src/Pure/proofterm.ML
Wed, 02 May 2018 23:34:40 +0200 wenzelm tuned -- slightly smaller future closure size;
Tue, 24 Apr 2018 18:10:08 +0200 wenzelm more ambitious parallelism (in contrast to a8ee8e4884ec): pri = 1 ensures that internal proof tasks are executed before the already forked theory outline with pri = 0;
Tue, 24 Apr 2018 11:07:18 +0200 wenzelm less ambitious parallelism: avoid problems with HOL-Proofs and threads=2 (congestion with many thousands futures and rather dense heap);
Mon, 19 Feb 2018 18:12:28 +0100 wenzelm tuned: more parallel;
Thu, 22 Jun 2017 21:10:13 +0200 wenzelm consolidate proofs more simultaneously;
Thu, 22 Jun 2017 15:20:32 +0200 wenzelm more informative task_statistics;
Sun, 09 Apr 2017 19:03:55 +0200 wenzelm tuned signature -- prefer qualified names;
Fri, 16 Dec 2016 19:07:16 +0100 wenzelm consolidate nested thms with persistent result, for improved performance;
Fri, 16 Dec 2016 14:06:31 +0100 wenzelm tuned signature -- more abstract type thm_node;
Thu, 15 Dec 2016 22:22:45 +0100 wenzelm tuned signature;
Wed, 14 Dec 2016 18:22:18 +0100 wenzelm more careful derivation_closed / close_derivation;
Wed, 14 Dec 2016 15:48:18 +0100 wenzelm tuned whitespace;
Mon, 12 Sep 2016 20:24:42 +0200 wenzelm tuned -- fewer warnings;
Sat, 06 Aug 2016 17:39:21 +0200 wenzelm tuned signature;
Thu, 07 Apr 2016 12:08:02 +0200 wenzelm prefer regular context update, to allow continuous editing of Pure;
Tue, 08 Mar 2016 21:07:47 +0100 haftmann provide explicit hint concering uniqueness of derivation
Mon, 23 Mar 2015 19:43:03 +0100 wenzelm local fixes may depend on goal params;
Wed, 26 Nov 2014 20:05:34 +0100 wenzelm renamed "pairself" to "apply2", in accordance to @{apply 2};
Fri, 21 Mar 2014 20:33:56 +0100 wenzelm more qualified names;
Thu, 18 Jul 2013 13:12:54 +0200 wenzelm immutable theory values with full stamp record of every update (increase of stamp size for HOL: 20000 -> 100000, JinjaThreads: 65000 -> 300000) -- minimal measurable impact on inference kernel performance;
Sun, 30 Jun 2013 11:37:34 +0200 wenzelm backout dedd7952a62c: static "proofs" value within theory prevents later inferencing with different configuration;
Thu, 27 Jun 2013 23:17:26 +0200 wenzelm manage option "proofs" within theory context -- with minor overhead for primitive inferences;
Sun, 23 Jun 2013 16:47:45 +0200 wenzelm support for XML data representation of proof terms;
Fri, 12 Apr 2013 14:54:14 +0200 wenzelm tuned signature;
Wed, 13 Feb 2013 15:40:59 +0100 wenzelm discontinued home-made sharing for proof terms -- leave memory management to the Poly/ML RTS;
Thu, 24 Jan 2013 21:18:30 +0100 wenzelm avoid extra fork for fulfill_proof_future whenever possible -- without proof terms it merely doubles the number of proof tasks redundantly, by piggy-backing another 10 microseconds task;
Mon, 07 Jan 2013 21:49:59 +0100 wenzelm no fork from draft thy -- avoid potential for crash via classrel_proof/arity_proof;
Mon, 19 Nov 2012 20:23:47 +0100 wenzelm theorem status about oracles/futures is no longer printed by default;
Sun, 16 Oct 2011 18:48:30 +0200 wenzelm added Term.dummy_pattern conveniences;
