From taube at uiuc.edu Mon Aug 1 05:01:38 2005 From: taube at uiuc.edu (Rick Taube) Date: Mon, 1 Aug 2005 07:01:38 -0500 Subject: [CM] ILU tutorials added Message-ID: <6b4e4c826d947164576e6b9c81b75cf0@uiuc.edu> Ive added several tutorials i gave at the ILU conference in june to CVS. http://commonmusic.sf.net/etc/tutorials If you already have a CVS tree of CM you get the tutorials by doing cd /path/to/cm cvs -dAP update (If you dont add the -dAP business you wont get the new directory, thank you tobias for pointing this out) -rick From taube at uiuc.edu Mon Aug 1 05:09:21 2005 From: taube at uiuc.edu (Rick Taube) Date: Mon, 1 Aug 2005 07:09:21 -0500 Subject: [CM] ILU tutorials added In-Reply-To: <6b4e4c826d947164576e6b9c81b75cf0@uiuc.edu> References: <6b4e4c826d947164576e6b9c81b75cf0@uiuc.edu> Message-ID: <8d88c6471beaf4feb1231393426156c0@uiuc.edu> oops i meant: cvs update -dAP On Aug 1, 2005, at 7:01 AM, Rick Taube wrote: > Ive added several tutorials i gave at the ILU conference in june to > CVS. > > http://commonmusic.sf.net/etc/tutorials > > > If you already have a CVS tree of CM you get the tutorials by doing > cd /path/to/cm > cvs -dAP update > > (If you dont add the -dAP business you wont get the new directory, > thank you tobias for pointing this out) > > -rick > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From taube at uiuc.edu Mon Aug 1 15:43:59 2005 From: taube at uiuc.edu (Rick Taube) Date: Mon, 1 Aug 2005 17:43:59 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1122935105.11196.11.camel@strawberri.maginvent.org> References: <1122935105.11196.11.camel@strawberri.maginvent.org> Message-ID: Hi Juan, ok thansk Ill take a look. in the meantime just dont save the image -- Once you compile gtk it only takes seconds to load from the fasls. cm -l sbcl -e emacs & Note that SBCL is a bit of problem for me right now, it seems that: 1. sbcl 0.9.n doesnt support callbacks so GTK and CM gui cant be used 2. sbcl 0.8.n doesnt support threads so realtime wont be usefulg on the positive side, Ken reports that in the Planet CCRMA distro using sbcl 0.9.0 the new real time stuff seem to works well and is fast. the only problem so far seems to be that if you run with slime you will eventually get a n error when you receive. --rick On Aug 1, 2005, at 5:25 PM, Juan Reyes wrote: > > Hi Rick, > > I am trying to compile cm-2.7 on Nando's PlanetCCRMA Linux SBCL > 0.8.16. > > Compilation with midishare and portmidi goes fine until I try to save > the image with your instructions for loading "make.lisp" which also > seem > to be working according to the web page. > > However, if I try to run the script on ./bin/cm.sh SBCL will give me an > error. > > Here are the commands and listing: > > * (load "src/make") > > T > * (make-cm :save-image t) > > ; Installation directory: "/home/juanig/lisp/cm/" > ; Loading "bin/sbcl_0.8.16_linux-i686/gtkffi-cmusbcl.fasl" > ; Loading "bin/sbcl_0.8.16_linux-i686/clm-stubs.fasl" > . > . > . > ; Loading "bin/sbcl_0.8.16_linux-i686/eventio.fasl" > ; Removing fasl files. > ; Garbage collecting. > ; Saving application image. > ; Bye! > [doing purification: roots handlers stack bindings static cleanup done] > [undoing binding stack and other enclosing state... done] > [saving current Lisp image > into /home/juanig/lisp/cm/bin/sbcl_0.8.16_linux-i686/cm.img: > writing 27459304 bytes from the read-only space at 0x01000000 > writing 10609256 bytes from the static space at 0x05000000 > writing 4096 bytes from the dynamic space at 0x09000000 > done] > > <> ./cm.sh > > debugger invoked on a UNDEFINED-FUNCTION in thread 11105: > The function SB-POSIX::GETENV is undefined. > > You can type HELP for debugger help, or (SB-EXT:QUIT) to exit from > SBCL. > > (no restarts: If you didn't do this on purpose, please report it as a > bug.) > (SB-KERNEL::UNDEFINED-FUN-ERROR-HANDLER > 4 > # > #.(SB-SYS:INT-SAP #XB7D97EE0) > # (SB- > ALIEN:STRUCT > SB-VM::OS- > CONTEXT-T-STRUCT))> > (14))[:EXTERNAL] > 0] (SB-EXT:QUIT) > > > > > From taube at uiuc.edu Mon Aug 1 18:24:58 2005 From: taube at uiuc.edu (Rick Taube) Date: Mon, 1 Aug 2005 20:24:58 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1122935105.11196.11.camel@strawberri.maginvent.org> References: <1122935105.11196.11.camel@strawberri.maginvent.org> Message-ID: <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> > debugger invoked on a UNDEFINED-FUNCTION in thread 11105: > The function SB-POSIX::GETENV is undefined. juan -- the sbcl posix support changed somewhere between 0.8.16 (which ccrma has) and 0.9.0: in 0.8.16 its sb-ext:posix-getenv and in 0.9.0 its sb-posix:getenv. ive fixed the problem in rev 1.26 of sbcl.lisp (it will take a while for the changes to migrate to anonymous cvs...). but you still have another problem: the ccrma sbcl 0.8.16 has not been patched for callbacks so cm still wont build in it without you first either (1) adding callbacks into sbcl (its easy, see http://commonmusic.sf.net/doc/install.html#gtk_sbcl ) or (2) disabling gtk loading by making the function plotter-can-build? in make.lisp return false. if you decide you dont need/want the gui (option 2) then really you should just move up to sbcl 0.9.0 -- it already refuses to build gtk (since callbacks dont work there at all) BUT you can then try out the realtime code if you want (sans slime for now). From mjkoskin at koti.soon.fi Tue Aug 2 06:21:50 2005 From: mjkoskin at koti.soon.fi (Matti Koskinen) Date: Tue, 02 Aug 2005 16:21:50 +0300 Subject: [CM] snd and fc4 Message-ID: <42EF736E.80404@koti.soon.fi> hi, I installed a clean install of fedora core 4, and now snd (or any other recording program) doesn't work. First I had an old snd from RH8.0 and that works for playback but not for recording. Then I installed snd from sources I had (snd-7) and that doesn't record either. The record button turns to red, but time counter sticks at 0.0 and nothing gets recorded. I can't even get the VU-meters show anything. Is this a known issue in fc4? A quick search with google didn't reveal anything. arecord records only silence and k3guitune's oscilloscope doesn't show anything. None of the programs show any errors, dmesg shows that snd uses obsolete OSS audio interface. Where to start searching for cause of this? TIA -matti From juanig at ccrma.Stanford.EDU Tue Aug 2 09:57:00 2005 From: juanig at ccrma.Stanford.EDU (Juan Reyes) Date: Tue, 02 Aug 2005 09:57:00 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> Message-ID: <1123001821.14784.61.camel@strawberri.maginvent.org> Thanks a lot Rick > > if you decide you dont need/want the gui (option 2) then really you > should just move up to sbcl 0.9.0 -- it already refuses to build gtk > (since callbacks dont work there at all) BUT you can then try out the > realtime code if you want (sans slime for now). > I decided to install sbcl-0.9.0.18-1 RPM binary from the SBCL sourceforge site because on planetCCRMA?s 0.8.16, I had already compiled a new sbcl.core with the callback support alien function package. On 0.8.16 the graphic gtk stuff compiles but as you said the sbcl posix support has changed. Also because on sbcl-0.9.0.18-1, I was able to compile and load CMN. The make.lisp function saves CMN with CM to be run with the bin/cm.sh script. Therefore if I am lucky I think I will have CMN, midishare, portmidi running on this cm image (know on wood). --* Juan From taube at uiuc.edu Tue Aug 2 11:31:07 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 2 Aug 2005 13:31:07 -0500 Subject: [CM] recording Message-ID: ive added portmidi-record! to cvs (rev 1.30 pm.scm). once you've recorded data into a seq you can do pretty much what you want with it; the second tutorial shows how to edit data using the map-objects function i added earlier this summer. doc is at http://commonmusic.sf.net/doc/dict/portmidi-topic.html#portmidi-record --rick From rdlatimer at tjhsst.edu Tue Aug 2 11:31:15 2005 From: rdlatimer at tjhsst.edu (rdlatimer at tjhsst.edu) Date: Tue, 2 Aug 2005 14:31:15 -0400 (EDT) Subject: [CM] Intro Program - 2-part Invention In-Reply-To: References: <1122935105.11196.11.camel@strawberri.maginvent.org> Message-ID: <2352.151.188.16.41.1123007475.squirrel@mail.tjhsst.edu> I'm a newcomer to Common Music. I'm using CLisp and WindowsXP. Rick's Intro programs are running fine and sound good. Also to get started, I'd like to enter in some music - like for instance the Bach F major 2-part invention. Any advice on a sensible way to do this? Thanks Randy Latimer rdlatimer at tjhsst.edu From taube at uiuc.edu Tue Aug 2 11:43:51 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 2 Aug 2005 13:43:51 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1123001821.14784.61.camel@strawberri.maginvent.org> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123001821.14784.61.camel@strawberri.maginvent.org> Message-ID: <07a5707d8087b5f5c9f8ca3213550d72@uiuc.edu> > Also because on sbcl-0.9.0.18-1, I was able to compile and load CMN. > The > make.lisp function saves CMN with CM to be run with the bin/cm.sh > script. Therefore if I am lucky I think I will have CMN, midishare, > portmidi running on this cm image (know on wood). juan - you might consider NOT saving an image until the dust settles with sbcl and real time, im not sure how the lib initialziation will work in saveed images and you will undoubtably have save new images often foor the next week or so. for example, just committed a facility for recording midi data. if its a question of wanting to always starting up with the same software, then create a ~/.cminit.lisp file and add load statements tere: ;; my ~/.cminit.lisp file (load "/path/to/clm-3/all.lisp") (load "/path/to/cmn/cmn-all.lisp") (load "/path/to/fomus/load.lisp") and so on. From taube at uiuc.edu Tue Aug 2 11:58:54 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 2 Aug 2005 13:58:54 -0500 Subject: [CM] Intro Program - 2-part Invention In-Reply-To: <2352.151.188.16.41.1123007475.squirrel@mail.tjhsst.