Pd-dev July 2010

pd-dev@lists.iem.at

20 participants
139 discussions

[ pure-data-Patches-1933800 ] fix --disable_fat on OSX/Intel (0.41-4)
by SourceForge.net 25 Jul '10

25 Jul '10

Patches item #1933800, was opened at 2008-04-03 14:29 Message generated for change (Comment added) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1933800&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata-dev Group: feature Status: Open Resolution: None Priority: 5 Private: No Submitted By: Thomas Grill (xovo) Assigned to: Miller Puckette (millerpuckette) Summary: fix --disable_fat on OSX/Intel (0.41-4) Initial Comment: It's currently very misleading to use the configure flag --disable-fat on OSX/Intel, because configure.in defines PA_BIG_ENDIAN and the default extension for externals in d_ppc. It is sometimes necessary to build non-fat binaries because not all compilers (especially the experimental gcc-4.3.0) support multiple architectures. The patches introduce a .d_i386 extension and add necessary stuff to the makefiles. The patch must be applied from the pd root (not src), since it includes src/configure.in, src/s_loader.c, extra/makefile and extra/expr~/makefile The patch also fixes minor issues apart from i386 ... the bundled externals are now really called *.d_ppc or *.d_i386, and not *.pd_darwin. ---------------------------------------------------------------------- >Comment By: Miller Puckette (millerpuckette) Date: 2010-07-24 21:03 Message: I'm not sure this is still needed (GCC still not supporting FAT? I doubt it :) ---------------------------------------------------------------------- Comment By: Thomas Grill (xovo) Date: 2008-04-03 15:49 Message: Logged In: YES user_id=350252 Originator: YES a small additional note: this is still a workaround somehow: non-fat building depends on -DMACOSX3 which is really ugly ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1933800&group_…

1 0

[ pure-data-Patches-1933679 ] FFTW3 interface for pd-0.41-4
by SourceForge.net 25 Jul '10

25 Jul '10

Patches item #1933679, was opened at 2008-04-03 12:19 Message generated for change (Comment added) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1933679&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata-dev Group: feature >Status: Closed >Resolution: Accepted Priority: 5 Private: No Submitted By: Thomas Grill (xovo) Assigned to: Miller Puckette (millerpuckette) Summary: FFTW3 interface for pd-0.41-4 Initial Comment: The attached patch implements the necessary functionality to emulate the mayer_fft with FFTW3. It's a strange approach and not ultimately efficient, but it's better than the plain mayer_fft in any case. ---------------------------------------------------------------------- >Comment By: Miller Puckette (millerpuckette) Date: 2010-07-24 21:00 Message: Took it -- but probably won't test it - caveat emptor ---------------------------------------------------------------------- Comment By: Rich E (reakin) Date: 2008-05-21 13:19 Message: Logged In: YES user_id=1586621 Originator: NO info->in = (fftwf_complex*) fftwf_malloc(sizeof(fftw_complex) * n); info->out = (fftwf_complex*) fftwf_malloc(sizeof(fftw_complex) * n); shouldn't this be fftwf_malloc(sizeof(fftwf_complex) *n)? I think you allocating for doubles instead of floats. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1933679&group_…

1 0

[ pure-data-Patches-1931686 ] functionality for 128-bit-aligned memory (against pd-0.41-4)
by SourceForge.net 25 Jul '10

