This reduces the size of io.m a bit. The other reason for moving the
code is to allow modules that occur in a dependency cycle with io.m
to read results from io.error_util.trans_opt, whereas they may have
been prevented from reading io.trans_opt.
library/io.m:
library/io.error_util.m:
Move is_error, throw_on_error and other similar predicates to a
new undocumented submodule of io.m.
Also move the support functions/predicates used by those predicates.
Make io.make_io_error_from_system_error and
io.make_io_error_from_windows_error call the implementations in
io.error_util.m.
library/MODULES_UNDOC:
library/library.m:
List the new submodule as undocumented.
library/benchmarking.m:
library/bitmap.m:
library/dir.m:
library/io.call_system.m:
library/io.file.m:
Import the new submodule.
compiler/generate_dep_d_files.m:
When the new code to handle the new --trans-opt-deps-spec option
found an error, it printed out an error message and then stopped.
Replace this with code that
- continues processing after each error, to allow more than one error
to be reported by a single compiler invocation, and
- does not any print error messages, but instead returns
an error_spec for each error.
In addition, add code to look for and report
- any references in the trans_opt_deps_spec to module names
that do not occur in the deps graph that the trans_opt_deps_spec
is intended to apply to; and
- any duplicate entries in module name lists.
To make the above possible, modify the representation of
trans_opt_deps_specs to include information that is needed
only for the generation of meaningful error messages.
Note some potential problems with the code that outputs the deps graph
as a .dot file.
compiler/error_spec.m:
Add a mechanism to gather all the contexts in error_specs, for use
by new code in generate_dep_d_files.
compiler/mercury_compile_main.m:
Write out the error_specs (if any) returned by generate_dep_d_files.m.
This option lets the user provide a file containing information to
remove some edges from the trans-opt dependency graph, i.e. the graph
used to determine which .trans_opt files to read when making a module's
own .trans_opt file.
The reason to remove edges from the graph is to break dependency cycles.
.trans_opt files for modules within an SCC have to be made one after
another, instead of in parallel. For example, the standard library
contains one large SCC due to circular imports, so making .trans_opt
files turns out to be a bottleneck when building the standard library
(on a machine with sufficient parallelism available).
Furthermore, the user had no control over which modules in an SCC
could read the .trans_opt files of other modules in the same SCC.
If a and b happened to import each other, the compiler would always
break the cycle by allowing a to read b.trans_opt, but not allow b to
read a.trans_opt, simply based on module names. The new option lets the
user break the cycle in a way that may improve analysis results.
compiler/options.m:
Add the --trans-opt-deps-spec option.
compiler/generate_dep_d_files.m:
If the option --trans-opt-deps-spec FILE is used, use the
information given in the file to remove some edges from the
trans-opt dependency graph.
If --generate-module-order is passed, also output the module order
computed from the trans-opt dependency graph to a file.
Users may find this a useful starting point when writing their own
spec file.
compiler/write_deps_file.m:
Add a field to intermod_deps to hold the (user-adjusted)
trans-opt dependencies.
Add a type, maybe_include_trans_opt_rule, to indicate whether or not
to include a trans_opt_deps rule in a mmake dependency file.
Separate the case when a trans_opt_deps rule is to be written,
indicating where the dependencies come from.
When automatically rewriting a .d file after producing target code,
etc., write the trans_opt_deps rule with the same list of dependencies
as the old .d file.
compiler/mercury_compile_make_hlds.m:
Conform to maybe_include_trans_opt_rule change.
Rename some variables for clarity.
library/std_util.m:
Make std_util.pow/3 throw an exception if it is called with a negative
integer as its second argument.
tests/hard_coded/Mmakefile:
tests/hard_coded/func_exp.{m,exp}:
Add a test of pow/3.
NEWS:
Announce the above change.
Convert it to Markdown.
README.Java:
Add a .md extension.
Update many of the details in this file.
Add a table-of-contents.
Break up the FAQ into separate named sections.
README.md:
Conform to the above change of name.
The implementation of digraph.rtc was incorrect (as demonstrated in the
new test case), which meant that digraph.tc was also incorrect.
library/digraph.m:
Fix the implementation of rtc (reflexive transitive closure):
- Following the algorithm used in digraph.cliques, it needs to
traverse the graph G in *reverse* depth-first order.
- To find the clique containing a vertex X, it needs to do a DFS on
the *reversed* graph to find the vertices with a path to X.
The vertices that were previously unvisited will be members of
the same clique as X.
- Previously it found the "followers" of the elements of the clique,
and the followers of those followers, then added edges from the
members of the current clique to those followers. However, that
only includes vertices two steps removed from the clique.
I have fixed it to add edges to *all* vertices reachable from
members of the clique.
Add straightforward implementations of tc and rtc for comparison.
Add some comments.
tests/hard_coded/Mmakefile:
tests/hard_coded/digraph_tc.exp:
tests/hard_coded/digraph_tc.inp:
tests/hard_coded/digraph_tc.m:
Add test case.