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <2352.151.188.16.41.1123007475.squirrel@mail.tjhsst.edu> Message-ID: <7f5e7abe6d769cb5c049816b7d6aa531@uiuc.edu> > I'm a newcomer to Common Music. I'm using CLisp and WindowsXP. Rick's > Intro programs are running fine and sound good. Hi im glad to hear that, i have almost no chance to boot windows to test. > Also to get started, I'd like to enter in some music - like for > instance > the Bach F major 2-part invention. > Any advice on a sensible way to do this? i would probably use import-events to load data in without typing anything, here is doc http://commonmusic.sf.net/doc/dict/import-events-fn.html also look at the Example 2 of Fomus-file, which imports and notates: http://commonmusic.sf.net/doc/dict/fomus-file-cls.html the 6-page pdf that resulted looked really good. if you are using the cvs version of cm i just added some tutorials yestarday cm/etc/tutorials > Thanks > Randy Latimer rdlatimer at tjhsst.edu > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From juanig at ccrma.Stanford.EDU Tue Aug 2 12:59:20 2005 From: juanig at ccrma.Stanford.EDU (Juan Reyes) Date: Tue, 02 Aug 2005 12:59:20 -0700 Subject: [CM] cm and timidity In-Reply-To: <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> Message-ID: <1123012760.16810.6.camel@strawberri.maginvent.org> Hi Rick, CM and MIDI stuff in Linux have gone a long way. May I suggest the following options for using timidity with ALSA: (setf *midi-player* "timidity -Os") I have found the following switches good on timidity: timidity -Os -A130 -p3 -Ei5 -module=2 -q20 -T140 test.midi -Os Use Alsa -A100 Scale amplitude by factor(100) -p3 3 voice polyphone -Ei5 Sound (patch) 5 -module=2 Roland 88 behaviour -q20 Audio Buffer size of 20/100 default 5/100 -T140 Tempo of 140 bps From juanig at ccrma.Stanford.EDU Tue Aug 2 13:17:45 2005 From: juanig at ccrma.Stanford.EDU (Juan Reyes) Date: Tue, 02 Aug 2005 13:17:45 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <07a5707d8087b5f5c9f8ca3213550d72@uiuc.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123001821.14784.61.camel@strawberri.maginvent.org> <07a5707d8087b5f5c9f8ca3213550d72@uiuc.edu> Message-ID: <1123013865.16810.16.camel@strawberri.maginvent.org> > > ;; my ~/.cminit.lisp file > (load "/path/to/clm-3/all.lisp") > (load "/path/to/cmn/cmn-all.lisp") > (load "/path/to/fomus/load.lisp") > > and so on. > This is a good idea just for now with the exception of CMN (which takes a while loading) with SBCL 0.9.0.x on Linux. XEmacs side and Slime are not so ready now for this SBCL either. Honestly speaking I am curious about the Fomus solution to notation and CMN. Thanks a lot Rick, --* Juan From taube at uiuc.edu Tue Aug 2 14:42:50 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 2 Aug 2005 16:42:50 -0500 Subject: [CM] cm and timidity In-Reply-To: <1123012760.16810.6.camel@strawberri.maginvent.org> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123012760.16810.6.camel@strawberri.maginvent.org> Message-ID: <171059b247380a86c002a6fcfc00893b@uiuc.edu> > I have found the following switches good on timidity: > timidity -Os -A130 -p3 -Ei5 -module=2 -q20 -T140 test.midi gee now why didnt i think of that! to make this the default add this to your .cminit.lisp file: (in-package :cm) (setq *midi-player* "timidity -Os -A130 -p3 -Ei5 -module=2 -q20 -T140") BTW, I would also recommend using Timidity on OS X over qtplay or QuickTime Player. In my opinion the default Timidity sounds better than the default QT sound set and it also seems to have more polyphony available. From nando at ccrma.Stanford.EDU Wed Aug 3 16:46:58 2005 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: 03 Aug 2005 16:46:58 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> Message-ID: <1123112818.16989.424.camel@cmn37.stanford.edu> On Mon, 2005-08-01 at 18:24, Rick Taube wrote: > > debugger invoked on a UNDEFINED-FUNCTION in thread 11105: > > The function SB-POSIX::GETENV is undefined. > > juan -- the sbcl posix support changed somewhere between 0.8.16 (which > ccrma has) and 0.9.0: in 0.8.16 its sb-ext:posix-getenv and in 0.9.0 > its sb-posix:getenv. ive fixed the problem in rev 1.26 of sbcl.lisp (it > will take a while for the changes to migrate to anonymous cvs...). How about 0.9.2? Does if have callbacks? That's the latest test build version[*] I have done at Planet CCRMA (not available yet). Or: how do you know if a given sbcl has callbacks? -- Fernando [*] I just checked and there's a 0.9.3. This is source, not the binary download. From taube at uiuc.edu Wed Aug 3 18:29:35 2005 From: taube at uiuc.edu (Rick Taube) Date: Wed, 3 Aug 2005 20:29:35 -0500 Subject: [CM] midishare revisited Message-ID: Ive reimplemented midishare so that it uses rts and set-receiver! for "real time" work rather than using midishare alarms and callbacks that dont seem to work well anyway. the midishare docs are completely reorganized as well, the single dictionary entry is now http://commonmusic.sf.net/doc/dict/midishare-topic.html I tested the examples in Openmcl only. since the midishare stream now behaves more like portmidi and supercollider i will add an sbcl ffi for it at some point. From taube at uiuc.edu Wed Aug 3 19:15:57 2005 From: taube at uiuc.edu (Rick Taube) Date: Wed, 3 Aug 2005 21:15:57 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1123112818.16989.424.camel@cmn37.stanford.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> Message-ID: <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> > > Or: how do you know if a given sbcl has callbacks? this is a good question. I havent tried to build sbcl from sources and so i dont know if callbacks are in 0.9.2. But one other gotcha -- the thread api seems to have changed a bit between 0.9.0 and 0.9.2. its no big deal in itself but its really impossible for me to try to account for all the differences in the minor releases since so many (good) things are happening in sbcl now. I think maybe cm should hold at the binary release 0.9.0 until there is another binary release that contains their x86 callback support (so gtk can run again) and, of course, their new thread stuff just my opinion. --rick From mlang at delysid.org Wed Aug 3 05:15:37 2005 From: mlang at delysid.org (Mario Lang) Date: Wed, 03 Aug 2005 14:15:37 +0200 Subject: [CM] [PATCH] midi sequencer event import Message-ID: <87wtn3dxna.fsf@lexx.delysid.org> Hi. Since I just tripped over this while trying to IMPORT-EVENTS one of the midi files lying around on my hard disk, here is a patch to make sequencer event import correctly: Index: src/midi3.scm =================================================================== RCS file: /cvsroot/commonmusic/cm/src/midi3.scm,v retrieving revision 1.32 diff -u -r1.32 midi3.scm --- src/midi3.scm 5 Jul 2005 13:08:13 -0000 1.32 +++ src/midi3.scm 3 Aug 2005 12:11:31 -0000 @@ -254,6 +254,10 @@ :time time :offset (loop for i from 1 to 5 collect (vector-ref data i)))) + ((sequencer-event-p m) + (make + :time time + :data data)) ;; FIX! add classes.., ;;((midi-channel-p m) ) (else -- CYa, Mario From taube at uiuc.edu Thu Aug 4 06:09:50 2005 From: taube at uiuc.edu (Rick Taube) Date: Thu, 4 Aug 2005 08:09:50 -0500 Subject: [CM] [PATCH] midi sequencer event import In-Reply-To: <87wtn3dxna.fsf@lexx.delysid.org> References: <87wtn3dxna.fsf@lexx.delysid.org> Message-ID: <02e85a86e98bcb02faeca210786af4d4@uiuc.edu> thanks! its always nice to get a patch ive made the fix in rev 1.33 of midi3.scm On Aug 3, 2005, at 7:15 AM, Mario Lang wrote: > Hi. > > Since I just tripped over this while trying to IMPORT-EVENTS one > of the midi files lying around on my hard disk, here is a patch to make > sequencer event import correctly: > > Index: src/midi3.scm > =================================================================== > RCS file: /cvsroot/commonmusic/cm/src/midi3.scm,v > retrieving revision 1.32 > diff -u -r1.32 midi3.scm > --- src/midi3.scm 5 Jul 2005 13:08:13 -0000 1.32 > +++ src/midi3.scm 3 Aug 2005 12:11:31 -0000 > @@ -254,6 +254,10 @@ > :time time > :offset (loop for i from 1 to 5 > collect (vector-ref data i)))) > + ((sequencer-event-p m) > + (make > + :time time > + :data data)) > ;; FIX! add classes.., > ;;((midi-channel-p m) ) > (else > > -- > CYa, > Mario > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From mlang at delysid.org Thu Aug 4 06:31:51 2005 From: mlang at delysid.org (Mario Lang) Date: Thu, 04 Aug 2005 15:31:51 +0200 Subject: [CM] The Emacs saga continues: Music editing and programming Message-ID: <87y87hakvs.fsf@lexx.delysid.org> Hi. IIRC, I already told of my plans to write