mirror of
https://github.com/Mercury-Language/mercury.git
synced 2026-04-18 19:03:45 +00:00
083d376e6598628362ee91c2da170febd83590f4
17 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Zoltan Somogyi
|
07f877bc3f |
Carve term_context.m out of term.m.
library/term.m:
library/term_context.m:
As above.
Rename the term.context type as term_context.term_context, with
term.context now being defined as an equivalence type.
Replace the context_init function and predicate and the dummy_context_init
function with just one function: dummy_context. This name includes
the important part (the fact that it return a *dummy* context) and deletes
the nonimportant part (dummy contexts are just about never updated,
so the function does not really "initialize" them).
Reduce function/predicate pairs that do the same thing to just a function.
library/MODULES_DOC:
library/library.m:
Add the new module to the list of standard library modules.
NEWS:
Mention the new module, and the obsoleting of the moved predicates
and functions in term.m.
compiler/*.m:
library/*.m:
Conform to the changes above.
|
||
|
Zoltan Somogyi
|
d787ee9355 |
Store var_tables in proc_infos.
This fixes the performance problem reported in Mantis bug #562. compiler/hlds_pred.m: Instead of storing a varset and a vartypes in each proc_info, store just a var_table. Update the predicates that create or clone procedures accordingly. Where we had operations on proc_infos that had two versions, one operating on a varset/vartypes pair and one operating on var_table, keep only the latter, with the (shorter) name of the former. Delete the arity argument of proc_info_init, because the only valid value of that argument is the length of the list of the argument types. (In other words, this arg has been redundant all along.) Change the operations that create new variables in a procedure to get the caller to specify the (base) name of the new variable up front. Delete the unused predicate proc_info_ensure_unique_names. compiler/type_util.m: Due to the change above, we now construct var_tables during the construction of the HLDS. The code that does that needs to fill in the field that says whether the type of each variable in the table is a dummy type or not. However, at this time, the pass that decides type representations has not been run yet. The code of is_type_a_dummy used to throw an exception in such situations. Change this so that in such situations, is_type_a_dummy returns a placeholder, not-guaranteed-to-be-correct value. Document why this is ok. compiler/post_typecheck.m: Replace the placeholder values in vte_is_dummy fields in all the entries in the var_tables in all (valid) predicates with valid data. (If there are any invalid predicates, the compilation will fail anyway.) The clause_to_proc pass will copy these updated var_tables to be the initial var_tables in procedures. compiler/make_goal.m: Change the operations that create new variables in a procedure to get the caller to specify the (base) name of the new variable up front. This is simpler than the old method, which created new variables without a name, and had the caller give them a name as a separate operation. And since var_tables need this info, get the caller to also specify whether the type is a dummy, if the type is not a builtin type which is known not to be a dummy. compiler/var_table.m: Document the times when the types and is_dummy fields in var_table entries become meaningful. Fix a potential bug: when performing type substitutions in var_table entries, updating a variable's type may change whether that variable is a dummy or not, so recompute that info. It is quite possible that we *never* replace a nondummy type with a dummy type or vice versa, but in the absence of a convincing correctness argument for that proposition, better safe than sorry. Export the previously-private predicate transform_var_table to post_typecheck. Add code to implement the unused predicate deleted from hlds_pred.m: at the time I wrote it, I haven't yet realised that it was unused. The code I wrote here is therefore unused as well, so it is commented out. I did not delete it, because it may be useful later on. compiler/direct_arg_in_out.m: Don't make and split var_tables, since it is no longer needed. compiler/accumulator.m: compiler/add_class.m: compiler/add_clause.m: compiler/add_heap_ops.m: compiler/add_pred.m: compiler/add_special_pred.m: compiler/add_trail_ops.m: compiler/arg_info.m: compiler/build_mode_constraints.m: compiler/bytecode_gen.m: compiler/check_typeclass.m: compiler/clause_to_proc.m: compiler/closure_analysis.m: compiler/code_gen.m: compiler/code_loc_dep.m: compiler/complexity.m: compiler/continuation_info.m: compiler/cse_detection.m: compiler/ctgc.livedata.m: compiler/deep_profiling.m: compiler/default_func_mode.m: compiler/deforest.m: compiler/delay_construct.m: compiler/delay_partial_inst.m: compiler/dep_par_conj.m: compiler/det_analysis.m: compiler/det_report.m: compiler/distance_granularity.m: compiler/equiv_type_hlds.m: compiler/exception_analysis.m: compiler/float_regs.m: compiler/follow_code.m: compiler/goal_mode.m: compiler/goal_path.m: compiler/higher_order.m: compiler/hlds_out_pred.m: compiler/hlds_rtti.m: compiler/hlds_statistics.m: compiler/inlining.m: compiler/intermod.m: compiler/intermod_analysis.m: compiler/introduce_exists_casts.m: compiler/introduce_parallelism.m: compiler/lambda.m: compiler/lco.m: compiler/live_vars.m: compiler/liveness.m: compiler/loop_inv.m: compiler/mark_tail_calls.m: compiler/ml_accurate_gc.m: compiler/ml_args_util.m: compiler/ml_closure_gen.m: compiler/ml_gen_info.m: compiler/ml_proc_gen.m: compiler/mode_errors.m: compiler/mode_info.m: compiler/modecheck_goal.m: compiler/par_loop_control.m: compiler/pd_debug.m: compiler/pd_info.m: compiler/pd_util.m: compiler/polymorphism_info.m: compiler/post_typecheck.m: compiler/proc_gen.m: compiler/proc_requests.m: compiler/purity.m: compiler/push_goals_together.m: compiler/quantification.m: compiler/rbmm.add_rbmm_goal_infos.m: compiler/rbmm.live_variable_analysis.m: compiler/rbmm.points_to_analysis.m: compiler/rbmm.points_to_graph.m: compiler/rbmm.points_to_info.m: compiler/rbmm.region_liveness_info.m: compiler/rbmm.region_transformation.m: compiler/recompute_instmap_deltas.m: compiler/saved_vars.m: compiler/simplify_goal_unify.m: compiler/simplify_info.m: compiler/simplify_proc.m: compiler/size_prof.m: compiler/ssdebug.m: compiler/stack_alloc.m: compiler/stack_layout.m: compiler/stack_opt.m: compiler/stm_expand.m: compiler/store_alloc.m: compiler/structure_reuse.analysis.m: compiler/structure_reuse.direct.choose_reuse.m: compiler/structure_reuse.direct.detect_garbage.m: compiler/structure_reuse.domain.m: compiler/structure_reuse.indirect.m: compiler/structure_reuse.lbu.m: compiler/structure_reuse.lfu.m: compiler/structure_reuse.versions.m: compiler/structure_sharing.analysis.m: compiler/structure_sharing.domain.m: compiler/switch_detection.m: compiler/table_gen.m: compiler/tabling_analysis.m: compiler/term_constr_build.m: compiler/term_constr_initial.m: compiler/term_errors.m: compiler/term_pass1.m: compiler/term_pass2.m: compiler/trace_gen.m: compiler/trailing_analysis.m: compiler/try_expand.m: compiler/tupling.m: compiler/unneeded_code.m: compiler/untupling.m: compiler/unused_args.m: compiler/unused_imports.m: Conform to the changes above. Mostly this means - not passing a module_info to get a var_table out of a proc_info, but - having to pass a module_info to code that either constructs a var_table, or adds entries to a var_table (since we now need the type table to figure out whether variables' types are dummies). |
||
|
Zoltan Somogyi
|
4c9d04434a |
Classify pred_names into four categories.
