Thu, 21 Feb 2013 18:21:40 +0100 wenzelm more explicit session dependency, for improved parallel performance of HOL-UNITY test session -- NB: separate 'theories' sections are sequential;
Thu, 21 Feb 2013 16:00:48 +0100 wenzelm merged
Thu, 21 Feb 2013 15:35:09 +0100 wenzelm removed obsolete tags;
Thu, 21 Feb 2013 10:52:14 +0100 wenzelm highest priority for proofs with unknown / very short timing -- recover original scheduling with parallel_proofs_reuse_timing = false;
Thu, 21 Feb 2013 16:36:19 +0100 blanchet tuned misleading message
Thu, 21 Feb 2013 15:10:04 +0100 wenzelm merged
Wed, 20 Feb 2013 19:57:17 +0100 wenzelm option parallel_proofs_reuse_timing controls reuse of log information -- since it is not always beneficial for performance;
Wed, 20 Feb 2013 18:04:44 +0100 wenzelm prefer outdegree in comparison again (cf. 88c96e836ed6) -- NB: big jobs might hide behind small ones in this naive queuing scheme;
Wed, 20 Feb 2013 15:22:22 +0100 wenzelm more tight representation of command timing;
Wed, 20 Feb 2013 13:03:50 +0100 wenzelm prefer comparison of session timing, if this is known already;
Wed, 20 Feb 2013 11:40:30 +0100 wenzelm proper check of Proof.is_relevant (again, cf. c3e99efacb67 and df8fc0567a3d);
Wed, 20 Feb 2013 00:00:42 +0100 wenzelm support nested Thy_Syntax.element;
Tue, 19 Feb 2013 21:44:37 +0100 wenzelm back to non-schematic 'sublocale' and 'interpretation' (despite df8fc0567a3d) for more potential parallelism;
Tue, 19 Feb 2013 20:19:21 +0100 wenzelm help JVM to cope with large symbolic structures;
Tue, 19 Feb 2013 17:55:26 +0100 wenzelm improved scheduling of forked proofs, based on elapsed time estimate (from last run via session log file);
Tue, 19 Feb 2013 17:02:52 +0100 wenzelm read logs from failed sessions as well;
Tue, 19 Feb 2013 16:49:40 +0100 wenzelm recover timing information from old log files;
Tue, 19 Feb 2013 14:47:57 +0100 wenzelm suppress timing message in full PIDE protocol -- this is for batch build;
Tue, 19 Feb 2013 13:57:13 +0100 wenzelm support for build passing timings from Scala to ML;
Tue, 19 Feb 2013 12:58:32 +0100 wenzelm support for prescient timing information within command transactions;
Tue, 19 Feb 2013 10:55:11 +0100 wenzelm emit command_timing properties into build log;
Thu, 21 Feb 2013 12:22:26 +0100 blanchet generate Isar proof if Metis appears to be too slow
Thu, 21 Feb 2013 12:22:26 +0100 blanchet swap slices so that the last slice is more complete (for minimization)
Thu, 21 Feb 2013 12:22:26 +0100 blanchet tuning
Wed, 20 Feb 2013 17:42:20 +0100 blanchet ensure all conjecture clauses are in the graph -- to prevent exceptions later
Wed, 20 Feb 2013 17:31:28 +0100 blanchet generalize syntax of SPASS proofs
Wed, 20 Feb 2013 17:15:06 +0100 blanchet tweaked hack some more
Wed, 20 Feb 2013 17:12:21 +0100 blanchet more simplifying constructors
Wed, 20 Feb 2013 17:05:24 +0100 blanchet remove needless steps from refutation graph -- these confuse the proof redirection algorithm (and are needless)
Wed, 20 Feb 2013 16:21:04 +0100 blanchet more precise error
Wed, 20 Feb 2013 15:43:51 +0100 blanchet improved hack
Wed, 20 Feb 2013 15:26:19 +0100 blanchet upgraded to Alt-Ergo 0.95
Wed, 20 Feb 2013 15:12:38 +0100 blanchet don't pass chained facts directly to SMT solvers -- this breaks various invariants and is never necessary
Wed, 20 Feb 2013 14:47:19 +0100 blanchet trust preplayed proof in Mirabelle
Wed, 20 Feb 2013 14:44:00 +0100 blanchet added case taken out by mistake
Wed, 20 Feb 2013 14:21:17 +0100 blanchet tuning (removed redundant datatype)
Wed, 20 Feb 2013 14:10:51 +0100 blanchet minimize SMT proofs with E if Isar proofs are desired and Metis managed to preplay
Wed, 20 Feb 2013 13:04:03 +0100 blanchet honor linearization option also in the evaluation driver
Wed, 20 Feb 2013 10:54:13 +0100 blanchet got rid of rump support for Vampire definitions
Wed, 20 Feb 2013 10:45:23 +0100 blanchet optimize Isar output some more
Wed, 20 Feb 2013 10:45:01 +0100 blanchet turn off more evil Vampire options to facilitate Isar proof generation
Wed, 20 Feb 2013 09:58:28 +0100 blanchet fixed typo in option name
Wed, 20 Feb 2013 08:56:34 +0100 blanchet tuning
Wed, 20 Feb 2013 08:44:24 +0100 blanchet use new skolemizer only if some skolems have two or more arguments -- otherwise the old skolemizer cannot get the arg order wrong
Wed, 20 Feb 2013 08:44:24 +0100 blanchet slacker comparison for Skolems, to avoid trivial equations
Wed, 20 Feb 2013 08:44:24 +0100 blanchet auto-minimizer should respect "isar_proofs = true"
Wed, 20 Feb 2013 08:44:24 +0100 blanchet made "isar_proofs" a 3-way option, to provide a way to totally disable isar_proofs if desired
Wed, 20 Feb 2013 08:44:24 +0100 blanchet alias for people like me
Tue, 19 Feb 2013 19:44:10 +0100 haftmann dropped spurious left-over from 0a2371e7ced3
Tue, 19 Feb 2013 17:01:06 +0100 blanchet avoid using "smt" for minimization -- better use the prover itself, since then Sledgehammer gets to try metis again and gives the opportunity to output an Isar proof -- and show Isar proof as fallback for SMT proofs
Tue, 19 Feb 2013 15:37:42 +0100 blanchet interpret "max_facts" argument in a slice-dependent fashion, instead of forcing the same number of facts to all slices
Tue, 19 Feb 2013 15:03:36 +0100 kuncar delete also predicates on relations when hiding an implementation of an abstract type
Tue, 19 Feb 2013 13:37:07 +0100 blanchet reintroduced crucial sorting accidentally lost in 962190eab40d
Tue, 19 Feb 2013 13:27:33 +0100 blanchet compile
Tue, 19 Feb 2013 13:21:49 +0100 blanchet provide two modes for MaSh driver: linearized or real visibility
Mon, 18 Feb 2013 18:34:55 +0100 blanchet implement (more) accurate computation of parents
Mon, 18 Feb 2013 18:34:54 +0100 blanchet tuning
Mon, 18 Feb 2013 12:16:27 +0100 smolkas split isar_step into isar_step, fix, assms; made isar_proof explicit; register fixed variables in ctxt and auto_fix terms to avoid superfluous annotations
Mon, 18 Feb 2013 12:16:02 +0100 smolkas simplified byline, isar_qualifier
Mon, 18 Feb 2013 11:33:43 +0100 blanchet tuned code: factored out parent computation
(0) -30000 -10000 -3000 -1000 -300 -100 -60 +60 +100 +300 +1000 +3000 +10000 tip