Snd 24.7:
s7: added complex-vectors. These are directly compatible with
gsl_vector_complex.
make-iterator's optional third argument can be #t which tells s7
to choose the carrier based on the iterated sequence.
s7_immutable has been deprecated, replaced by s7_set_immutable which
has t
Thanks very much for the bug report! That was a tricky one -- I managed
to overoptimize op_x_aa, using a constant list where it got stepped on.
A quick fix, around line 90626, change
sc->value = c_function_call(f)(sc, with_list_t2(fx_call(sc,
cdr(code)), fx_call(sc, cddr(code;
to
sc-
Good luck with Goldfish Scheme! Massimiliano Gubinelli has been
poking me to check out the Mogan project (to deal with an s7
problem), but I never seem to get around to it (he is probably
grinding his teeth, causing his dentist to shake his head and
mutter).
_
Snd 24.6:
Kjetil Metheussen made a new version of s7webserver that works in
Python3 as well as Python2
and updated the s7webserver Makefile
johnm donated gsl_vector_size and several helper functions to libgsl.scm
s7: dynamic-unwind has changed slightly (unwinder gets the arg list, not
just
You can load libc.scm, which ties most of libc into s7,
then call mkdir as (*libc* 'mkdir):
((*libc* 'mkdir) "new-dir")
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
framples and srate refer to the currently opened sound if
none is passed as an argument, so the open-sound merely
provides a fallback, but leaves the file open, and has no
effect on sndwarp, unless I'm misreading that code.
The file argument to sndwarp worked in the single-file
case because you ga
I think you don't need the open-sound, but the
file passed to sndwarp should be (string-append
indir file), not just file (the latter defaults
to the current directory).
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mail
README.Snd has a list of the packages you need when you
use the --with-motif switch. I haven't checked it since
about version 22.04 of Ubuntu. You'll need xpm and maybe
xt, and probably one header file, shape.h from libext.
Also the x11-fonts-misc package.
__
You were misled by the word "allocated"? The message is
reporting overall GC stats.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
This is probably a typo, but just in case...
I open a .csm file
but you've set the autoload-file for .cms files?
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
s7.html has a section toward the end of the file about
repl.scm and nrepl.acm. These have a function drop-into-repl
which can be used to poke around in the current context.
I think you could call this when an error is hit, or
if you're in a "raw" terminal, when you hit some key
like C-C. The fil
Thanks for pointing this out -- I have updated the
homepage.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 24.5:
in s7, added (settable) symbol-initial-value (the #_... value)
changed invalid-escape-function to invalid-exit-function
new *s7* fields: symbol-quote?, symbol-printer, and make-function
added function-arglist
in sublet, varlet and s7_load_with_environment, ()
That's an interesting project! I don't know anything
about SDL, so I can't offer any sage advice. On the
matrix library, there's also the libgsl stuff in the
s7 tarball. If you're writing the matrix functions in Scheme
and notice anything that seems unnaturally slow,
please let me know. Good L
:SB-PACKAGE-LOCKS :SB-THREAD :SB-UNICODE :SBCL :UNIX)
* (load "all.lisp")
; using existing configuration file mus-config.h
; Compiling "/home/bil/clm/xen.c"
; Compiling "/home/bil/clm/io.c"
; Compiling "/home/bil/clm/headers.c"
; Compiling "/hom
It looks like rm-radians is not used in new-effects.scm, so that code
is multiplying the input samples by an oscil at rm-frequency. This
is the same as the code in examp.scm without the envelope. ring-modulate
is just a multiply, so I think you want (ring-modulate mysnd (oscil os))
where the "os"
Iain, thanks a million for the kind words about s7 in reddit and Hacker
News.
When you're tired and down, a pat on the back makes a big difference.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 24.4:
s7: added (settable) port-string
checked: sbcl 2.4.4
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 24.3:
s7: format now always returns a string (where it used to return #f it
now returns "")
sam.c: bugfixes and improvements thanks to David Jaffe
checked: sbcl 2.4.3
Thanks!: Michael Edwards, Norman Gray, David Jaffe
___
Cmdist mailing list
Snd 24.2
mostly work on optimizations in s7
checked: sbcl 2.4.2
Thanks!: James Hearon
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 24.1
More optimizations, minor bug fixes, and rewrites.
checked: sbcl 2.4.1
Thanks!: Norman Gray, Andreas Enge
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
I do not know what the problem is; that code has not
changed in many years. Perhaps you're running Wayland
and it is mixing colors (via a sort of transparency
effect I guess, like happens on a Mac).