25 Jul '10

Patches item #1931686, was opened at 2008-04-01 14:54 Message generated for change (Comment added) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1931686&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: feature Status: Open Resolution: None Priority: 5 Private: No Submitted By: Thomas Grill (xovo) Assigned to: Miller Puckette (millerpuckette) Summary: functionality for 128-bit-aligned memory (against pd-0.41-4) Initial Comment: This patch implements public functions for allocation of 128-bit aligned memory. For allocation of all signals and buffers within PD these functions are consequently used. 128-bit alignment can speed up vector calculations, esp. if used with SIMD (also with autovectorization techniques). The infrastructure has been part of devel_0_39 and has now been ported to vanilla pd. ---------------------------------------------------------------------- >Comment By: Miller Puckette (millerpuckette) Date: 2010-07-24 20:55 Message: I need to stud this more -- probably posix_memalign() can be used. But how do "modern" compilers treat this? Maybe we can get vecorization w/o special code now? ---------------------------------------------------------------------- Comment By: Thomas Grill (xovo) Date: 2008-04-14 14:47 Message: Logged In: YES user_id=350252 Originator: YES Here's an incremental patch which checks for null pointers in freealignedbytes and resizealignedbytes. It's obviously necessary to do that. File Added: x-0.41.4-128bit_alignment-2.diff ---------------------------------------------------------------------- Comment By: Thomas Grill (xovo) Date: 2008-04-03 07:15 Message: Logged In: YES user_id=350252 Originator: YES Yes, you are right, but the current implementation wastes a bit of memory which might not be acceptable (although it's only 4-19 bytes per block). The implementation is portable because it relies on the existing functions, but there's also the possibility to rewrite getbytes() in a way that the system provides allocated memory right from the start. Under OS X it seems that plain malloc is ok for that (http://developer.apple.com/documentation/Darwin/Reference/ManPages/man3/mal…) under general POSIX there seems to be memalign (http://man-wiki.net/index.php/3:posix_memalign) , under Windows there's _aligned_malloc (http://msdn2.microsoft.com/en-us/library/8z34s9c6.aspx). However, i didn't want to deal with platform issues. ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2008-04-03 06:45 Message: Logged In: YES user_id=27104 Originator: NO This sounds like a very useful addition. I have one minor question about the patch implementation. It seems that it would be more manageable if the new functions had the same name as the originals, then the original and the optimized versions could be swapped for each other with a single #ifdef which could be controlled by a ./configure flag. Is there some code where you don't want to use the aligned functions? Redefining the original names would automatically give all the externals aligned memory as well. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1931686&group_…

1 0

[ pure-data-Patches-1930769 ] 0.41-4 fix memory leak with array sinesum/cosinesum
by SourceForge.net 25 Jul '10

25 Jul '10

Patches item #1930769, was opened at 2008-03-31 19:09 Message generated for change (Settings changed) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1930769&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: bugfix >Status: Closed >Resolution: Accepted Priority: 7 Private: No Submitted By: Thomas Grill (xovo) Assigned to: Miller Puckette (millerpuckette) Summary: 0.41-4 fix memory leak with array sinesum/cosinesum Initial Comment: svecs are allocated twice.... --- ../src.ori/g_array.c 2008-03-15 01:03:00.000000000 +0100 +++ g_array.c 2008-04-01 04:05:50.000000000 +0200 @@ -1221,7 +1221,7 @@ static void garray_sinesum(t_garray *x, t_symbol *s, int argc, t_atom *argv) { - t_float *svec = (t_float *)t_getbytes(sizeof(t_float) * argc); + t_float *svec; int npoints, i; if (argc < 2) { @@ -1244,7 +1244,7 @@ static void garray_cosinesum(t_garray *x, t_symbol *s, int argc, t_atom *argv) { - t_float *svec = (t_float *)t_getbytes(sizeof(t_float) * argc); + t_float *svec; int npoints, i; if (argc < 2) { ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2008-04-15 16:51 Message: Logged In: YES user_id=27104 Originator: NO checked this into branches/pd-extended/0.40 ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1930769&group_…

1 0

[ pure-data-Patches-1914851 ] menubar - window filename update
by SourceForge.net 25 Jul '10

25 Jul '10

Patches item #1914851, was opened at 2008-03-15 08:45 Message generated for change (Settings changed) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1914851&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: None >Status: Closed >Resolution: Accepted Priority: 5 Private: No Submitted By: marius schebella (mariusschebella) Assigned to: Miller Puckette (millerpuckette) Summary: menubar - window filename update Initial Comment: hi, I found a minor bug. when you open a patch (or create new patch) and then "save as" (or "save" with a new patch) then that name is not updated in the menubar-window menu. the patch is still listed under the old name or "untitled". maybe that's only on mac 10.5. I tried pd-extended 0.40.3(080222) and the current pd 0.41-2 from millers site. marius. ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2008-04-28 13:45 Message: Logged In: YES user_id=27104 Originator: NO Ok, found the source of the problem. I changed it so whenever saving, update the window list for the Window menu in case the save was a "Save As" that changed the name of the window. File Added: update_window_list_on_saveas-0.41.4.patch ---------------------------------------------------------------------- Comment By: marius schebella (mariusschebella) Date: 2008-04-28 05:09 Message: Logged In: YES user_id=1856112 Originator: YES sorry, if I was not clear. to reproduce create a new patch and look in the menubar under "Windows". It will be listed as Untitled-1 or something. then save it as somename.pd and look again under "Windows". it is still listed as untitled. this does not get updated until you open a new patch. ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2008-04-27 21:57 Message: Logged In: YES user_id=27104 Originator: NO I cannot reproduce his bug on Pd-0.40.3-extended-20080427 or Pd-0.41-4 on Mac OS X 10.4.11. If you saveas an abstraction that is in use, it will not change the title , and it will indeed remain as that abstraction, and just save a copy, IIRC. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1914851&group_…

