[m-dev.] for review: exposing the table data structure for benchmarking

[Tyson Dowd]

On 24-Nov-1998, Zoltan Somogyi <zs at cs.mu.OZ.AU> wrote:
> 
> This is for Tyson.
> 
> Estimated hours taken: 16
> 
> Reorganize the handling of global data structures, and expose the table
> pointers for tabled predicates to allow the tables to be reset to empty
> by hand-written C code for benchmarking purposes.
> 
> compiler/hlds_module.m:
> 	Introduce a new submodule for dealing with global data.
> 	At the moment it deals with (a) layout structures and (b)
> 	tabling pointers.
> 
> compiler/continuation_info.m:
> 	Remove the data structures and predicates whose equivalents are now
> 	in hlds_module.
> 
> compiler/llds.m:
> 	Introduce a new kind of item, that of compiler-generated variables
> 	whose value is not defined by rvals. At the moment, the only such
> 	item is a tabling pointer variable.
> 
> compiler/mercury_compile.m:
> 	Include the new kind of item in the generated LLDS.
> 
> compiler/table_gen.m:
> 	Use the new kind of item instead of a static variable declaration
> 	in pragma C code to hold the tabling pointer.
> 
> 	Remove lots of spaces at ends of lines, since many of these screw up
> 	paragraph commands in vi.
> 
> compiler/base_type_layout.m:
> compiler/bytecode_gen.m:
> compiler/code_gen.m:
> compiler/code_util.m:
> compiler/dependency_graph.m:
> compiler/hlds_out.m:
> compiler/llds_out.m:
> compiler/mercury_to_mercury.m:
> compiler/switch_gen.m:
> compiler/transform_llds.m:
> compiler/unify_gen.m:
> 	Handle the changes in the data structures.

Looks fine.

-- 

Because I dislike being quoted I lie almost constantly when talking 
about my work.
		-- Terry Gilliam

Tyson Dowd   <tyson at tyse.net>   http://tyse.net/