecco.de (Isidor Zeuner) Date: Mon, 3 Sep 2012 14:31:11 +0200 (CEST) Subject: [CM] New user - a few questions In-Reply-To: <50449938.1080904@peterodoherty.net> References: <50449938.1080904@peterodoherty.net> <50446756.5030707@peterodoherty.net> <20120903112520.M37264@ccrma.Stanford.EDU> Message-ID: <20120903123111.5068988822C@quidecco.de> Hi Peter, [...] > g++ -MMD -D "JUCE_IOS=0" -D "JUCE_QUICKTIME=0" -D "JUCE_OPENGL=0" -D > "JUCE_USE_FLAC=0" -D "JUCE_USE_OGGVORBIS=0" -D "JUCE_USE_CDBURNER=0" -D > "JUCE_USE_CDREADER=0" -D "JUCE_WEBBROWSER=0" -D > "JUCE_CHECK_MEMORY_LEAKS=0" -D "JUCE_JACK=1" -D "COMMONMUSIC=1" -D > "WITH_SNDLIB=1" -D "SVNVERSION=\"`/usr/bin/svnversion`\"" -D "LINUX=1" > -D "NDEBUG=1" -I "src" -I "../sndlib/" -O2 -o > "obj/cm/Release/CmSupport.o" -c "src/CmSupport.cpp" > src/CmSupport.cpp: In function ?void mp_send_note(s7_pointer, > s7_pointer, s7_pointer, s7_pointer, s7_pointer)?: > src/CmSupport.cpp:1457:32: error: cannot convert ?s7_pointer {aka > s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double > s7_number_to_real(s7_scheme*, s7_pointer)? > src/CmSupport.cpp:1464:31: error: cannot convert ?s7_pointer {aka > s7_cell*}? to ?s7_scheme*? for argument ?1? to ?s7_Double > s7_number_to_real(s7_scheme*, s7_pointer)? [...] I ran into the same problem, which I assume is due to an interface change in s7 which requires to pass a pointer to the scheme instance now. I could compile it using the attached patch. I'm not yet too experienced in working with the sndlib/cm sources, but so I could successfully work with the resulting binaries so far. Best regards, Isidor -------------- next part -------------- A non-text attachment was scrubbed... Name: commonmusic-s7.diff.gz Type: application/gzip Size: 2235 bytes Desc: not available URL: From mail at peterodoherty.net Mon Sep 3 05:55:14 2012 From: mail at peterodoherty.net (Peter O'Doherty) Date: Mon, 03 Sep 2012 14:55:14 +0200 Subject: [CM] New user - a few questions In-Reply-To: <20120903112520.M37264@ccrma.Stanford.EDU> References: <50446756.5030707@peterodoherty.net> <20120903112520.M37264@ccrma.Stanford.EDU> Message-ID: <5044A8B2.2030607@peterodoherty.net> On 09/03/2012 01:32 PM, Bill Schottstaedt wrote: > cmn is still supported in the sense that I'll fix bugs that look > easy to fix, and I make sure it works in sbcl as new versions of > that lisp come along. It's not part of the SAL/s7 version > of Common Music (I think Fomus is what you want). Someday. > I might translate it to scheme, but I hesitate -- it makes > heavy use of CLOS which I now consider a huge mistake, > and I really only intended it to serve the composer's need for > a quick view of an algorithm's output. As I got more and > more requests for calligraphic enhancements, I got less > and less interested. > > I understand that cmn is not actively supported/ developed, but how willing are you to answer questions which are not covered in the user manual? (for example - is it possible to use midi numbers or other representation of notes instead of pitch names, eg. c4?) Many thanks, Peter -- //============================= -> Peter O'Doherty -> http://www.peterodoherty.net -> mail at peterodoherty.net -> https://joindiaspora.com/people/70716 //============================= From bil at ccrma.Stanford.EDU Mon Sep 3 13:50:42 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Mon, 3 Sep 2012 13:50:42 -0700 Subject: [CM] New user - a few questions In-Reply-To: <5044A8B2.2030607@peterodoherty.net> References: <50446756.5030707@peterodoherty.net> <20120903112520.M37264@ccrma.Stanford.EDU> <5044A8B2.2030607@peterodoherty.net> Message-ID: <20120903204846.M25046@ccrma.Stanford.EDU> > questions which are not covered in the user > manual? > > (for example - is it possible to use midi numbers or other > representation of notes instead of pitch names, eg. c4?) that one is covered, but maybe it's hard to find -- see add-notes, or display-notes in the section about note list display. You could write a midi->hz function and use these directly. From cmdist at nospam.frostglow.com Thu Sep 6 19:26:28 2012 From: cmdist at nospam.frostglow.com (cmdist at nospam.frostglow.com) Date: Thu, 6 Sep 2012 22:26:28 -0400 Subject: [CM] cm 3.8.0 - slecting audio/midi in/out from the repl Message-ID: <20120907022628.GA31342@frostglow.com> Hello, I just started to learn Common Music 3.8.0 and was wondering if there was a way to specify the audio and midi input and output settings directly from the cm repl, or must I work in Grace in order to do that? Thanks, --Sergey From dlphillips at woh.rr.com Mon Sep 10 09:18:59 2012 From: dlphillips at woh.rr.com (Dave Phillips) Date: Mon, 10 Sep 2012 12:18:59 -0400 Subject: [CM] build failure on Arch 64 Message-ID: <504E12F3.2060308@woh.rr.com> Greetings, I'm trying to build GRACE/CM on an Arch 64 system. I get this far : [dlphilp at BigBlack cm-3.8.0]$ premake --verbose --target gnu --sndlib ../sndlib Generating GNU makefiles: ...juce ...cm ...grace [dlphilp at BigBlack cm-3.8.0]$ make ... ... g++ -MMD -D "JUCE_IOS=0" -D "JUCE_QUICKTIME=0" -D "JUCE_OPENGL=0" -D "JUCE_USE_FLAC=0" -D "JUCE_USE_OGGVORBIS=0" -D "JUCE_USE_CDBURNER=0" -D "JUCE_USE_CDREADER=0" -D "JUCE_WEBBROWSER=0" -D "JUCE_CHECK_MEMORY_LEAKS=0" -D "JUCE_JACK=1" -D "COMMONMUSIC=1" -D "WITH_SNDLIB=1" -D "SVNVERSION=\"`/usr/bin/svnversion`\"" -D "LINUX=1" -D "NDEBUG=1" -I "src" -I "../sndlib/" -O2 -o "obj/cm/Release/CmSupport.o" -c "src/CmSupport.cpp" src/CmSupport.cpp: In function 'void mp_send_note(s7_pointer, s7_pointer, s7_pointer, s7_pointer, s7_pointer)': src/CmSupport.cpp:1457:32: error: cannot convert 's7_pointer {aka s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double s7_number_to_real(s7_scheme*, s7_pointer)' src/CmSupport.cpp:1464:31: error: cannot convert 's7_pointer {aka s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double s7_number_to_real(s7_scheme*, s7_pointer)' src/CmSupport.cpp:1471:31: error: cannot convert 's7_pointer {aka s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double s7_number_to_real(s7_scheme*, s7_pointer)' src/CmSupport.cpp:1478:31: error: cannot convert 's7_pointer {aka s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double s7_number_to_real(s7_scheme*, s7_pointer)' src/CmSupport.cpp:1485:32: error: cannot convert 's7_pointer {aka s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double s7_number_to_real(s7_scheme*, s7_pointer)' make[1]: *** [obj/cm/Release/CmSupport.o] Error 1 make: *** [cm] Error 2 I retrieved and built sndlib today, I hope I'm up-to-date. Any suggestions ? Best, dp From dlphillips at woh.rr.com Mon Sep 10 09:36:10 2012 From: dlphillips at woh.rr.com (Dave Phillips) Date: Mon, 10 Sep 2012 12:36:10 -0400 Subject: [CM] build failure on Arch 64 In-Reply-To: <504E12F3.2060308@woh.rr.com> References: <504E12F3.2060308@woh.rr.com> Message-ID: <504E16FA.6030101@woh.rr.com> Greetings, Never mind, I found the previous thread on this topic. I used the sndlib-may22 tarball, the GRACE/CM build is moving along nicely now. Sorry for the noise. Best, dp On 09/10/12 12:18, Dave Phillips wrote: > Greetings, > > I'm trying to build GRACE/CM on an Arch 64 system. I get this far : > > [dlphilp at BigBlack cm-3.8.0]$ premake --verbose --target gnu --sndlib > ../sndlib > Generating GNU makefiles: > ...juce > ...cm > ...grace > [dlphilp at BigBlack cm-3.8.0]$ make > ... > ... > g++ -MMD -D "JUCE_IOS=0" -D "JUCE_QUICKTIME=0" -D "JUCE_OPENGL=0" -D > "JUCE_USE_FLAC=0" -D "JUCE_USE_OGGVORBIS=0" -D "JUCE_USE_CDBURNER=0" -D > "JUCE_USE_CDREADER=0" -D "JUCE_WEBBROWSER=0" -D > "JUCE_CHECK_MEMORY_LEAKS=0" -D "JUCE_JACK=1" -D "COMMONMUSIC=1" -D > "WITH_SNDLIB=1" -D "SVNVERSION=\"`/usr/bin/svnversion`\"" -D "LINUX=1" > -D "NDEBUG=1" -I "src" -I "../sndlib/" -O2 -o > "obj/cm/Release/CmSupport.o" -c "src/CmSupport.cpp" > src/CmSupport.cpp: In function 'void mp_send_note(s7_pointer, > s7_pointer, s7_pointer, s7_pointer, s7_pointer)': > src/CmSupport.cpp:1457:32: error: cannot convert 's7_pointer {aka > s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double > s7_number_to_real(s7_scheme*, s7_pointer)' > src/CmSupport.cpp:1464:31: error: cannot convert 's7_pointer {aka > s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double > s7_number_to_real(s7_scheme*, s7_pointer)' > src/CmSupport.cpp:1471:31: error: cannot convert 's7_pointer {aka > s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double > s7_number_to_real(s7_scheme*, s7_pointer)' > src/CmSupport.cpp:1478:31: error: cannot convert 's7_pointer {aka > s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double > s7_number_to_real(s7_scheme*, s7_pointer)' > src/CmSupport.cpp:1485:32: error: cannot convert 's7_pointer {aka > s7_cell*}' to 's7_scheme*' for argument '1' to 's7_Double > s7_number_to_real(s7_scheme*, s7_pointer)' > make[1]: *** [obj/cm/Release/CmSupport.o] Error 1 > make: *** [cm] Error 2 > > > I retrieved and built sndlib today, I hope I'm up-to-date. Any suggestions ? > > Best, > > dp > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From dlphillips at woh.rr.com Mon Sep 10 10:03:41 2012 From: dlphillips at woh.rr.com (Dave Phillips) Date: Mon, 10 Sep 2012 13:03:41 -0400 Subject: [CM] build failure on Arch 64 In-Reply-To: <20120910165420.M84122@ccrma.Stanford.EDU> References: <504E12F3.2060308@woh.rr.com> <504E16FA.6030101@woh.rr.com> <20120910165420.M84122@ccrma.Stanford.EDU> Message-ID: <504E1D6D.2030903@woh.rr.com> On 09/10/12 12:54, Bill Schottstaedt wrote: > Sorry