schedcat.git/native/src, branch master

Implement LP-based OMIP blocking analysis

2013-07-12T12:19:22+00:00

Precision fix: user deliberate rounding in LP result conversion

2013-02-12T11:55:16+00:00

Use proper rounding when converting blocking terms to avoid rounding
issues.  This could cause some (very rare) off-by-one bugs when
incorrectly truncating an objective result very close to the next
largest integer.

Add LP-based blocking analysis for FMLP+

2013-02-12T11:55:16+00:00

Add LP-based blocking analysis for MPCP

2013-02-12T11:55:16+00:00

Add generic LP support for shared-memory protocols

2013-02-12T11:55:16+00:00

Allow LPs in which not all variables are part of the objective function

2013-02-12T11:55:16+00:00

Add support for dumping LPs to iostreams

2013-02-12T11:55:16+00:00

Quite handy when debugging LPs...

C++ LP analysis: add LP debugging helper

2013-02-12T11:55:16+00:00

Dump a LP solution to a stream for inspection.

Disable creation of new variables after creation of the objective function

2013-02-12T11:55:16+00:00

This can reveal bugs in constraints that accidentally include Ti.

Implement CPLEX integration via C API

2013-02-12T11:55:15+00:00

Using the plain, old C API seems to be *much* faster.

single:
lp_dpcp_bounds::cpu_costs: total=19569.9ms last=93.2767ms average=88.9543ms count=220
lp_dflp_bounds::cpu_costs: total=21537.3ms last=102.459ms average=97.8969ms count=220