If a type T is used directly (i.e, not through a pointer or reference)
as a function parameter or as a base class, a change in T should never
be marked as redundant in that context. Otherwise, the change in that
context might be filtered out, possibly hiding real ABI incompatible
changes.
This patch implements this policy.
Also, it turned out in some circumstances, we where marking the first
visited diff node of a given class of equivalence of nodes as being
redundant, while we should only mark the *subsequently* visited nodes
of that class of equivalence as visited. The patch also fixes that.
* include/abg-comparison.h (pointer_map): Make this be a map of
{size_t, size_t} pairs, rather than {size_t, bool}, so that each
pointer in the map can be associated to another one.
(diff_context::diff_has_been_visited): Return the pointer to the
first diff node of the equivalence class that has been visited.
* src/abg-comparison.cc (is_pointer_diff, is_reference_diff)
(is_reference_or_pointer_diff, is_fn_parm_diff, is_base_diff)
(is_child_node_of_function_parm_diff, is_child_node_of_base_diff):
Define new static functions.
(diff_context::diff_has_been_visited): Return the pointer to the
first diff node of the equivalence class that has been visited.
(diff_context::mark_diff_as_visited): Save the pointer to the
first diff node of a given class of equivalence that has been
visited.
(redundancy_marking_visitor::visit_begin): If a diff node is a
child node of a function parameter diff or base diff node and if
it's not a pointer or reference diff node, then do not mark it as
redundant. Also, make sure to not mark the first diff node of a
given class of equivalence that has been visited, as redundant;
only the other subsequent nodes should be marked redundant; we
were hitting this case because of an optimization that makes
equivalent class diff nodes to share their private (pimpl) data.
* tests/data/test-diff-filter/test29-finer-redundancy-marking-v{0,1}.o:
New test input binaries.
* tests/data/test-diff-filter/test29-finer-redundancy-marking-v{0,1}.cc:
Source code of the new test input binaries above.
* tests/data/test-diff-filter/test29-finer-redundancy-marking-report-0.txt:
New test input.
* tests/data/Makefile.am: Add the new test material above to the
source distribution.
* tests/test-diff-filter.cc (in_out_specs): Make this test harness
run over the additional test input above.
* tests/data/test-diff-suppr/test5-fn-suppr-report-0.txt: Adjust.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It appears we were flagging too many base class changes as local.
That was preventing some change category propagation through base
class diff nodes. This patch fixes that.
* abg-ir.cc (equals): In the overload of class_decl::base_spec, if
the underlying class carries changes, then do not flag these
changes as local for the class_decl::base_spec.
* tests/data/test-diff-dwarf/test27-local-base-diff-v{0,1}.o: New
test input binaries.
* tests/data/test-diff-dwarf/test27-local-base-diff-v{0,1}.cc: Source
code for the test input binaries above.
* tests/data/test-diff-dwarf/test27-local-base-diff-report.txt:
New test input.
* tests/data/Makefile.am: Add the test inputs above to source
distribution.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
While working on something else, I noticed that the code for handling
copying symbols (and their aliases) was broken, and so comparing two
symbols which main name were different by which had aliases that were
equal was wrongly resulting in the two symbol being different. I think
we shouldn't actually copy symbols and their aliases. Once a symbol
is allocated, interested code should just manipulate that symbol by
address rather than by value an thus do away with the copying.
The patch does that, essentially. In the implementation of a symbol,
the aliases as well as the main symbol are now weak pointers, rather
than naked pointers. Numerous API entry points that were taking
containers of elf_symbol (and were copying elf_symbols over) are not
taking containers of smart pointers to elf_symbol. Copying of
instances of elf_symbol is now thus disabled.
As a result many tests that were exercising elf_symbols (with alias)
comparison have been updated.
As a result, many empty sub-result of PR libabigail/PR17948 are now
fixed.
* include/abg-ir.h (elf_symbol_wptr): New typedef.
(elf_symbol): Make the constructors and assignment operator
private. The type can neither be copied nor created with the new
operator.
(elf_symbol::create): New static member function.
(elf_symbol::{get_main_symbol, get_next_alias, add_alias}):
Adjust.
( compute_aliases_for_elf_symbol): Likewise.
(elf_symbol::operator=): Make this private.
(elf_symbol::get_alias_which_equals): Declare new member function.
* src/abg-comp-filter.cc (function_name_changed_but_not_symbol):
Adjust.
* src/abg-comparison.cc
(class_diff::ensure_lookup_tables_populated): Adjust.
* src/abg-corpus.cc
(corpus::priv::build_unreferenced_symbols_tables): Likewise.
* include/abg-dwarf-reader.h (lookup_symbol_from_elf)
(lookup_public_function_symbol_from_elf): Adjust.
* src/abg-dwarf-reader.cc (lookup_symbol_from_sysv_hash_tab)
(lookup_symbol_from_gnu_hash_tab, lookup_symbol_from_elf_hash_tab)
(lookup_symbol_from_symtab, lookup_symbol_from_elf)
(lookup_public_function_symbol_from_elf)
(lookup_public_variable_symbol_from_elf): Adjust.
(read_context::lookup_elf_symbol_from_index): Likewise.
(read_context::lookup_elf_fn_symbol_from_address): Likewise.
(read_context::lookup_elf_var_symbol_from_address): Likewise.
(read_context::lookup_public_function_symbol_from_elf): Likewise.
(read_context::lookup_public_variable_symbol_from_elf): Likewise.
(read_context::load_symbol_maps): Likewise.
(build_var_decl, build_function_decl): Likewise.
* src/abg-ir.cc (elf_symbol::priv::{main_symbol_, next_alias_}):
Change the type of these from elf_symbol* to elf_symbol_wptr.
(elf_symbol::priv::priv): Adjust.
(elf_symbol::{create, get_alias_which_equals}): Define new functions.
(textually_equals): Likewise.
(elf_symbol::{get_main_symbol, is_main_symbol, get_next_alias,
add_alias}): Adjust to return or take elf_symbol_sptr type, rather
than a elf_symbol* one.
(elf_symbol::{get_aliases_id_string, does_alias}): Adjust.
(compute_alias_for_elf_symbol): Likewise.
(elf_symbol::operator==): Two symbols A and B are now equal if A
has at least one alias that is textually equal to B.