about that -- I could have sworn I checked CM before making > that change! > > > Hi Bill, No problem with the May 22 tarball, but I'll gladly try a rebuild with current sndlib sources. I've hit a different snag though, one for Rick I think, it looks like GRACE problem. I can't switch the Audio setting from its default ALSA to my preferred JACK. When I select JACK from the options GRACE segfaults. Any advice from anyone ? Best, dp From taube at illinois.edu Mon Sep 10 11:12:05 2012 From: taube at illinois.edu (Heinrich Taube) Date: Mon, 10 Sep 2012 13:12:05 -0500 Subject: [CM] build failure on Arch 64 In-Reply-To: <504E1D6D.2030903@woh.rr.com> References: <504E12F3.2060308@woh.rr.com> <504E16FA.6030101@woh.rr.com> <20120910165420.M84122@ccrma.Stanford.EDU> <504E1D6D.2030903@woh.rr.com> Message-ID: <0048E51E-8A81-46AB-A465-3ED92489CBF0@illinois.edu> dave im in the process of moving to the latest sndlib. one thing you might need to do is enable jack support in the juce compilation?? ill take a look at this once i can build cm On Sep 10, 2012, at 12:03 PM, Dave Phillips wrote: > On 09/10/12 12:54, Bill Schottstaedt wrote: >> Sorry about that -- I could have sworn I checked CM before making >> that change! >> >> >> > > Hi Bill, > > No problem with the May 22 tarball, but I'll gladly try a rebuild with > current sndlib sources. > > I've hit a different snag though, one for Rick I think, it looks like > GRACE problem. I can't switch the Audio setting from its default ALSA to > my preferred JACK. When I select JACK from the options GRACE segfaults. > > Any advice from anyone ? > > Best, > > dp > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From taube at illinois.edu Tue Sep 11 07:04:33 2012 From: taube at illinois.edu (Heinrich Taube) Date: Tue, 11 Sep 2012 09:04:33 -0500 Subject: [CM] build failure on Arch 64 In-Reply-To: <504E1D6D.2030903@woh.rr.com> References: <504E12F3.2060308@woh.rr.com> <504E16FA.6030101@woh.rr.com> <20120910165420.M84122@ccrma.Stanford.EDU> <504E1D6D.2030903@woh.rr.com> Message-ID: I've checked in the changes to get cm working in the latest sndlib (2.11 6-Aug-12) . it was a bit more work than i thought because since mayI'm on a new machine and I've switched over to the clang compiler...but things at least seem to be working, e.g. reich example runs and fm-violin makes its toot. at this point i can try building on linux -- this will take me a while because I've got to reinstall the whole ubuntu in virtual box and then build the development toolchain... From taube at illinois.edu Tue Sep 11 09:32:36 2012 From: taube at illinois.edu (Heinrich Taube) Date: Tue, 11 Sep 2012 11:32:36 -0500 Subject: [CM] build failure on Arch 64 In-Reply-To: <504E1D6D.2030903@woh.rr.com> References: <504E12F3.2060308@woh.rr.com> <504E16FA.6030101@woh.rr.com> <20120910165420.M84122@ccrma.Stanford.EDU> <504E1D6D.2030903@woh.rr.com> Message-ID: <05A9F20A-9DAD-49FB-BB3F-4D25547E7153@illinois.edu> ok with a few changes (already in svn...) i was able to build everything on ubuntu and selecting jack instead of also in the audio settings do not crash the app. however, i don't seem to be able to start jack using qjackctl :/ after installing jack like this: sudo apt-get install libjack-dev sudo apt-get install qjackclt when i run it either sudo or as myself it does seem to start, but I'm not sure if i need to do something in Setup before I press Start, but pressing start yields this: sudo qjackctl <> 11:25:02.243 Patchbay deactivated. 11:25:02.244 Statistics reset. 11:25:02.396 ALSA connection change. 11:25:02.456 D-BUS: Service not available (org.jackaudio.service aka jackdbus). 11:25:02.512 ALSA connection graph change. 11:25:04.849 JACK is starting... 11:25:04.851 /usr/bin/jackd -dalsa -dhw:0 -r44100 -p1024 -n2 jackd 0.121.0 Copyright 2001-2009 Paul Davis, Stephane Letz, Jack O'Quinn, Torben Hohn and others. jackd comes with ABSOLUTELY NO WARRANTY This is free software, and you are welcome to redistribute it under certain conditions; see the file COPYING for details 11:25:04.880 JACK was started with PID=15894. JACK compiled with System V SHM support. loading driver .. SSE2 detected apparent rate = 44100 creating alsa driver ... hw:0|hw:0|1024|2|44100|0|0|nomon|swmeter|-|32bit control device hw:0 configuring for 44100Hz, period = 1024 frames (23.2 ms), buffer = 2 periods ALSA: final selected sample format for capture: 16bit little-endian ALSA: use 2 periods for capture ALSA: final selected sample format for playback: 16bit little-endian ALSA: use 2 periods for playback 11:25:07.756 JACK connection change. 11:25:07.760 Server configuration saved to "/home/hkt/.jackdrc". 11:25:07.760 Statistics reset. 11:25:07.903 Client activated. 11:25:07.952 Buffer size change (1024). 11:25:07.953 XRUN callback (1). SSE2 detected 11:25:10.180 XRUN callback (36 skipped). 11:25:12.343 XRUN callback (34 skipped). 11:25:14.598 XRUN callback (34 skipped). jackd watchdog: timeout - killing jackd no message buffer overruns 11:25:15.628 Shutdown notification. 11:25:15.629 Client deactivated. 11:25:15.632 JACK is being forced... cannot read server event (Success) cannot continue execution of the processing graph (Bad file descriptor) zombified - calling shutdown handler cannot send request type 7 to server cannot read result for request type 7 from server (Broken pipe) cannot send request type 7 to server cannot read result for request type 7 from server (Broken pipe) 11:25:15.832 JACK was stopped with exit status=1. qjackclt <> 11:22:36.744 Patchbay deactivated. 11:22:36.827 Statistics reset. 11:22:36.867 ALSA connection change. 11:22:37.096 D-BUS: Service is available (org.jackaudio.service aka jackdbus). 11:22:37.156 ALSA connection graph change. 11:22:41.527 D-BUS: JACK server could not be started. Sorry Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: driver "alsa" selected Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Saving settings to "/home/hkt/.config/jack/conf.xml" ... Tue Sep 11 11:22:41 2012: Starting jack server... Tue Sep 11 11:22:41 2012: JACK server starting in realtime mode with priority 10 Tue Sep 11 11:22:41 2012: ERROR: Cannot lock down memory area (Cannot allocate memory) Tue Sep 11 11:22:41 2012: ERROR: can't load "/usr/lib/x86_64-linux-gnu/jack/jack_alsa.so": /usr/lib/x86_64-linux-gnu/jack/jack_alsa.so: undefined symbol: jack_driver_nt_init Tue Sep 11 11:22:41 2012: ERROR: Cannot initialize driver Tue Sep 11 11:22:41 2012: ERROR: JackServer::Open() failed with -1 Tue Sep 11 11:22:41 2012: ERROR: Failed to open server 11:22:46.329 Could not connect to JACK server as client. - Overall operation failed. - Unable to connect to server. Please check the messages window for more info. From taube at illinois.edu Sun Sep 16 07:01:44 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sun, 16 Sep 2012 09:01:44 -0500 Subject: [CM] cm in juce 2.0, jack Message-ID: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> I've moved CM to the latest sndlib and JUCE2. this has triggered some changes to building (all for the better I hope), including a move to premake4. to build cm from sources now you do: cd cm premake4 make to build the app with jack support on linux you would do premake4 --jack make the biggest change in the building process is that the new cm uses its own versions of juce and sndlib under its top directory (cm/juce and cm/sndlib) these systems are NOT stored in the cm repo but ill ship them with every release so that the system doesn't go out of sync with later versions of sndlib (unless you force it) tif you checkout cm from svn the build process will automatically download and install both juce and sndlib for you as part of its build (you'll need git and curl on your path...) -- The system builds on both maces and linux, but JUCE2 has some major differences with JUCE1 and I haven't had any time yet for testing, except to say that on linux choosing jack no longer segfaults :). it would be very helpful if you try to build it on osx or linux and let me know what is working and what is not. i am also taking this opportunity to refactor the code, removing stale files and cleaning up the code, including purging all "using namespace juce". i would really like to move to mac menu style, but this will be a lot of work.i will add back the "optional" packages like fomus, sdif and osc once the dust settles and i have a bit of time. so the next release of cm i plan will be a stable version of cm working with the new code base. From taube at illinois.edu Sun Sep 16 11:36:38 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sun, 16 Sep 2012 13:36:38 -0500 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <20120916171210.GA17310@frostglow.com> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> Message-ID: thanks for your detailed report! ill take a look at the editor issue tomorrow. > run "(mp:receiver (lambda args (print args)))" from the editor, I get an > error: ">>> Error: mp:receiver: unbound variable" despite "mp:receiver" > being documented here: > i think its actually called (mp:receive ?) i thought i fixed the docs, ill do that tomorrow too. you can always use Audio>Midi In>Trace Input to get a printout of incoming messages to the console? > Finally, it looks like only the "Grace" binary was built, and there is > no "cm" command line REPL, as in the stable 3.8.0 version. yes i will do it, but only after i refactor the code. the current bin/cm is a bit of a mess to compile and has been cause all sorts #ifdefs in the code. in the meantime i think could compiled a s7 only repl for you if that would be useful, let me know From cmdist at nospam.frostglow.com Sun Sep 16 12:29:34 2012 From: cmdist at nospam.frostglow.com (cmdist at nospam.frostglow.com) Date: Sun, 16 Sep 2012 15:29:34 -0400 Subject: [CM] cm in juce 2.0, jack In-Reply-To: References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> Message-ID: <20120916192934.GA17557@frostglow.com> On Sun, Sep 16, 2012 at 01:36:38PM -0500, Heinrich Taube wrote: > > thanks for your detailed report! ill take a look at the editor issue tomorrow. Oops. Looks like I had accidentally replied just to you instead of to both you the mailinglist. I'll just repost my original message below, for posterity. --Sergey > > Thank you for the new version of Common Music! > > > > I just built it successfully on amd64 gentoo linux. The full build log > > can be seen here: http://pastie.org/4733080 > > > > Grace starts up fine, and reports: http://pastie.org/4733091 > > > > I was able to connect Grace to yoshimi by selecting "Audio->Midi > > Out...