Commit Graph

139 Commits

Author SHA1 Message Date
Ricardo Rocha
4df6ed1394 extra built-in predicate for frozen choice point support 2010-05-18 17:34:52 +01:00
Vítor Santos Costa
be2a3a635e L constants are evil in 64 bits 2010-05-11 12:25:49 +01:00
Ricardo Rocha
16c1056a01 tabling support for the common global trie is now available by default
(macro GLOBAL_TRIE was removed). Use 'yap_flag(tabling_mode,MODE)' to
set the storing mode for all tabled predicates to MODE ('local_trie' or
'global_trie). Alternatively, use 'tabling_mode(PRED,MODE)' to set the
default tabling mode of predicate PRED to MODE ('local_trie' or
'global_trie'). The default mode for all tabled predicates is 'local_trie'.
2010-04-23 04:22:10 +01:00
Ricardo Rocha
7eee5a5b56 new built-in predicates to access OPTYap statistics related to memory
usage.
2010-04-20 03:59:48 +01:00
Ricardo Rocha
c221c2ff85 renaming some OPTYap files in order to have a more consistent name organization 2010-04-19 02:34:08 +01:00
Ricardo Rocha
73d210c991 reducing the number of macros for error checking in the OPTYap files 2010-04-18 04:59:11 +01:00
Ricardo Rocha
467dd91b32 new built-in predicate global_trie_statistics/0. 2010-04-16 03:07:37 +01:00
Ricardo Rocha
ddb1cd8604 adding two new modes to the yap_flag/2 and tabling_mode/2 built-in
predicates: 'local_trie' mode and 'global_trie' mode.
2010-04-16 02:08:06 +01:00
Ricardo Rocha
e122f2ca8d Global trie support: atomic terms (vars, integers and atoms) are now stored
in the local tries (and not in the global trie). This required major changes
to the trie instructions in order to unify the use of the auxiliary stack
organization for the terms in the local tries and in the global trie.
2010-04-15 01:09:59 +01:00
Vítor Santos Costa II
8efec0b26e put ThOr back in shape. 2010-04-09 11:56:07 +01:00
Ricardo Rocha
ab8fee4e08 common global trie for subtems (seems to be working correctly !) 2010-04-05 03:34:02 +01:00
Ricardo Rocha
9c3bb73bab common global trie (version GT-ST: global trie for subterms) 2010-04-05 03:31:12 +01:00
Ricardo Rocha
6c222b8e87 use a macro (stack_terms_limit) to delimit the top of the auxiliary term
stack
2010-04-03 16:05:38 +01:00
Ricardo Rocha
b3e2c562a2 misplaced #endif declaration 2010-04-03 06:31:34 +01:00
Ricardo Rocha
52a2b49263 new support for the common global trie (version GT-ST: global trie for subterms) 2010-04-03 05:58:14 +01:00
Ricardo Rocha
55855d8d90 fix abolish_table/1 abolish_all_tables/0 for predicates with arity 0 2010-03-29 15:41:59 +01:00
Ricardo Rocha
3742f09c10 missing brackets in condition using macro new_dependency_frame() 2010-03-15 14:21:16 +00:00
Ricardo Rocha
ee0022bb47 checking for frozen segments in the trail stack in now done using the
macro IN_BETWEEN(Yap_TrailBase, ref, Yap_TrailTop)
2010-03-12 20:02:32 +00:00
Ricardo Rocha
6dbc6f09fa new built-in predicate abolish_all_frozen_choice_points/0 2010-03-09 14:48:20 +00:00
Vítor Santos Costa
500d6ac453 fix LOCAL with tabling only. 2010-02-12 16:25:57 +00:00
Vítor Santos Costa
5937a72b76 try to get rid of a warning. 2010-02-12 16:25:43 +00:00
Vitor Santos Costa
95e596a8a1 make LOCAL a register 2010-02-10 09:22:20 +00:00
Vitor Santos Costa
0e85839517 remove debugging info. 2010-02-10 09:21:04 +00:00
vsc
c95c89cfe3 oops... 2010-02-04 00:00:33 +00:00
vsc
eb6ac94e45 use mapfield in current directory, not in / 2010-02-03 22:54:24 +00:00
vsc
9c100ae09d fix INCREMENTAL_COPY
delay P until Q has copied trailed bindings (should not be needed, but hey!)
don't trust TR when copying bindings.