compiler/pred_name.m:
Group pred_origins into four categories, giving each category its own
function symbol. The categories are
- predicates that contain, or may contain, code directly written
by the user;
- predicates created wholly by the compiler,
- predicates created by the compiler by transforming a predicate, and
- predicates created by the compiler by transforming a procedure.
We distinguish between the first two because we want to report
errors only in user-written code, since reporting an error that
the user is powerless to fix is not a good idea.
We distinguish between the last two just to allow us to store
the proc_id of the transformed procedure next to its pred_id.
compiler/higher_order.m:
Replace a boolean with value of a bespoke type.
Rename some predicates to avoid ambiguity.
compiler/*.m:
Conform to the changes above.
|
||
|
Zoltan Somogyi
|
1eb1e239b3 |
Move the pred_origin type to pred_name.m.
compiler/pred_name.m:
Add the pred_origin type and its component types, moved here from
hlds_pred.m and hlds_rtti.m.
Also, add the functions and predicates that convert pred_origins to
strings, moved here from hlds_pred.m, hlds_out_util.m and layout_out.m.
compiler/hlds_pred.m:
Delete the pred_origin type moved to pred_name.m.
compiler/hlds_rtti.m:
Delete the instance_method_constraints type moved to pred_name.m.
It has nothing to do with RTTI, and was not used in hlds_rtti.m itself.
The only data structure it is part of is pred_origin, though
it is used as temporary data by a few other modules.
compiler/hlds_out_util.m:
compiler/hlds_pred.m:
compiler/layout_out.m:
Delete the functions and predicates moved to pred_name.m.
compiler/*.m:
Conform to the change above, mostly by adding imports of pred_name.m,
in a few cases replacing hlds_pred.m or hlds_rtti.m.
|
||
|
Zoltan Somogyi
|
814aae5bb7 |
Add a distinguishing prefix to recompilation items.
compiler/recompilation.m:
Add a distinguishing prefix to the names of both types relating to
recompilation items, and to the names of their function symbols,
in order to avoid confusion between them and the items defined in
prog_item.m.
Update the names of the related functions.
compiler/add_pragma_type_spec.m:
compiler/equiv_type.m:
compiler/equiv_type_hlds.m:
compiler/module_qual.id_set.m:
compiler/module_qual.qualify_items.m:
compiler/proc_requests.m:
compiler/prog_item.m:
compiler/qual_info.m:
compiler/recompilation.check.m:
compiler/recompilation.usage.m:
compiler/recompilation.used_file.m:
compiler/recompilation.version.m:
Conform to the change above.
|
||
|
Zoltan Somogyi
|
5cada10369 |
Rename pred_table to pred_id_table.
compiler/pred_table.m:
As above. This addresses half of an old XXX, which is that the two types
defined in this module, predicate_table and pred_table (as it was then)
should have names that (a) say what they do, and (b) are distinct.
Addressing the other half requires a more descriptive but not-too-long
name to replace "predicate_table".
Rename the predicates that operate on the type to follow the name change.
Add a distinguishing prefix to the names of the fields of the
predicate_table type.
compiler/add_clause.m:
compiler/add_foreign_proc.m:
compiler/add_pragma.m:
compiler/add_pragma_tabling.m:
compiler/add_pragma_type_spec.m:
compiler/add_pred.m:
compiler/add_special_pred.m:
compiler/arg_info.m:
compiler/bytecode_gen.m:
compiler/clause_to_proc.m:
compiler/closure_gen.m:
compiler/cse_detection.m:
compiler/dead_proc_elim.m:
compiler/deep_profiling.m:
compiler/default_func_mode.m:
compiler/det_analysis.m:
compiler/det_util.m:
compiler/direct_arg_in_out.m:
compiler/distance_granularity.m:
compiler/export.m:
compiler/float_regs.m:
compiler/goal_mode.m:
compiler/granularity.m:
compiler/hlds_defns.m:
compiler/hlds_error_util.m:
compiler/hlds_module.m:
compiler/hlds_out_goal.m:
compiler/hlds_out_module.m:
compiler/hlds_out_util.m:
compiler/hlds_pred.m:
compiler/hlds_statistics.m:
compiler/implementation_defined_literals.m:
compiler/inlining.m:
compiler/intermod.m:
compiler/introduce_exists_casts.m:
compiler/introduce_parallelism.m:
compiler/lambda.m:
compiler/lco.m:
compiler/make_hlds_passes.m:
compiler/mark_tail_calls.m:
compiler/mercury_compile_llds_back_end.m:
compiler/ml_proc_gen.m:
compiler/mode_info.m:
compiler/modecheck_call.m:
compiler/modes.m:
compiler/oisu_check.m:
compiler/old_type_constraints.m:
compiler/passes_aux.m:
compiler/polymorphism.m:
compiler/polymorphism_post_copy.m:
compiler/post_typecheck.m:
compiler/pre_typecheck.m:
compiler/proc_gen.m:
compiler/proc_requests.m:
compiler/simplify_proc.m:
compiler/stm_expand.m:
compiler/structure_reuse.analysis.m:
compiler/structure_reuse.direct.m:
compiler/structure_sharing.analysis.m:
compiler/switch_detection.m:
compiler/table_gen.m:
compiler/term_constr_build.m:
compiler/term_constr_initial.m:
compiler/term_constr_util.m:
compiler/term_util.m:
compiler/termination.m:
compiler/typecheck.m:
compiler/typecheck_errors.m:
compiler/typecheck_info.m:
compiler/unused_args.m:
compiler/unused_imports.m:
compiler/xml_documentation.m:
Conform to the change in pred_table.m.