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
htt
Thanks for the bug report. I have fixed extsnd.html.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 24.0:
in s7, fixed various bugs, made a few more optimizations.
checked: sbcl 2.3.11, sbcl 2.4.0
Thanks!: Norman Gray, James Hearon
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
I think this is the same problem that Kenneth Flak
reported in July. Here's a portion of my reply:
"... In snd-motif.scm, there are two uses
of the font "9x15" -- this is a very old (but very good)
fixed width font that may now be relegated to the
X miscellaneous fonts package. You may have
to
I'd need to see you sampler code to see what the problem
is, but this sequence works:
;; write a square-wave to test.snd
(with-sound (:output "test.snd" :to-snd #f)
(let ((g (make-square-wave 440.0)))
(do ((i 0 (+ i 1)))
((= i 44100))
(outa i (square-wave g)
;; read it us
Is that chdir (haven't had my coffee yet) -- if so,
maybe ((*libc* 'chdir) "new-dir")?
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 23.9:
s7: added optional let argument to immutable? and immutable!
finally added error checks to the tree-* functions
see lint.scm for the previous versions
s7_make_c_pointer_wrapper_with_type
' (apostrophe) now is (#_quote ...) and similarly
for the quasiquote helpers
Thanks! I added that to the s7 README.md file,
and also added a short section to s7.html.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
I mostly have bill's wonderful old 'birds.scm' file chirping
away in python. It made me laugh!!
When Jonathan Harvey used file that in a piece (almost unedited!)
a reviewer said they were recordings. Can't win sometimes.
___
Cmdist mailing list
C
I can speak to the CLM/Snd side of this which was closely associated
with Common Music.
I followed Rick's lead, but I was already searching for an alternative
to
Common Lisp. The sound editor, Snd, was a C version of an earlier
editor named
Dpysnd, written in SAIL at SAIL in the 70's and 80's.
Any s7 questions are ok in this list. I can't tell what you're doing,
but
this doesn't strike me as a GC problem. If you can send me enough of
your
program that I can run it, I'm willing to try to see what is wrong.
___
Cmdist mailing list
Cmdist@cc
I think you forgot to load "all.lisp" -- you need
the clm code loaded before trying to compile v.ins.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 23.8:
s7: error handling, reader-cond, various (*s7* 'max-*) checks
checked: sbcl 2.3.9
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Snd 23.7:
various minor optimizations and bugfixes in s7
checked: sbcl 2.3.8
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
I think you want to use a multidimensional float-vector:
(let* ((fsize 10)
(im (make-float-vector (list fsize 2)))
(x 1)
(y 1.0)
)
(do ((k 0 (+ k 1)))
((>= k fsize))
(set! (im k 0) x) ; this will be turned into a float
(set! (im k 1) y)
(format #t "
Snd 23.6:
various small improvements and bugfixes in Snd and s7.
checked: sbcl 2.3.7
Thanks!: Kenneth Flak
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
One way is to loop through the channels calling save-sound-as on each,
something like:
(do ((i 0 (+ i 1)))
((= i (channels snd)))
(save-sound-as filename snd :channel i))
where filename needs to be different for each channel.
___
Cmd
Snd no longer has any recording ability. Originally
there was a simple recorder (although it had great VU
meters), and I could maintain it. But at that time audio
hardware was changing rapidly, especially adding many
more channels. I got many requests for support for
new devices which I did not
I changed this to use default-output-header-type. Snd
used mus-next because it stores data as doubles, and it's
simpler to write/read double samples, but mus-riff does not
support that sample type (or didn't the last time I looked).
So you'll need to use:
(save-region r (format #f "test~A.wav"
In new-effects.scm the only font mentioned that might
be a problem is the "built-in" X font (for stuff like
XC_crosshair). In snd-motif.scm, there are two uses
of the font "9x15" -- this is a very old (but very good)
fixed width font that may now be relegated to the
X miscellaneous fonts package.
One way is to call snd-spectrum using channel->float-vector
to get the data. If you already have the fft data in snd,
you can use the peaks function to write it either to the
listener or a text file.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
ht
Dave Phillips wrote a tutorial for Snd a long time
ago -- for Snd 6.0 (2003), I think. It eventually got
out of sync with the program, but I still have
a copy -- I can send it to you if you like.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https:
I was using mouse selection to test it. I might not
say anything in the Snd docs because it's an X/Motif
thing (i.e. the "clipboard" I think it's called.
Look for "interclient communication" in the X docs,
and XA_PRIMARY). It doesn't work for me on a Mac.