(equals): In the overload for function_decls, in the part where we
compare the decl_base part of the functions without considering
their decl names, we now also omit considering their linkage
names, because we compared they symbols before.
* tools/abisym.cc (main): Adjust.
* tests/data/test-diff-dwarf/test12-report.txt: Adjust.
* tests/data/test-diff-dwarf/test12-report.txt: Adjust.
* tests/data/test-diff-dwarf/test18-alias-sym-report-0.txt: Adjust.
* tests/data/test-diff-dwarf/test8-report.txt: Adjust.
* tests/data/test-diff-filter/test10-report.txt: Adjust.
* tests/data/test-diff-filter/test13-report.txt: Adjust.
* tests/data/test-diff-filter/test2-report.txt: Adjust.
* tests/data/test-diff-filter/test20-inline-report-0.txt: Adjust.
* tests/data/test-diff-filter/test20-inline-report-1.txt: Adjust.
* tests/data/test-diff-filter/test9-report.txt: Adjust.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This patch is for supporting this kind of things:
[suppress_type]
name = S
has_data_member_inserted_between = {8, end}
or:
[suppress_type]
name = S
has_data_members_inserted_between = {{8, 31}, {64, end}}
or:
[suppress_type]
name = S
has_data_members_inserted_at = offset_after(member0)
How cool is that, heh?
Anyway, to do this, the patch adds support for tuple values (i.e,
lists of values) in INI files.
Then on top of that the patch adds support for the specific
has_data_member_inserted_between, has_data_members_inserted_between
and has_data_members_inserted_at properties.
* include/abg-comparison.h (type_suppression::insertion_range):
Declare new type.
(type_suppression::insertion_ranges): Declare new typedef.
(type_suppression::{s,g}et_data_member_insertion_ranges): Declare
new member functions.
(is_integer_boundary, is_fn_call_expr_boundary): Declare new
functions.
(type_suppression::insertion_range::{boundary, integer_boundary,
fn_call_expr_boundary}): Define new types.
* src/abg-comparison.cc:
(struct type_suppression::insertion_range::priv): New type.
(type_suppression::insertion_range::{insertion_range, begin,
end}): Define new member functions.
(type_suppression::priv::insertion_ranges_): Add data member.
(type_suppression::{s,g}et_data_member_insertion_ranges): Define
new member functions.
(type_suppression::insertion_range::boundary::priv): Define new
type.
(type_suppression::insertion_range::boundary::{boundary,
~boundary}): Define new member functions.
(type_suppression::insertion_range::integer_boundary::priv):
Define new type.
(type_suppression::insertion_range::integer_boundary::{integer_boundary,
as_integer, operator int, ~integer_boundary}): Define member
functions.
(type_suppression::insertion_range::fn_call_expr_boundary::priv):
Define new type.
(type_suppression::insertion_range::fn_call_expr_boundary::{fn_call_expr_boundary,
as_function_call_expr, operator ini::function_call_expr_sptr}):
Define new member functions.
(type_suppression::insertion_range::{create_integer_boundary,
type_suppression::insertion_range::create_fn_call_expr_boundary,
type_suppression::insertion_range::eval_boundary}): Define new
member functions.
(is_integer_boundary, is_fn_call_expr_boundary): Define new
functions.
(read_type_suppression, read_function_suppression)
(read_variable_suppression): Support the new kinds of
property-related types. Aslo, in read_type_suppression, support
the new properties has_data_member_inserted_at,
has_data_member_inserted_between and
has_data_members_inserted_between.
(type_suppression::suppresses_diff): If we are looking at a type
diff node that has inserted data members, evaluate the insertion
ranges of the current type_suppression and see if they match the
inserted data members.
* include/abg-ini.h (property, simple_property, property_value)
(string_property_value, tuple_property_value, function_call_expr):
Declare new types.
(property_sptr, property_value_sptr, string_property_value_sptr)
(tuple_property_value_sptr): Declare new typedefs.
(is_string_property_value, is_tuple_property_value)
(is_simple_property, is_tuple_property, read_function_call_expr):
Declare new functions.
* src/abg-ini.cc (char_is_white_space, char_is_comment_start)
(char_is_delimiter, char_is_property_value_char)
(char_is_section_name_char, char_is_property_name_char)
(char_is_comment_start, char_is_white_space)
(remove_trailing_white_spaces, is_string_property_value)
(is_tuple_property_value, is_simple_property, is_tuple_property)
(write_property_value, char_is_function_name_char)
(char_is_function_argument_char): Define new functions.
(property::priv, tuple_property_value::priv)
(simple_property::priv, tuple_property::priv): Define new types.
(property::{property, get_name, set_name, ~property}): Define new
member functions.
(struct property_value::priv): Define new type.
(property_value::{property_value, get_kind, operator const
string&(), ~property_value}): Define new member functions.
(struct string_property_value::priv): Define new type.
(string_property_value::{string_property_value, set_content,
as_string, operator string()}, ~string_property_value): Define new
member functions.
(tuple_property_value::{tuple_property_value, get_value_items,
~tuple_property_value, as_string}): Likewise.
(simple_property::{simple_property, get_value, set_value,
~simple_property}): Likewise.
(tuple_property::{tuple_property, set_value, get_value}):
Likewise.
(config::section::find_property): Adjust return type.
(read_context::{char_is_delimiter, char_is_property_value_char,
char_is_section_name_char, char_is_property_name_char,
char_is_comment_start, char_is_white_space}): Remove these from
here as they got moved them to be non-member functions above.
(read_context::read_property_value): Return a property_value_sptr
and do not take any parameter anymore.
(read_context::{read_string_property_value,
read_tuple_property_value, read_function_name,
read_function_argument, read_function_call_expr}): Define new
member functions.
(read_context::read_property): Adjust return type. Also, change to read
the different new kinds of properties values.
(function_call_expr::priv): Define new type.
(function_call_expr::{function_call_expr, get_name,
get_arguments}): New member functions.
(read_context::read_section): Adjust.
(write_property, write_section): Adjust.
* tests/data/test-diff-suppr/libtest{11,12}-add-data-member-v{0,1}.so:
New test input binaries.