->jack_midi" in Grace and using patchage to connect the > > system:midi_capture_8 port to yoshimi's "midi_in" port. > > > > Here is what my jack connections look like in patchage: > > > > http://img1.imagilive.com/0912/cm-svn-patchage.png > > > > When I selected "Audio->Midi Out...->Test Output", I could hear notes > > come from yoshimi. > > > > Then I in "Audio->Audio Settings...", I selected "audio device type" of > > "JACK" (screenshot of all the settings can be seen here: > > http://img1.imagilive.com/0912/grace-audio-settings.png ) and when I > > pressed the "Test" button, I could hear a tone. > > > > So far so good. > > > > Unfortunately, there seems to be some kind of screen refresh issue in > > the editor. When I switch from one theme to another, the editor window > > does not fully refresh. For example, here is what it looks like when I > > switch from the "Deep Blue" theme to "Clarity and Beauty": > > http://img1.imagilive.com/0912/grace-editor-refresh-issue.png > > > > As you can see, only the portion of the editor window that was under the > > "Options->Theme" menu got refreshed and switched to the "Clarity and > > Beauty" theme. The rest of the editor window remains in the previous > > theme colors until I do somethign like switch workspaces in my window > > manager and then switch back (or until I switch to another window and > > switch back to the editor window). > > > > Also, note how some of the text in the "Clarity and Beauty" theme > > remains black, so that it's effectively invisible and unreadable on the > > black background of the "Clarity and Beauty" theme. These refresh > > issues and the black text issue are new for the svn version of Grace, > > and were not present in the stable 3.8.0 version. > > > > Another problem is that I can't seem to get midi input to work. When I > > run "(mp:receiver (lambda args (print args)))" from the editor, I get an > > error: ">>> Error: mp:receiver: unbound variable" despite "mp:receiver" > > being documented here: > > > > http://commonmusic.sourceforge.net/cm/res/doc/cm.html#midi-in > > > > Finally, it looks like only the "Grace" binary was built, and there is > > no "cm" command line REPL, as in the stable 3.8.0 version. > > > > That's all I've been able to test so far. > > > > --Sergey From cmdist at nospam.frostglow.com Sun Sep 16 12:51:15 2012 From: cmdist at nospam.frostglow.com (cmdist at nospam.frostglow.com) Date: Sun, 16 Sep 2012 15:51:15 -0400 Subject: [CM] vim2grace - vim (or other external editor) integration with Grace Message-ID: <20120916195115.GA17630@frostglow.com> I needed a simple way to pass blocks of source code from within my editor of choice (vim) to Grace/Common Music for automatic execution. To this end, I've written a small, simple program that you can download from: http://www.vim.org/scripts/script.php?script_id=4218 The program should be able to work without modification for any editor, not just vim. All your editor has to do is write the blocks of source code you want to send to Grace to a fifo (which is just like writing to a regular file, so every editor should be capable of doing this). Detailed instructions for using vim2grace can be found at the above link. --Sergey From dlphillips at woh.rr.com Sun Sep 16 14:40:15 2012 From: dlphillips at woh.rr.com (Dave Phillips) Date: Sun, 16 Sep 2012 17:40:15 -0400 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <20120916192934.GA17557@frostglow.com> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <20120916192934.GA17557@frostglow.com> Message-ID: <5056473F.6010702@woh.rr.com> Greetings, Grace at r1999 compiles cleanly on Arch 64, but the binary still segfaults when I select JACK in the Audio Settings. No other error message on stdout. :( Suggestions ? Best, dp From taube at illinois.edu Sun Sep 16 15:03:28 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sun, 16 Sep 2012 17:03:28 -0500 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <5056473F.6010702@woh.rr.com> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <20120916192934.GA17557@frostglow.com> <5056473F.6010702@woh.rr.com> Message-ID: <70DD6FCF-7BB2-43F4-944B-493DB9AC5AB3@illinois.edu> arrg I've no idea, its not segfaulting on ubuntu and sergey is able to select connections and make sound. perhaps you need to upgrade the jack? alternately you could try building in debug mode and see if there are any assertions before the segfault. On Sep 16, 2012, at 4:40 PM, Dave Phillips wrote: > Greetings, > > Grace at r1999 compiles cleanly on Arch 64, but the binary still > segfaults when I select JACK in the Audio Settings. No other error > message on stdout. :( > > Suggestions ? > > Best, > > dp > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From phil at philthomson.ca Sun Sep 16 23:28:29 2012 From: phil at philthomson.ca (Phil Thomson) Date: Sun, 16 Sep 2012 23:28:29 -0700 Subject: [CM] vim2grace - vim (or other external editor) integration with Grace In-Reply-To: <20120916195115.GA17630@frostglow.com> References: <20120916195115.GA17630@frostglow.com> Message-ID: <5056C30D.5090006@philthomson.ca> I'm very excited to try this! Phil Thomson http://philthomson.ca/ On 12-09-16 12:51 PM, cmdist at nospam.frostglow.com wrote: > I needed a simple way to pass blocks of source code from within > my editor of choice (vim) to Grace/Common Music for automatic execution. > > To this end, I've written a small, simple program that you can > download from: > > http://www.vim.org/scripts/script.php?script_id=4218 > > The program should be able to work without modification for any editor, > not just vim. All your editor has to do is write the blocks of source > code you want to send to Grace to a fifo (which is just like writing > to a regular file, so every editor should be capable of doing this). > > Detailed instructions for using vim2grace can be found at the above > link. > > --Sergey > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From nando at ccrma.Stanford.EDU Mon Sep 17 10:19:10 2012 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: Mon, 17 Sep 2012 10:19:10 -0700 Subject: [CM] cm in juce 2.0, jack In-Reply-To: References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> Message-ID: <50575B8E.5010706@localhost> On 09/16/2012 11:36 AM, Heinrich Taube wrote: >> Finally, it looks like only the "Grace" binary was built, and there is >> no "cm" command line REPL, as in the stable 3.8.0 version. > > yes i will do it, but only after i refactor the code. the current bin/cm is a bit of a mess to compile and has been cause all sorts #ifdefs in the code. > in the meantime i think could compiled a s7 only repl for you if that would be useful, let me know That would be useful I think (cm would be better), I'm trying to bring old code up and running and it needs patterns ("make-item-stream"!). It does not look like I can get patterns.scm to load on a plain s7... -- Fernando From andersvi at notam02.no Tue Sep 18 01:42:44 2012 From: andersvi at notam02.no (andersvi at notam02.no) Date: Tue, 18 Sep 2012 10:42:44 +0200 Subject: [CM] snd: control-panel blocks playback Message-ID: <87k3vroh6j.fsf@notam02.no> Hi. Im getting a strange error with snd on a freshly setup machine here. Opening the control-panel stops playing of audio. Closing the control-panel resumes playing. If control-panel is opened before playing, playing wont start. Ive tried the recent packaged snd-13* (the current snd-motif for fc17 seems to be using gtk as well...?) + built my own: various previous versions from cvs (back to snd-12.8), w & wo jack-support, but same behavior. Machine is a 64bit, running fc17. Both stock-kernel (3.5.3-1) and planetccrma-kernel (3.4.10-1.rt18.1) shows the same behaviour vs. snd & control-panel. Might this be something with gtk? pkg-config --modversion gtk+-3.0 => 3.4.4. The display-driver (nouveau)? Any clues? Id be happy for suggestions vs. debugging this. -anders From andersvi at notam02.no Tue Sep 18 02:05:17 2012 From: andersvi at notam02.no (andersvi at notam02.no) Date: Tue, 18 Sep 2012 11:05:17 +0200 Subject: [CM] snd: control-panel blocks playback References: <87k3vroh6j.fsf@notam02.no> Message-ID: <87fw6fog4y.fsf@notam02.no> >>>>> "a" == andersvi: a> Opening the control-panel stops playing of audio. Closing the a> control-panel resumes playing. If control-panel is opened before a> playing, playing wont