1 0

[ pure-data-Patches-1881907 ] more t_float/t_sample fixes
by SourceForge.net 25 Jul '10

25 Jul '10

Patches item #1881907, was opened at 2008-01-29 08:24 Message generated for change (Comment added) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1881907&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: None >Status: Closed >Resolution: Accepted Priority: 5 Private: No Submitted By: IOhannes m zmlnig (zmoelnig) Assigned to: Miller Puckette (millerpuckette) Summary: more t_float/t_sample fixes Initial Comment: pd-0.41 has a lot of cleanup regarding the types of numbers (see patch-#1822001 'use of t_float/t_sample instead of "float"') however, some things have been left out by the original patches, here are some more: m_pd.h: seems like the t_float-enabling of postfloat() was forgotten in the declaration d_math.c: here it is the other way round: while the declarations in m_pd.h use t_float, the definitions do not g_array.c g_template.c: here my patch was a bit rash: scanf() doesn't like to get pointers to double (if "t_float" is set to "double") if the format string is really "%f" (or "%g"); the new patches use an internal _double_ variable and the format string "%lf" ("%lg"); the _double_ variable is then cast to (t_float). this should be on the save side; the alternatives would have been using an internal _float_ variable, and/or to change the format-string depending on the used type. d_fftroutine.c: this needed a bit more changes, but luckily enough, the original code already was aware of possible type-changes; i have currently used "t_float" (in line with the m_pd.h declaration of "pd_fft()"), but i am not sure whether this really should be "t_sample". changes should be rather simple: just change the defines of FLOAT and SAMPLE in the beginning... (and don't forget the declaration in m_pd.h) with these changes, Pd compiles without (additional) warnings with t_float/t_floatarg and t_sample set to "double". (there is still work to do in d_ugen.c though, to make it really usable...) ---------------------------------------------------------------------- >Comment By: Miller Puckette (millerpuckette) Date: 2010-07-24 19:17 Message: I adopted changes where they looked right.. I then tried actually running Pd in 64 bit and found lots of problems. So I'll mark this 'accepted' but there's more to do here. ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2010-01-01 17:43 Message: it seems that some of this stuff has been accepted and some not, perhaps they should be reformated? ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-01-29 09:54 Message: Logged In: YES user_id=564396 Originator: YES File Added: sigmund~.c.diff ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-01-29 09:12 Message: Logged In: YES user_id=564396 Originator: YES added a fix for extra/sigmung~ File Added: sigmund~.c.diff ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-01-29 09:11 Message: Logged In: YES user_id=564396 Originator: YES added a fix for pique File Added: pique.c.diff ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-01-29 09:09 Message: Logged In: YES user_id=564396 Originator: YES added a fix for extra/lrshift~ File Added: lrshift~.c.diff ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-01-29 09:08 Message: Logged In: YES user_id=564396 Originator: YES added a fix for extra/loop~ (the core code originally was not dependent on Pd; i therefore introduced an additional "#define t_sample float" in case it is not compiled for Pd) File Added: loop~.c.diff ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-01-29 09:06 Message: Logged In: YES user_id=564396 Originator: YES added a fix for extra/choice File Added: choice.c.diff ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1881907&group_…

1 0

[ pure-data-Patches-3032844 ] Sort method tables by symbol and use binary search
by SourceForge.net 22 Jul '10