* tests/data/test-diff-suppr/test{11,12}-add-data-member-{0,1}.suppr:
New input suppression files.
* tests/data/test-diff-suppr/test11-add-data-member-{2,3,4}.suppr:
Add new test input files.
* tests/data/test-diff-suppr/test{11,12}-add-data-member-report-{0,1}.txt:
New reference output files.
* tests/data/test-diff-suppr/test12-add-data-member-report-2.txt:
Likewise.
* tests/data/test-diff-suppr/test{11,12}-add-data-member-v{0,1}.cc:
Source code for the new binaries above.
* tests/test-diff-suppr.cc (in_out_specs): Add new test inputs.
* tests/data/Makefile.am: Add the new test related files above to
source distribution.
* doc/manuals/libabigail-concepts.rst: Document the new properties
has_data_member_inserted_at, has_data_member_inserted_between and
has_data_members_inserted_between.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Types read from DWARF don't have any alignment information so we
shouldn't try to guess it, especially for structures. So this patch
sets the alignment to zero in that case. This helps remove some
spurious alignment changes detected by abidiff just because in some
cases we fail to guess that.
In the process, I noticed that when calculating the hash value of a
given data member, we were not including the hash value of its
context. This led to mistakenly considering some data member changes
as redundant. So the patch fixes that too.
* src/abg-dwarf-reader.cc (build_type_decl)
(build_class_type_and_add_to_ir, build_pointer_type_def)
(build_reference_type, build_function_decl): Set the alignment for
native types, class, reference and function type to zero,
effectively meaning that they don't have alignment information.
* src/abg-hash.cc (var_decl:#️⃣:operator): Take the hash value
of the data member context in account when computing the hash
value of a given data member.
* tests/data/test-diff-dwarf/test-23-diff-arch-report-0.txt:
Adjust.
* tests/data/test-diff-dwarf/test10-report.txt: Likewise.
* tests/data/test-diff-dwarf/test13-report.txt: Likewise.
* tests/data/test-diff-dwarf/test22-changed-parm-c-report-0.txt: Likewise.
* tests/data/test-diff-dwarf/test26-added-parms-before-variadic-report.txt: Likewise.
* tests/data/test-diff-dwarf/test8-report.txt: Likewise.
* tests/data/test-diff-dwarf/test9-report.txt: Likewise.
* tests/data/test-diff-filter/test13-report.txt: Likewise.
* tests/data/test-diff-filter/test6-report.txt: Likewise.
* tests/data/test-diff-suppr/test9-changed-parm-c-report-0.txt: Likewise.
* tests/data/test-read-dwarf/test0.abi: Likewise.
* tests/data/test-read-dwarf/test1.abi: Likewise.
* tests/data/test-read-dwarf/test2.so.abi: Likewise.
* tests/data/test-read-dwarf/test3.so.abi: Likewise.
* tests/data/test-read-dwarf/test4.so.abi: Likewise.
* tests/data/test-read-dwarf/test5.o.abi: Likewise.
* tests/data/test-read-dwarf/test6.so.abi: Likewise.
* tests/data/test-read-dwarf/test7.so.abi: Likewise.
* tests/data/test-read-dwarf/test8-qualified-this-pointer.so.abi: Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Now the name ID of a function parameter is going to be
"parameter-<index>". I think it's clearer an simpler.
* src/abg-ir.cc (function_decl::parameter::get_name_id): Make this
be "parameter-<index>".
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Just looking at if the name of the changed type hasn't changed is not
enough for detecting a sub-type change; that will be fooled by
compatible changes (changes involving typedefs). So this patch looks
through compatible changes for that matter.
* include/abg-fwd.h (type_has_sub_type_changes): Declare new
function.
* src/abg-ir.cc (type_has_sub_type_changes): Define it.
* src/abg-comparison.cc (fn_parm_diff::report): Use the new
function type_has_sub_type_changes() instead of just looking at
name changes.
* tests/data/test-diff-dwarf/test4-report.txt: Adjust this
reference test output.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* include/abg-fwd.h (get_pretty_representation): Declare new
overload for type_or_decl_base*.
* src/abg-ir.cc (get_pretty_representation): Define it and express
the previous overload for type_or_decl_base_sptr in terms of this
new one.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is just a small optimization in the passing
* src/abg-ir.h (type_decl::get_void_type_decl): Return a reference
to the smart pointer initially returned.
* src/abg-ir.cc (type_decl::get_void_type_decl): Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In the IR built from DWARF, a variadic variadic parameter has an empty
type. Later during type comparison, comparing an empty (NULL) type
with other types proves to be troublesome.
This patch handles the issue by creating a new kind of
abigail::type_decl type specifically for variadic parameters. This is
like what is done for void types.
After that it appears that the categorizing sub-system flags a change
of variadic type to non-variadic type as redundant if that change
appears several times on different functions. We don't want that
because it can hide important changes we want to see. The patch fixes
that too.
* include/abg-fwd.h (is_array_type): New overload for a naked
pointer.
* include/abg-ir.h (type_decl::get_variadic_parameter_type_decl): Declare new
static function.
* src/abg-ir.cc (is_array_type): Define new function overload for
naked pointers
(type_decl::get_variadic_parameter_type_decl): Define new static
function.
* src/abg-dwarf-reader.cc (build_function_decl): The type of
variadic parameter is now a special type_decl.
* include/abg-comparison.h (is_diff_of_variadic_parameter_type)
(is_diff_of_variadic_parameter): New function declarations.
* src/abg-comparison.cc (is_diff_of_variadic_parameter_type)
(is_diff_of_variadic_parameter): Define new functions.
(compute_diff): Refuse to return a NULL
diff for types. Assert that the parameters are non-NULL.
(report_size_and_alignment_changes): We are comparing arrays only
if the two parameters are arrays.
(fn_parm_diff::fn_parm_diff): Refuse that type diff for this diff
node is non empty.
(fn_parm_diff::report): Strengthen an assert. Cleanup a comment.
(redundancy_marking_visitor::visit_begin): Do not mark function
type and variadic parms diff nodes as redundant for local changes.
* tests/data/test-diff-dwarf/libtest26-added-parms-before-variadic-v{0,1}.so:
New test input binaries.
* tests/data/test-diff-dwarf/test26-added-parms-before-variadic-report.txt:
New test output reference.