NEWS:
Announce the fixes.
tools/bootcheck:
Allow the executable name used for sed to be overridden
in the environment. (This is useful for using the GNU
version of sed on macOS instead of the one it ships with.)
tools/bootcheck.java:
Use $(...) in preference to `...` for command substitutions.
Quote variables.
Abort if we attempt to rm / due to a variable not being set.
Rename mmc and mgnuc options that set this grade component to --c-debug-grade.
Let the options named --c-debug of both mmc and mgnuc enable C level debugging
of only the module being compiled.
runtime/mercury_grade.h:
Rename the .ll_debug grade component to .c_debug. Also rename the C macro
that controls the presence or absence of this grade component
from MR_LL_DEBUG to MR_C_DEBUG_GRADE.
runtime/mercury_conf_param.h:
runtime/mercury_debug.c:
runtime/mercury_debug.h:
runtime/mercury_engine.c:
runtime/mercury_label.c:
runtime/mercury_memory_zones.c:
runtime/mercury_memory_zones.h:
runtime/mercury_overflow.c:
runtime/mercury_std.h:
runtime/mercury_wrapper.c:
Rename the MR_LOWLEVEL_DEBUG macro to MR_DEBUG_THE_RUNTIME.
Previously, the name of this macro wrongly implied that it had
something to do with the old .ll_debug grade component, even though
- the MR_LOWLEVEL_DEBUG macro was designed to debug LLDS grades,
since only these existed when it was created, while
- the .ll_debug grade component (now .c_debug) is useful only for
MLDS grades targeting C.
compiler/options.m:
Rename the old confusingly named low_level_debug option to c_debug_grade.
Move it to the list of grade options, and fix its documentation, which
was completely wrong:
- code in compile_target_code.m treated it as being a synonym of
the .ll_debug (now .c_debug) grade component, while
- its (commented out) documentation here in options.m said it called for
the enabling of what is now MR_DEBUG_THE_RUNTIME.
compiler/compile_target_code.m:
Conform to the rename just above.
Define MR_C_DEBUG_GRADE instead of MR_LL_DEBUG if c_debug_grade is enabled.
Pass -g to the C compiler if either c_debug_grade or target_debug
is enabled.
Add an XXX about a missing safety check for an obsolete experimental
feature.
compiler/compute_grade.m:
When given a grade with a .c_debug grade component, set only the
c_debug_grade option; don't set the target_debug option, which is NOT
a grade option. The change to compile_target_code.m above handles the
only situation in which this implication was formerly required.
scripts/canonical_grade.sh-subr:
scripts/init_grade_options.sh-subr:
scripts/parse_grade_options.sh-subr:
Look for and process the .c_debug grade component instead of .ll_debug.
Use a sh variable named c_debug_grade to record its absence/presence.
Look for and process the --c-debug-grade grade-component option,
setting the same sh variable, c_debug_grade. (All grade components
can be set piecemeal using sh options to the scripts using these
subroutines.) This replaces the old, confusingly named option
--low-level-debug.
scripts/mgnuc.in:
scripts/mgnuc_file_opts.sh-subr:
Consistently use the sh variable c_debug to record the presence of
the (non-grade) --c-debug option to mgnuc, and the sh variable
c_debug_grade to record the presence of the .c_debug grade component.
Stop looking for and handling the --low-level-debug option, which
mgnuc used to document, even though this duplicated the same documentation
in init_grade_options.sh-subr, which mgnuc includes. The difference was
that init_grade_options.sh-subr meant it to represent the old .ll_debug
MLDS grade component, while mgnuc treated it as specifying what is now
MR_DEBUG_THE_RUNTIME for LLDS grades. It didn't help that two sh variables
with quite different semantics had names that differed only in an
underscore: LLDEBUG_OPTS vs LL_DEBUG_OPTS.
scripts/Mmakefile:
Add a missing dependency to force the rebuild of mgnuc after each update
of its sh subroutine mgnuc_file_ops.sh-subr.
doc/user_guide.texi:
Document the --c-debug-grade option of mmc. This option was not publicly
documented under its original misleading name (--low-level-debug), but
its documentation is now possible without contorted dancing around the
name.
Clarify the documentation of mgnuc's --c-debug option.
README.sanitizers:
configure.ac:
Conform to the rename of the grade component.
grade_lib/grade_spec.m:
grade_lib/grade_string.m:
grade_lib/grade_structure.m:
grade_lib/try_all_grade_structs.m:
Conform to the rename of the grade component .ll_debug to .c_debug.
Don't allow the .c_debug grade component in LLDS grades.
In grade_string.m, add some obvious implications of some grade components.
grade_lib/choose_grade.m:
grade_lib/grade_lib.m:
grade_lib/test_grades.m:
grade_lib/var_value_names.m:
Fix white space.
scripts/ml.in:
tools/lmc.in:
tools/test_mercury:
Conform to the change in compile_target_code.m to the naming of
Boehm gc library variants.
compiler/term_pass2.m:
Fix a compiler abort that would occur if we replaced a call
to array.lookup with a call to array.unsafe_lookup in array_to_doc_loop
in pretty_printer.m.
compiler/options.m:
Provide a way to check for the presence of the fix.