To get all the current names, use (symbol-table)
in scheme, or s7_symbol_table and s7_for_each_symbol
in C. I can send you the output of
(format #f "~W" (symbol-table))
It's too long to include here (4266 names, 57k bytes).
I guess you already know this, but snd-completion.c
and snd-help.c h
It works both directions and in both the Motif and
terminal versions for me (I tried it with a terminal
and emacs). The notcurses repl may not work in
this regard -- I'll check later today.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-
Snd 23.5
s7: various small bugs and optimizations.
checked: sbcl 2.3.5|6
Thanks!: Anders Vinjar, Todd Ingalls, johnm, Kjetil Matheussen
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
No, I don't know how emacs does it. It used to be going
through the emacs file comint.el, if I remember right.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
This appears to be some pipe issue with stdin -- if
you run Snd from a terminal, then type expressions
in that terminal, get_stdin_string is called and
the expressions are evaluated, etc. If you call
Snd as a target of a pipe (or whatever the correct
terminology is), it works once, but then stdin
0x76e5bc0f in __GI___poll (fds=0x7fffd790, nfds=1,
timeout=5)
at ../sysdeps/unix/sysv/linux/poll.c:29
I think that you're in XtAppMainLoop waiting for an event;
that is, snd is waiting for you to send input etc.
___
Cmdist mailing li
[I forgot to "reply all"]:
In the motif version, XtAppAddInput in snd-motif.c sets get_stdin_string
as the function to call when input is available on stdin. That function
tries to get a full string, then evals it. It doesn't return the value
of the incoming expression -- you have to use the wr
I just tried it with a file 80 seconds long, and it
worked fine. If you want to use it with a multichannel
file, set channel layout to "separate" -- you can use
the channel layout item in the view menu, or
(set! (channel-style) channels-separate) in the
listener. Or maybe uncheck the "unite" but
The short answer is that it's an historical artifact, and
I try not to change s7.h unless it's absolutely necessary.
The long form: the s7_p_* procedures came first when I was
developing the second level of the optimizer, and at that
time I thought I'd always need the s7_scheme* argument, so
ther
Everything seems to work for me. I'm running on Ubuntu,
using X, not Wayland I think (but I hope that doesn't
matter). I built snd --with-motif --with-gl, opened
a file (oboe.snd), clicked 'f' (for fft) and unclicked
'w' (for time domain wave), went to the Options:Transform options
dialog, set s
Does snd spit out anything like a prefix or exit code or
something through stdout when evaluating an expression,
in addition to the return value?
I don't think so. Is snd still running after you get that
value? Currently snd blocks SIGTTIN and SIGTTOU.
I'm glad you found that -- I was puzzling my way through the
conjure docs -- I know nothing about conjure or neovim.
Another possible problem might be which snd repl you're
using. If you're running the no-gui snd, it's the code
in repl.scm, but I'm not sure which repl is running in
this case -- th
Snd 23.4
fixed a move-formants bug which also affected bullfrog and leopard-frog
in animals.scm.
checked: FC 38, Ubuntu 23.04, sbcl 2.3.4
Thanks!: johnm, Todd Ingalls
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/ma
Ouch! I optimized formant-bank at some point, and broke
that instrument. I think this slower version works:
(definstrument (move-formants start file amp radius move-env
num-formants)
(let* ((frms (make-vector num-formants))
(beg (seconds->samples start))
(dur (mus-sound-fr
That's very impressive! I'll add a pointer to it to s7's README.md.
Thanks!
(The libarb support in s7 was really just for my amusement -- I wanted
to play with the Bessel functions).
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.st
Do you have any suggestions for what one can do to cut GC times down
on a code level when there is a lot of code going?
Avoid creating things unnecessarily, and don't keep large structures
around when they're no longer needed. The (*s7* 'memory-usage)
function can show how many objects of each
To keep an environment from the being garbage collected,
assign it to a global variable -- as long as the variable
holds the environment, it will be safe. This won't speed
up the GC. It's possible to make s7 strings that are not
seen at all by the GC (make_permament_string), but it's much
tricki
Snd 23.3
s7: autoload bugfix
checked: sbcl 2.3.3
Thanks!: john M693, Todd Ingalls
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Sounds good -- whenever you're ready, let me know
what files to include in sndlib, or whatever you want.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Wow! That's a lot of work! What would be the best way to include this
in sndlib? I don't have a git site for that library -- just the
libraries
at the ccrma ftp site. I don't often change the sndlib files (except
the s7 files).
___
Cmdist mailing
Snd 23.2
the usual minor bugfixes and whatnot.
checked: sbcl 2.3.2
Thanks!: Woody Douglass
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
I think some of the newer Motif packages (like post-2012)
need both /usr/include/X11/extensions/Print.h and
/usr/include/X11/bitmaps/gray -- in the latter case I
sometimes have to create the bitmaps directory.