Sat, 20 Aug 2011 20:00:55 +0200 wenzelm more direct balanced version Ord_List.unions;
Sat, 20 Aug 2011 19:21:03 +0200 wenzelm reverted to join_bodies/join_proofs based on fold_body_thms to regain performance (escpecially of HOL-Proofs) -- see also aa9c1e9ef2ce and 4e2abb045eac;
Sat, 20 Aug 2011 18:11:17 +0200 wenzelm tuned future priorities (again);
Sat, 20 Aug 2011 16:06:27 +0200 wenzelm clarified fulfill_norm_proof: no join_thms yet;
Sat, 20 Aug 2011 15:52:29 +0200 wenzelm added Future.joins convenience;
Fri, 19 Aug 2011 21:40:52 +0200 wenzelm incremental Proofterm.join_body, with join_thms step in fulfill_norm_proof -- avoid full graph traversal of former Proofterm.join_bodies;
Fri, 19 Aug 2011 18:01:23 +0200 wenzelm tuned;
Wed, 10 Aug 2011 19:45:57 +0200 wenzelm avoid OldTerm operations -- with subtle changes of semantics;
Wed, 10 Aug 2011 19:21:28 +0200 wenzelm avoid OldTerm operations -- with subtle changes of semantics;
Wed, 10 Aug 2011 16:05:14 +0200 wenzelm future_job: explicit indication of interrupts;
Tue, 09 Aug 2011 22:30:33 +0200 wenzelm misc tuning and clarification;
Tue, 09 Aug 2011 21:48:36 +0200 wenzelm tuned whitespace;
Wed, 13 Jul 2011 21:44:15 +0200 wenzelm recovered some runtime sharing from d6b6c74a8bcf, without the global memory bottleneck;
Mon, 11 Jul 2011 22:55:47 +0200 wenzelm tuned signature -- corresponding to Scala version;
Thu, 09 Jun 2011 20:22:22 +0200 wenzelm tuned signature: Name.invent and Name.invent_names;
Thu, 09 Jun 2011 17:51:49 +0200 wenzelm simplified Name.variant -- discontinued builtin fold_map;
Thu, 09 Jun 2011 16:34:49 +0200 wenzelm discontinued Name.variant to emphasize that this is old-style / indirect;
Wed, 08 Jun 2011 15:56:57 +0200 wenzelm more robust exception pattern General.Subscript;
Thu, 03 Feb 2011 19:27:04 +0100 wenzelm tuned comments;
Thu, 03 Feb 2011 19:21:12 +0100 wenzelm clarified Proofterm.proofs_enabled;
Mon, 31 Jan 2011 23:02:53 +0100 wenzelm tuned signature;
Mon, 31 Jan 2011 22:57:01 +0100 wenzelm support named tasks, for improved tracing;
Mon, 31 Jan 2011 21:54:49 +0100 wenzelm more direct Future.bulk, which potentially reduces overhead for Par_List;
Fri, 24 Sep 2010 15:53:13 +0200 wenzelm modernized structure Ord_List;
Wed, 01 Sep 2010 15:33:59 +0200 haftmann replaced Table.map' by Table.map
Wed, 02 Jun 2010 21:39:35 +0200 wenzelm always unconstrain thm proofs;
Wed, 02 Jun 2010 11:09:26 +0200 wenzelm normalize and postprocess proof body in a separate future, taking care of platforms without multithreading (greately improves parallelization in general without the overhead of promised proofs, cf. usedir -q 0);
Tue, 01 Jun 2010 11:30:57 +0200 berghofe merged
Tue, 01 Jun 2010 11:01:16 +0200 berghofe - outer_constraints with original variable names, to ensure that argsP is consistent with args
Mon, 31 May 2010 21:06:57 +0200 wenzelm modernized some structure names, keeping a few legacy aliases;
Fri, 14 May 2010 19:53:36 +0200 wenzelm added Proofterm.unconstrain_thm_proof for Thm.unconstrainT -- loosely based on the version by krauss/schropp;
less more (0) -100 -60 tip