start. a> Machine is a 64bit, running fc17. Both stock-kernel (3.5.3-1) a> and planetccrma-kernel (3.4.10-1.rt18.1) shows the same behaviour a> vs. snd & control-panel. a> Might this be something with gtk? pkg-config --modversion a> gtk+-3.0 => 3.4.4. The display-driver (nouveau)? Just checked with snd-13 on an older machine, but same behavior vs. control-panel: (on older machine) $ snd --version This is Snd version 13.1 of 18-Sep-12: s7: 2.12 (22-Aug-12), Xen: 3.15 ALSA 1.0.24 Sndlib 22.1 (16-July-12, double samples) CLM 5.2 (8-Aug-12) GSL 1.14 fftw-3.2.2 Gtk+ 2.22.0, Glib 2.26.0, Pango 1.28.1, Cairo 1.10.2 LADSPA: 1.1 Compiled Sep 18 2012 10:47:09 C: 4.5.1 20100924 (Red Hat 4.5.1-4) Libc: 2.13.stable host: i686-pc-linux-gnu (on newer machine) $ snd --version This is Snd version 13.1 of 18-Sep-12: s7: 2.12 (22-Aug-12), Xen: 3.15 Jack: 1.9.8 Sndlib 22.1 (16-July-12, double samples) CLM 5.2 (8-Aug-12) fftw-3.3.1 Gtk+ 3.4.4, Glib 2.32.4, Pango 1.30.0, Cairo 1.10.2 Compiled Sep 18 2012 11:01:14 C: 4.7.0 20120507 (Red Hat 4.7.0-5) Libc: 2.15.stable host: x86_64-unknown-linux-gnu From bil at ccrma.Stanford.EDU Tue Sep 18 03:36:43 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 18 Sep 2012 03:36:43 -0700 Subject: [CM] snd: control-panel blocks playback In-Reply-To: <87fw6fog4y.fsf@notam02.no> References: <87k3vroh6j.fsf@notam02.no> <87fw6fog4y.fsf@notam02.no> Message-ID: <20120918103332.M77341@ccrma.Stanford.EDU> Thanks very much for the bug report. It appears that something in the control panel is generating such a heavy stream of events that the background play process never gets to run. If I force it to run in the foreground, it gets interrupted a lot. Closing the panel turns off this stream of events, so a pending play can finally run. I'll have to poke at this... From bil at ccrma.Stanford.EDU Tue Sep 18 03:53:23 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 18 Sep 2012 03:53:23 -0700 Subject: [CM] snd: control-panel blocks playback In-Reply-To: <87fw6fog4y.fsf@notam02.no> References: <87k3vroh6j.fsf@notam02.no> <87fw6fog4y.fsf@notam02.no> Message-ID: <20120918105234.M25812@ccrma.Stanford.EDU> The culprit is the arrow icon on the speed control -- why would it go berserk? From dlphillips at woh.rr.com Tue Sep 18 05:39:25 2012 From: dlphillips at woh.rr.com (Dave Phillips) Date: Tue, 18 Sep 2012 08:39:25 -0400 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <70DD6FCF-7BB2-43F4-944B-493DB9AC5AB3@illinois.edu> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <20120916192934.GA17557@frostglow.com> <5056473F.6010702@woh.rr.com> <70DD6FCF-7BB2-43F4-944B-493DB9AC5AB3@illinois.edu> Message-ID: <50586B7D.2000601@woh.rr.com> On 09/16/2012 06:03 PM, Heinrich Taube wrote: > arrg I've no idea, its not segfaulting on ubuntu and sergey is able to select connections and make sound. perhaps you need to upgrade the jack? alternately you could try building in debug mode and see if there are any assertions before the segfault. > Hi Rick, I've yet to try the debug build but I thought you might like (?) to know that a freshly-build Grace binary segfaults on my Debian Squeeze system too (AVLinux 5.0.1). I'll do debug builds on both boxes today. I also have an Ubuntu 10.04 machine that I haven't yet tested for Grace, will do so this morning. One of them has to work, right ? :) Btw, JACK1 on both systems is at v. 0.120.2. Are you using JACK 1 or 2 ? Best, dp From taube at illinois.edu Tue Sep 18 06:07:51 2012 From: taube at illinois.edu (Heinrich Taube) Date: Tue, 18 Sep 2012 08:07:51 -0500 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <50586B7D.2000601@woh.rr.com> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <20120916192934.GA17557@frostglow.com> <5056473F.6010702@woh.rr.com> <70DD6FCF-7BB2-43F4-944B-493DB9AC5AB3@illinois.edu> <50586B7D.2000601@woh.rr.com> Message-ID: <946C62DC-143C-4396-8D63-FBB595FC8DE7@illinois.edu> > > Btw, JACK1 on both systems is at v. 0.120.2. Are you using JACK 1 or 2 ? > hi dave it looks like I'm running jack1: dpkg --get-selections | grep jack jackd install jackd1 install jackd1-firewire install jackd2 deinstall libjack-dev install libjack-jackd2-0 deinstall libjack0 install qjackctl install hkt at hkt-VirtualBox:~$ c > a freshly-build Grace binary segfaults on my Debian Squeeze system > too (AVLinux 5.0.1). I'll do debug builds on both boxes today. I also > h well sheesh! I'm just running a vanilla ubuntu with just the depends needed to compile juce. the list of linux dependencies i installed is here: http://www.rawmaterialsoftware.com/viewtopic.php?f=5&t=6502&hilit=juce+linux+install its strange as cm doesn't even know about jack, it just opening the juce audio settings dialog and you say those are working in the JuceDemo. From dlphillips at woh.rr.com Tue Sep 18 06:49:59 2012 From: dlphillips at woh.rr.com (Dave Phillips) Date: Tue, 18 Sep 2012 09:49:59 -0400 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <946C62DC-143C-4396-8D63-FBB595FC8DE7@illinois.edu> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <20120916192934.GA17557@frostglow.com> <5056473F.6010702@woh.rr.com> <70DD6FCF-7BB2-43F4-944B-493DB9AC5AB3@illinois.edu> <50586B7D.2000601@woh.rr.com> <946C62DC-143C-4396-8D63-FBB595FC8DE7@illinois.edu> Message-ID: <50587C07.8030203@woh.rr.com> On 09/18/12 09:07, Heinrich Taube wrote: > hi dave it looks like I'm running jack1: dpkg --get-selections | grep > jack jackd install jackd1 install jackd1-firewire install jackd2 > deinstall libjack-dev install libjack-jackd2-0 deinstall libjack0 > install qjackctl install hkt at hkt-VirtualBox:~$ c >> a freshly-build Grace binary segfaults on my Debian Squeeze system >> too (AVLinux 5.0.1). I'll do debug builds on both boxes today. I also >> h > well sheesh! I'm just running a vanilla ubuntu with just the depends needed to compile juce. the list of linux dependencies i installed is here: > > http://www.rawmaterialsoftware.com/viewtopic.php?f=5&t=6502&hilit=juce+linux+install > > > its strange as cm doesn't even know about jack, it just opening the juce audio settings dialog and you say those are working in the JuceDemo. > Hi Rick, I haven't built the JuceDemo. I'm building it now on the Ubunut 10.04 machine, where - surprise - another fresh build of Grace segfaulted when I selected JACK in the Audio Settings. I'll start debug builds today. Something is seriously awry if it's not behaving on three different systems (Debian Squeeze, Ubuntu 10.04, and Arch 64). Best, dp From cmdist at nospam.frostglow.com Tue Sep 18 07:21:45 2012 From: cmdist at nospam.frostglow.com (cmdist at nospam.frostglow.com) Date: Tue, 18 Sep 2012 10:21:45 -0400 Subject: [CM] vim2grace - vim (or other external editor) integration with Grace In-Reply-To: <20120916195115.GA17630@frostglow.com> References: <20120916195115.GA17630@frostglow.com> Message-ID: <20120918142145.GA3863@frostglow.com> I've had a chance to test vim2grace a bit more thoroughly, and found a serious problem. It turns out that procedures that do background processing, like (mp:receive) and (sprout), don't work when sent from vim to Grace via the vim2grace fifo. I'm not exactly sure why they don't work, but I suspect it's because as soon as vim2grace uses (load) to read and execute the code it gets from vim, it immediately starts reading from the fifo again, and that will block until vim makes another write to the fifo (which may not happen for a long time, or ever). So, I think while Grace is blocked like that, it won't run any background tasks that have been scheduled with (mp:receive) or (sprout). I'm don't know how to fix this problem, within the bounds of Grace and s7. I suspect I'll have to abandon the simple expedient of reading from a fifo and use some other method of communicating from vim to Grace, but I'm at a loss as to what method to use. Suggestions are very welcome! --Sergey From taube at illinois.edu Tue Sep 18 07:32:33 2012 From: taube at illinois.edu (Heinrich Taube) Date: Tue, 18 Sep 2012 09:32:33 -0500 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <505880D6.40707@woh.rr.com> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <20120916192934.GA17557@frostglow.com> <5056473F.6010702@woh.rr.com> <70DD6FCF-7BB2-43F4-944B-493DB9AC5AB3@illinois.edu> <50586B7D.2000601@woh.rr.com> <946C62DC-143C-4396-8D63-FBB595FC8DE7@illinois.edu> <505880D6.40707@woh.rr.com> Message-ID: <86B980CF-410D-47B5-99FE-8BC1577D062F@illinois.edu> On Sep 18, 2012, at 9:10 AM, Dave Phillips wrote: > Hi Rick, > > A quick note to mention that I built and ran the JuceDemo on my Ubuntu box. It works fine, but there's no JACK selection in the Audio Tests tab. your demo probably doesn't have jack turned on (JUCE_JACK) but don't worry about this, just try rebuilding the latest cm from scratch and see if it segfaults on ubuntu, i just tried it and it does not segfault do this cd cm svn update premake4 --clean premake4 --jack make at that point you will have Grace with juce enabled for Jack support. Then switching from Alsa to Jack in Grace's Audio Settings panel does not segfault for me. but i don't have jack connection to test as i cannot figure jack out. but sergey had even that working i think: >> Here is what my jack connections look like in patchage: >> >> http://img1.imagilive.com/0912/cm-svn-patchage.png >> > > How do I specify a debug build for the CM/Grace trunk ? > premake4 clean premake4 --jack make config=debug that'll compile juce and grace with assertions to