I can send you copies of these files if needed.
There was a time when everyone was sayi
I think s7_call_with_catch needs to set a jump point for the
sc->longjmp_ok case, but I haven't figured out yet how to
capture the error handler result in all cases.
Here's what I have now:
(ca line 51216)
else
{
declare_jump_info();
TRACK(sc);
store_jump_info(sc);
Snd 23.1:
clm: Michael Edwards fixed a bug in scentroid.ins.
s7: David St-Hilaire integrated s7 into tic-80.
checked: FC 37, sbcl 2.3.1
Thanks!: David St-Hilaire, Michael Edwards
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stan
Original Message
Subject: Re: [CM] S7 integrated in TIC-80, and bug reports
Date: 2023-02-03 10:52
From: b...@ccrma.stanford.edu
To: David St-Hilaire
Would there be a way to give a better error message?
The error message is the third argument to error_nr.
One simple change
(sorry about the chaotic state of cmdist -- all my fault).
Original Message
Subject: Re: [CM] S7 integrated in TIC-80, and bug reports
Date: 2023-02-03 10:10
From: David St-Hilaire
To: b...@ccrma.stanford.edu
Hi bil! I implemented your changes and it's working very we
Here's a first stab at it -- put this code at line 1430 or thereabouts
in s7.c (after the error_nr forward declaration), then build s7 with
-DDISABLE_FILE_OUTPUT=1:
#ifndef DISABLE_FILE_OUTPUT
#define DISABLE_FILE_OUTPUT 0
#endif
#if DISABLE_FILE_OUTPUT
static FILE *old_fopen(const char *pathn
Do you need to disallow reading a file? If it's just
creating or altering a file that needs to be blocked,
you could redirect fopen and fwrite (in s7.c) to
functions that raise an error. I don't think s7 uses
creat, open (except with O_RDONLY), or write. Also
build it with WITH_C_LOADER=0 (to d
Defining noreturn to nothing is fine; in gcc it makes a
measurable difference in speed, but it's just a minor
optimization. I can't find a compiler feature macro
to tell me I'm running C23.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-m
I don't think standard scheme allows set! with a target that is a pair,
or at least I can't find anything about it in r7rs.pdf. In s7,
the target needs a setter.
The jmp_bufs are needed to unwind the C stack if an error is raised.
It might work to make longjmp a no-op -- the stack will grow as
e
I forgot to say that I get this from the repl:
(define-macro (test x . args) `(list ,@(map car args)))
(test 10 ('x 1) ('y 2))
'(x y)
On the noreturn problem, I'm using this:
#ifdef _MSC_VER
#define noreturn _Noreturn /* deprecated in C23 */
...
which I got from somewhere online. I wonder
On the first issue, it looks to me like you have
your own map function -- scheme_map is not an s7
function. It is passing a null pointer to s7_integer.
As s7.html explains (under "FFI notes"), the C-side
functions do very little error checking.
___
Cmd
Thanks for the info and bug reports. Are you catching
scheme errors or using a repl? In the repl, I get:
(define-macro (test x . args) (list ,@(map car args)))
(test 10 ('x 1) ('y 2))
;unquote (',') occurred outside quasiquote: (test 10 ('x 1) ('y 2))
;(test 10 ('x 1) ('y 2))
; test: (lis
This is probably the "pipewire problem"; apparently
the issue (at pipewire) was closed and labelled that
it was no longer a problem -- someone at pipewire
is embarrassed about unfixed bugs, I guess.
The cmdist archives for early September has:
The workaround for the pipewire problem in Snd
(or re
Are these modification containing 3rd library port welcomed to snd
project?
I don't know in advance -- I need to see the proposed changes.
Thanks very much for looking into this.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanfo
Snd 23.0:
In Snd, drag and drop can handle "percent-encoded" file names,
thanks to Tito Latini.
In clm, Orm Finnendahl got Juan Pampin's ATS system running:
https://github.com/ormf/cl-ats.
In s7: reverted s7.h to previous version; use tools/s7.def instead.
checked: Ubuntu 22.10 (gcc 12.2.
warning C4013: 'lookup' undefined; assuming extern returning int
I was trying to get rid of some #if's and forgot to include
the lookup forward declaration in the msvc case; here's
a change that ought to fix that problem:
at line 4061, add the declaration
static inline s7_pointer lookup(s7_sch
The common lisp clm didn't support that kind of
setf in the run loop. I think you need to do it
this way:
(defparameter *test* (make-double-array 100 :initial-element (double
0.0)))
(definstrument get-samples (arr)
(let ((num (length arr)))
(file->array "oboe.snd" 0 0 num arr)
arr))
Wow, that goes back about 20 years! Why can't you
use definstrument and a compiler? I looked at the
lisp code, and I think in-any returns a file->sample
generator, but all these IO functions expect you're
using the run macro to set up the buffers in C.