Refer to values of the renamed type using a consistent naming scheme.
When the affected code repeats the body of existing helper predicates
for lookup up a pred_info or proc_info, or updating a proc_info
inside a pred_info, or updating a pred_info inside a module_info,
call the helper predicate instead. This makes code shorter and less
cluttered, and the use of the helper predicates for updates automatically
ensures that we don't accidentally update a stale version of the relevant
table. (This has sometimes been a problem in the past.)
In a few places, carve a new predicate for processing one element
of a list out of an existing predicate for processing all list elements.
|
||
|
Zoltan Somogyi
|
d76c7bf617 |
Break up inst_util.m and mode_util.m.
This step significantly improves module cohesion.
compiler/inst_abstract_unify.m:
New module carved out of inst_util.m, which does abstract unifications
on insts.
compiler/inst_merge.m:
New module carved out of inst_util.m, which merges insts.
compiler/inst_lookup.m:
New module carved partly out of inst_util.m and partly out of mode_util.m,
which looks up insts in the module_info, and then possibly expands out
the result.
compiler/mode_test.m:
New module carved out of mode_util.m, whose predicates
perform tests on modes.
compiler/mode_top_functor.m:
New module carved out of mode_util.m, which computes top_functor_modes
from modes.
compiler/inst_mode_type_prop.m:
New module carved out of mode_util.m, which propagates type information
into both insts and modes.
compiler/recompute_instmap_deltas.m:
New module carved out of mode_util.m, which recomputes goals'
instmap_deltas.
compiler/inst_test.m:
Move here the predicates in inst_util.m that perform tests on insts.
compiler/inst_util.m:
compiler/mode_util.m:
Delete the code that this diff moves to other modules.
compiler/check_hlds.m:
Add the new modules to the check_hlds package, the package that also
contains inst_util.m and mode_util.m. (Some of these modules could
be argued to fit better in the hlds package, but moving them there
would not be desirable while they depend on code that is still in the
check_hlds package.)
compiler/notes/compiler_design.html:
Document the new modules.
compiler/add_pragma_tabling.m:
compiler/arg_info.m:
compiler/bytecode_gen.m:
compiler/closure_analysis.m:
compiler/complexity.m:
compiler/deep_profiling.m:
compiler/deforest.m:
compiler/dep_par_conj.m:
compiler/det_report.m:
compiler/direct_arg_in_out.m:
compiler/distance_granularity.m:
compiler/equiv_type_hlds.m:
compiler/error_msg_inst.m:
compiler/fact_table.m:
compiler/float_regs.m:
compiler/follow_code.m:
compiler/goal_util.m:
compiler/higher_order.m:
compiler/hlds_pred.m:
compiler/hlds_rtti.m:
compiler/inlining.m:
compiler/inst_match.m:
compiler/inst_user.m:
compiler/instmap.m:
compiler/intermod.m:
compiler/interval.m:
compiler/introduce_exists_casts.m:
compiler/lambda.m:
compiler/lco.m:
compiler/liveness.m:
compiler/lookup_util.m:
compiler/loop_inv.m:
compiler/mark_tail_calls.m:
compiler/ml_args_util.m:
compiler/ml_code_util.m:
compiler/ml_foreign_proc_gen.m:
compiler/ml_unify_gen_construct.m:
compiler/ml_unify_gen_util.m:
compiler/mode_constraints.m:
compiler/mode_errors.m:
compiler/modecheck_call.m:
compiler/modecheck_coerce.m:
compiler/modecheck_goal.m:
compiler/modecheck_unify.m:
compiler/modecheck_util.m:
compiler/modes.m:
compiler/oisu_check.m:
compiler/par_conj_gen.m:
compiler/pd_util.m:
compiler/post_typecheck.m:
compiler/pragma_c_gen.m:
compiler/proc_requests.m:
compiler/prog_rep.m:
compiler/push_goals_together.m:
compiler/rbmm.region_transformation.m:
compiler/saved_vars.m:
compiler/simplify_goal_switch.m:
compiler/simplify_proc.m:
compiler/size_prof.m:
compiler/ssdebug.m:
compiler/stack_opt.m:
compiler/stm_expand.m:
compiler/stratify.m:
compiler/structure_reuse.versions.m:
compiler/structure_sharing.domain.m:
compiler/superhomogeneous.m:
compiler/table_gen.m:
compiler/term_constr_build.m:
compiler/term_pass2.m:
compiler/term_util.m:
compiler/tupling.m:
compiler/unify_gen_construct.m:
compiler/unify_gen_util.m:
compiler/unique_modes.m:
compiler/unneeded_code.m:
compiler/untupling.m:
compiler/unused_args.m:
Conform to the changes above by importing the required new modules,
sometimes in addition to inst_util.m or mode_util.m, but more usually
instead of them.
|
||
|
Zoltan Somogyi
|
1f328fe18b |
More recompilation type specializations.
compiler/recompilation.m:
Introduce a new enum type, used_item_type, which has equivalents
of those item_types for which we collect used item sets. This allows us
to delete several calls to unexpected.