* tests/data/test-diff-dwarf/test26-added-parms-before-variadic-v{0,1}.c:
Source code of the new test input binaries above.
* tests/data/Makefile.am: Add the new test stuff to source
distribution.
* tests/test-diff-dwarf.cc (in_out_specs): Add the new test inputs
above to the set of input to run this test harness over.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
The runtestwritereadarchive test is failing on Rawhide (Fedora 23) for
me now. It appears it was because the content to write to the zip
archive is in a buffer which is stored in a vector of buffers.
When the vector grows, the buffers are potentially copied over, and
the old buffers are destroyed, making the address of the old buffers
being stalled. But then those old address are used by the code that
write stuff in the zip archive. Ooops.
This patch essentially replaces the vector of buffer with a list, so
that growing the list doesn't invalidate the buffers. The patch also
does away with using deprecated APIs of libzip.
* configure.ac: Require libzip 0.10.1 at least.
* src/abg-writer.cc (archive_write_ctxt::serialized_tus): Make
this be a list<string>, rather than a vector<string>.
(create_archive_write_context): Truncate the archive if it exists
already.
(write_translation_unit_to_archive): Do not use the deprecated
zip_add() function anymore. Rather, use zip_file_add().
* tests/test-write-read-archive.cc (main): Double check if the
translation unit we read is empty or not.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
The library is distributed under the terms of LGPLV3+. We shouldn't
forget the '+' in the COPYING file.
* COPYING: Update to a more complete description.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* COPYING: Of course we know the licence we want to use now :-)
* COPYING-GPLV3: Add the text of GPLv3.
* Makefile.am: Add the file COPYING-GPLV3 above to source
distribution.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In the changed functions/variables section of the abidiff report, when
function parameters were added or removed, they were not properly
sorted. This patch fixes that.
* src/abg-comparison.cc (sort_string_parm_map): Define new static
function.
(struct parm_comp): Define new type.
(function_type_diff::priv::{sorted_deleted_parms_,
sorted_added_parms_}): New data members that hold sorted
deleted/added parameters.
(function_type_diff::ensure_lookup_tables_populated): Initialize
the two new data members above.
(function_type_diff::report): For the report of parameters that
got added/removed, use the sorted set of added/removed parameters
above.
* tests/data/test-diff-dwarf/test24-added-fn-parms-report-0.txt:
New test input.
* tests/data/test-diff-dwarf/libtest24-added-fn-parms-v{0,1}.so:
Likewise.
* tests/data/test-diff-dwarf/test25-removed-fn-parms-report-0.txt:
Likewise.
* tests/data/test-diff-dwarf/libtest25-removed-fn-parms-v{0,1}.so:
Likewise.
* tests/data/test-diff-dwarf/test24-added-fn-parms-v{0,1}.c:
Likewise.
* tests/data/test-diff-dwarf/test25-removed-fn-parms-v{0,1}.c:
Likewise.
* tests/data/Makefile.am: Add the new test material above to the
source distribution.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
As per https://sourceware.org/bugzilla/show_bug.cgi?id=18146, abidiff
the exit code of abidiff and abicompat is now a bit field that can be
inspected to know if the ABI change reported is incompatible for sure,
or if it needs user review of the output to decide.
This patch also updates the documentation.
* doc/manuals/abicompat.rst: Update documentation for abicompat
exit codes.
* doc/manuals/abidiff.rst: Likewise for abidiff exit codes.
* include/abg-tools-utils.h (enum abidiff_status): Declare new
enum.
(operator{|,&,|=}): Declare new operators for the new enum
abidiff_status.
(abidiff_status_has_error, abidiff_status_has_abi_change)
(abidiff_status_has_incompatible_abi_change): Declare new
functions.
* src/abg-tools-utils.cc (operator{|,&,|=}): Define these new
operators.
(abidiff_status_has_error, abidiff_status_has_abi_change)
(abidiff_status_has_incompatible_abi_change): Define new
functions.
* tests/test-diff-filter.cc (main): Adjust for the new exit code
of abidiff.
* tests/test-diff-suppr.cc (main): Likewise.
* tests/test-abicompat.cc (main): Likewise.
* tools/abicompat.cc (enum abicompat_status): Remove.
(operator{|,&,|=}): Remove these operators for enum
abicompat_status.
(perform_compat_check_in_normal_mode)
(perform_compat_check_in_weak_mode): Return abidiff_status instead
of abicompat_status. Adjust therefore.
(main): Adjust to return abidiff_status now, instead of a just
zero for all non-error cases.
* tools/abidiff.cc (main): Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This patch implements the weak mode of abicompat. In this mode, just
the application and the new version of the library are provided. The
types of functions and variables of the library that are consumed by
the application are compared to the types of the functions and
variables expected by the application. The goal is to check if the
types of the declarations consumed by the application and provided by
the library are compatible with what the application expects.
The abicompat first gets the set of symbols undefined in the
application and exported by the library. It then builds the set of
declarations exported by the library that have those symbols. We call
these the set of declarations of the library that are consumed by the
application.
Note that the debug information for the application does not contain
the declarations of the functions/variables whose symbols are
undefined. So we can not just read them to compare them to
declarations exported by the library.
But the *types* of the variables and the *sub-types* of the functions
whose symbols are undefined in the application are present in the
debug information of the application.
So in the weak mode, abicompat compare the *types* of the declarations
consumed by the application as expected by the application (described
by the debug information of the application) with the types of the
declarations exported by the library.
To do this a number of changes were necessary.
The patch builds a representation of all the types found in the
application's debug info. Before that, only the types that are
reachable from exported declarations were represented.
The abidw tool got a new --load-all-types to test this new ability of
loading all types.
The patch also adds support for looking a type, not by name, but by
its internal representation.
In the comparison engine, function_type_diff is introduced to
represent changes between two function types. For this, a new class
type_or_decl_base has been introduced in the IR. It's now the base
class for both decl_base and type_base. And abigail::comparison::diff
now takes two pointers of type_or_decl, not decl_base anymore. So
function_type_diff can take two function_type now; not that a
function_type has no declaration so it doesn't inherit decl_base. A
bunch of changes got made just to adjust to this modification.