22 Jul '10

Patches item #3032844, was opened at 2010-07-21 23:06 Message generated for change (Tracker Item Submitted) made by claudiusmaximus You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=3032844&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: None Status: Open Resolution: None Priority: 5 Private: No Submitted By: ClaudiusMaximus (claudiusmaximus) Assigned to: Nobody/Anonymous (nobody) Summary: Sort method tables by symbol and use binary search Initial Comment: Sort method tables by symbol and use O(log(N)) binary search instead of an O(N) linear search. This should speed up calling existing methods but it might slow down adding new methods. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=3032844&group_…

1 0

sys_expandpath() implementation notes
by Hans-Christoph Steiner 21 Jul '10

21 Jul '10

Hey Miller, I just saw that you accepted and changed the sys_expandpath() stuff, thanks for that. There are two little related issues that I thought I'd bring up: - in commit c78d6b22941725368af014fd0863b00e96f0dde7 "bug fix to sys_expandpath", you added some curly braces around the code at the end in "#ifdef _WIN32 section. They don't seem to do anything and could make the code confusing, i.e. the ExpandEnvironmentStrings stuff at the end is not part of the "if ((strlen(from) == 1 &&..." code block. - second, MAXPDSTRING (1000) is bigger than FILENAME_MAX on Win32 (255). That is a source of crashes if anyone makes a string/symbol longer than 255 on Windows. I think we should be using the POSIX macro FILENAME_MAX rather than MAXPDSTRING anytime we are dealing with paths and filenames for that reason. Plus I think FILENAME_MAX is part of the C89 spec, so it should be very widely implemented. http://www.mofeel.net/957-microsoft-public-vc-language/11149.aspx .hc

1 0

[ pure-data-Patches-1862178 ] [feature] make makefilename accept floats in %s
by SourceForge.net 21 Jul '10

21 Jul '10

Patches item #1862178, was opened at 2008-01-02 05:40 Message generated for change (Settings changed) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1862178&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: feature >Status: Closed >Resolution: Accepted Priority: 1 Private: No Submitted By: IOhannes m zmlnig (zmoelnig) Assigned to: Miller Puckette (millerpuckette) Summary: [feature] make makefilename accept floats in %s Initial Comment: while fiddling with the bug-fix for the missing "%c" in [makefilename] (see #1862168), i found that [makefilename %s] will safely ignore any number that it receives. in theory this is nice and consistent behaviour. in practice i find it weird that [32( | [makefilename %s] outputs nothing(!). and [32( | [makefilename _%s+] outputs "_+" the attached patch uses "%g" to convert a number into a string which is then inserted as %s. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1862178&group_…

1 0

[ pure-data-Patches-1851193 ] [symbolatom\ : doesn't allow to type spaces (ascii 32)
by SourceForge.net 21 Jul '10

21 Jul '10

Patches item #1851193, was opened at 2007-12-14 21:23 Message generated for change (Settings changed) made by millerpuckette You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1851193&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: puredata Group: bugfix >Status: Closed >Resolution: Accepted Priority: 9 Private: No Submitted By: Roman Haefeli (reduzent) Assigned to: Miller Puckette (millerpuckette) Summary: [symbolatom\ : doesn't allow to type spaces (ascii 32) Initial Comment: i would like to read a textfile into an object [textfile] by using the symbolatom as a gui to specify the filename. however, it fails when the filename of the file contains a space. example patch: [symbolatom\ | [read $1( | [textfile] ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2010-07-15 15:32 Message: I'm elevating the priority since this is a trivial fix and has been in Pd-extended since 2007. ---------------------------------------------------------------------- Comment By: Hans-Christoph Steiner (eighthave) Date: 2009-03-20 20:16 Message: Wow, such a simple fix, I wish I thought of trying years ago. This does make me think that we should have a discussion at PdCon just about spaces and symbols. I included this change in Pd-extended 0.41. Here's the commit: http://pure-data.svn.sourceforge.net/viewvc/pure-data?view=rev&revision=108… ---------------------------------------------------------------------- Comment By: IOhannes m zmlnig (zmoelnig) Date: 2008-04-06 10:42 Message: Logged In: YES user_id=564396 Originator: NO the fix for this is rather trivial (see attached patch) i'm wondering why spaces have been disabled in the first place... File Added: spacesymbol.diff ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1851193&group_…

1 0

← Newer
1
2
3
4
5
6
7
8
...
14
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pd-dev July 2010