stdio. lookout! -------------- next part -------------- An HTML attachment was scrubbed... URL: From taube at illinois.edu Tue Sep 18 07:37:23 2012 From: taube at illinois.edu (Heinrich Taube) Date: Tue, 18 Sep 2012 09:37:23 -0500 Subject: [CM] vim2grace - vim (or other external editor) integration with Grace In-Reply-To: <20120918142145.GA3863@frostglow.com> References: <20120916195115.GA17630@frostglow.com> <20120918142145.GA3863@frostglow.com> Message-ID: <417F373D-3BAF-445B-BFC4-D8EC279F922E@illinois.edu> On Sep 18, 2012, at 9:21 AM, wrote: > I'm don't know how to fix this problem, within the bounds of Grace and > s7. I suspect I'll have to abandon the simple expedient of reading from > a fifo and use some other method of communicating from vim to Grace, but > I'm at a loss as to what method to use. Suggestions are very welcome! im wondering if -- instead of providing a separate 'cm' binary for console work -- i instead add an 'eval server' to Grace? that way one could start Grace and then send text to its port for evalling from other apps. It would sure make things a lot more robust as it would be exactly like typing into grace's editor. and certainly less difficult to maintain from my end! juce has a InterprocessCommunication thread that i can be used to do this. Todd Inagalls and I had it working quite nicely for sending/receiving stuff from Grace to a common lisp (sbcl i think) -------------- next part -------------- An HTML attachment was scrubbed... URL: From cmdist at nospam.frostglow.com Tue Sep 18 09:52:28 2012 From: cmdist at nospam.frostglow.com (cmdist at nospam.frostglow.com) Date: Tue, 18 Sep 2012 12:52:28 -0400 Subject: [CM] vim2grace - vim (or other external editor) integration with Grace In-Reply-To: <417F373D-3BAF-445B-BFC4-D8EC279F922E@illinois.edu> References: <20120916195115.GA17630@frostglow.com> <20120918142145.GA3863@frostglow.com> <417F373D-3BAF-445B-BFC4-D8EC279F922E@illinois.edu> Message-ID: <20120918165228.GA3955@frostglow.com> On Tue, Sep 18, 2012 at 09:37:23AM -0500, Heinrich Taube wrote: > > im wondering if -- instead of providing a separate 'cm' binary for > console work -- i instead add an 'eval server' to Grace? This sounds like it should solve the vim2grace background processing problem I just described. And since I think I'd mostly be using Grace through the intermediary of a program like vim2grace anyway, I would be pretty happy with that. Those users who don't want to use vim2grace would still need some other way to connect to Grace's eval server if they wanted to use Grace from the console. One solution could be to have a the cm binary be just a simple console-only frontend to Grace's eval server. > that way one could start Grace and then send text to its port for > evalling from other apps. What kind of port do you envision using? A TCP/IP port? A unix socket? Something else? > It would sure make things a lot more robust as it would be exactly > like typing into grace's editor. and certainly less difficult to > maintain from my end! I think this is a legitimate architecture, and it makes a lot of sense if your main focus of development is Grace. However, for those users who prefer to use a console-only app, it might make more sense to have the core features be present in cm, and have the Grace GUI be an optional addon built on top of cm. In that model, it would probably make more sense to have cm provide an eval server that's used by Grace, rather than the other way around. But I can appreciate how implementing this latter architecture would be more work, and can understand your reluctance to go that route. --Sergey From andersvi at notam02.no Tue Sep 18 10:00:47 2012 From: andersvi at notam02.no (andersvi at notam02.no) Date: Tue, 18 Sep 2012 19:00:47 +0200 Subject: [CM] snd: control-panel blocks playback References: <87k3vroh6j.fsf@notam02.no> <87fw6fog4y.fsf@notam02.no> <20120918105234.M25812@ccrma.Stanford.EDU> Message-ID: <87vcfbz2o0.fsf@notam02.no> Just to notify the list as well: Bill fixed the speed-arrow-bug at 5 o'clock this morning :-) From bil at ccrma.Stanford.EDU Tue Sep 18 10:09:57 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 18 Sep 2012 10:09:57 -0700 Subject: [CM] snd: control-panel blocks playback In-Reply-To: <87vcfbz2o0.fsf@notam02.no> References: <87k3vroh6j.fsf@notam02.no> <87fw6fog4y.fsf@notam02.no> <20120918105234.M25812@ccrma.Stanford.EDU> <87vcfbz2o0.fsf@notam02.no> Message-ID: <20120918170738.M92212@ccrma.Stanford.EDU> > Just to notify the list as well: Bill fixed the speed-arrow-bug at 5 > o'clock this morning :-) I was too embarrassed about it to say anything... (sleepy coder, yowling cats, gtk...) From taube at illinois.edu Tue Sep 18 10:17:59 2012 From: taube at illinois.edu (Heinrich Taube) Date: Tue, 18 Sep 2012 12:17:59 -0500 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <50575B8E.5010706@localhost> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <50575B8E.5010706@localhost> Message-ID: <1C9A5627-3263-47A1-AEA5-FA338FD21BDC@illinois.edu> > > That would be useful I think (cm would be better) I have the code base cleaned up quite a bit now, at least in terms of #ifdefs and also not using juice namespace and getting rid of T() and juice::String() calls (which used to me mandatory). i will continue refactoring as i can and also upgrading some features. > , I'm trying to bring old code up and running and it needs patterns ("make-item-stream"!). It does not look like I can get patterns.scm to load on a plain s7? make-item-stream! wow that really is old!! :) hmm cm2 has (new ?) maybe you mean that? in the best of all worlds one could simply load up cm2 code into s7/grace, but cm2 used a tiny-clos'ish object system that chicken, gauche and stklos managed to provide in one form or another. if i could mimic this with s7's object system that would work, but i don't know yet if thats possible or what sort of speed issues would arise. its on my todo list. From andersvi at notam02.no Tue Sep 18 11:40:14 2012 From: andersvi at notam02.no (andersvi at notam02.no) Date: Tue, 18 Sep 2012 20:40:14 +0200 Subject: [CM] snd: control-panel blocks playback In-Reply-To: <20120918170738.M92212@ccrma.Stanford.EDU> (Bill Schottstaedt's message of "Tue, 18 Sep 2012 10:09:57 -0700") References: <87k3vroh6j.fsf@notam02.no> <87fw6fog4y.fsf@notam02.no> <20120918105234.M25812@ccrma.Stanford.EDU> <87vcfbz2o0.fsf@notam02.no> <20120918170738.M92212@ccrma.Stanford.EDU> Message-ID: <87r4pzyy29.fsf@notam02.no> B> I was too embarrassed about it to say anything... (sleepy coder, B> yowling cats, gtk...) A short little story (for the cmdist archives), about a young musicology student back in 1992, drawn into the confusing combination of lisp and dsp mantras, square-eyed trying to make sense of what someone called the debugger output (more like assembler code) coming from some linked-in 56k-code, wondering where it all went wrong: finally, after daring to push C-Alt-x, a long nights wait (12 hrs or more) for the nextstation (turbo!) to compute a 2 minute (stereo!) soundfile, and finally some sound to play! Only silence coming out.. Many attempts during the weeks all ended in vain. I was surely misunderstanding something very basic, but noone around to help... This was in 1992, before altavista and all that, but we had ftp and email (and archie!). The README file suggested to contact the programmer himself - bil at ccrma... What! the guy who wrote those articles about counterpoint and things? Nope, he's surely got better things to do. Finally i got a clumsy email sent along, describing the troubles as far as possible. After 1 or 2 hours a reply came back from .edu, starting with (the now familiar) "thanks for the bug-report!", and a notice to get a fresh tarball from ccrma. What!? - he's suggesting its the programs fault, not mine! You all can guess the rest: next morning, after the Next had spun for another night, sound came out as expected. After that, the idea of closed-source and special support-deals as something good seemed strange, as it indeed still does. Thanks again! -anders From nando at ccrma.Stanford.EDU Tue Sep 18 20:29:10 2012 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: Tue, 18 Sep 2012 20:29:10 -0700 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <1C9A5627-3263-47A1-AEA5-FA338FD21BDC@illinois.edu> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <50575B8E.5010706@localhost> <1C9A5627-3263-47A1-AEA5-FA338FD21BDC@illinois.edu> Message-ID: <50593C06.8050001@localhost> On 09/18/2012 10:17 AM, Heinrich Taube wrote: >> , I'm trying to bring old code up and running and it needs patterns ("make-item-stream"!). It does not look like I can get patterns.scm to load on a plain s7? > > make-item-stream! wow that really is old!! :) hmm cm2 has (new ?) maybe you mean that? Nope, make-item-stream 'items 'heap '( things ...) :-) Or just (items x y z in heap) Very old. > in the best of all worlds one could simply load up cm2 code into s7/grace, but cm2 used a tiny-clos'ish object system that chicken, gauche and stklos managed to provide in one form or another. if i could mimic this with s7's object system that would work, but i don't know yet if thats possible or what sort of speed issues would arise. its on my todo list. I ended up hacking patterns.scm enough so it loads and runs (very limited testing just for the patterns I was using) in an s7 interpreter. It was not that difficult. It would be good to have an "official" version that loads into