a midi.el for Emacs Lisp to handle MIDI file data in Emacs directly on LAD. This idea has been recently supersceeded by something much more powerful, the marriage of CommonMusic and Emacs. SLIME is the key, SLIME makes it possible! I've been recently learning CommonMusic again, and discovered that it already can nicely handle MIDI file data (and much more). So reimplementing the wheel in yet another lisp dialect doesn't seem like a very useful exercise. What I can do, actually, is bring Emacs and CM a little more together, using SLIME. This effectively means that now, I can control SC and CM from Emacs, very easily. Its all a bit of puzzlework for now, but as soon as we/I worked out some ways to make things stick more together, I think this is going to be a very powerful combination. ;; Example Emacs Lisp code: (require 'slime) (defun cm-eval (form &optional callback) "Eval FORM in the CM package. If CALLBACK is a function, pass result to it as soon as evaluation is finished (async). Otherwise, return result." (setq form `(swank:interactive-eval ,(format "%S" form))) (if callback (slime-eval-async form `(lambda (arg) (funcall ,callback (read arg))) "CM") (read (slime-eval form "CM"))))) (defun cm-get-midi-notes (filename) "Return all notes in a midi file in sequencial order." ;; Not very useful in real world, just a demonstration (cm-eval `(note (nreverse (fold-objects 'cons (events (import-events ,(expand-file-name filename)) (new seq)) () :slot 'keynum :class 'midi)))))) ;(cm-get-midi-notes "~/music/Bach_Suite_no4_BWV1006a_Prelude.mid") => ; (E2 E5 EF5 E5 B4 AF4 B4 E4 FS4 E4 EF4 E4 ...) ;(length (cm-get-midi-notes "~/music/Bach_Suite_no4_BWV1006a_Prelude.mid")) => ; 1741 From immanuell at enfocus.be Fri Aug 5 03:06:09 2005 From: immanuell at enfocus.be (Immanuel Litzroth) Date: Fri, 05 Aug 2005 11:06:09 +0100 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> (Rick Taube's message of "Wed, 3 Aug 2005 21:15:57 -0500") References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> Message-ID: <64uk66lq.fsf@immanuelinux.site> "Rick Taube" writes: >> >> Or: how do you know if a given sbcl has callbacks? > > this is a good question. I havent tried to build sbcl from sources and > so i dont know if callbacks are in 0.9.2. sbcl 9.3 has the callback code (in code/target-alieneval.lisp). It does say in one place > ;;;; See "Foreign Linkage / Callbacks" in the SBCL Internals manual. and a little bit further > ;;;; interface (not public, yet) for alien callbacks The manual doesn't mention callbacks. I tried to get it to work yesterday for 30 min. but I gave up. > But one other gotcha -- the > thread api seems to have changed a bit between 0.9.0 and 0.9.2. its no > big deal in itself but its really impossible for me to try to account > for all the differences in the minor releases since so many (good) > things are happening in sbcl now. > > I think maybe cm should hold at the binary release 0.9.0 until there > is another binary release that contains their x86 callback support > (so gtk can run again) and, of course, their new thread stuff > just my opinion. > > --rick > From csr21 at cam.ac.uk Fri Aug 5 03:16:01 2005 From: csr21 at cam.ac.uk (Christophe Rhodes) Date: Fri, 05 Aug 2005 11:16:01 +0100 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <64uk66lq.fsf@immanuelinux.site> (Immanuel Litzroth's message of "Fri, 05 Aug 2005 11:06:09 +0100") References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <64uk66lq.fsf@immanuelinux.site> Message-ID: Immanuel Litzroth writes: > "Rick Taube" writes: > >>> >>> Or: how do you know if a given sbcl has callbacks? >> >> this is a good question. I havent tried to build sbcl from sources and >> so i dont know if callbacks are in 0.9.2. > > sbcl 9.3 has the callback code (in code/target-alieneval.lisp). It > does say in one place >> ;;;; See "Foreign Linkage / Callbacks" in the SBCL Internals manual. > > and a little bit further > > ;;;; interface (not public, yet) for alien callbacks > > The manual doesn't mention callbacks. I tried to get it to work > yesterday for 30 min. but I gave up. As I understand it -- and bear in mind that despite my position in sbcl development, I don't understand it all that well -- the infrastructure for callbacks is present on x86 and ppc/Darwin platforms; there is discussion of the implementation details in the Internals Manual (as distinct from the User Manual), and that there isn't (yet) a finalized public interface interface. The file tests/callback.impure.lisp in the SBCL source distribution demonstrates the current status of the callback implementation (on those platforms on which it is supported at all): there are tests for the usual qsort call, and that floats and doubles get passed correctly from one end to the other. Nikodemus Siivola, who has been driving this development direction, is at present on holiday, so things may not move terribly fast until he returns. Cheers, Christophe From immanuell at enfocus.be Fri Aug 5 03:43:10 2005 From: immanuell at enfocus.be (Immanuel Litzroth) Date: Fri, 05 Aug 2005 11:43:10 +0100 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: (Christophe Rhodes's message of "Fri, 05 Aug 2005 11:16:01 +0100") References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <64uk66lq.fsf@immanuelinux.site> Message-ID: "Christophe Rhodes" writes: > > > Immanuel Litzroth writes: > >> "Rick Taube" writes: >> >>>> >>>> Or: how do you know if a given sbcl has callbacks? >>> >>> this is a good question. I havent tried to build sbcl from sources and >>> so i dont know if callbacks are in 0.9.2. >> >> sbcl 9.3 has the callback code (in code/target-alieneval.lisp). It >> does say in one place >>> ;;;; See "Foreign Linkage / Callbacks" in the SBCL Internals manual. >> >> and a little bit further >> > ;;;; interface (not public, yet) for alien callbacks >> >> The manual doesn't mention callbacks. I tried to get it to work >> yesterday for 30 min. but I gave up. > > As I understand it -- and bear in mind that despite my position in > sbcl development, I don't understand it all that well -- the > infrastructure for callbacks is present on x86 and ppc/Darwin > platforms; there is discussion of the implementation details in the > Internals Manual (as distinct from the User Manual), and that there > isn't (yet) a finalized public interface interface. > > The file tests/callback.impure.lisp in the SBCL source distribution > demonstrates the current status of the callback implementation (on > those platforms on which it is supported at all): there are tests for > the usual qsort call, and that floats and doubles get passed correctly >>From one end to the other. > > Nikodemus Siivola, who has been driving this development direction, is > at present on holiday, so things may not move terribly fast until he > returns. > > > > Cheers, > > Christophe Any idea where I can get a version of this internals manual? I've been doing a lot of sbcl development lately, and I would like to take a look at this. Thanks Immanuel From pkw at lupulin.net Fri Aug 12 19:03:12 2005 From: pkw at lupulin.net (paul wisehart) Date: Fri, 12 Aug 2005 22:03:12 -0400 Subject: [CM] problem with some basic CM functions Message-ID: <20050813020312.GA27725@onion.blobot.mine.nu> I think I'm missing something basic in my setup of cm. I'm using cmucl 19b, and emacs/slime fro the interface. Hers's an exmaple of me loading cm, and trying to play a midi file. How do I get the play function to be defined? This is happening on some other cm functions. But some of the stuff works. For ex. I can "(new midi)" , and it will return successfully. any ideas? . thanks PS oh yeah I installed cm with asdf-install if that could casue any funniness. ; SLIME 2005-08-12 CL-USER> (load "/usr/local/asdf-install/site/cm-2.6.0/src/cm") ; Loading #P"/usr/local/asdf-install/site/cm-2.6.0/src/cm.x86f". ; Installation directory: "/usr/local/asdf-install/site/cm-2.6.0/" ; Loading "bin/cmucl_19b_linux-i686/gtkffi-cmusbcl.x86f" ; Loading "bin/cmucl_19b_linux-i686/clm-stubs.x86f" ; Loading "bin/cmucl_19b_linux-i686/cmn-stubs.x86f" ; Loading "bin/cmucl_19b_linux-i686/midishare-stubs.x86f" ; Loading "bin/cmucl_19b_linux-i686/pkg.x86f" ; Loading "bin/cmucl_19b_linux-i686/cmu.x86f" ; Loading "bin/cmucl_19b_linux-i686/clos.x86f" ; Loading "bin/cmucl_19b_linux-i686/iter.x86f" ; Loading "bin/cmucl_19b_linux-i686/level1.x86f" ; Loading "bin/cmucl_19b_linux-i686/scheme.x86f" ; Loading "bin/cmucl_19b_linux-i686/utils.x86f" ; Loading "bin/cmucl_19b_linux-i686/mop.x86f" ; Loading "bin/cmucl_19b_linux-i686/objects.x86f" ; Loading "bin/cmucl_19b_linux-i686/data.x86f" ; Loading "bin/cmucl_19b_linux-i686/scales.x86f" ; Loading "bin/cmucl_19b_linux-i686/spectral.x86f" ; Loading "bin/cmucl_19b_linux-i686/patterns.x86f" ; Loading "bin/cmucl_19b_linux-i686/io.x86f" ; Loading "bin/cmucl_19b_linux-i686/scheduler.x86f" ; Loading "bin/cmucl_19b_linux-i686/sco.x86f" ; Loading "bin/cmucl_19b_linux-i686/clm.x86f" ; Loading "bin/cmucl_19b_linux-i686/clm2.x86f" ; Loading "bin/cmucl_19b_linux-i686/midi1.x86f" ; Loading "bin/cmucl_19b_linux-i686/midi2.x86f" ; Loading "bin/cmucl_19b_linux-i686/midi3.x86f" ; Loading "bin/cmucl_19b_linux-i686/midishare.x86f" ; Loading "bin/cmucl_19b_linux-i686/player.x86f" ; Loading "bin/cmucl_19b_linux-i686/cmn.x86f" ; Loading "bin/cmucl_19b_linux-i686/plotter.x86f" ; Loading "bin/cmucl_19b_linux-i686/support.x86f" ; Loading "bin/cmucl_19b_linux-i686/widgets.x86f" ; Loading "bin/cmucl_19b_linux-i686/editing.x86f" ; Loading "bin/cmucl_19b_linux-i686/drawing.x86f" ; Loading "bin/cmucl_19b_linux-i686/eventio.x86f" ; Garbage collecting. T CL-USER> (play "/home/wise/killing.mid") ; ; Warning: This function is undefined: ; PLAY ; ; No value CL-USER> ; No value CL-USER> -- paul \ / wisehart >/ |\|\|\ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From taube at uiuc.edu Sat Aug 13 05:29:38 2005 From: taube at uiuc.edu (Rick Taube) Date: Sat, 13 Aug 2005 07:29:38 -0500 Subject: [CM] problem with some basic CM functions In-Reply-To: <20050813020312.GA27725@onion.blobot.mine.nu> References: <20050813020312.GA27725@onion.blobot.mine.nu> Message-ID: <422e02de0d085aeecf5ab4e00540d483@uiuc.edu> > CL-USER> (play "/home/wise/killing.mid") Its easiest if you are in the CM package when you work. Do this: CL-USER> (load "/usr/local/asdf-install/site/cm-2.6.0/src/cm") CL-USER> (cm) /\\\ ---\\\--------- ----\\\-------- ----/\\\------- Common Music 2.7.0 ---/--\\\------ --/----\\\----- / \\\/ CM> (play "killing.mid") --- alternately you can start up using cm.sh $ cm/bin/cm.sh -e emacs it should start cm under emacs using slime if your .emacs loads slime. From taube at uiuc.edu Sat Aug 13 05:53:13 2005 From: taube at uiuc.edu (Rick Taube) Date: Sat, 13 Aug 2005 07:53:13 -0500 Subject: [CM] problem with some basic CM functions In-Reply-To: <20050813020312.GA27725@onion.blobot.mine.nu> References: <20050813020312.GA27725@onion.blobot.mine.nu> Message-ID: actually as i think about it im not sure if 2.6.0 even has 'play', or if it works if it is there. you might consider swithing to cm-2.7.0 in cvs head, that 'play' is documented and should work. if you stick with 2.6.0 and play doesn work you can always do something like: (shell "timidity killing.mid") On Aug 12, 2005, at 9:03 PM, paul wisehart wrote: > > I think I'm missing something basic in my setup of cm. > I'm using cmucl 19b, and emacs/slime fro the interface. From k.s.matheussen at notam02.no Sat Aug 13 08:03:06 2005 From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen) Date: Sat, 13 Aug 2005 17:03:06 +0200 (CEST) Subject: [CM] [ANN] snd-ls-0.9.5.0_beta Message-ID: At the usual place: http://www.notam02.no/arkiv/src/snd/ I haven't tested it very much yet, so I mark it as beta. 0.9.4.3 -> 0.9.5.0_beta -Upgraded snd to V7.15 from 12.8.2005. Many important changes. -Various other things. -- From nando at ccrma.Stanford.EDU Mon Aug 15 18:07:31 2005 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: 15 Aug 2005 18:07:31 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> Message-ID: <1124154451.8013.69.camel@cmn37.stanford.edu> On Wed, 2005-08-03 at 19:15, Rick Taube wrote: > > > > Or: how do you know if a given sbcl has callbacks? > > this is a good question. I havent tried to build sbcl from sources and > so i dont know if callbacks are in 0.9.2. But one other gotcha -- the > thread api seems to have changed a bit between 0.9.0 and 0.9.2. its no > big deal in itself but its really impossible for me to try to account > for all the differences in the minor releases since so many (good) > things are happening in sbcl now. > > I think maybe cm should hold at the binary release 0.9.0 until there is > another binary release that contains their x86 callback support (so > gtk can run again) and, of course, their new thread stuff > just my opinion. Well, I'm building from source and 0.9.0 won't build out of the box with fc4/gcc4... I guess I'll have to somehow use 0.9.3 for now. -- Fernando From taube at uiuc.edu Mon Aug 15 19:21:25 2005 From: taube at uiuc.edu (Rick Taube) Date: Mon, 15 Aug 2005 21:21:25 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1124154451.8013.69.camel@cmn37.stanford.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> Message-ID: <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> ill be going into my office later this week and will try to build cm with 0.9.3. Im not sure what surprises it will hold other than I know the real time thread code will have to be changed slightly to account for changes in their thread api since 0.9.0. i dont know if callbacks are supposed to be working or not in 0.9.3, but probably not since i dont see it listed in their news page. if not then perhaps thomas burdicks callback code could be made to run in 0.9.3 but i wont have time for this until mid sept at the earliest. portmidi From bil at ccrma.Stanford.EDU Tue Aug 16 04:30:07 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 16 Aug 2005 04:30:07 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> Message-ID: <4301CE3F.50600@ccrma> > ill be going into my office later this week and will try to build cm with 0.9.3. > Im not sure what surprises it will hold I haven't gotten 0.9.3 to run except in its build directory -- both the sources and rpm version die with some incomprehensible error otherwise. 0.9.2 works fine, so I doubt it's an environment problem, unless that too has changed without warning. From znmeb at cesmail.net Tue Aug 16 06:41:22 2005 From: znmeb at cesmail.net (M. Edward (Ed) Borasky) Date: Tue, 16 Aug 2005 06:41:22 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4301CE3F.50600@ccrma> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> Message-ID: <4301ED02.7010302@cesmail.net> I gave up on the SBCL/CM combo after Gentoo deleted the callback patch. Last I heard on the SBCL mailing list they were still arguing about the "right way" to do callbacks. CM/CMUCL 19a works just fine on all my Linux systems. I'm not sure where that leaves Windows folks, though. Will SBCL 1.0 run on Windows? Will there be a native installer? Bill Schottstaedt wrote: > > ill be going into my office later this week and will try to build cm > with 0.9.3. > > Im not sure what surprises it will hold > > I haven't gotten 0.9.3 to run except in its build directory -- both the > sources and rpm version die with some incomprehensible error otherwise. > 0.9.2 works fine, so I doubt it's an environment problem, unless that > too has changed without warning. > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From taube at uiuc.edu Tue Aug 16 07:41:11 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 16 Aug 2005 09:41:11 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4301ED02.7010302@cesmail.net> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <4301ED02.7010302@cesmail.net> Message-ID: yes, sbcl certainly seems to be in an "in between" state right now as they (apparently) add in callback and threading support. cmucl 19b will be fine except that i dont know how useful the "realtime" stuff will be there: cmucl only has lisp threads so the rts scheduler will use a 1ms periodic task and will have to share the slice with receiving. it might work ok for small amounts of data. > I'm not sure where that leaves Windows folks, though. clisp and acl seem be what people use on windows. the LispWorks "personal edition" does have native threads and I ported cm to it sevearl months ago but I dont think anyone is using it. From nando at ccrma.Stanford.EDU Tue Aug 16 13:14:12 2005 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: 16 Aug 2005 13:14:12 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4301CE3F.50600@ccrma> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> Message-ID: <1124223251.21951.240.camel@cmn37.stanford.edu> On Tue, 2005-08-16 at 04:30, Bill Schottstaedt wrote: > > ill be going into my office later this week and will try to build cm with 0.9.3. > > Im not sure what surprises it will hold > > I haven't gotten 0.9.3 to run except in its build directory -- both the > sources and rpm version die with some incomprehensible error otherwise. > 0.9.2 works fine, so I doubt it's an environment problem, unless that > too has changed without warning. I have 0.9.3 running, packaged from source for FC4, on which distro were you trying? "Running" means CLM (3) compiles, builds, installs and the comprehensive test suit - that is, (with-sound()(fm-violin 0 1 440 0.1) - works fine :-) -- Fernando From bil at ccrma.Stanford.EDU Wed Aug 17 03:57:31 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 17 Aug 2005 03:57:31 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1124223251.21951.240.camel@cmn37.stanford.edu> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> Message-ID: <4303181B.5000209@ccrma> > I have 0.9.3 running, packaged from source for FC4, on which distro were > you trying? FC4 -- I guess the answer is to keep trying -- here I am typing in a shell, with interspersed unprintable commentary: /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core This is SBCL 0.9.3, an implementation of ANSI Common Lisp. More information about SBCL is available at . SBCL is free software, provided as is, with absolutely no warranty. It is mostly in the public domain; some portions are provided under BSD-style licenses. See the CREDITS and COPYING files in the distribution for more information. debugger invoked on a SIMPLE-ERROR in thread #: Invalid external-format internal error #29 SC: 14, Offset: 4 lispobj 0x50ba5ff fatal error encountered in SBCL