Rename type_abstract_item to type_name_item and type_body_item to
type_defn_item.
compiler/recompilation.used_file.m:
Tighten the definition of the resolved_functor type by changing the types
of some fields from generic (e.g. item_name) to specific (e.g. type_ctor).
Specify that the arity recorded for predicates and functions is
pred_form_arity, since that is what the code that actually resolves
references in the pred table uses, even though some other parts of
the code took it to be user arity, which is different for functions.
Mark one of these differences with XXXs, since they are likely to be bugs.
Fix another, since in that case it is *clear* that it was a bug.
Also give some of the function symbols clearer names, and put their
arguments in their conventional order.
compiler/prog_data.m:
Add a type for use in recompilation.used_file.m.
compiler/equiv_type.m:
compiler/equiv_type_hlds.m:
compiler/module_qual.id_set.m:
compiler/module_qual.qualify_items.m:
compiler/proc_requests.m:
compiler/qual_info.m:
compiler/recompilation.check.m:
compiler/recompilation.usage.m:
compiler/recompilation.version.m:
Conform to the changes above.
In some cases, give some predicates and/or variables more meaningful names,
or convert clause lists into explicit disjunctions. (In some cases,
the latter revealed arguments that all clauses ignored.)
|
||
|
Zoltan Somogyi
|
99334e8469 |
Switch to structured item sequence numbers.
compiler/prog_data.m:
Define a data type that encodes the distinction between valid
and dummy item sequence numbers in the type. Previously, different
parts of the compiler expressed this distinction in two different ways,
either using "-1" to represent a dummy sequence number, or using "no"
in a maybe(int) type.
compiler/prog_item.m:
Use the new type instead of plain "int" as the sequence number in items.
compiler/hlds_pred.m:
Use the new type instead of plain "int" as the sequence number
in cur_user_decl_infos. Document the fact that the presence of a
cur_user_decl_info in a pred_info does NOT guarantee a valid
item sequence number.
compiler/add_foreign_proc.m:
When adding a foreign_proc to the HLDS, pass a whole item_pragma_info,
not its components. (This change is what this diff started as, before
the item_seq_num changes overwhelmed it.)
compiler/accumulator.m:
compiler/add_class.m:
compiler/add_clause.m:
compiler/add_mutable_aux_preds.m:
compiler/add_pragma.m:
compiler/add_pragma_tabling.m:
compiler/add_pred.m:
compiler/add_solver.m:
compiler/add_special_pred.m:
compiler/check_typeclass.m:
compiler/comp_unit_interface.m:
compiler/decide_type_repn.m:
compiler/default_func_mode.m:
compiler/hlds_clauses.m:
compiler/intermod.m:
compiler/item_util.m:
compiler/lambda.m:
compiler/make_hlds_passes.m:
compiler/par_loop_control.m:
compiler/parse_class.m:
compiler/parse_dcg_goal.m:
compiler/parse_inst_mode_defn.m:
compiler/parse_item.m:
compiler/parse_module.m:
compiler/parse_mutable.m:
compiler/parse_pragma.m:
compiler/parse_pragma_analysis.m:
compiler/parse_pragma_foreign.m:
compiler/parse_pragma_tabling.m:
compiler/parse_type_defn.m:
compiler/parse_type_repn.m:
compiler/parse_types.m:
compiler/proc_requests.m:
compiler/prog_mutable.m:
compiler/split_parse_tree_src.m:
compiler/stm_expand.m:
compiler/style_checks.m:
compiler/table_gen.m:
Conform to the changes above.
|
||
|
Zoltan Somogyi
|
df9420c3e6 |
Flatten the unify_mode structure.
compiler/hlds_goal.m:
Change the unify_mode structure from
unify_modes_lhs_rhs(from_to_insts(LI, LF), from_to_insts(RI, RF))
to
unify_modes_li_lf_ri_rf(LI, LF, RI, RF)
This requires fewer memory allocations (1 vs 3) and less memory
(4 words vs 6), though the performance improvement is too small
to measure.
It should also require writing fewer function symbols in code.
compiler/instmap.m:
compiler/mode_util.m:
For each utility predicate that works with from_to_insts, provide
a version that works with the separate insts contained in it.
Delete the from_to_insts version if no longer needed.
compiler/prog_mode.m:
Delete utility predicates on from_to_insts that are not needed anymore.
compiler/accumulator.m:
compiler/add_pred.m:
compiler/bytecode_gen.m:
compiler/common.m:
compiler/const_prop.m:
compiler/deep_profiling.m:
compiler/delay_partial_inst.m:
compiler/dep_par_conj.m:
compiler/equiv_type_hlds.m:
compiler/erl_unify_gen.m:
compiler/float_regs.m:
compiler/format_call.m:
compiler/goal_util.m:
compiler/higher_order.m:
compiler/hlds_out_goal.m:
compiler/hlds_out_mode.m:
compiler/interval.m:
compiler/lambda.m:
compiler/lco.m:
compiler/make_goal.m:
compiler/ml_unify_gen_construct.m:
compiler/ml_unify_gen_util.m:
compiler/modecheck_goal.m:
compiler/modecheck_unify.m:
compiler/polymorphism.m:
compiler/proc_requests.m:
compiler/prog_rep.m:
compiler/rbmm.region_transformation.m:
compiler/simplify_goal_call.m:
compiler/simplify_goal_scope.m:
compiler/simplify_goal_switch.m:
compiler/size_prof.m:
compiler/stm_expand.m:
compiler/term_util.m:
compiler/unify_gen_construct.m:
compiler/unify_gen_util.m:
compiler/unused_args.m:
Conform to the changes above.
|
||
|
Zoltan Somogyi
|
f82a2082e1 |
Unify and compare terms as MR_Unsigned when possible.
compiler/unify_proc.m:
We have long generated HLDS code that led to C code such as
succeeded = ((MR_Integer) X) == ((MR_Integer) Y)
when the shared type of X and Y is an enum. Generate similar code,
only with MR_Unsigned, when this type is NOT an enum, but nevertheless
all values of the type fit into one word, because all the non-constant
function symbols have all their arguments packed next to the primary tag.
