75986bd | weis | 23 February 2006, 08:31:07 UTC | options becomes command_options. | 23 February 2006, 08:31:07 UTC |
1d7b8c8 | weis | 23 February 2006, 07:38:24 UTC | More up-to-date .cvsignore files. Renamed module options to command_options to avoid naming conflict with applications written with camlp3l. | 23 February 2006, 07:38:24 UTC |
b63b5f4 | weis | 20 February 2006, 12:58:53 UTC | Definition insidemake could loop for ever creating a non limited number of processes. | 20 February 2006, 12:58:53 UTC |
11f988b | zli | 07 February 2006, 21:05:50 UTC | ignore ocamlp3lcc | 07 February 2006, 21:05:50 UTC |
9fee4ff | zli | 07 February 2006, 20:59:31 UTC | *** empty log message *** | 07 February 2006, 20:59:31 UTC |
927fd08 | zli | 07 February 2006, 19:46:23 UTC | *** empty log message *** | 07 February 2006, 19:46:23 UTC |
c44b119 | zli | 07 February 2006, 14:43:09 UTC | *** empty log message *** | 07 February 2006, 14:43:09 UTC |
96a4b74 | zli | 07 February 2006, 14:42:15 UTC | *** empty log message *** | 07 February 2006, 14:42:15 UTC |
452faa4 | zli | 07 February 2006, 14:41:11 UTC | Bug fix: error in computing color value of seq skeleton. | 07 February 2006, 14:41:11 UTC |
60f8037 | zli | 07 February 2006, 14:38:59 UTC | A serious enhancement to avoide Stack Overflow problem when deep recursion happens in sequetial model. Now the exception for the end of stream becomes P3lstream.End_of_stream, to follow the standard OCaml naming convention and avoid memorize effort. | 07 February 2006, 14:38:59 UTC |
cb3df36 | zli | 07 February 2006, 14:36:20 UTC | Now the exception for the end of stream becomes P3lstream.End_of_stream, to follow the standard OCaml naming convention and avoid memorize effort. | 07 February 2006, 14:36:20 UTC |
c1c3cfc | zli | 07 February 2006, 14:32:49 UTC | Reducevector has never been implemented for graphics semantics (why?), now we have it! Bug fix: the captions of emittor/collector nodes. | 07 February 2006, 14:32:49 UTC |
7072fb6 | zli | 07 February 2006, 14:30:46 UTC | Eliminate a unused variable, thanks to the new warning feature of OCaml 3.09 | 07 February 2006, 14:30:46 UTC |
7d326e8 | zli | 07 February 2006, 14:26:31 UTC | Changes corresponding to the new OcamlP3l compilers | 07 February 2006, 14:26:31 UTC |
b53001b | zli | 07 February 2006, 14:21:44 UTC | New refinement of Makefiles, more generic and short :) | 07 February 2006, 14:21:44 UTC |
5e68e81 | zli | 07 February 2006, 14:17:34 UTC | The initial version of OcamlP3l basic toolset written in OCaml, as well as various improvements and bug-fix. | 07 February 2006, 14:17:34 UTC |
647d5b2 | weis | 19 January 2006, 17:19:12 UTC | New lazy stream semantics of OCamlP3l. | 19 January 2006, 17:19:12 UTC |
8dbaafe | fclement | 19 January 2006, 10:19:08 UTC | Add debug option to ocamlc. | 19 January 2006, 10:19:08 UTC |
d7d50d5 | zli | 21 November 2005, 16:20:20 UTC | A better (efficient) implementation of the tag mechanics in the unif function. It should be more efficient and more generic if I could got more time, but not now :) | 21 November 2005, 16:20:20 UTC |
c0cfd6e | zli | 21 November 2005, 16:16:47 UTC | Previously, you may raise any kind of exception when using P3lstream.of_fun to identify this is the end of produced stream. However this is not safe at all. By catching any kind of exception, it covers both the expected ending and those unexpected, and exit normally when something bad really happens. This will be difficult for the programmer to note a wrong answer and even more difficult to trace the reason. Now we require an explicit P3lstream.End_Of_Stream exception to identify the stream end when programming the argument function of P3lstream.of_fun. When running old source that doesn't follow this convention, you will probably see the raising of the exceptions defined by yourself to identify the stream end. Fortunately, it will also be accompanied by detailed explanation and suggestion. | 21 November 2005, 16:16:47 UTC |