pid 19526(tid 1075396352): internal error too early in init, can't recover The system is too badly corrupted or confused to continue at the Lisp level. If the system had been compiled with the SB-LDB feature, we'd drop into the LDB low-level debugger now. But there's no LDB in this build, so we can't really do anything but just exit, sorry. /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core This is SBCL 0.9.3, an implementation of ANSI Common Lisp. More information about SBCL is available at . SBCL is free software, provided as is, with absolutely no warranty. It is mostly in the public domain; some portions are provided under BSD-style licenses. See the CREDITS and COPYING files in the distribution for more information. debugger invoked on a SIMPLE-ERROR in thread #: Invalid external-format internal error #29 SC: 14, Offset: 4 lispobj 0x50ba5ff fatal error encountered in SBCL pid 19530(tid 1075396352): internal error too early in init, can't recover The system is too badly corrupted or confused to continue at the Lisp level. If the system had been compiled with the SB-LDB feature, we'd drop into the LDB low-level debugger now. But there's no LDB in this build, so we can't really do anything but just exit, sorry. /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core This is SBCL 0.9.3, an implementation of ANSI Common Lisp. More information about SBCL is available at . SBCL is free software, provided as is, with absolutely no warranty. It is mostly in the public domain; some portions are provided under BSD-style licenses. See the CREDITS and COPYING files in the distribution for more information. * (load "cmn-all.lisp") Ta Da! I'm glad I don't depend on sbcl for anything. From immanuell at enfocus.be Wed Aug 17 05:05:29 2005 From: immanuell at enfocus.be (Immanuel Litzroth) Date: Wed, 17 Aug 2005 13:05:29 +0100 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4303181B.5000209@ccrma> (Bill Schottstaedt's message of "Wed, 17 Aug 2005 03:57:31 -0700") References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> <4303181B.5000209@ccrma> Message-ID: "Bill Schottstaedt" writes: > FC4 -- I guess the answer is to keep trying -- here I am > typing in a shell, with interspersed unprintable commentary: If your feel adventurous you should be able to debug your problem with gdb, see http://sbcl-internals.cliki.net/gdb My personal experience with 0.9.3 is also not so good but I have the impression that the current cvs is better (maybe they fixed some things). Immanuel From znmeb at cesmail.net Wed Aug 17 06:41:12 2005 From: znmeb at cesmail.net (M. Edward (Ed) Borasky) Date: Wed, 17 Aug 2005 06:41:12 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4303181B.5000209@ccrma> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> <4303181B.5000209@ccrma> Message-ID: <43033E78.5090604@cesmail.net> Bill, Fernando, Rick -- are you on the SBCL list? I dropped off before I went on vacation and haven't gone back yet. I'm pretty sure they care as much about being able to run the CCRMA software as they do about, say, MAXIMA. Algorithmic composition is an important application in the Lisp world. It's not just CCRMA -- David Cope and a few others also choose to work in Lisp. Bill Schottstaedt wrote: > > I have 0.9.3 running, packaged from source for FC4, on which distro > were > > you trying? > > FC4 -- I guess the answer is to keep trying -- here I am > typing in a shell, with interspersed unprintable commentary: > > > > /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl > --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core > This is SBCL 0.9.3, an implementation of ANSI Common Lisp. > More information about SBCL is available at . > > SBCL is free software, provided as is, with absolutely no warranty. > It is mostly in the public domain; some portions are provided under > BSD-style licenses. See the CREDITS and COPYING files in the > distribution for more information. > > debugger invoked on a SIMPLE-ERROR in thread # {90043E9}>: > Invalid external-format > internal error #29 > SC: 14, Offset: 4 lispobj 0x50ba5ff > fatal error encountered in SBCL pid 19526(tid 1075396352): > internal error too early in init, can't recover > The system is too badly corrupted or confused to continue at the Lisp > level. If the system had been compiled with the SB-LDB feature, we'd drop > into the LDB low-level debugger now. But there's no LDB in this build, so > we can't really do anything but just exit, sorry. > > /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl > --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core > This is SBCL 0.9.3, an implementation of ANSI Common Lisp. > More information about SBCL is available at . > > SBCL is free software, provided as is, with absolutely no warranty. > It is mostly in the public domain; some portions are provided under > BSD-style licenses. See the CREDITS and COPYING files in the > distribution for more information. > > debugger invoked on a SIMPLE-ERROR in thread # {90043E9}>: > Invalid external-format > internal error #29 > SC: 14, Offset: 4 lispobj 0x50ba5ff > fatal error encountered in SBCL pid 19530(tid 1075396352): > internal error too early in init, can't recover > The system is too badly corrupted or confused to continue at the Lisp > level. If the system had been compiled with the SB-LDB feature, we'd drop > into the LDB low-level debugger now. But there's no LDB in this build, so > we can't really do anything but just exit, sorry. > > /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl > --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core > This is SBCL 0.9.3, an implementation of ANSI Common Lisp. > More information about SBCL is available at . > > SBCL is free software, provided as is, with absolutely no warranty. > It is mostly in the public domain; some portions are provided under > BSD-style licenses. See the CREDITS and COPYING files in the > distribution for more information. > * (load "cmn-all.lisp") > > > Ta Da! I'm glad I don't depend on sbcl for anything. > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From taube at uiuc.edu Wed Aug 17 08:25:28 2005 From: taube at uiuc.edu (Rick Taube) Date: Wed, 17 Aug 2005 10:25:28 -0500 Subject: [CM] Re: cannot run cm script on Linux SBCL Message-ID: <299b94c291def3377d8df84b3c6c9061@uiuc.edu> > FC4 -- I guess the answer is to keep trying -- here I am > typing in a shell, with interspersed unprintable commentary: I wonder if you are being bitten by a memory randomization issue in fc4, see below (there seems to be lots of postings about problems in 0.9.3) --- From: jsnell at iki.fi Subject: [Sbcl-devel] Re: problem building 0.9.3 under Fedora Core 4 using CMUCL 19b Date: August 17, 2005 10:14:00 AM CDT To: patrick.may at intamission.com Cc: sbcl-devel at lists.sourceforge.net On Wed, Aug 17, 2005 at 12:09:45AM +0100, Patrick May wrote: > Near as I can tell, I've got some bad craziness related to threading > going on. Can anyone lend me a clue? Are you compiling with gcc 4? There are some problems with threading and gcc4 (possibly solved in HEAD). Using gcc 3 instead might help. > *** glibc detected *** sbcl: malloc(): memory corruption: 0x096ee058 > *** The FC4 kernel has some patches for memory map randomization, which (unlike the earlier versions) is automatically turned on for all applications. For full details see the "Memory randomization problems coming" thread on sbcl-devel. For now, either of the following solutions should work: Disable randomization globally "echo 0 > /proc/sys/kernel/randomize_va_space" Start sbcl with "setarch i386 -R sbcl" -- Juho Snellman ------------------------------------------------------- SF.Net email is Sponsored by the Better Software Conference & EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf _______________________________________________ Sbcl-devel mailing list Sbcl-devel at lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sbcl-devel From nando at ccrma.Stanford.EDU Wed Aug 17 11:00:19 2005 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: 17 Aug 2005 11:00:19 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <4303181B.5000209@ccrma> References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> <4303181B.5000209@ccrma> Message-ID: <1124301618.2339.146.camel@cmn37.stanford.edu> On Wed, 2005-08-17 at 03:57, Bill Schottstaedt wrote: > > I have 0.9.3 running, packaged from source for FC4, on which distro were > > you trying? > > FC4 -- I guess the answer is to keep trying -- here I am > typing in a shell, with interspersed unprintable commentary: Yuck. Works for me. I just released sbcl 0.9.3 and cmucl 2005.08 (from the binary snapshot - no use trying to build cmucl :-) packages for Fedora Core 4 on Planet CCRMA, you should be able to apt-get them (also included in the release is clm 2004.08.14, cmn 2005.07.12 and slime 1.2.1 for both sbcl and cmucl, all seem to be working fine so far). Let me know if the lisp packages work in your setup. -- Fernando > /home/bil/cmn/ /home/bil/test/sbcl-0.9.3-x86-linux/src/runtime/sbcl --core /home/bil/test/sbcl-0.9.3-x86-linux/output/sbcl.core > This is SBCL 0.9.3, an implementation of ANSI Common Lisp. > More information about SBCL is available at . > > SBCL is free software, provided as is, with absolutely