Likewise, we have long generated HLDS code that compared two values of
an enum type by casting them to an integer and comparing the integers.
Do this (again with MR_Unsigned) for a non-enum du type if either
- it consists of exactly one non-constant functor, all of whose args
are packed next to the ptag and are comparable as unsigned, or
- it consists of exactly one constant functor and one non-constant functor,
with the same condition holding for the args of the latter, if the
representation of the constant is zero, the representation of
the non-constant cannot be zero, and the non-constant follows the
constant in the desired comparison order.
These are the cases in which cast-to-unsigned-and-compare is guaranteed
to yield the same results as the code we used to generate. Document why
the same comparison technique would not work in other cases.
compiler/clause_to_proc.m:
Both the new optimization and the previous change to unify_proc.m
do bulk unifications. These require not just that both arguments
be ground (which we did test), but also that we are in the standard
<in,in> mode of unification, mode id 0, in which we do NOT know anything
about the arguments beyond the fact that they are ground.
If we *do* know e.g. that a field in X must have value Xf but that
same field in Y must have value Yf, and Xf != Yf, then the semidet
bulk unification code we generate is wrong; we should generate code
whose determinism is failure.
Fix this bug (revealed by thinking about the applicability of the new
optimization) by using the bulk-comparison version only for mode id 0.
Don't pass around the module_info, since with the fix we do not need it.
compiler/proc_requests.m:
Don't pass the module_info to clause_to_proc.m.
|
||
|
Zoltan Somogyi
|
8303b503b6 |
Unify and compare packed args in bulk when possible.
compiler/unify_proc.m:
Try to optimize the code we generate for unification and comparison
predicates when a function symbol's arguments include sub-word-sized
arguments packed together into a word.
For unify predicates, generate code to test whether the two words
at the same offset in the terms being unified are equal. This works
regardless of whether the arguments are signed or unsigned.
For compare predicates, generate code to compare the two words
at the same offset in the terms being compared *if* all the arguments
in the terms being compared are unsigned. This works because we put
the earlier arguments in the more significant bit positions. But if
some of the arguments are signed, then divide the argument word
in sequences of zero or more unsigned arguments separated by signed
arguments. We then generate code that compares any contiguous sequences
of unsigned arguments in bulk, while comparing each signed field
separately.
Do the bulk unification and comparison via foreign_proc goals generated
inline. This works only when we are generating C, but this is ok because
we pack sub-word-sized arguments into a word only when generating C.
We do the comparison of signed sub-word-sized fields (int8, int16 or int32)
via foreign_proc goals generated inline as well. Doing them using unify
goals would work as well, but would be less efficient in general. This is
because having N such arguments in a function symbols requires storing
only one value across calls for each term being compared (the term itself)
when generating foreign_procs, but would require storing N values across
calls (the values of the sub-word-sized signed arguments) when generating
unifications. Generating inline foreign_procs is effectively a manual
application of the optimization implemented by saved_vars.m.
library/private_builtin.m:
Add the builtin predicates that unify_proc.m now generates calls to.
We should never need their bodies, but the compiler does need to know
the declarations of all predicates mentioned in inline foreign_procs.
configure.ac:
runtime/mercury_conf.h.in:
Define either MR_MERCURY_IS_32_BITS or MR_MERCURY_IS_64_BITS depending
on the word size. Make the configured value of MR_BITS_PER_WORD available
to C code.
mdbcomp/program_representation.m:
Register the new builtin predicates as no_typeinfo_builtins, i.e.
builtins whose arguments' types contain type variables, that nevertheless
should *not* be passed the typeinfos of the actual types bound to those
type variables.
compiler/hlds_clauses.m:
Bulk unification of arguments works only when all the arguments involved
are initially ground. The optimized unification clauses we can now generate
are thus appropriate only for <in,in> unifications. (Technically, they
*would* work for unifications for which the function symbol arguments
involved in bulk unify operations are ground even if some other arguments
are initially free, but that distinction is too hard to make, compared
to the extremely small performance gain that would be available
if we *could* make that distinction.)
Provide a way for unify_proc.m to mark a clause as being for use either
in the <in,in> modes of unifications (for the optimized version using bulk
unifications), or as in all other modes of unifications (for a version in
which that optimization has been disabled).
Replace two boolean fields in clauses_infos with bespoke types, for
greater readability and reliability. These are a remnant of a different
way to differentiate <in,in> vs non-<in,in> clauses that I ultimately
decided against. These bespoke types are independent of the main change
in this diff, but there is no reason to undo their use.
compiler/clause_to_proc.m:
When copying clauses to procedure bodies inside type-specific unify
predicates, pay attention to the markers that unify_proc.m put on
those clauses about which are for <in,in> modes and which are for
non-<in,in> modes.
To make this possible, make our callers pass us extra information.
compiler/options.m:
Add a bootstrapping option that governs whether unify_proc.m should
try to apply the new optimization.
Give an option that governs comparisons of function symbols for Erlang
a name that reflects that fact.
compiler/hlds_pred.m:
Fix a misleading predicate name.
compiler/add_class.m:
compiler/add_clause.m:
compiler/add_foreign_proc.m:
compiler/add_pragma_type_spec.m:
compiler/add_pred.m:
compiler/dead_proc_elim.m:
compiler/det_report.m:
compiler/erl_code_gen.m:
compiler/handle_options.m:
compiler/higher_order.m:
compiler/hlds_out_module.m:
compiler/hlds_out_pred.m:
compiler/hlds_statistics.m:
compiler/intermod.m:
compiler/mercury_compile_front_end.m:
compiler/ml_proc_gen.m:
compiler/modecheck_unify.m:
compiler/proc_gen.m:
compiler/proc_requests.m:
compiler/purity.m:
compiler/resolve_unify_functor.m:
compiler/structure_reuse.indirect.m:
compiler/structure_sharing.analysis.m:
compiler/tabling_analysis.m:
compiler/type_constraints.m:
compiler/typecheck.m:
compiler/unused_args.m:
Conform to the changes above.
|
||
|
Zoltan Somogyi
|
5d5c167d1d |
Make unify_proc.m more cohesive.
compiler/unify_proc.m:
compiler/add_special_pred.m:
Move the predicates in unify_proc.m that were not concerned with
the creation of clauses for unify, index and compare predicates
to add_special_pred.m. Since they *were* concerned with the creation
of special predicates, that is where they belonged.