6e81d9b | zli | 21 November 2005, 16:01:12 UTC | A bug is reported by Charles Bouillaguet on reduce skeleton. This bug is due to the reason that: reduce(seq) optimization has never been fully implemented, but there exists some residual code which attempts to do that is not commented away. Also note that the reduce skeleton itself is currently buggy, try to use mapvector when necessary. | 21 November 2005, 16:01:12 UTC |
addbeef | fclement | 14 November 2005, 18:12:22 UTC | Replace the use of P3lstream.of_list and P3lstream.to_list by P3lstream.of_fun and P3lstream.iter. | 14 November 2005, 18:12:22 UTC |
3df0de2 | fclement | 09 November 2005, 17:02:15 UTC | Rm. | 09 November 2005, 17:02:15 UTC |
ca73f86 | zli | 07 November 2005, 16:10:56 UTC | Add CVS "id" mark. | 07 November 2005, 16:10:56 UTC |
d4d3015 | zli | 07 November 2005, 16:07:32 UTC | *** empty log message *** | 07 November 2005, 16:07:32 UTC |
a7a4778 | zli | 07 November 2005, 15:52:40 UTC | To complete CVS "id" marks for several files. | 07 November 2005, 15:52:40 UTC |
1691c47 | zli | 07 November 2005, 15:50:48 UTC | The default port number is now 7749 to avoid confliction with various network applications (e.g. IRC, D&D, mldonkey etc.). | 07 November 2005, 15:50:48 UTC |
d578ece | zli | 07 November 2005, 15:35:29 UTC | The following bug fixed: The p3lroot node acts as both the main-stream sender and receiver. In the previous interpretation, this sending/receiving is executed sequently in one process. It will cause buffering problem (program suspend) if given data amount is sufficiently large. Hence we need to spawn a extra process(thread) for the p3lroot. | 07 November 2005, 15:35:29 UTC |
474bddc | zli | 28 October 2005, 09:37:31 UTC | New lazy stream model for sequential semantics. | 28 October 2005, 09:37:31 UTC |
9b94fe5 | zli | 03 September 2005, 00:11:51 UTC | Eliminate all pattern matching warnings during system building. | 03 September 2005, 00:11:51 UTC |
799caa1 | zli | 03 September 2005, 00:09:26 UTC | full refinement of makefiles | 03 September 2005, 00:09:26 UTC |
f56db61 | fclement | 05 August 2005, 13:38:28 UTC | Switched back to the fixed configuration file "config" since we have to set the number of workers before any parfun is applied to a mapvector or farm skeleton, any parfun application should be done outside the pardo scope, and options are only parsed as the first instruction of the pardo skeleton! | 05 August 2005, 13:38:28 UTC |
448ccc3 | fclement | 28 July 2005, 15:03:51 UTC | Last(?) bug: Meta-data as the number of workers for farm or mapvector skeletons must be known when defining the parfun's. Is this a bug from OCamlP3l? | 28 July 2005, 15:03:51 UTC |
3f1b6dc | fclement | 28 July 2005, 15:00:50 UTC | The official version is still the old one. | 28 July 2005, 15:00:50 UTC |
d02fecf | fclement | 28 July 2005, 14:58:37 UTC | Add aref. | 28 July 2005, 14:58:37 UTC |
4f73ea3 | fclement | 28 July 2005, 14:57:47 UTC | Banner. | 28 July 2005, 14:57:47 UTC |
bf81c3c | fclement | 28 July 2005, 14:56:31 UTC | *** empty log message *** | 28 July 2005, 14:56:31 UTC |
2683e57 | fclement | 25 July 2005, 18:20:18 UTC | *** empty log message *** | 25 July 2005, 18:20:18 UTC |
02f1f42 | fclement | 25 July 2005, 18:18:06 UTC | The new version compiles, but is still bugged! | 25 July 2005, 18:18:06 UTC |
9c46ff9 | fclement | 22 July 2005, 17:22:44 UTC | *** empty log message *** | 22 July 2005, 17:22:44 UTC |
2b72e02 | fclement | 22 July 2005, 17:22:02 UTC | Simplifying the code. | 22 July 2005, 17:22:02 UTC |
d66e12b | fclement | 22 July 2005, 17:21:19 UTC | Mutable fields can advantageously be replace by the use of "with" in this case. | 22 July 2005, 17:21:19 UTC |
60dd512 | fclement | 22 July 2005, 17:20:12 UTC | Nil is not a proper name for a contructor. | 22 July 2005, 17:20:12 UTC |