A number of fixes were made too, to make this work, like adding
missing comparison operators, removing asserts that too strong, etc..
The patch also adjust the test suite as well as the documentation.
* include/abg-fwd.h (class type_or_decl_base): Forward declare
this.
(is_decl, is_type, is_function_type, get_name, get_type_name)
(get_function_type_name, get_pretty_representation)
(lookup_function_type_in_corpus, lookup_type_in_translation_unit)
(lookup_function_type_in_translation_unit)
(synthesize_function_type_from_translation_unit)
(hash_type_or_decl): New function declarations.
* src/abg-corpus.cc (lookup_type_in_corpus)
(lookup_function_type_in_corpus): Define new functions.
* include/abg-ir.h
(translation_unit::lookup_function_type_in_translation_unit):
Declare new friend function.
(class type_or_decl_base): Declare this.
(operator==(const type_or_decl_base&, const type_or_decl_base&)):
Declare new operator.
(operator==(const type_or_decl_base_sptr&, const
type_or_decl_base_sptr&)): Likewise.
(class {decl_base, type_base}): Make these class inherit
type_or_decl_base.
(decl_base::get_member_scopes): New const overload.
(bool operator==(const function_decl::parameter_sptr&,
const function_decl::parameter_sptr&)): New operator.
(function_type::get_parameters): Remove the non-const overload.
(function_type::get_pretty_representation): Declare new member
function.
(method_type::get_pretty_representation): Likewise.
* src/abg-ir.cc (bool operator==(const type_or_decl_base&, const
type_or_decl_base&)): Define new equality operator.
(bool operator==(const type_or_decl_base_sptr&, const
type_or_decl_base_sptr&)): Likewise.
(strip_typedef): Do not expect canonicalized types anymore. Now
the system accepts (and expects) canonicalized types in certain
cases. For instance, non-complete types and aggregated types that
contain non-complete sub-types.
(get_name, get_function_type_name, get_type_name)
(get_pretty_representation, is_decl, is_type, is_function_type)
(lookup_function_type_in_translation_unit)
(synthesize_function_type_from_translation_unit)
(lookup_type_in_scope, lookup_type_in_translation_unit): Define
new functions or new overloads.
(bool operator==(const function_decl::parameter_sptr&,
const function_decl::parameter_sptr& r)): Define
new operator.
(function_type::get_parameters): Remove non-const overload.
(function_type::get_pretty_representation): Define new function.
(function_type::traverse): Adjust.
(method_type::get_pretty_representation): Likewise.
(function_decl::get_pretty_representation): Avoid emitting the
type of cdtors.
(hash_type_or_decl): Define new function.
* include/abg-dwarf-reader.h (create_read_context)
(read_corpus_from_elf): Take a new 'read_all_types' flag.
* src/abg-dwarf-reader.cc (read_context::load_all_types_): New
flag.
(read_context::read_context): Initialize it.
(read_context::canonical_types_scheduled): If some types still
have non-canonicalized sub-types, then do not canonicalize them.
(read_context::load_all_types): New member functions.
(build_function_decl): Do not represent void return type like
empty type anymore, rather, represent it like a void type node.
(build_ir_node_from_die): When asked, load all types
including those that are not reachable from an exported
declaration.
(create_read_context, read_corpus_from_elf): Take a new
'load_all_types' flag and honour it.
* src/abg-reader.cc (read_context::type_is_from_translation_unit):
Support looking up function types in the current translation unit,
now that we now how to lookup function types.
* include/abg-comparison.h (diff_context::{has_diff_for, add_diff,
set_canonical_diff_for, set_or_get_canonical_diff_for,
get_canonical_diff_for}): Make these take instances of
type_or_decl_base_sptr, instead of decl_base_sptr.
(diff::diff): Likewise.
(diff::{first_subject, second_subject}): Make these return
type_or_decl_base_sptr instead of decl_base_sptr.
(type_diff_base::type_diff_base): Make these take instances of
type_or_decl_base_sptr instead of decl_base_sptr.
(distinct_diff::distinct_diff): Likewise.
(distinct_diff::{first, second}): Make these return
type_or_decl_base_sptr instead of decl_base_sptr.
(distinct_diff::entities_are_of_distinct_kinds): Make these take
instances of type_or_decl_base_sptr instead of decl_base_sptr.
(class function_type_diff): Create this new type. It's a
factorization of the function_decl_diff type.
* src/abg-comparison.cc ():
* src/abg-comp-filter.cc ({harmless, harmful}_filter::visit):
Adjust as diff::{first,second}_subject() now returns a
type_or_decl_base_sptr, no more a decl_base_sptr.
(decls_type, decls_diff_map_type): Remove these typedefs and replace it with ...
(types_or_decls_type, types_or_decls_diff_map_type): ... these.
(struct {decls_hash, decls_equals): Remove these type sand replace them with ...
(struct {types_or_decls_hash, types_or_decls_equals}): ... these.
({type_suppression, variable_suppression}::suppresses_diff):
Adjust.
(diff_context::priv::decls_diff_map): Replace this with ...
(diff_context::priv::types_or_decls_diff_map): ... this.
(diff_context::{has_diff_for, add_diff, get_canonical_diff_for,
set_canonical_diff_for, set_or_get_canonical_diff_for}): Take
type_or_decl_base_sptr instead of decl_base_sptr.
(diff::priv::{first, second}_subject): Make the type of these be
type_or_decl_base_sptr, no more decl_base_sptr.
(diff::priv::priv): Adjust for the subjects of the diff being of
type type_or_decl_sptr now, no more decl_base_sptr.
(diff_less_than_functor::operator()(const diff_sptr, const
diff_sptr) const): Adjust.
(diff::diff): djust for the subjects of the diff being of type
type_or_decl_sptr now, no more decl_base_sptr.
(diff::{first,second}_subject): Make the type of these be
type_or_decl_base_sptr, no more decl_base_sptr.
(report_size_and_alignment_changes): Likewise.
(type_diff_base::type_diff_base): Make the type of this be
type_or_decl_base_sptr instead of type_base_sptr.
(distinct_diff::distinct_diff): Make this take instances of
type_or_decl_base_sptr instead of decl_base_sptr.
(distinct_diff::{first, second, entities_are_of_distinct_kinds}):
Likewise.