plain s7... The piece (from 1994) is starting to make noises again with much help from Bill. -- Fernando From taube at illinois.edu Wed Sep 19 03:17:10 2012 From: taube at illinois.edu (Heinrich Taube) Date: Wed, 19 Sep 2012 05:17:10 -0500 Subject: [CM] cm in juce 2.0, jack In-Reply-To: <50593C06.8050001@localhost> References: <952D389F-A6E6-4636-BA08-1F4FDA9D40A4@illinois.edu> <20120916171210.GA17310@frostglow.com> <50575B8E.5010706@localhost> <1C9A5627-3263-47A1-AEA5-FA338FD21BDC@illinois.edu> <50593C06.8050001@localhost> Message-ID: <022A93B5-C981-46F2-A33F-4D6A9496401E@illinois.edu> i think patterns.scm should be a pure scheme except for record definitions that (at least at one point) had to be added from the c side. any code you convert to pure scheme ill add into the repo. On Sep 18, 2012, at 10:29 PM, Fernando Lopez-Lezcano wrote: > On 09/18/2012 10:17 AM, Heinrich Taube wrote: >>> , I'm trying to bring old code up and running and it needs patterns ("make-item-stream"!). It does not look like I can get patterns.scm to load on a plain s7? >> >> make-item-stream! wow that really is old!! :) hmm cm2 has (new ?) maybe you mean that? > > Nope, make-item-stream 'items 'heap '( things ...) > :-) > Or just (items x y z in heap) > Very old. > >> in the best of all worlds one could simply load up cm2 code into s7/grace, but cm2 used a tiny-clos'ish object system that chicken, gauche and stklos managed to provide in one form or another. if i could mimic this with s7's object system that would work, but i don't know yet if thats possible or what sort of speed issues would arise. its on my todo list. > > I ended up hacking patterns.scm enough so it loads and runs (very limited testing just for the patterns I was using) in an s7 interpreter. It was not that difficult. It would be good to have an "official" version that loads into plain s7... > > The piece (from 1994) is starting to make noises again with much help from Bill. > -- Fernando From bil at ccrma.Stanford.EDU Wed Sep 19 16:12:46 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 19 Sep 2012 16:12:46 -0700 Subject: [CM] Snd 13.1 Message-ID: <20120919231117.M47094@ccrma.Stanford.EDU> Snd 13.1: cload.scm to automatically link in a C function from a C shared library this file and lint.scm are now also included in the s7 tarball. symbol->dynamic-value checked: gtk 3.5.12|14|16|18, clang 3.1, mpc 1.0.1 Thanks!: Fernando, Mike, Rick, Anders. From cmdist at nospam.frostglow.com Sat Sep 22 08:33:19 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 11:33:19 -0400 Subject: [CM] Grace freezes on sprout Message-ID: <20120922153318.GA9853@frostglow.com> When I run the following code in Grace: (define (foo) (let ((sequence '(0 1 2))) (process for i from 0 to (length sequence) for current-element = (nth sequence i) do (print current-element) (wait 1)))) (sprout (foo)) grace will print: 0 1 2 and then will go in to what seems like some kind of infinite loop, spiking my CPU usage to 100%, and slowing everything on the system down to a crawl. If I let it run for longer than maybe 30 seconds, my system will completely freeze (except for disk activity, as I can see the disk drive light flash), with no mouse movement or keyboard input of any kind having any effect and no more screen refresh happening. The only solution at this point is a hard reboot. Grace has had the same behavior in numerous other cases as well, but this is just one of the simplest reproduceable cases I've been able to find. Another thing I'd like to report that even when doing absolutely nothing, Grace uses about 34% of my CPU. That's what happens when I run a fresh instance of Grace, without loading any files, nor opening Grace's editor or doing anything else. That's pretty significant CPU use for a process that's doing virtually nothing. Is there any way to make Grace more efficient? --Sergey From taube at illinois.edu Sat Sep 22 09:22:09 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 11:22:09 -0500 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922153318.GA9853@frostglow.com> References: <20120922153318.GA9853@frostglow.com> Message-ID: <2BC94897-722B-47AE-B8DD-F8352F19357A@illinois.edu> this might be an issue with the new error handing. let me try? On Sep 22, 2012, at 10:33 AM, Sergey wrote: > When I run the following code in Grace: > > (define (foo) > (let ((sequence '(0 1 2))) > (process > for i from 0 to (length sequence) > for current-element = (nth sequence i) > do > (print current-element) > (wait 1)))) > > (sprout (foo)) > > grace will print: > > 0 > 1 > 2 > > and then will go in to what seems like some kind of infinite loop, > spiking my CPU usage to 100%, and slowing everything on the system down > to a crawl. If I let it run for longer than maybe 30 seconds, my > system will completely freeze (except for disk activity, as I can > see the disk drive light flash), with no mouse movement or keyboard > input of any kind having any effect and no more screen refresh > happening. The only solution at this point is a hard reboot. > > Grace has had the same behavior in numerous other cases as well, but > this is just one of the simplest reproduceable cases I've been able to > find. > > Another thing I'd like to report that even when doing absolutely > nothing, Grace uses about 34% of my CPU. That's what happens when > I run a fresh instance of Grace, without loading any files, nor > opening Grace's editor or doing anything else. That's pretty > significant CPU use for a process that's doing virtually nothing. > Is there any way to make Grace more efficient? > > --Sergey > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From taube at illinois.edu Sat Sep 22 09:37:32 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 11:37:32 -0500 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922153318.GA9853@frostglow.com> References: <20120922153318.GA9853@frostglow.com> Message-ID: <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> > > Another thing I'd like to report that even when doing absolutely > nothing, Grace uses about 34% of my CPU. That's what happens when > I run a fresh instance of Grace, without loading any files, nor On OSX grace hovers around 1.2 pct of cpu which is total fine, ill test it on linux once i can figure out the crash. does switching to format stop the crash? (define (foo) (let ((sequence '(0 1 2))) (process for i from 0 to (length sequence) for current-element = (nth sequence i) do (format #t "~A~%" current-element) (wait 1)))) (sprout (foo)) From taube at illinois.edu Sat Sep 22 10:13:02 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 12:13:02 -0500 Subject: [CM] Grace freezes on sprout In-Reply-To: <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> Message-ID: <2DCDEAC5-E3E0-47A0-A5E9-AD68386EC935@illinois.edu> ok for sure you need to do an svn update as i had not actually committed the new error hook code for the latest sndlib!!! please do this so we are on the same page: cd cm rm -rf sndlin premake4 clean premake4 make that will pull the latest sndlib tar ball and compile everything. the start the app and try that example, perhaps it won't crash. the reason that grace is probably racing is that it is in a tight loop because the C side never sees the error return value. let me know if this fixes it or not. lots has changed so you should expect issues to come up. I'm not adding any new features until its as stable as last spring. if it ever does this you can go to the Eval menu and select Interrupt Scheme (assuming you will get enough cycles to use the mouse.) on OSX you can press Command-K and this will do it more quickly. also when the queue is empty it will print a "queue is empty" to std:out -- if you don't see that after an error or when all processes are done then its probably racing. From cmdist at nospam.frostglow.com Sat Sep 22 10:14:51 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 13:14:51 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> Message-ID: <20120922171451.GA9994@frostglow.com> On Sat, Sep 22, 2012 at 11:37:32AM -0500, Heinrich Taube wrote: > > > > Another thing I'd like to report that even when doing absolutely > > nothing, Grace uses about 34% of my CPU. That's what happens when > > I run a fresh instance of Grace, without loading any files, nor > > On OSX grace hovers around 1.2 pct of cpu which is total fine, ill test it on linux once i can figure out the crash. Actually, it looks like on a freshly rebooted Linux system, Grace only takes up about 14% of the CPU. That's a lot better than the 34% I saw earlier, but still a far cry from 1.2% I should mention that I'm running Grace on a single-core amd64 3000+ CPU, with 4 gigs of ram, and a 3.3.8-gentoo kernel with the PREEMPT patch. > does switching to format stop the crash? > > > (define (foo) > (let ((sequence '(0 1 2))) > (process > for i from 0 to (length sequence) > for current-element = (nth sequence i) > do > (format #t "~A~%" current-element) > (wait 1)))) > > (sprout (foo)) Nope. Grace reacts identically when running this code as it did when running the code I posted in my original message. Also, I don't think this behavior is a consequence of the new error handling code, because the old 3.8.0 version of Grace exhibits the same behavior. Thanks for looking in to this, --Sergey From cmdist at nospam.frostglow.com Sat Sep 22 10:50:07 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 13:50:07 