fcdf8a1 | fclement | 22 July 2005, 10:16:19 UTC | Truly unify the printing of the command line for all modes ;). | 22 July 2005, 10:16:19 UTC |
94666fa | fclement | 22 July 2005, 10:12:11 UTC | Bug fix. | 22 July 2005, 10:12:11 UTC |
6bb4b06 | fclement | 22 July 2005, 10:11:38 UTC | Unify the printing of the command line for all modes. | 22 July 2005, 10:11:38 UTC |
37750b9 | fclement | 22 July 2005, 10:10:43 UTC | Made generic ;). | 22 July 2005, 10:10:43 UTC |
82cb2ea | fclement | 22 July 2005, 10:09:48 UTC | Add a task to do about debugging color annotations. | 22 July 2005, 10:09:48 UTC |
a58fc46 | fclement | 22 July 2005, 09:40:42 UTC | Change version number. | 22 July 2005, 09:40:42 UTC |
fd70d64 | fclement | 21 July 2005, 22:00:32 UTC | Make some fields of type subdomain mutable and define extraction and insertion functions. | 21 July 2005, 22:00:32 UTC |
3fc4395 | weis | 21 July 2005, 18:39:45 UTC | New version 2.0. | 21 July 2005, 18:39:45 UTC |
925cff9 | weis | 21 July 2005, 18:38:44 UTC | Adding a default definition to OCAMLC and OCAMLOPT. | 21 July 2005, 18:38:44 UTC |
0407e82 | weis | 21 July 2005, 18:37:08 UTC | Now loc is _loc. | 21 July 2005, 18:37:08 UTC |
bfc867c | weis | 21 July 2005, 18:36:10 UTC | Better error treatment of unknown options: an error message is generated instead of silently ignoring them! | 21 July 2005, 18:36:10 UTC |
4ef1cbc | weis | 21 July 2005, 18:32:12 UTC | Sum rewriting and simplifications. | 21 July 2005, 18:32:12 UTC |
b46ac8a | fclement | 21 July 2005, 14:12:42 UTC | On the way to reorganize the coupling code. | 21 July 2005, 14:12:42 UTC |
85b0526 | fclement | 20 July 2005, 17:46:17 UTC | Add a temporary default config file (waiting for the correct implementation using Options module). | 20 July 2005, 17:46:17 UTC |
3d2d521 | fclement | 20 July 2005, 17:45:13 UTC | Sys.argv is not the correct way to catch arguments, it is better to use Options module, but it implies a complete reorganization of the coupling code (still to be done). | 20 July 2005, 17:45:13 UTC |
8a65d26 | weis | 20 July 2005, 12:41:33 UTC | New version. | 20 July 2005, 12:41:33 UTC |
76894f7 | weis | 20 July 2005, 11:46:47 UTC | A truely old ocamlp3lcc is suppressed. A brend new .cvsignore is added. | 20 July 2005, 11:46:47 UTC |
7bb6240 | fclement | 19 July 2005, 17:43:35 UTC | Change the computation of the complexity in the mk_config_files.sh script (bc may not be a good idea). | 19 July 2005, 17:43:35 UTC |
ede523a | fclement | 19 July 2005, 16:20:57 UTC | Simplify the computation of the complexity. | 19 July 2005, 16:20:57 UTC |
bd5eabc | fclement | 19 July 2005, 15:28:10 UTC | Change the display of the complexity. | 19 July 2005, 15:28:10 UTC |
a83a3ed | fclement | 19 July 2005, 15:17:20 UTC | Automatically generated configuration files should now be remove by "make clean-all". | 19 July 2005, 15:17:20 UTC |
8f4db0a | fclement | 19 July 2005, 15:11:53 UTC | Configuration files can now be automatically generated. | 19 July 2005, 15:11:53 UTC |
b3d1e09 | fclement | 19 July 2005, 15:10:58 UTC | Explain how generate configuration files. | 19 July 2005, 15:10:58 UTC |
baa12f1 | fclement | 19 July 2005, 15:10:34 UTC | Add a script to generate configuration files. | 19 July 2005, 15:10:34 UTC |
1c09e40 | fclement | 19 July 2005, 15:09:22 UTC | Add the iteration number into the stream data type. The config filename is now an optional command-line argument. | 19 July 2005, 15:09:22 UTC |
b683bcd | fclement | 19 July 2005, 15:06:15 UTC | Debug the numerical scheme (when dx <> dy). Change the computation of errors. | 19 July 2005, 15:06:15 UTC |
114712b | fclement | 18 July 2005, 17:51:03 UTC | Debug configure (version number is a float). | 18 July 2005, 17:51:03 UTC |
996e163 | fclement | 18 July 2005, 17:47:22 UTC | Update the README file. | 18 July 2005, 17:47:22 UTC |
647ea48 | fclement | 18 July 2005, 17:39:57 UTC | Add the computation (in poisson.c) and printing (in coupling.ml) of the relative and absolute errors. These can be used to tune the different configuration nmbers... Add new config files. | 18 July 2005, 17:39:57 UTC |