(distinct_diff::has_changes): Simplify logic.
(distinct_diff::report): Adjust.
(compute_diff_for_types): Add an additional case to support the
new function_type.
(report_size_and_alignment_changes): Make this take instances of
type_or_decl_base_sptr instead of decl_base_sptr.
(class_diff::priv::member_type_has_changed): Return an instance of
type_or_decl_base_sptr rather than a decl_base_sptr.
(class_diff::report): Adjust.
(diff_comp::operator()(const diff&, diff&) const): Adjust.
(enum function_decl_diff::priv::Flags): Remove.
(function_decl_diff::priv::{first_fn_flags_, second_fn_flags_,
fn_flags_changes_}): Remove.
(function_decl_diff::priv::{fn_is_declared_inline_to_flag,
fn_binding_to_flag}): Remove.
(function_decl_diff::{deleted_parameter_at,
inserted_parameter_at}): Remove.
(function_decl_diff::ensure_lookup_tables_populated): Empty this.
(function_decl_diff::chain_into_hierarchy): Adjust.
(function_decl_diff::function_decl_diff): This now only takes the
subjects. It's body is now empty.
(function_decl_diff::{return_type_diff, subtype_changed_parms,
removed_parms, added_parms, type_diff}): Remove these member
functions.
(function_decl_diff::type_diff): Define new member function.
(function_decl_diff::report): Simplify logic by using the
reporting of the child type diff node.
(compute_diff): Likewise, in the overload for function_decl_sptr
simplify logic by using the child type diff object.
(function_type_diff::priv): Define new type.
(function_type_diff::{function_type_diff,
ensure_lookup_tables_populated, deleted_parameter_at,
inserted_parameter_at, finish_diff_type, first_function_type,
second_function_type, return_type_diff, subtype_changed_parms,
removed_parms, added_parms, get_pretty_representation,
has_changes, has_local_changes, report, chain_into_hierarchy}):
Define new functions.
(compute_diff): Define new overload for function_type_sptr.
* tools/abicompat.cc (options::weak_mode): New data member.
(options::options): Initialize it.
(enum abicompat_status): New enum
(abicompat_status operator|(abicompat_status, abicompat_status))
(abicompat_status& operator|=(abicompat_status &, abicompat_status))
(abicompat_status operator&(abicompat_status, abicompat_status)):
New operators to manipulate the abicompat_status enum.
(display_usage): Add help string for the new --weak-mode option.
(parse_command_line): Add the new --weak-mode command line
argument. If the tool is called with just the application and one
library then assume that we are in the weak mode.
(perform_compat_check_in_normal_mode): Define new function, factorized
from what was in the main function.
(perform_compat_check_in_weak_mode): Define new function.
(struct {fn,var}_change): Define new types.
(main): Use perform_compat_check_in_weak_mode() and
perform_compat_check_in_normal_mode().
* tools/abidiff.cc (main): Adjust.
* tools/abidw.cc: (options::load_all_types): Add new data member.
(options::options): Initialize it.
(display_usage): New help string for --load-all-types.
(parse_command_line): Support the new --load-all-types option.
(main): Adjust and honour the --load-all-types option.
* tools/abilint.cc (main): Adjust.
* doc/manuals/abicompat.rst: Update documentation for the new weak
mode. Also provide stuff that was missing from the examples
provided.
* doc/manuals/abidw.rst: Update documentation for the new
--load-all-types option.
* tests/print-diff-tree.cc (main): Adjust.
* tests/test-diff-dwarf.cc (main): Likewise.
* tests/test-read-dwarf.cc (main): Likewise.
* tests/data/test-abicompat/test0-fn-changed-app: Recompile this.
* tests/data/test-abicompat/libtest5-fn-changed-libapp-v{0,1}.so:
New new test input binaries
* tests/data/test-abicompat/test5-fn-changed-app: Likewise.
* tests/data/test-abicompat/test6-var-changed-app: Likewise.
* tests/data/test-abicompat/libtest6-var-changed-libapp-v{0,1}.so:
Likewise.
* tests/data/test-abicompat/test5-fn-changed-report-0.txt:
Reference output for one test above.
* tests/data/test-abicompat/test6-var-changed-report-0.txt:
Likewise.
* tests/data/test-abicompat/test5-fn-changed-app.cc: Source file
for a binary above.
* tests/data/test-abicompat/test5-fn-changed-libapp-v{0,1}.{h,cc}:
Likewise.
* tests/data/test-abicompat/test6-var-changed-libapp-v{0,1}.{cc,h}:
Likewise.
* tests/data/test-abicompat/test6-var-changed-app.cc: Likewise.
* tests/data/Makefile.am: Add the test related files above to the
source distribution.
* tests/test-abicompat.cc (in_out_spec): Add the new test input
above to the list of inputs to feed to this test harness.
(main): Support taking just the app and one library.
* tests/data/test-read-dwarf/test{0, 1, 2.so, 3.so, 5.o,
8-qualified-this-pointer.so,}.abi: Adjust for void type being
really emitted now, as opposed to just being an empty type.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
When the DIE of a class has its 'byte_size' property set, it means the
class is complete, even if the 'declaration' property is set to true
on it. The DWARF reader was not honouring this in all cases and
sometimes some classes were unduly considered as non-complete. The
binary the problem was exhibited on was generated by llvm.
This patch fixes that.
I couldn't produce a binary with similar DWARF output, so this patch
doesn't have a regression test associated to it :( I guess at some
point we should have another git repository with binaries and an
associated test harness, in which we'd stash binaries for which we
haven't wrote the sources ourselves.
* src/abg-dwarf-reader.cc (build_class_type_and_add_to_ir): When
the size of the class is provided then the class is complete, no
matter if this function called to update the class or to build it
for the first time.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Macros EM_AARCH64, EM_TILEPRO and EM_TILEGX were defined in diffrent
commits and release of glibc which is glibc 2.16 and 2.17 in elf/elf.h
file. Compiling libabigail was failing duing to undefined behaviour of
above macros in older glibc release. To solve it, configure.ac checks
whether these macros are defined or not and sets accordinlgy
HAVE_EM_AARCH64_MACRO, HAVE_EM_TILEPRO_MACRO and HAVE_EM_TILEGX_MACRO
macros.