NEWS:
Mention the fix.
compiler/term_pass2.m:
Replace some uses of pairs with bespoke types.
Put type definitions into a top-down order.
Eliminate some redundant map lookups.
Give some predicates more meaningful names.
Put loop-invariant input arguments before loop-varying input arguments.
Move an initial det computation out of an if-then-else's condition.
library/array.m:
library/char.m:
library/float.m:
library/int.m:
library/int16.m:
library/int32.m:
library/int64.m:
library/int8.m:
library/list.m:
library/one_or_more.m:
library/string.m:
library/tree234.m:
library/uint.m:
library/uint16.m:
library/uint32.m:
library/uint64.m:
library/uint8.m:
library/version_array.m:
Mark the X_to_doc function in each of these modules as obsolete,
and make it a forwarding function to the actual implementation
in pretty_printer.m. The intention is that when these forwarding
functions are eventually removed, this will also remove the dependency
of these modules on pretty_printer.m. This should help at least some
of these modules escape the giant SCC in the library's dependency graph.
(It does not make sense that a library module that adds code to increment
an int thereby becomes dependent on pretty_printer.m through int.m.)
library/pretty_printer.m:
Move all the X_to_doc functions from the above modules here.
Fix the one_or_more_to_doc function, which was
- missing the comma between the two arguments of the one_or_more
function symbol, and
- would print "..., ...]" instead of just "...]" at the end of the
tail list when that list exceeded the limits of the specified pp_params.
Rename one of the moved types along with its function symbols,
to reduce ambiguity.
Put arrays before their indexes in the argument lists of some of
the moved functions.
Some of the moved X_to_doc functions for compound types returned
a doc that had an indent wrapper. These indents differed between the
various X_to_doc functions without any visible reason, but they are
also redundant. The callers can trivially add such wrappers if they
want to, but taking them off, if they want them off, is harder.
Eliminate the problem by deleting all such indent wrappers.
Add formatters for the intN, uintN and one_or_more types to the
default formatter map. Their previous absence was an oversight.
Add a function, get_formatter_map_entry_types, that returns the ids
of the types in the formatter_map given to the function. It is intended
for tests/hard_coded/test_pretty_printer_defaults.m, but is exported
for anyone to use.
tests/hard_coded/test_pretty_printer_defaults.{m,exp}:
Use get_formatter_map_entry_types to print the default formatter map
in a format that is much more easily readable.
NEWS:
Announce all the user-visible changes above.
The recent change to sparse_bitsets broke the lex library in extras.
Specifically, we now now need to make characters an instance of the
uenum typeclass. This diff does so.
library/char.m:
Add predicates and functions for converting between unsigned integers
and characters.
Make characters an instance of the uenum typeclass.
tests/hard_coded/Mmakefile:
tests/hard_coded/char_uint_conv.{m,exp,exp2}:
Add a test of the above conversions.
NEWS:
Announce the additions.
extras/lex/lex.m:
Conform to recent changes.
library/bitmap.m:
Factor out repeated code.
Use predicate versions of operations over function versions.
Avoid advertising field access notation.
Fix module name in exception messages.
When writing out a trans_opt_deps rule in a .d file,
always write each dependency on a separate line. If the rule is
written "horizontally" because there is only one dependency, e.g.
foo.trans_opt_date : bar.trans_opt
then maybe_read_d_file_for_trans_opt_deps will not be able to parse the
list, and the .trans_opt file will not be used.
compiler/mmakefiles.m:
As above.
compiler/handle_options.m:
compiler/intermod_analysis.m:
compiler/lookup_switch.m:
compiler/ml_top_gen.m:
compiler/modecheck_goal.m:
compiler/options_file.m:
compiler/string_switch.m:
compiler/term_constr_build.m:
compiler/term_constr_data.m:
compiler/term_constr_initial.m:
compiler/term_constr_util.m:
grade_lib/grade_setup.m:
Replace the use of some predicates from the std_util module where more
direct alternatives now exist.
Update copyright notices.
compiler/make.dependencies.m:
Inline target_dependencies into find_target_dependencies_of_modules.
For several arms of the inlined switch (i.e. for several target types),
replace the use of FindDeps with direct, first-order code.
In some of these arms, the replacement is partial: it is done for some
dependencies, but not all (yet).
Give a type and its function symbols more meaningfule names.
compiler/make.dependencies.m:
Create find_target_dependencies_of_modules, a version of the
deps_set_foldl3_maybe_stop_at_error_find_union_fi predicate
that is specialized to the call site make.module_target.m
that passes make.dependencies.m's target_depencies function
as its FindDeps argument.
Make some exported types/insts private, now that this is possible.
Eliminate an unnecessary recursive call. Put the different target types
into an approximately ascending order of complexity.
compiler/make.module_target.m:
Call the new specialized predicate.
Make some code more readable.