72da0a2 | fclement | 18 July 2005, 12:40:36 UTC | Add files to be ignored. | 18 July 2005, 12:40:36 UTC |
c359c7c | fclement | 18 July 2005, 12:39:38 UTC | Add Gnuplot module to take care of the gnuplot version number (syntax is slightly modified...). Update the banners. | 18 July 2005, 12:39:38 UTC |
82ec6bd | zli | 19 June 2005, 06:03:19 UTC | Sec.3 due. | 19 June 2005, 06:03:19 UTC |
77698e2 | zli | 18 June 2005, 22:54:21 UTC | Lexical errors fixed. | 18 June 2005, 22:54:21 UTC |
0b697ea | zli | 18 June 2005, 18:55:28 UTC | Sec.2, Sec.5.5 due. Probably can't finish the whole paper, but try to step as further as I can. | 18 June 2005, 18:55:28 UTC |
2ee7180 | zli | 18 June 2005, 10:35:08 UTC | First section due. | 18 June 2005, 10:35:08 UTC |
c902dfe | dicosmo | 17 June 2005, 19:47:57 UTC | Roberto's final touch. | 17 June 2005, 19:47:57 UTC |
5fab017 | zli | 30 May 2005, 11:41:40 UTC | Checking through the color section. | 30 May 2005, 11:41:40 UTC |
799118c | dicosmo | 26 May 2005, 18:44:17 UTC | first pass up to and excluding colors. | 26 May 2005, 18:44:17 UTC |
86072a5 | pelagatt | 21 April 2005, 14:35:43 UTC | Added a few files for figures in new manual | 21 April 2005, 14:35:43 UTC |
9966611 | pelagatt | 21 April 2005, 14:32:05 UTC | Updated manual to 2.0 version of Ocamlp3l CVS ---------------------------------------------------------------------- | 21 April 2005, 14:32:05 UTC |
6badd7e | fclement | 13 April 2005, 17:19:44 UTC | *** empty log message *** | 13 April 2005, 17:19:44 UTC |
a51cdc1 | fclement | 13 April 2005, 17:19:05 UTC | Added a configure script to handle the .depend file. | 13 April 2005, 17:19:05 UTC |
22e54ba | fclement | 13 April 2005, 06:21:05 UTC | Bug fixed in the Makefile, comment about gnuplot added in the README. | 13 April 2005, 06:21:05 UTC |
c59ed89 | fclement | 12 April 2005, 10:25:24 UTC | Added a banner in all important files. | 12 April 2005, 10:25:24 UTC |
7ce8265 | fclement | 12 April 2005, 09:53:54 UTC | Added .depend file to make sure it is there. | 12 April 2005, 09:53:54 UTC |
fd1dc68 | fclement | 12 April 2005, 09:37:43 UTC | Added a new example directory for a Domain Decomposition application. Once it is validated, the outdated MultiDomains version should be removed. | 12 April 2005, 09:37:43 UTC |
0de00ec | pelagatt | 08 April 2005, 13:03:24 UTC | Final files for submission HLPP05 | 08 April 2005, 13:03:24 UTC |
ca74bdc | pelagatt | 07 April 2005, 07:08:42 UTC | *** empty log message *** | 07 April 2005, 07:08:42 UTC |
492c71c | pelagatt | 07 April 2005, 07:03:00 UTC | Files regarding HLP05 paper soubmission | 07 April 2005, 07:03:00 UTC |
a509db0 | zli | 28 September 2004, 14:55:58 UTC | Bug reported by Martial Mancip, Thanks! | 28 September 2004, 14:55:58 UTC |
27b58a6 | dicosmo | 28 September 2004, 09:03:21 UTC | Fixe ocamlp3lcc | 28 September 2004, 09:03:21 UTC |
e0b24cf | zli | 20 September 2004, 15:08:45 UTC | A little patch to the problem caused by options compiling | 20 September 2004, 15:08:45 UTC |
c288c3b | dicosmo | 17 September 2004, 09:34:38 UTC | Added options.ml and options.mli to allow the user to insert option handling in the OcamlP3l programs. | 17 September 2004, 09:34:38 UTC |
b17de4c | dicosmo | 16 September 2004, 23:00:19 UTC | Added the possibility for the user to add command line parameters by using the Options module (stealed from Advi). Look at Examples/Tests/testoptions.ml for an example. | 16 September 2004, 23:00:19 UTC |
3752e59 | zli | 09 April 2004, 13:34:18 UTC | A small bug fixed. | 09 April 2004, 13:34:18 UTC |
5620fc3 | zli | 05 April 2004, 14:17:10 UTC | Remove a bug in the display sequence of farm-like skeleton. | 05 April 2004, 14:17:10 UTC |
1cce5b3 | zli | 05 April 2004, 12:21:45 UTC | Example Streams adapted. | 05 April 2004, 12:21:45 UTC |