Delete some no-longer-relevant comments in add_special_pred.m.
compiler/higher_order.m:
compiler/proc_requests.m:
Conform to the change above.
|
||
|
Zoltan Somogyi
|
fb97df69ed |
Make "compute type representations" a separate pass.
The ultimate purpose of this diff is to prepare for future improvements
in type representations, allowing values of some data types to be represented
more compactly than up to now.
The main way this diff does that is by creating a separate pass for deciding
how values of each type should be represented. We have traditionally decided
data representations for each type as its type definition was processed
during the make_hlds pass, but these decisions were always tentative,
and could be overridden later, e.g. when we processed foreign_type or
foreign_enum pragmas for the type. This dispersed decision making algorithm
is hard to understand, and therefore to change.
This diff centralizes decisions about type representations in a separate
pass that does nothing else. It leaves the algorithm distributed among
several files (du_type_layout.m, make_tags.m, and add_foreign_enum.m) for now,
to make reviewing this diff easier, but soon after it is committed I intend
to move all the relevant code to du_type_layout.m, to centralize the decision
code in "space" as well as in time.
For the reason why this pass runs before any of the semantic analysis
passes, instead of after all of them as I originally intended and as we
discussed on m-dev in late october 2017, see the big comment at the start of
du_type_layout.m.
As per another part of that same discussion on m-dev, this diff
makes a start on implementing a new type of item, the type_repn item,
which is intended *only* to be used in compiler-generated interface files,
*not* in source files. It is only a start because we can use these items
only *after* the creation of a separate type representation decision pass,
and this diff is already very big. The code for making the compiler understand
these items will be added later. The code for generating them will be added
later still, once the code for understanding them has been installed on
all our systems.
Since I was going to be working on the affected code anyway, this diff
also carries out two other decisions that came out of that discussion:
- the deletion of the ability to reserve a tag in a type for HAL,
either via a compiler option or via a pragma, and
- the deletion of the ability to represent a functor using the address
of a statically allocated object (which we haven't used and won't use,
because it slows down accesses to *all the other functors* of the type).
compiler/mercury_compile_front_end.m:
Invoke the new pass for making decisions about type representations
after the make_hlds pass. (We used to do only the final part of it then.)
Fix a bad dump stage name.
Add an extra check for what it means for a module to be error free.
Make a sub-switch explicit.
compiler/hlds.m:
compiler/make_hlds.m:
Move the modules that implement the new pass from the make_hlds package
to the hlds package, to give the compiler's top level access to them.
Make the same move for the modules that the new pass's modules need.
Since they are now part of hlds, they cannot reach into make_hlds,
and I think this is a cleaner solution than forwarding predicates.
Delete some forwarding predicates that are no longer needed.
compiler/notes/compiler_design.html:
Document the updated location of the moved modules.
Add an XXX to note a place where the documentation has not been
updated in the past.
compiler/du_type_layout.m:
Add code to implement the new pass.
Keep the algorithm for deciding type representations as close
to the previously used algorithm as possible, since this diff
is already big enough. (The previous algorithm was scattered across
add_type.m, add_foreign_enum.m, and make_hlds_passes.m.)
Simplifications and optimizations will come later, after this module
is merged with make_tags.m and with (at least) the foreign_enum half of
add_foreign_enum.m.
compiler/make_tags.m:
Keep the functionality of this module, which does both the first part
of deciding type representations (tentatively assigning tags to functors,
an assignment that may be overridden later), and the last part (packing
multiple adjacent less-than-word-sized enum args into a single word,
if possible.), but simplify it where possible, and note possibilities
for further improvements.
compiler/add_foreign_enum.m:
This module has two halves, one dealing with foreign_enum pragmas
and one dealing with foreign_export_enum pragmas.
Change the half that deals with foreign_enum pragmas to just build
a data structure that du_type_layout.m will need to make its decisions,
this structure being a map from type_ctors to the foreign enum
specification applicable to the current target language. Include
in this structure a component that add_foreign_enum.m itself can use
to report better error messages for duplicate foreign_enum pragmas;
this component records, for each type_ctor and language, the context
of the previous foreign_enum pragma for that combo.
Change the input for the half that deals with foreign_export_enum pragmas
to reflect the fact that it is invoked by du_type_layout.m after all
decisions about type representations have already been made.
compiler/add_special_pred.m:
Move this module from the make_hlds package to the hlds package,
since the code that adds special preds for type is now called from
du_type_layout.m.
Change the names of predicates to make clear whether they add
only the declaration of a predicate, only its definition, or both.
Don't try to pre-guess whether the implementation of a type's
compare predicate will need an index predicate. Let the code
that generates calls to the index predicate both declare and define
the index predicate. This change removes the potential for
inconsistencies between the two pieces of code.
compiler/add_pred.m:
Move this module from the make_hlds package to the hlds package,
since add_special_pred.m needs access to it.
compiler/add_type.m:
When adding a type definition to the HLDS, don't try to decide
its representation. Any such decision was tentative anyway, due
to the possibility of e.g. the later processing of foreign_type
or foreign_enum pragmas for the type. Likewise, don't try to
create the special (unify, compare) predicates for the type.
Leave both tasks to the du_type_layout pass.
Likewise, don't try to pack the representation of types, or record
no_tag types in the table of no_tag types, during the post-processing
pass either; leave both of these to du_type_layout as well.
Rename the predicate that post_processes type definitions to reflect
the two tasks left for it to do.
compiler/prog_data.m:
Do not store width information about the arguments of those data
constructors in the parse tree. That information is not computed
until later; until then, it was always filled in with dummy values.
(But see hlds_data.m below.)