* config.h.in: Generated autoheader by configure.ac
for added macros
* configure.ac: Defining HAVE_EM_AARCH64_MACRO,
HAVE_EM_TILEPRO_MACRO and HAVE_EM_TILEGX_MACRO to check
whether EM_AARCH64, EM_TILEPRO and EM_TILEGX macros are defined
in elf.h or not
* src/abg-dwarf-reader.cc (e_machine_to_string): Look for
EM_AARCH64, EM_TILEPRO and EM_TILEGX macros only
if they are defined in elf.h
Signed-off-by: Sinny Kumari <sinny@redhat.com>
commit eea9ce11eabfda96b2192341b50e9dcc27653369
Author: Dodji Seketeli <dodji@redhat.com>
Date: Mon Mar 30 22:02:52 2015 +0200
18180 - runtestreadwrite fails on i386
The issue here is that unresolved decl-only classes and resolved
decl-only classes have the same hash value (zero), and that changes
the type-id of the resolved decl-only class.
This patch ensures that the hash value of a decl-only class is zero
only for non-resolved classes. The patch also fixes an error in an
input XML file.
* src/abg-hash.cc (class_decl:#️⃣:operator()(const class_decl&)
const): Return zero only for class declarations that are not
resolved.
* tests/data/test-read-write/test20.xml: Fix the output to make a
class definition to reference its declaration, when there was a
forward declaration for it.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
At non-complete class resolution time, it appears that some type
lookup can return non-class type (for instance typedef types, when we
are expecting class types.
This patch implements class type lookup specifically (rather than
broad types lookup) and uses that during the non-complete class
resolution process.
* include/abg-fwd.h (lookup_class_type_in_corpus)
(lookup_class_type_in_translation_unit): Declare new functions.
* src/abg-ir.cc (lookup_class_type_in_translation_unit): Define
new function.
(get_node, convert_node_to_decl): Define new specializations for
the class_decl type.
* src/abg-corpus.cc (lookup_class_type_in_corpus): Define new
function.
* src/abg-dwarf-reader.cc
(read_context::resolve_declaration_only_classes): Lookup class
types specifically.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In the previous commit b9fa3b7, I forgot to initialize the new
read_context::m_exported_decls_builder_ data member in abg-reader.cc. This patch fixes
that.
* src/abg-reader.cc (read_context::read_context): Initialize the
new m_exported_decls_builder_ data member.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It turns out the support for reading the native libabigail XML format
is falling short since the work on (late) type canonicalizing.
The type ID -> XML node map is wrongly considered as being per corpus
data while it should be per translation unit data; type IDs are unique
only for a given translation unit.
The code to walk a XML sub-tree to perform the ID -> node mapping is
wrongly walking all the XML node *after* the sub-tree node too; that
is, it walks the entire corpus file starting from the XML sub-tree
node it's given. It should only walk the sub-tree it's given.
These two issues normally solve the crash reported here. But then
there are other related issues too.
The native XML format reader doesn't populate the set of exported
declarations for the current corpus it's building.
This patch addresses all these issues and makes tests/test-abidiff.cc
supports corpus files for a new regression test.
* src/abg-reader.cc (read_context::m_exported_decls_builder_): New
data member.
(read_context::read_context): Initialize it.
(read_context::{type_is_from_translation_unit,
get_exported_decls_builder, set_exported_decls_builder,
maybe_add_fn_to_exported_decls, maybe_add_fn_to_exported_decls,
type_id_new_in_translation_unit}): New member functions.
(read_context::clear_per_translation_unit_data): Clear id->xml
node map here ...
(read_context::clear_per_corpus_data): ... not here.
(read_context::walk_xml_node_to_map_type_ids): Only walk the
sub-tree we are asked to walk.
(read_translation_unit_from_input): Cleanup.
(read_corpus_from_input): Wire populating of exported declarations
of the current corpus.
(build_function_decl, build_var_decl): Populate exported
declarations of the current corpus here.
(build_type_decl, build_qualified_type_decl)
(build_pointer_type_def, build_reference_type_def)
(build_array_type_def, build_enum_type_decl, build_type_decl)
(build_template_tparameter): Adjust assert on ID to make sure
it's the first type it's being defined in the current translation
unit.
* tests/data/test-abidiff/test-corpus0-report0.txt: New test
reference output.
* tests/data/test-abidiff/test-corpus0-v{0,1}.so.abi: New test
input.
* tests/test-abidiff.cc (specs): Add the test inputs above to the
list of inputs over which to run the test harness.
(main): Support reading corpora too, as this test harness was
reading just translation units before.
(tests/data/Makefile.am): Add test material above to source
distribution.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
I have just added information for how to checkout the web site source
code.
* CONTRIBUTING: How to check out the web pages.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Update the website to include a link to the mailing list archive.
Also cleanup up the markup a little bit.
* doc/website/mainpage.txt: Update the web page.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-ir.cc (type_base::get_canonical_type_for): Cleanup the
logic here. Basically since we are not trying to cache the result
of type hashing anymore, this can be simpler.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Until now, if a diff node N has a local change, even if all of its
children nodes are redundant, N is not considered as being redundant.
This is an issue if the local changes of N are filtered out; in that
case, N should be considered redundant.
This patch fixes that. It introduces a second category bitmap on the
diff node that stores the categorization of the diff node that does
*NOT* take in account the categories inherited from its children
nodes. That way, it's possible to know if the *local changes* of a
given node have been filtered out.
* include/abg-comparison.h (diff::{get_local_category,
add_to_local_category, add_to_local_and_inherited_categories,
remove_from_local_category, set_local_category,
is_filtered_out_wrt_non_inherited_categories,
has_local_changes_to_be_reported}): Declare new member functions.
* src/abg-comp-filter.cc ({harmless, harmful}_filter::{visit,
visit_end}): Update local category too.
* src/abg-comparison.cc (diff::priv::local_category_): Add new
data member.
(diff::priv::priv): Initialize it.
(diff::priv::is_filtered_out): Add new member function. This is
factorized out of diff::is_filtered_out().
(diff::is_filtered_out): Re-write in terms of
diff::priv::is_filtered_out().