no warranty. > It is mostly in the public domain; some portions are provided under > BSD-style licenses. See the CREDITS and COPYING files in the > distribution for more information. > > debugger invoked on a SIMPLE-ERROR in thread #: > Invalid external-format > internal error #29 > SC: 14, Offset: 4 lispobj 0x50ba5ff > fatal error encountered in SBCL pid 19526(tid 1075396352): > internal error too early in init, can't recover > The system is too badly corrupted or confused to continue at the Lisp > level. If the system had been compiled with the SB-LDB feature, we'd drop > into the LDB low-level debugger now. But there's no LDB in this build, so > we can't really do anything but just exit, sorry. From nando at ccrma.Stanford.EDU Wed Aug 17 11:04:55 2005 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: 17 Aug 2005 11:04:55 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <299b94c291def3377d8df84b3c6c9061@uiuc.edu> References: <299b94c291def3377d8df84b3c6c9061@uiuc.edu> Message-ID: <1124301895.2339.156.camel@cmn37.stanford.edu> On Wed, 2005-08-17 at 08:25, Rick Taube wrote: > > FC4 -- I guess the answer is to keep trying -- here I am > > typing in a shell, with interspersed unprintable commentary: > > I wonder if you are being bitten by a memory randomization issue in > fc4, see below (there seems to be lots of postings about problems in > 0.9.3) > > --- > From: jsnell at iki.fi > Subject: [Sbcl-devel] Re: problem building 0.9.3 under Fedora Core 4 > using CMUCL 19b I'm building using the 0.9.0 binary image. > Date: August 17, 2005 10:14:00 AM CDT > To: patrick.may at intamission.com > Cc: sbcl-devel at lists.sourceforge.net > > On Wed, Aug 17, 2005 at 12:09:45AM +0100, Patrick May wrote: > > Near as I can tell, I've got some bad craziness related to threading > > going on. Can anyone lend me a clue? > > Are you compiling with gcc 4? There are some problems with threading > and gcc4 (possibly solved in HEAD). Using gcc 3 instead might help. > > > *** glibc detected *** sbcl: malloc(): memory corruption: 0x096ee058 > > *** > > The FC4 kernel has some patches for memory map randomization, which > (unlike the earlier versions) is automatically turned on for all > applications. For full details see the "Memory randomization problems > coming" thread on sbcl-devel. For now, either of the following solutions > should work: > > Disable randomization globally "echo 0 > > /proc/sys/kernel/randomize_va_space" > Start sbcl with "setarch i386 -R sbcl" Hmmm, interesting... I was running on the Planet CCRMA low latency kernel, probably does not have those patches. I should try to run on the "normal" Fedora kernel and see what happens. Sigh, I will probably have to release another version with the setarch workaround so it works on all kernels. -- Fernando From nando at ccrma.Stanford.EDU Wed Aug 17 11:14:10 2005 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: 17 Aug 2005 11:14:10 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <1124301895.2339.156.camel@cmn37.stanford.edu> References: <299b94c291def3377d8df84b3c6c9061@uiuc.edu> <1124301895.2339.156.camel@cmn37.stanford.edu> Message-ID: <1124302449.2339.169.camel@cmn37.stanford.edu> On Wed, 2005-08-17 at 11:04, Fernando Lopez-Lezcano wrote: > On Wed, 2005-08-17 at 08:25, Rick Taube wrote: > > > FC4 -- I guess the answer is to keep trying -- here I am > > > typing in a shell, with interspersed unprintable commentary: > > > > I wonder if you are being bitten by a memory randomization issue in > > fc4, see below (there seems to be lots of postings about problems in > > 0.9.3) > > > > --- > > From: jsnell at iki.fi > > Subject: [Sbcl-devel] Re: problem building 0.9.3 under Fedora Core 4 > > using CMUCL 19b > > I'm building using the 0.9.0 binary image. (and with gcc4) > > Date: August 17, 2005 10:14:00 AM CDT > > To: patrick.may at intamission.com > > Cc: sbcl-devel at lists.sourceforge.net > > > > On Wed, Aug 17, 2005 at 12:09:45AM +0100, Patrick May wrote: > > > Near as I can tell, I've got some bad craziness related to threading > > > going on. Can anyone lend me a clue? > > > > Are you compiling with gcc 4? There are some problems with threading > > and gcc4 (possibly solved in HEAD). Using gcc 3 instead might help. > > > > > *** glibc detected *** sbcl: malloc(): memory corruption: 0x096ee058 > > > *** > > > > The FC4 kernel has some patches for memory map randomization, which > > (unlike the earlier versions) is automatically turned on for all > > applications. For full details see the "Memory randomization problems > > coming" thread on sbcl-devel. For now, either of the following solutions > > should work: > > > > Disable randomization globally "echo 0 > > > /proc/sys/kernel/randomize_va_space" > > Start sbcl with "setarch i386 -R sbcl" > > Hmmm, interesting... I was running on the Planet CCRMA low latency > kernel, probably does not have those patches. Seems like it does have that feature (running 2.6.12-0.21.rdt on fc4), /proc/sys/kernel/randomize_va_space exists and is set to "1". Must be some other issue then. -- Fernando From k.s.matheussen at notam02.no Wed Aug 17 14:06:42 2005 From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen) Date: Wed, 17 Aug 2005 23:06:42 +0200 (CEST) Subject: [CM] [ANN] snd-ls-0.9.5.1 Message-ID: Download from ************* http://www.notam02.no/arkiv/src/snd/ About ***** Snd-ls is a distribution of the sound editor Snd. Its target is people that don't know scheme very well, and don't want to spend too much time configuring Snd. It can also serve as a quick introduction to Snd and how it can be set up. Changes ******* 0.9.4.0 -> 0.9.5.1 ------------------ -Upgraded SND to V7.15 from 17.8.2005. Many important changes. -Various improvements in the user-interface. -Upgraded snd to V7.15 from 12.8.2005. Many important changes. -Various other things. -Upgraded various rt-stuff. -Upgraded various rt-stuff++ -Removed jack_set_server_dir guile-binding from rt-engine.scm, because its removed from the newer versions of jack. -- From bil at ccrma.Stanford.EDU Thu Aug 18 04:12:38 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 18 Aug 2005 04:12:38 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> <4303181B.5000209@ccrma> Message-ID: <43046D26.2040909@ccrma> > If your feel adventurous you should be able to debug your problem > with gdb, see > http://sbcl-internals.cliki.net/gdb Ok: /home/bil/cmn/ sbcl [... delete blathering ...] debugger invoked on a SIMPLE-ERROR in thread #: Invalid external-format internal error #29 SC: 14, Offset: 4 lispobj 0x50ba5ff fatal error encountered in SBCL pid 3454(tid 1075396352): internal error too early in init, can't recover The system is too badly corrupted or confused to continue at the Lisp level. If the system had been compiled with the SB-LDB feature, we'd drop into the LDB low-level debugger now. But there's no LDB in this build, so we can't really do anything but just exit, sorry. error 29 is apparently "unseen throw tag error" /home/bil/cmn/ gdb /home/bil/test/sbcl-0.9.3/src/runtime/sbcl Using host libthread_db library "/lib/libthread_db.so.1". (gdb) break call_into_c Breakpoint 1 at 0x805aa30: file x86-assem.S, line 77. (gdb) run --core /home/bil/test/sbcl-0.9.3/output/cold-sbcl.core Starting program: /home/bil/test/sbcl-0.9.3/src/runtime/sbcl --core /home/bil/test/sbcl-0.9.3/output/cold-sbcl.core Reading symbols from shared object read from target memory...done. Loaded system supplied DSO at 0x4001c000 [...] Program received signal SIGSEGV, Segmentation fault. 0x0a0d8d8f in ?? () (gdb) (gdb) where #0 0x0a0d8d8f in ?? () #1 0x00000018 in ?? () #2 0x0943a24f in ?? () #3 0x00000148 in ?? () #4 0x0000004f in ?? () #5 0x0de90a17 in ?? () #6 0x0a230c77 in ?? () #7 0x00000008 in ?? () #8 0x0a230c77 in ?? () #9 0x0a031855 in ?? () #10 0x405b7eb8 in ?? () #11 0x405b8028 in ?? () #12 0x00000000 in ?? () (gdb) x/8x 0x50ba5f0 0x50ba5f0: 0x00000000 0x00000000 0x00000000 0x00000000 0x50ba600: 0x00000000 0x00000000 0x00000000 0x00000000 A dead-end, at least for me. I looked into including :sb-ldb from /base-target-features.lisp-expr, but I don't have a working sbcl, so I can't (easily) rebuild it. And I can't see what good it would do -- I know nothing about cmu's ldb or sbcl internals. From csr21 at cam.ac.uk Wed Aug 17 08:06:43 2005 From: csr21 at cam.ac.uk (Christophe Rhodes) Date: Wed, 17 Aug 2005 16:06:43 +0100 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: <43033E78.5090604@cesmail.net> (M. Edward Borasky's message of "Wed, 17 Aug 2005 06:41:12 -0700") References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> <4303181B.5000209@ccrma> <43033E78.5090604@cesmail.net> Message-ID: "M. Edward (Ed) Borasky" writes: > Bill, Fernando, Rick -- are you on the SBCL list? I dropped off before I > went on vacation and haven't gone back yet. I'm pretty sure they care as > much about being able to run the CCRMA software as they do about, say, > MAXIMA. Algorithmic composition is an important application in the Lisp > world. It's not just CCRMA -- David Cope and a few others also choose to > work in Lisp. We care, in as much as a bunch of volunteers can care about anything outside what counts as "fun"... There is a bug in sbcl 0.9.3 such that attempts to run in an "invalid" locale (more on