Use bespoke types to represent the presence or absence of user-specified
unify and compare predicates.
Change the representation of data constructors to use a single "maybe"
type, not two lists, to denote the presence or absence of existentially
typed arguments.
Give the HLDS the ability to hold representation information about
abstract types that in the future we will get from type_repn items
in the defining modules' interface files.
Delete the uses_reserved_tag type, since we never use reserved tags
anymore.
compiler/prog_item.m:
Add the new type_repn item type, which is not used yet.
Delete the reserve_tag pragma.
Fix an earlier mistake in the wording of a context message.
compiler/hlds_data.m:
Put all the fields of hlds_du_type (the type definition variant dealing
with discriminated union types) that deal with type representation
issues in a single "maybe" field that is set to "no" before the
type representation decision pass has been run.
Add new type, constructor_repn, that stores the same information as the old
constructor type (defined in prog_data.m), PLUS the information
describing how terms with that data constructor are stored.
Likewise, add a new type ctor_arg_rep, which likewise stores
the widths of each constructor argument. When we implement
argument reordering, we would store the offset of the arg as well.
Since the parse tree representations of constructors and their arguments
don't store representation information anymore, the cons_table they
are stored in doesn't either. Make the lookup of representation information
for a given constructor possible by adding a map to the new "maybe" field
of hlds_du_type.
Provide some utility predicates.
Optimize some existing predicates.
Rename some types to better reflect their meaning.
compiler/hlds_module.m:
Provide a slot in the module_info for storing the information
gathered by make_hlds.m that is needed by the new pass.
compiler/make_hlds_separate_items.m:
When we see either a foreign_enum or a foreign_export_enum pragma,
return values of a bespoke type for them (a type defined in
hlds_module.m), instead of an item_pragma. This makes handling them
considerably easier.
compiler/make_hlds_passes.m:
With the changes in this diff, adding a type to the HLDS won't
decide its representation. Therefore delete the code that used
to loop over foreign_export_enum pragmas; in the absence of
the final type representation information, it won't work right.
Record the information that the du_type_layout pass will need
in the module_info.
compiler/add_pragma.m:
Delete the code for passing on foreign_enum and foreign_export_enum
pragmas to add_foreign_enum.m; they are now passed to add_foreign_enum.m
by du_type_layout.m.
Move a utility predicate to make_hlds_error.m, to allow add_foreign_enum.m
to call it.
compiler/make_hlds_error.m:
Add the utility predicate moved from add_pragma.m.
Move the module from the make_hlds to the hlds package.
compiler/module_qual.m:
Provide a mechanism for recording error messages about e.g. undefined
types without recording that we found an undefined type. This sounds
strange, but there is a valid use case.
When a type definition declares a functor's argument to be of an
undefined type, that error is usually fatal; we stop the compiler
from proceeding even to typechecking, since the typechecker will
probably abort with a map lookup failure. Most other references
to undefined types are similarly fatal for the same reason. However,
if e.g. a foreign_export_enum pragma refers to an undefined type,
that error *won't* be visible to the typechecker, and therefore
won't crash it. The error will still cause the compiler to exit
without generating any target language code, but at least it will be
able to run the typechecker and other semantic analysis passes.
Without this change, the compiler will report only one error in
the ee_invalid.m test case; with it, it reports *every* error
in the test case expected output.
compiler/module_qual.qualify_items.m:
Use the capability describe above for undefined types in
foreign_export_enum pragmas.
compiler/module_qual.qual_errors.m:
Delete a (somewhat incorrect) copy of a predicate in prog_item.m,
to reduce code duplication.
compiler/prog_type.m:
Add ways to represent abstract types whose representations are nevertheless
known (from type_repn items in the defining modules' interface files)
to be notag or dummy types. This will be needed to fix Mantis bug #441,
a fix that will probably be one of the first later changes to build
on this diff.
Delete a type moved to type_util.m.
compiler/type_util.m:
Provide extra versions of some predicates, with the difference between
the old and the new versions being that one requires type representations
to have been decided already, and the other one does not.
Move the definition of the ctor_defn type here from prog_type.m,
since prog_type.m itself does not use it, but type_util.m does.
Give some predicates more meaningful names.
compiler/parse_type_defn.m:
Simplify the code for parsing type definitions, to make it easier
to reuse to parse type_repn items.
Add a sanity check that requires existential constraints to have
*some* existential variables to apply to.
Allow "type_is_representable_in_n_bits" as a synonym for
"type_is_abstract_enum", since in the future we want to be able to pack
e.g. multiple int8s, not just multiple enums, into a single word.
Generate more specific error messages for some classes of malformed input.
compiler/parse_type_repn.m:
New module to parse type_repn items.
compiler/polymorphism.m:
Make some predicates that operate on type constructors take
the type constructors themselves as input arguments, not a whole type
*using* that type constructor. Put the arguments of those predicates
in a more standard order.
Note that some predicates don't belong in this module.
compiler/special_pred.m:
Make the code that decides whether a special predicate for a type
constructor can be defined lazily avoid using type representation
information. (Actually, we now make decisions about lazy vs eager
definitions after type representation is available, but that was
not so in an earlier version of this change, and the new code
is more robust.)
compiler/unify_proc.m:
When we decide to generate code for a compare predicate that needs
the type to have an index predicate, don't presume that the index
predicate has already been declared and defined; instead, declare
and define it then and there. (Index predicates are *never* called
from anywhere else.)
Pack the information needed to define a special predicate
into a single structure, to simplify the above.
Since the creation of a clause for a compare predicate may now require
the declaration and definition of an index predicate, the module_info
field of the unify_proc_info is now a writeable field.
Give some predicates and function symbols more meaningful names.