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <2DCDEAC5-E3E0-47A0-A5E9-AD68386EC935@illinois.edu> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <2DCDEAC5-E3E0-47A0-A5E9-AD68386EC935@illinois.edu> Message-ID: <20120922175007.GA10082@frostglow.com> On Sat, Sep 22, 2012 at 12:13:02PM -0500, Heinrich Taube wrote: > > ok for sure you need to do an svn update as i had not actually committed the new error hook code for the latest sndlib!!! > > please do this so we are on the same page: > > cd cm > rm -rf sndlin > premake4 clean > premake4 > make Ok. I did the above (modifying "sndlin" to "sndlib", and the last "premake4" to "premake4 --jack" It seemed to compile fine, with just a few warnings. The full compile log is here: http://pastie.org/4781177 Then, when I ran Grace, in my terminal I saw the following errors: http://pastie.org/4781141 When I ran either the code I posted in my original message or your suggested modification with "format" instead of "print", Grace reported replied "syntax-error" when I executed the "(sprout (foo))" line, though no complaints when executing the definition of foo. Also, at the same time it reported the syntax-error in the Grace window, on my terminal I saw Grace output: ;sprout: unbound variable ; (sprout (foo)) This is definitely better than going in to an infinite loop and hanging my machine, but now I think I'm a bit at a loss as far as understanding what's going on here. Which variable does Grace think is unbound? And what is the source of the syntax-error? Thanks, --Sergey > that will pull the latest sndlib tar ball and compile everything. the start the app and try that example, perhaps it won't crash. the reason that grace is probably racing is that it is in a tight loop because the C side never sees the error return value. > > let me know if this fixes it or not. lots has changed so you should expect issues to come up. I'm not adding any new features until its as stable as last spring. > if it ever does this you can go to the Eval menu and select Interrupt Scheme (assuming you will get enough cycles to use the mouse.) on OSX you can press Command-K and this will do it more quickly. > > also when the queue is empty it will print a "queue is empty" to std:out -- if you don't see that after an error or when all processes are done then its probably racing. > > > From cmdist at nospam.frostglow.com Sat Sep 22 10:57:42 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 13:57:42 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922173002.M87196@ccrma.Stanford.EDU> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <20120922171451.GA9994@frostglow.com> <20120922173002.M87196@ccrma.Stanford.EDU> Message-ID: <20120922175742.GB10082@frostglow.com> On Sat, Sep 22, 2012 at 10:30:48AM -0700, Bill Schottstaedt wrote: > > > for i from 0 to (length sequence) > > do you mean "below" not "to"? I forget how loop works, or nth. If I'm interpreting the Common Music docs correctly, I should be able to use "to": for var [from expr] [to | below expr] [by expr] increments var from an on optional starting value to an ending value by an optional amount stopping statements That's from: http://commonmusic.sourceforge.net/cm/res/doc/cm.html#iteration And nth works like this: > (nth '(A B C) 0) A > (nth '(A B C) 1) B etc.. --Sergey From taube at illinois.edu Sat Sep 22 10:57:42 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 12:57:42 -0500 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922175007.GA10082@frostglow.com> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <2DCDEAC5-E3E0-47A0-A5E9-AD68386EC935@illinois.edu> <20120922175007.GA10082@frostglow.com> Message-ID: <4D0A3938-A29C-487D-9983-833B4DDDF488@illinois.edu> wow, the thing isn't loading I've never seen that! ill have to do a fresh install and see whats going on, i was doing some scheme stuff with bill and maybe i managed to hammer lisp loading. On Sep 22, 2012, at 12:50 PM, Sergey wrote: > On Sat, Sep 22, 2012 at 12:13:02PM -0500, Heinrich Taube wrote: >> >> ok for sure you need to do an svn update as i had not actually committed the new error hook code for the latest sndlib!!! >> >> please do this so we are on the same page: >> >> cd cm >> rm -rf sndlin >> premake4 clean >> premake4 >> make > > Ok. I did the above (modifying "sndlin" to "sndlib", and the last > "premake4" to "premake4 --jack" > > It seemed to compile fine, with just a few warnings. The full compile > log is here: > > http://pastie.org/4781177 > > Then, when I ran Grace, in my terminal I saw the following errors: > > http://pastie.org/4781141 > > When I ran either the code I posted in my original message or your > suggested modification with "format" instead of "print", Grace reported > replied "syntax-error" when I executed the "(sprout (foo))" line, though > no complaints when executing the definition of foo. > > Also, at the same time it reported the syntax-error in the Grace window, > on my terminal I saw Grace output: > > ;sprout: unbound variable > ; (sprout (foo)) > > This is definitely better than going in to an infinite loop and hanging > my machine, but now I think I'm a bit at a loss as far as understanding > what's going on here. Which variable does Grace think is unbound? And > what is the source of the syntax-error? > > Thanks, > > --Sergey > >> that will pull the latest sndlib tar ball and compile everything. the start the app and try that example, perhaps it won't crash. the reason that grace is probably racing is that it is in a tight loop because the C side never sees the error return value. >> >> let me know if this fixes it or not. lots has changed so you should expect issues to come up. I'm not adding any new features until its as stable as last spring. >> if it ever does this you can go to the Eval menu and select Interrupt Scheme (assuming you will get enough cycles to use the mouse.) on OSX you can press Command-K and this will do it more quickly. >> >> also when the queue is empty it will print a "queue is empty" to std:out -- if you don't see that after an error or when all processes are done then its probably racing. >> >> >> From bil at ccrma.Stanford.EDU Sat Sep 22 10:59:45 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 22 Sep 2012 10:59:45 -0700 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922175007.GA10082@frostglow.com> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <2DCDEAC5-E3E0-47A0-A5E9-AD68386EC935@illinois.edu> <20120922175007.GA10082@frostglow.com> Message-ID: <20120922175751.M56152@ccrma.Stanford.EDU> That's the new sndlib, but the old cm/scm/s7.scm. I was going to wait to make the change to sndlib, but then something unrelated came up, and the new error-hook handling accidentally got moved into the current sndlib. From bil at ccrma.Stanford.EDU Sat Sep 22 11:01:55 2012 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 22 Sep 2012 11:01:55 -0700 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922175742.GB10082@frostglow.com> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <20120922171451.GA9994@frostglow.com> <20120922173002.M87196@ccrma.Stanford.EDU> <20120922175742.GB10082@frostglow.com> Message-ID: <20120922180047.M50610@ccrma.Stanford.EDU> yes, but "to (length sequence)" here counts to 3, including the 3, and (nth sequence 3) is an error, hence, I think, Rick's question about the error handler. From cmdist at nospam.frostglow.com Sat Sep 22 11:19:30 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 14:19:30 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922180047.M50610@ccrma.Stanford.EDU> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <20120922171451.GA9994@frostglow.com> <20120922173002.M87196@ccrma.Stanford.EDU> <20120922175742.GB10082@frostglow.com> <20120922180047.M50610@ccrma.Stanford.EDU> Message-ID: <20120922181930.GC10082@frostglow.com> On Sat, Sep 22, 2012 at 11:01:55AM -0700, Bill Schottstaedt wrote: > > yes, but "to (length sequence)" here counts to 3, including > the 3, and (nth sequence 3) is an error, hence, I think, Ahh.. so "to" is inclusive and "below" is exclusive? I guess I've bitten myself with the dreaded off-by-one bug. --Sergey > Rick's question about the error handler. > > From taube at illinois.edu Sat Sep 22 11:29:54 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 13:29:54 -0500 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922180047.M50610@ccrma.Stanford.EDU> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <20120922171451.GA9994@frostglow.com> <20120922173002.M87196@ccrma.Stanford.EDU> <20120922175742.GB10082@frostglow.com> <20120922180047.M50610@ccrma.Stanford.EDU> Message-ID: <48E2045D-700D-406A-A911-540A499588B8@illinois.edu> i did a completely fresh checkout, built it and works fine. you might try the same, ill try linux but i can't see what the difference would be. On Sep 22, 2012, at 1:01 PM, Bill Schottstaedt wrote: > yes, but "to (length sequence)" here counts to 3, including > the 3, and (nth sequence 3) is an error, hence, I think, > Rick's question about the error handler. > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From taube at illinois.edu Sat Sep 22 11:56:45 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 13:56:45 -0500 Subject: [CM] Grace freezes on sprout Message-ID: <48A4CA90-852F-4FB1-B0AF-04BBBE46DBBB@illinois.edu> i did a fresh install on ubuntu and its working there too (doesn't crash doesn't spin) : rm -r cm svn co http://commonmusic.svn.sf.net/svnroot/commonmusic/trunk cm cd cm premake4 make bin/Grace & top my top shows 12% in idle going up to 30% when it does anything. I've no idea why its this high, you can grep for 'setPriority' settings in cm's sources and change them to smaller values, but I'm not sure if this will help or not, i think the os mitt do what it wants. also Open File and Save don't work for me on linux, i don't get a modal dialog and instead I see the following in the console: GThread-ERROR **: file /build/buildd/glib2.0-2.30.0/./gthread/gthread-posix.c: line 348 (g_thread_create_posix_impl): error 'Invalid argument' during 'pthread_attr_setschedparam (&attr, &sched)' i think this probably isn't due to cm directly, maybe the 'native' open file dialog i ask juce for isn't working on linux. ill have to poke around?. From cmdist at nospam.frostglow.com Sat Sep 22 12:13:02 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 15:13:02 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <48E2045D-700D-406A-A911-540A499588B8@illinois.edu> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <20120922171451.GA9994@frostglow.com> <20120922173002.M87196@ccrma.Stanford.EDU> <20120922175742.GB10082@frostglow.com> <20120922180047.M50610@ccrma.Stanford.EDU> <48E2045D-700D-406A-A911-540A499588B8@illinois.edu> Message-ID: <20120922191302.GD10082@frostglow.com> On Sat, Sep 22, 2012 at 01:29:54PM -0500, Heinrich Taube wrote: > > i did a completely fresh checkout, built it and works fine. you might > try the same, ill try linux but i can't see what the difference would > be. Once I rebuilt Grace from a fresh checkout it works fine here as well. No more errors are reported when Grace starts up. It just says "Queue is empty". And, once I run my program, Grace reports: # 0 1 2 >>> Error: list-ref argument 2, 3, is out of range (should be less than list length) Changing the "to" to "below" in the (process) initialization fixes the off-by-one bug I had, and the program runs fine. Grace no longer goes in to the endless loop of death and I am very happy! Thank you for your help! --Sergey From cmdist at nospam.frostglow.com Sat Sep 22 12:22:46 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 15:22:46 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <48A4CA90-852F-4FB1-B0AF-04BBBE46DBBB@illinois.edu> References: <48A4CA90-852F-4FB1-B0AF-04BBBE46DBBB@illinois.edu> Message-ID: <20120922192246.GE10082@frostglow.com> On Sat, Sep 22, 2012 at 01:56:45PM -0500, Heinrich Taube wrote: > > my top shows 12% in idle going up to 30% when it does anything. I've > no idea why its this high, you can grep for 'setPriority' settings in > cm's sources and change them to smaller values, but I'm not sure if > this will help or not, i think the os mitt do what it wants. Even if lowering the priority did help, I'm not sure if it's the best solution, as in the past I've had to raise Grace's priority (via reducing its nice value to -1) in order to keep latency at an acceptable level. Before I did that, sometimes I'd hear gaps in playback when playing notes via mp:midi. I should note, however, that I observed the 14% to 34% CPU use at idle happened without any mucking around with priorities or nice levels. > also Open File and Save don't work for me on linux, i don't get a > modal dialog and instead I see the following in the console: > > GThread-ERROR **: file > /build/buildd/glib2.0-2.30.0/./gthread/gthread-posix.c: line 348 > (g_thread_create_posix_impl): error 'Invalid argument' during > 'pthread_attr_setschedparam (&attr, &sched)' > > i think this probably isn't due to cm directly, maybe the 'native' > open file dialog i ask juce for isn't working on linux. ill have to > poke around?. I have no problems getting the Open dialog or opening files using it on Gentoo Linux. I notice that the error message you pasted above mentions glib2.0-2.30.0 I'm using glibc-2.13-r4 on my system. Perhaps that accounts for the difference? I don't know what widget toolkit Grace uses to get the Open dialog on Linux, but perhaps there's a version difference for that between your Ubuntu system and my Gentoo system too. --Sergey From taube at illinois.edu Sat Sep 22 13:00:43 2012 From: taube at illinois.edu (Heinrich Taube) Date: Sat, 22 Sep 2012 15:00:43 -0500 Subject: [CM] Grace freezes on sprout In-Reply-To: <20120922191302.GD10082@frostglow.com> References: <20120922153318.GA9853@frostglow.com> <454BFC15-6A8B-4A3A-882A-C7CDFBE187C2@illinois.edu> <20120922171451.GA9994@frostglow.com> <20120922173002.M87196@ccrma.Stanford.EDU> <20120922175742.GB10082@frostglow.com> <20120922180047.M50610@ccrma.Stanford.EDU> <48E2045D-700D-406A-A911-540A499588B8@illinois.edu> <20120922191302.GD10082@frostglow.com> Message-ID: > I should note, however, that I observed the 14% to 34% CPU use at idle > happened without any mucking around with priorities or nice levels. i just checked tops on the JuceDemo in ubuntu and in its idle loop its running at 74% cpu. so given that I'm using the exact same code base as the demo app, think cm is actually doing pretty well :) > Grace no longer goes in to the endless loop of death and I am very happy! me too! -------------- next part -------------- An HTML attachment was scrubbed... URL: From cmdist at nospam.frostglow.com Sat Sep 22 13:53:27 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sat, 22 Sep 2012 16:53:27 -0400 Subject: [CM] Grace freezes on sprout In-Reply-To: <0DBD6576-B772-45A5-A569-F631E4E69BEC@illinois.edu> References: <48A4CA90-852F-4FB1-B0AF-04BBBE46DBBB@illinois.edu> <20120922192246.GE10082@frostglow.com> <0DBD6576-B772-45A5-A569-F631E4E69BEC@illinois.edu> Message-ID: <20120922205327.GA10340@frostglow.com> On Sat, Sep 22, 2012 at 03:04:29PM -0500, Heinrich Taube wrote: > > > I'm using glibc-2.13-r4 on my system. Perhaps that accounts for the > > difference? I don't know what widget toolkit Grace uses to get the Open > > what apt-get command i need to update my lib to what you are suing? Unfortunately, Gentoo (the version of Linux that I use) uses a completely different package management system from Ubuntu. So the commands for doing this are totally different on Gentoo (it doesn't use apt-get at all) and even the package names differ. I only very rarely use Ubuntu, and don't know enough about it to answer your question. My best guess is to try: # apt-get update # apt-get upgrade libc6 But I'm not sure if that'll upgrade the right package, and I think it'll only upgrade to whatever the recent version of libc6 is, not to a particular version. Perhaps someone else here who knows Ubuntu could help? --Sergey From cmdist at nospam.frostglow.com Sun Sep 23 13:25:07 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Sun, 23 Sep 2012 16:25:07 -0400 Subject: [CM] Grace segfaults while reading from a file Message-ID: <20120923202507.GA23470@frostglow.com> Seeing that neither Common Music nor s7 have any built-in procedures for reading in a file as a list of characters, nor as a list of strings, nor as a single string, I set out to write these procedures myself. While trying to write the first of these, I came up with this (incomplete) code: http://pastie.org/4787162 When the file "foo.txt" contains only the single word "test", executing line 10 in Grace works -- each of the individual characters is printed. It also works with one longer line of characters, and two lines. But, at some point, when the file grows to a certain size or there are enough characters or lines in the file, Grace segfaults. Here is a sample 501-line, 32565-character file that causes Grace to segfault on my system: http://pastie.org/4787221 Unfortunately, it seems that the Grace binary was stripped during the build process, so I am unable to get a useful backtrace. This is the best I can get: http://pastie.org/4787240 Here is the last bit of an strace: http://pastie.org/4787257 --Sergey From cmdist at nospam.frostglow.com Mon Sep 24 14:42:19 2012 From: cmdist at nospam.frostglow.com (Sergey) Date: Mon, 24 Sep 2012 17:42:19 -0400 Subject: [CM] Grace segfaults while reading from a file In-Reply-To: <20120923202507.GA23470@frostglow.com> References: <20120923202507.GA23470@frostglow.com> Message-ID: <20120924214219.GA32668@frostglow.com> Curiously, this works fine (well, it's slow, but at least it doesn't segfault): http://pastie.org/4793680 So it seems that it's only printing during the loop that segfaults. --Sergey On Sun, Sep 23, 2012 at 04:25:07PM -0400, Sergey wrote: > > Seeing that neither Common Music nor s7 have any built-in procedures for > reading in a file as a list of characters, nor as a list of strings, nor > as a single string, I set out to write these procedures myself. > > While trying to write the first of these, I came up with this > (incomplete) code: > > http://pastie.org/4787162 > > When the file "foo.txt" contains only the single word "test", executing > line 10 in Grace works -- each of the individual characters is printed. > > It also works with one longer line of characters, and two lines. > But, at some point, when the file grows to a certain size or there > are enough characters or lines in the file, Grace segfaults. > > Here is a sample 501-line, 32565-character file that causes Grace to > segfault on my system: > > http://pastie.org/4787221 > > Unfortunately, it seems that the Grace binary was stripped during the > build process, so I am unable to get a useful backtrace. This is the > best I can get: > > http://pastie.org/4787240 > > Here is the last bit of an strace: > > http://pastie.org/4787257 > > --Sergey