2010-02-03 22:52:45 +00:00
Vitor Santos Costa
9c85f67eee fix one more CP field to offset. 2010-02-02 23:42:15 +00:00
Vítor Manuel de Morais Santos Costa
9d1f0cd361 incremental copying with or-threads. 2010-02-01 20:05:01 +00:00
Vítor Manuel de Morais Santos Costa
d0fdcba231 make it compile
support non-incremental copy.
2010-01-29 12:28:23 +00:00
Vítor Manuel de Morais Santos Costa
4b8d655437 cut cps are different from everyone else? 2010-01-28 15:44:31 +00:00
Vítor Manuel de Morais Santos Costa
5b47d00cbc replace by macros wherever possible. 2010-01-28 15:43:52 +00:00
Vitor Santos Costa
995e8f6dc7 fix cut handling with threads
do not compile or.threadengine unless yapor on.
2010-01-27 11:01:28 +00:00
Vitor Santos Costa
d2bf8c717d make sure it compiles under YAPOR alone. 2010-01-22 22:30:11 +00:00
Vitor Santos Costa
6da35f6e7b count the number of alternatives left plus the actual choicepoint!! 2010-01-22 22:29:40 +00:00
Vitor Santos Costa
44a418bd04 small orp fixes. 2010-01-15 12:04:01 +00:00
Vitor Santos Costa
70c762c162 fix undefined predicate (made Logtalk loop). 2010-01-15 01:03:34 +00:00
Vítor Manuel de Morais Santos Costa
a498dc1265 icompile table+orp 2010-01-14 17:38:39 +00:00
Vítor Manuel de Morais Santos Costa
5ef31fbfd2 experiment in or-parallel threads. 2010-01-14 15:58:19 +00:00
Vitor Santos Costa
5f0f91208d fix compilation of threads+tabling 2009-12-18 00:19:49 -02:00
Vitor Santos Costa
d9016fa650 init_local should be called Yap_init_local to avoid clashesh. 2009-11-19 08:24:38 +00:00
Vitor Santos Costa
0179998d60 change Heap.h to YapHeap.h
rename $live to prolog/0 (for SWI compatibility).
2009-10-23 14:22:17 +01:00
Vítor Manuel de Morais Santos Costa
d810cbd41d we're moving down, not up, in both cases! 2009-10-19 16:39:40 +01:00
Ricardo Rocha
7120e79dd6 include the library shm.h only when needed 2009-09-29 00:56:58 +01:00
Ricardo Rocha
7d841065e9 support for the common global trie (version GT-T: global trie for terms) 2009-09-27 02:31:31 +01:00
Costa Vitor
1e16a9c339 modern machines do not guarantee sequential consistency. 2009-09-07 20:42:19 -05:00
Costa Vitor
f877e78785 make YAP compile with --enable-or-parallelism 2009-08-19 21:53:49 -05:00
Ricardo Rocha
9dc0618a12 make the support for compact lists the default configuration for the tabling engine 2009-08-12 14:32:12 +01:00
Ricardo Rocha
23d44f959d include cp_h field in deterministic choice points, otherwise we get in trouble 2009-08-09 03:38:07 +01:00
Ricardo Rocha
4f6676e5af fix garbage collector for deterministic tabling (still very
experimental)
2009-08-08 05:02:27 +01:00
Ricardo Rocha
4bdf8bf991 if the current generator choice point is the topmost choice point and
the current call is deterministic then the current answer is
deterministic and we can perform an early completion and remove the
current generator choice point
2009-08-08 03:53:51 +01:00