Note some problems with the existing code.
compiler/add_class.m:
compiler/add_clause.m:
compiler/add_foreign_proc.m:
compiler/add_mode.m:
compiler/add_mutable_aux_preds.m:
compiler/add_pragma_tabling.m:
compiler/add_pragma_type_spec.m:
compiler/add_solver.m:
compiler/check_typeclass.m:
compiler/code_info.m:
compiler/comp_unit_interface.m:
compiler/ctgc.selector.m:
compiler/ctgc.util.m:
compiler/default_func_mode.m:
compiler/det_report.m:
compiler/equiv_type.m:
compiler/equiv_type_hlds.m:
compiler/erl_code_gen.m:
compiler/export.m:
compiler/foreign.m:
compiler/get_dependencies.m:
compiler/goal_expr_to_goal.m:
compiler/hhf.m:
compiler/higher_order.m:
compiler/hlds_code_util.m:
compiler/hlds_out_module.m:
compiler/inst_check.m:
compiler/inst_test.m:
compiler/inst_util.m:
compiler/intermod.m:
compiler/item_util.m:
compiler/make_hlds_warn.m:
compiler/ml_accurate_gc.m:
compiler/ml_simplify_switch.m:
compiler/ml_type_gen.m:
compiler/ml_unify_gen.m:
compiler/mlds_to_cs.m:
compiler/mlds_to_java.m:
compiler/mode_util.m:
compiler/modecheck_goal.m:
compiler/module_qual.collect_mq_info.m:
compiler/modules.m:
compiler/parse_item.m:
compiler/parse_pragma.m:
compiler/parse_tree.m:
compiler/parse_tree_out.m:
compiler/parse_tree_out_pragma.m:
compiler/post_term_analysis.m:
compiler/proc_requests.m:
compiler/prog_item_stats.m:
compiler/qual_info.m:
compiler/recompilation.check.m:
compiler/recompilation.usage.m:
compiler/recompilation.version.m:
compiler/resolve_unify_functor.m:
compiler/rtti.m:
compiler/rtti_out.m:
compiler/rtti_to_mlds.m:
compiler/simplify_goal_ite.m:
compiler/stack_opt.m:
compiler/state_var.m:
compiler/structure_reuse.direct.choose_reuse.m:
compiler/superhomogeneous.m:
compiler/switch_gen.m:
compiler/switch_util.m:
compiler/table_gen.m:
compiler/term_constr_build.m:
compiler/term_norm.m:
compiler/trailing_analysis.m:
compiler/type_constraints.m:
compiler/type_ctor_info.m:
compiler/typecheck.m:
compiler/unify_gen.m:
compiler/untupling.m:
compiler/unused_imports.m:
compiler/write_module_interface_files.m:
compiler/xml_documentation.m:
Conform to the changes above.
tests/invalid/Mmakefile:
Disable the reserve_tag test case, as it is not applicable anymore.
tests/invalid/exported_foreign_enum.{m,err_exp}:
tests/invalid/pragma_qual_error.{m,err_exp}:
Delete reserve_tag pragmas from these test cases, and its effects
from the expected outputs.
tests/invalid/bad_foreign_type.err_exp:
tests/invalid/bigtest.err_exp:
tests/invalid/foreign_enum_invalid.err_exp:
tests/invalid/type_lhs_var.err_exp:
tests/invalid/uu_type.err_exp:
tests/invalid/where_abstract_enum.err_exp:
tests/invalid/where_direct_arg.err_exp:
Expect the updated messages for some errors.
tests/valid/Mmake.valid.common:
tests/valid/Mmakefile:
Disable any reserve_tag test cases, as they are not applicable anymore.
|
||
|
Zoltan Somogyi
|
12eb32ac1d |
Make clause_to_proc a separate pass.
Copying goals from clauses to procedures has to be done after type checking
(which operates on predicates) but before mode checking (which operates on
procedures). Previously, this copying was done within one of the passes
between type and mode checking, to reduce the overall number of passes.
For many years now, it was done within the polymorphism pass.
However, this arrangement has two undesirable consequences.
The first, which was the reason for this change, is that it makes it
much harder to see the effects of the polymorphism pass in HLDS dumps,
since the difference between HLDS dumps before and after the polymorphism
pass contains not just the effects of the polymorphism transformation itself,
but also the population of the goal fields of proc_infos.
The second, which I discovered while doing the change, is that the
polymorphism pass itself is split. It does the vast bulk of its work
before copying clauses to procs, but it does leave some tasks to be done
*after* copying clauses to procs. I don't *know* whether this is
by design or by accident, but I strongly suspect the latter.
compiler/mercury_compile_front_end.m:
Make copying clauses to procedures its own pass.
Make the part of polymorphism that runs *after* copying clauses
to procedures its own pass as well.
compiler/clause_to_proc.m:
Add a predicate that mercury_compile_front_end can call to do its job
on every predicate in the module. The setup we use for imported procedures
is the same as what polymorphism.m used to do for such procedures.
Make some predicate names more specific.
compiler/polymorphism.m:
Don't copy clauses to procs in the middle of the polymorphism
transformation.
Add a predicate that mercury_compile_front_end.m can call to do the
post-copying part of the polymorphism transformation.
compiler/introduce_exists_casts.m:
Add a predicate that polymorphism.m can call to do this module's part of
the post-copying part of the polymorphism transformation.
compiler/mode_ordering.m:
compiler/modes.m:
compiler/proc_requests.m:
Conform to the changes above.
|
||
|
Zoltan Somogyi
|
e107c3b959 | Give some predicates in clause_to_proc.m better names. | ||
|
Zoltan Somogyi
|
b0eac2e9a7 |
Carve proc_requests.m out of unify_proc.m.
compiler/unify_proc.m:
compiler/proc_requests.m:
The old unify_proc.m contained two kinds of predicates: those that
generated the source code for unify, index and compare predicates,
and those that managed the queue of procedure requests. The only link
between the two is that the latter calls one of the former.
This diff moves the predicates that deal with the queue of procedure
requests into a new module, proc_requests.m. This increases the cohesion
of both new modules.
compiler/check_hlds.m:
compiler/notes/compiler_design.html:
Include the new module in the check_hlds package.
compiler/hlds_module.m:
compiler/modecheck_call.m:
compiler/modecheck_unify.m:
compiler/modes.m:
compiler/simplify_goal_unify.m:
Import proc_requests.m instead of unify_proc.m.
|