which later) don't recover gracefully but instead die horribly with an uninformative error message. It may be that that's the cause of your problems; if it is, running sbcl with e.g. $ LANG=C sbcl stands a chance of fixing it. (Also, the version in HEAD CVS might fix it). "Invalid" here doesn't mean anything as straigtforward as invalid, though -- it means those for which nl_langinfo(CODESET) returns a string describing the character encoding which we haven't yet implemented support for (at present, we have support for ASCII, Latin-1, UTF-8, Latin-9, a few Russian encodings). If a particular codeset is important to you, we can at least point you to how to implement support for it, and possibly implement it ourselves. In general, we SBCL types do want to make using it not horrendously painful; this is balanced at the moment by the sub-1.0 version number -- we don't yet commit to stability of interfaces. In particular, I believe this has caused pain from the callback side of things; all I can suggest there (because that kind of thing is well outside my personal expertise) is that you (the cmdist people) talk to the sbcl people about what you need -- because it's a lot easier to get motivated about implementing something tricky if you know someone will use it (and ideally will give feedback over whether it's good or not :-) (It's many years since I worked with CM or satellite technologies -- I used CMN a little at one point -- but I am still at least partly a musician, as well as a lisp programmer, and I try to welcome users of the bits of software I'm responsible for. It is true to say that I don't have a large amount of time to devote purely to making SBCL better for users, and nor does anyone I know of on the SBCL development team; I can quite understand frustrations when things go wrong, and can only apologise. Sorry.) Cheers, Christophe From bil at ccrma.Stanford.EDU Sat Aug 20 04:38:47 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 20 Aug 2005 04:38:47 -0700 Subject: [CM] Re: cannot run cm script on Linux SBCL In-Reply-To: References: <1122935105.11196.11.camel@strawberri.maginvent.org> <4a3b9fb84e604f275285aebaf7f02c97@uiuc.edu> <1123112818.16989.424.camel@cmn37.stanford.edu> <3fdf0f52b4363b8d66a7a650fb6e10f1@uiuc.edu> <1124154451.8013.69.camel@cmn37.stanford.edu> <186dbe8988272409dc221b4eff5c8a12@uiuc.edu> <4301CE3F.50600@ccrma> <1124223251.21951.240.camel@cmn37.stanford.edu> <4303181B.5000209@ccrma> <43033E78.5090604@cesmail.net> Message-ID: <43071647.2080303@ccrma> > $ LANG=C sbcl > stands a chance of fixing it Yes! sbcl works now -- I was using en_US (my vague recollection is that this locale fixed some font problem in Redhat 7). From taube at uiuc.edu Sat Aug 20 09:02:35 2005 From: taube at uiuc.edu (Rick Taube) Date: Sat, 20 Aug 2005 11:02:35 -0500 Subject: [CM] CM in SBCL 0.9.3 and CLISP 2.34 Message-ID: CM's "realtime" scheduler seems to working in linux sbcl 0.9.3 now. I tested all the rt examples in the dictionary with portmidi, rts and set-receiver! and didnt encounter any problems. ditto the test suite. I also upgraded CM's clisp support to Clisp 2.34. This versin of clisp has better clos support (metaclasses, slot description api). rather than try to support all the other versions cm will now signal an error if you try to build it in a clisp eariler than 2.34. clisp has a real ffi now, callbacks, and i even saw the beginning of thread support in their ./configure script! clisp-2.34 hkt$ ./configure --help [...] --with-threads=FLAVOR support multiple threads in one CLISP image via OS threads [highly experimental - use at your own risk] FLAVORs = POSIX_THREADS POSIXOLD_THREADS SOLARIS_THREADS C_THREADS WIN32_THREADS From dlphillips at woh.rr.com Sun Aug 21 11:42:54 2005 From: dlphillips at woh.rr.com (Dave Phillips) Date: Sun, 21 Aug 2005 14:42:54 -0400 Subject: [CM] CM in SBCL 0.9.3 and CLISP 2.34 In-Reply-To: References: Message-ID: <4308CB2E.4080803@woh.rr.com> Hi Rick CM does appear to be working well in SBCL 0.9.3 under a Debian system (Demudi 1.2.1). However: WARNING: Cannot build GUI because GTK2 interface will not build in SBCL 0.9.3. :( Will that be possible soon ? Or am I missing something else ? Also, does any version of CMUCL support the whole CM enchilada (realtime, PortMIDI, MidiShare, GUI) ? Best, dp Rick Taube wrote: > CM's "realtime" scheduler seems to working in linux sbcl 0.9.3 now. I > tested all the rt examples in the dictionary with portmidi, rts and > set-receiver! and didnt encounter any problems. ditto the test suite. > > I also upgraded CM's clisp support to Clisp 2.34. This versin of clisp > has better clos support (metaclasses, slot description api). rather > than try to support all the other versions cm will now signal an error > if you try to build it in a clisp eariler than 2.34. clisp has a real > ffi now, callbacks, and i even saw the beginning of thread support in > their ./configure script! > > clisp-2.34 hkt$ ./configure --help > [...] > --with-threads=FLAVOR support multiple threads in one > CLISP image > via OS threads [highly experimental - use at your own > risk] > FLAVORs = POSIX_THREADS POSIXOLD_THREADS SOLARIS_THREADS > C_THREADS WIN32_THREADS > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From taube at uiuc.edu Sun Aug 21 12:35:58 2005 From: taube at uiuc.edu (Rick Taube) Date: Sun, 21 Aug 2005 14:35:58 -0500 Subject: [CM] CM in SBCL 0.9.3 and CLISP 2.34 In-Reply-To: <4308CB2E.4080803@woh.rr.com> References: <4308CB2E.4080803@woh.rr.com> Message-ID: <37114532ab5af071383581b652045ad6@uiuc.edu> > CM does appear to be working well in SBCL 0.9.3 under a Debian system > (Demudi 1.2.1). However: > WARNING: Cannot build GUI because GTK2 interface will not build in > SBCL 0.9.3. > :( Will that be possible soon ? Or am I missing something else ? i dont know, its out of my hands. i saw on the sbcl list yesterday that there is a 0.9.4 release planned for the end of this month but i dont know if callbacks will be part of it or not. thats all im waiting for i think. maybe ill send a note to the sbcl list and find out. > Also, does any version of CMUCL support the whole CM enchilada > (realtime, PortMIDI, MidiShare, GUI) ? used individually they should all work now, in one form or another. the main problem with cmucl is that there are no native threads -- at least that i can find -- which means that everything has to be done as a periodic task: the GTK event loop, rts scheduling and midi input receiving. as far as i can tell there is only one "global" periodic task manager so all this servicing will have to be done in the same loop. i can easily add this (the code is actually already there) but i dont think the results will be good. it also might be worthwhile testing cmucl's lisp threads to see if they would be responsive enough. the code is actually in place to do that as well you should be able to run rts and receiveing right now (without using the gui) and see how it works in cmucl. it might be good enough for interactive work that doesnt push alot of data, i dont know. as far as pushing data goes, we are sort of comparing midishare vs portmidi right now on macos. the advantage with midishare is that it has the noteType event, which means you only have to deal with half the data that you do with portmidi, which requires noteOn+noteOff transmissions. on the other hand, portmidi works with the underlying "host" midi system, and so if being ALSA-based is important to you then portmidi is the way to go on linux. another unknown is how the threaded rts implementation will work if slime is also running... my problem is i simply dont have enough time to really learn all the apps on linux so i can test all this stuff and see how good it really works there, ie see if one could actually do a concert using it. its easy for Todd and I to test on the mac becuase we have machines at home and at work and have midi Monitors and Max to help determine how good the scheduler is actually doing (and its looking good there). but i dont have a linux box at home and when im at work im basically teaching. so i test what i can but its really going to be up to people with linux chops to stress test. --rick From hkyburz at gmx.de Mon Aug 22 01:11:07 2005 From: hkyburz at gmx.de (Hanspeter Kyburz) Date: Mon, 22 Aug 2005 10:11:07 +0200 Subject: [CM] OpenMCL realtime scheduler Message-ID: Hi Rick, Todd's new OpenMCL "high resolution scheduler" seems to be unbelievably fast: When sending a chord of 1000 midimessages (5x100 noteons plus 5x100 noteoffs triggered by 5 keys) spying to output with "MIDI Monitor.app" shows a time difference between first and last message of .03 to .05 ms (30000 to 50000 nanoseconds!!!) . Is this realistic? Midishare (v.1.88) seems to be MUCH faster than Portmidi: when receiving chord trills while generating huge amounts of notes, there are no skips, all incoming events are received by MAX. Up to now Todd's new scheduler in openmcl.lisp (revision 1.39) works perfectly! And since rts and receive don't depend anymore on Midishare callbacks or on Midishare alarms, receiving seems to be stable also now. Many thanks to you both for your great help during the last days! Best, Hanspeter From taube at uiuc.edu Mon Aug 22 12:38:55 2005 From: taube at uiuc.edu (Rick Taube) Date: Mon, 22 Aug 2005 14:38:55 -0500 Subject: [CM] clm build problem on darwin Message-ID: <13b49ee08a10da8d70da925b71a458e1@uiuc.edu> When I compile todays tarball of CLM from ccrma ftp server I get this error: ? (load "/Lisp/adm/clm") ; running cd /Lisp/clm-3/ && ./configure --quiet --with-doubles --with-float-samples ; Compiling "/Lisp/clm-3/io.c" ; Compiling "/Lisp/clm-3/headers.c" ; Compiling "/Lisp/clm-3/audio.c" ; Compiling "/Lisp/clm-3/sound.c" ; Compiling "/Lisp/clm-3/clm.c" In file included from /Lisp/clm-3/clm.c:34: /Lisp/clm-3/clm.h:165:20: sndlib.h: No such file or directory ; Compiling "/Lisp/clm-3/sc.c" ; Compiling "/Lisp/clm-3/cmus.c" In file included from /Lisp/clm-3/cmus.h:14, from /Lisp/clm-3/cmus.c:34: /Lisp/clm-3/clm.h:165:20: sndlib.h: No such file or directory So I edited clm.h and changed "sndlib.h" to "_sndlib.h" deleted all .o and fasls and tried again -- this time clm loaded and v.ins compiled, loaded and wrote test.snd BUT sndplay doesnt play it. i tried sndplay in the terminal too but still nothing. however qtime will play the test.snd so output is ok. i dont notice any warnings or skipped files in the second build process so im stumped. From bil at ccrma.Stanford.EDU Tue Aug 23 04:17:17 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 23 Aug 2005 04:17:17 -0700 Subject: [CM] Re: clm build problem on darwin In-Reply-To: <13b49ee08a10da8d70da925b71a458e1@uiuc.edu> References: <13b49ee08a10da8d70da925b71a458e1@uiuc.edu> Message-ID: <430B05BD.9010302@ccrma> I think I forgot to update the clm tarball along with all the others when I changed the sndlib header stuff. sndlib.h is now created by the configure script, and the "internal" version is _sndlib.h -- this is to move all the sndlib-specific configuration macros out of the sndlib user's way. You also need to force configure to run again. I'd start over from scratch with today's clm-3.tar.gz -- I tested it on a Mac as well as various Linux systems. Actually, I'm not sure why sndplay didn't work for you, but this is a first step. Along the same lines, I removed all support for the old Mac OS -- the code is in old-mac-audio.c if someone still needs it. From taube at uiuc.edu Tue Aug 23 05:58:35 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 23 Aug 2005 07:58:35 -0500 Subject: [CM] Re: clm build problem on darwin In-Reply-To: <430B05BD.9010302@ccrma> References: <13b49ee08a10da8d70da925b71a458e1@uiuc.edu> <430B05BD.9010302@ccrma> Message-ID: > I tested it on a Mac as well as various Linux systems. Actually, I'm > not sure why sndplay didn't work for you, but this is a first step. you are right it is something else. i removed all fasls, reinstalled the clm-3 tree , it builds fine, libclm and sndplay are made, fm-violin writes a valid file, sndplay is called but doesnt make any sound (qtplay will play the file.): ? (dac "test.aiff") Calling (RUN-PROGRAM "/Lisp/bin/clm-3/sndplay" ("test.aiff") :OUTPUT T :WAIT NIL) RUN-PROGRAM returned # "test.aiff" here is a trace of the c stuff but it looks fine: ? (load "/Lisp/adm/clm") ; running cd /Lisp/clm-3/ && ./configure --quiet --with-doubles --with-float-samples ; Compiling "/Lisp/clm-3/io.c" ; Compiling "/Lisp/clm-3/headers.c" ; Compiling "/Lisp/clm-3/audio.c" ; Compiling "/Lisp/clm-3/sound.c" ; Compiling "/Lisp/clm-3/clm.c" ; Compiling "/Lisp/clm-3/sc.c" ; Compiling "/Lisp/clm-3/cmus.c" ; Creating "/Lisp/bin/clm-3/libclm.dylib" ;;cc -dynamiclib -o /Lisp/bin/clm-3/libclm.dylib /Lisp/bin/clm-3/headers.o /Lisp/bin/clm-3/audio.o /Lisp/bin/clm-3/io.o /Lisp/bin/clm-3/sound.o /Lisp/bin/clm-3/clm.o /Lisp/bin/clm-3/cmus.o /Lisp/bin/clm-3/sc.o -L/usr/X11R6/lib -lX11 -lm -lc -framework CoreAudio [...] Building sndplay program: "/Lisp/bin/clm-3/sndplay" [...] From taube at uiuc.edu Tue Aug 23 10:30:36 2005 From: taube at uiuc.edu (Rick Taube) Date: Tue, 23 Aug 2005 12:30:36 -0500 Subject: [CM] Re: sndplay In-Reply-To: <1fd65b5d050823080774c636ae@mail.gmail.com> References: <1fd65b5d050823080774c636ae@mail.gmail.com> Message-ID: > The next step is (mus-audio-describe) and see if there's any obvious > screw-up. ? (mus-audio-describe) NIL ? (quit) audio stubbed out its stubbed out -- for some reason MUS_MAC_OSX isnt getting defined in any .h (but config.log has "#define MUS_MAC_OSX 1") . Also tHe _sndlib.h file uses MUS_MAC_OSX but does not define it. if i add " -DMUS_MAC_OSX=1" to *cflags* and then build then it all works. :mac-osx is being defined as a lisp feature if thats an issue so somehing with the autodetection is screwed up i guess. From bil at ccrma.Stanford.EDU Wed Aug 24 04:10:39 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 24 Aug 2005 04:10:39 -0700 Subject: [CM] Re: sndplay In-Reply-To: References: <1fd65b5d050823080774c636ae@mail.gmail.com> Message-ID: <430C55AF.50100@ccrma> > so somehing with the autodetection is screwed up i guess. I think this is finally cleaned up -- I have 4 different files named config.h.in and can't keep anything straight anymore. From k.s.matheussen at notam02.no Fri Aug 26 04:06:02 2005 From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen) Date: Fri, 26 Aug 2005 13:06:02 +0200 (CEST) Subject: [CM] [ANN] Snd-ls V0.9.5.3 and Mammut V0.21 Message-ID: http://www.notam02.no/arkiv/src/ http://www.notam02.no/arkiv/src/snd/ Snd-ls ****** Snd-ls is a distribution of the sound editor Snd. Its target is people that don't know scheme very well, and don't want to spend too much time configuring Snd. It can also serve as a quick introduction to Snd and how it can be set up. 0.9.5.1 -> 0.9.5.3 ------------------ -Reduced the startup-time radically. -Reduced memory usage. -Realtime priority when playing when using jack. -Fixed window auto-sizing. -Turned off off-turning of mlockall(MCL_FUTURE) to avoid being zombified from jack. -Fixed some things in rt-compiler.scm and snd_conffile.scm Mammut ****** Mammut will FFT your sound in one single gigantic analysis (no windows). These spectral data, where the development in time is incorporated in mysterious ways, may then be transformed by different algorithms prior to resynthesis. An interesting aspect of Mammut is its completely non-intuitive sound transformation approach. 0.20 -> 0.21 ------------ -Set realtime priority for the player thread to ensure clickless playing. -Fixed some trouble when exiting. -- From bil at ccrma.Stanford.EDU Sun Aug 28 04:45:18 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sun, 28 Aug 2005 04:45:18 -0700 Subject: [CM] Snd 7.15 Message-ID: <4311A3CE.3030009@ccrma> Snd 7.15 View:Files dialog completely recast: removed current files list, added various ways to use the selected file(s) in the other list previous-files-* -> view-files-* (sort, select-hook) removed previous-files-sort-procedure added sort-files-by-name|size|date|entry added view-files-amp|speed|amp-env|speed-style|files|selected-files click of file name no longer opens that file -- it just adds it to the selected files list for the dialog view-files-select-hook return type no longer matters, and args are: dialog file-name preload-directory -> add-directory-to-view-files-list preload-file -> add-file-to-view-files-list optional dialog arg to view-files-sort (local set) added clear-minibuffer optional as-error arg to report-in-minibuffer sound-file-extensions as settable list before-exit|close-hook -init startup switch to set init filename as arg netBSD audio support File:Insert menu item and dialog, also insert-file-dialog read/write support OGG, Speex, and Flac files (via File:Save as) read support for Mpeg and Midi removed yes-or-no? with associated dialog (and snd-g|xerror.c) old Mac OS (8 or 9) support (old-mac-audio.c has the code), also MPW C (this affects CLM as well) File:Mix dialog no longer exits upon "mix" (removed keep-mix-dialog-open-upon-ok from snd-motif.scm) gtk version no longer defaults to happy button colors -- they clash with the fancy icons. as long as I was changing colors, I changed to an "autumnal" theme for the Motif side. added osx-10.4-snd at ccrma-ftp _sndlib.h is now an internal header, whereas sndlib.h is intended for programs that use sndlib -- this is an attempt to reduce name-space collisions by sequestering configuration info. checked: OSX 10.4 (gtk/motif/guile/ruby), gtk 2.7.3|4|5, gtk 2.8.0|1|2, NetBSD 2.0, Solaris 10 (x86 audio is broken), sbcl 0.9.3 with much help from: Orm Finnendahl, Michael Edwards, Mike Scholz, Thomas Klausner, Kjetil Matheussen, Petter Reinholdtsen, Rob Komar, Christophe Rhodes From andersvi at extern.uio.no Mon Aug 29 12:37:49 2005 From: andersvi at extern.uio.no (andersvi at extern.uio.no) Date: Mon, 29 Aug 2005 21:37:49 +0200 Subject: [CM] Snd 7.15 In-Reply-To: <4311A3CE.3030009@ccrma> (Bill Schottstaedt's message of "Sun, 28 Aug 2005 04:45:18 -0700") References: <4311A3CE.3030009@ccrma> Message-ID: Compiling snd-7/audio.c of recent snd-7 (as well as clm-3) breaks with ...audio.c:8090: error: `SIGKILL' undeclared ... putting in: #include somewhere above seems to fix it here at least. From bil at ccrma.Stanford.EDU Tue Aug 30 03:35:07 2005 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 30 Aug 2005 03:35:07 -0700 Subject: [CM] Snd 7.15 In-Reply-To: References: <4311A3CE.3030009@ccrma> Message-ID: <4314365B.2060008@ccrma> Thanks -- I added the include signal.h.