But the run macro only runs to completion a
I forgot to say you also need to set (*s7* 'print-length)
to be as big as the byte-vector -- perhaps
write-bytevector could use let-temporarily for this,
but you really don't want to write out a huge vector
to *stdout* inadvertently.
___
Cmdist mailing
I don't want to implement that count-value variable business --
I think it violates lexical scope by turning "count-value"
into a weird symbol macro equivalent to (let-ref experiment
'count-value).
So you have any number of count-value's floating around, and
if you set! one anywhere, all the res
I didn't notice that read-bytevector should return #
if no bytes are read. Here are changes to r7rs.scm that
make your code work I think (but I doubt it's faster):
127c127,129
< (if (= i start) # i))
---
(if (< i lim)
(copy (subvector bv 0 i))
bv))
130,13
argh I clicked the wrong damn button
(define (jiffies-per-second) 10)
(define (current-jiffy)
(with-let *libc*
(let ((res (clock_gettime CLOCK_MONOTONIC)))
(+ (* 10 (cadr res)) (caddr res)
(define (current-second) (* 1.0 ((*libc* 'time) (c-pointer 0
'time_t*
For awhile the thing currently called (*s7* 'cpu-time)
was returning the wall clock time by accident. I
think the r7rs.scm functions should be:
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
That looks like a bug in member (and assoc) to me --
I think I over-optimized environment handling -- I'll
make a new s7 tarball. The fix is to comment out line
1863 in s7.c.
Thanks!
(It would also work in the old version to use
(member (make-foo) (list (make-foo)) equivalent?))
__
I don't see how your work-around fits the paragraph you
quote; please send me a working example of the problem,
and I'll look at it. I trying very hard not to say what
I think of r7rs libraries...
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https
Snd 22.9
s7: added support for NaN "payloads": new functions nan and nan-payload
to create such a NaN or get its payload, read-time support for
+nan.123 (123 is the payload), and so on.
S7_EXPORT for MSVC thanks to vladimir florentino
checked: sbcl 2.2.10
Thanks!: Hyogeol Lee,
The error message (the first line) is the same in each case.
I don't know where the other output is coming from.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Thanks for the suggested code, but that is not how
I want to do it, if I do it at all. This is a minor
optimization, so it should not introduce compile-time
switches that split s7 into two incompatible versions;
I'd have to add stuff to ffitest.c and the documentation
and so on. I'll look at it
That was the way Rick Taube wanted it back around 2010 (he
was writing Grace, a GUI version of his CommonMusic). I
can probably add a string output choice -- will look at it
later.
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanf
Thanks very much! I'll merge those changes into my version.
(It may take a day or two).
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinfo/cmdist
Thanks very much for the bug report. If this is a recent
version of s7 (see S7_DATE in s7.h), I think the problem is
that the MSVC version of the code calls unbound_variable
where everyone else returns NULL. I don't have access to
a machine running MSVC, and I can't remember now why the
two case
Snd 22.8
A bit early, but I want to try something that doesn't fit at
the end of the cycle.
s7: s7_number_to_real_with_location. s7_wrong_type_error.
s7_make_string_wrapper_with_length. s7_make_semipermanent_string.
CL-style doc strings, (symbol...) settable.
checked: sbcl 2.2.9
__
I just noticed your s7_clone is not necessarily making a new
s7 instance -- in this case you'll have two views into
one s7; if these are in separate threads, the two s7's
will be colliding in their use of the evaluator -- it
is not thread safe in the sense that any number of
threads can be calling
To expand a bit on Elijah's note:
I don't think this will reduce cache misses or make objects smaller.
To evaluate anything, you have to access the things being evaluated.
Most of the points in the article you mention do not apply to s7 --
s7 manages its own memory, the cells are already packed t
Snd 22.7
s7: (*s7* 'number-separator) and heap-analyze|scan
and stomping on bugs, of course
checked: sbcl 2.2.8
Thanks!: Tito Latini, IOhannes Zmoelnig
___
Cmdist mailing list
Cmdist@ccrma.stanford.edu
https://cm-mail.stanford.edu/mailman/listinf
1 - 100 of 606 matches
Mail list logo