(diff::{get_local_category, add_to_local_category,
add_to_local_and_inherited_categories, remove_from_local_category,
set_local_category, is_filtered_out_wrt_non_inherited_categories,
has_local_changes_to_be_reported}): Define new member functions.
(suppression_categorization_visitor::visit_begin): Update local
categories too.
(redundancy_marking_visitor::visit_end): If all of the children
nodes of the a diff node N are redundant and if N has filtered-out
local changes, then N is redundant too.
* tests/data/test-diff-filter/libtest28-redundant-and-filtered-children-nodes-v{1,2}.so:
New binary test inputs.
* tests/data/test-diff-filter/test28-redundant-and-filtered-children-nodes-v{0,1}.cc:
Source code for the binary test inputs above.
* tests/data/test-diff-filter/test28-redundant-and-filtered-children-nodes-report-{0,1}.txt:
New test output references.
* tests/test-diff-filter.cc (in_out_specs): Add the test inputs
above to the set of inputs this test harness has to run over.
* tests/data/Makefile.am: Add the test materials above to the
source distribution.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This helps while debugging from GDB.
* include/abg-fwd.h (is_global_scope): Return a global_scope*.
* src/abg-ir.cc (is_global_scope): Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-comparison.cc (qualified_type_diff::report): Assert that
if the qualified type diff node has changes to be reported and no
local change, then its child node must have changes to be
reported.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It appears that the comparison engine reports deletion of *static*
data members in the context of the ABI changes of their parent
class. That is not right because static data member don't have any ABI
impact on their parent class. Fixed thus.
* src/abg-comparison.cc
(class_diff::priv::{get_deleted_non_static_data_members_number,
get_inserted_non_static_data_members_number}): Define new member
functions.
(class_diff::reports): Use the new functions above. Also, add
forgotten new lines where they belong.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
We started doing this earlier, but I forgot to update the code in
decl_base::get_hash().
* src/abg-ir.cc (decl_base::get_hash): Do not cache the hash
value.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-dwarf-reader.cc (build_class_type_and_add_to_ir): A type
that has its size defined is not non-complete. Same if it has a
method or a member type.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
From the DWARF emitted by GCC 4.4.7 for libstdc++ we encountered an
interesting construct.
A non-complete version of std::runtime_error is declared in
libstdc++-v3/src/functexcept.cc and is represented in DWARF as:
[ 37344] class_type
name (strp) "runtime_error"
declaration (flag)
Then a bit later, that *non-complete* class is used as a base class
for a class, *without* being fully defined! This shouldn't happen
but, well, it does:
[ 3b3a1] class_type
specification (ref4) [ 3733e]
byte_size (data1) 16
decl_file (data1) 5
decl_line (data1) 141
containing_type (ref4) [ 3734a]
sibling (ref4) [3b405]
[ 3b3b1] inheritance
type (ref4) [ 37344] <---- here.
The thing is that, later, in another translation unit
(libstdc++-v3/src/stdexcept.cc), that same class is defined fully:
[ 7e9f9] class_type
name (strp) "runtime_error"
declaration (flag)
[...]
[ 80c95] class_type
specification (ref4) [ 7e9f9]
byte_size (data1) 16
decl_file (data1) 4
decl_line (data1) 108
containing_type (ref4) [ 7e9ff]
sibling (ref4) [ 80d2b]
[...] <---------- and the definition goes here.
But then you see that the DIE offset of the "version" of the
runtime_error class that is "defined" libstdc++-v3/src/stdexcept.cc in
is different from the version that is only declared in
libstdc++-v3/src/functexcept.cc. But virtue of the "One Definition
Rule", we can assume that they designate the same type. But still,
runtime_error should have been defined in
libstdc++-v3/src/stdexcept.cc. Anyhow, libabigail needs to be able to
handle this. That is, it needs to wait until the entire ABI corpus is
loaded from DWARF, then lookup the definition of all the non-complete
types we have encountered.
And then only after that non-complete type resolution has taken place,
we can proceed with type canonicalizing, rather than doing it after
the loading of each translation unit like what we were doing
previously.
This is what this patch does.
* include/abg-fwd.h (lookup_type_in_corpus): Declare new function.
* src/abg-corpus.cc (lookup_type_in_corpus): Define new function
here.
* include/abg-ir.h (function_types_type): Declare new typedef.
(translation_unit::get_canonical_function_type): Remove member function.
(translation_unit::bind_function_type_life_time): Declare new
member function.
(classes_type): New typedef.
* src/abg-ir.cc
(translation_unit::priv::canonical_function_types_): Remove data
member.
(translation_unit::priv::function_types): New data member.
(translation_unit::get_canonical_function_type): Remove this
function definition.
(translation_unit::bind_function_type_life_time): New function
definition.
(lookup_node_in_scope): Ensure that the type returned is
complete.
* src/abg-dwarf-reader.cc (string_classes_map): New typedef.
(read_context::decl_only_classes_map_): New data member.
(read_context::declaration_only_classes): New accessor.
(read_context::{maybe_schedule_declaration_only_class_for_resolution,
is_decl_only_class_scheduled_for_resolution,
resolve_declaration_only_classes, current_elf_file_is_executable,
current_elf_file_is_dso}): Define new member functions.
(read_context::clear_per_translation_unit_data): Do not clear the
data structures that associate DIEs to decls/types or that contain
the types to canonicalize here. Rather, clear them ...
(read_context::clear_per_corpus_data): ... here instead.
(read_context::build_translation_unit_and_add_to_ir): Do not
perform late type canonicalizing here. Rather, do it ...
(read_debug_info_into_corpus): ... here instead. And before that,
call read_context::clear_per_corpus_data() and the new
read_context::resolve_declaration_only_classes() here.
(build_class_type_and_add_to_ir): Schedule the non-complete types
for resolution to complete types. Assert that base classes that
are non-complete are scheduled to be completed.
(build_function_decl): Do not try to canonicalize function types
this early, systematically. Now, all the non-complete types needs
to be completed before starting canonicalizing. So let function
types go through the normal processes of deciding when to
canonicalize them. But then, bind the life time of the function
type to the life time of the current translation unit.
(maybe_canonicalize_type): If a class type is non-complete,
schedule it for late canonicalizing.
* src/abg-hash.cc (class_decl:#️⃣:operator()(const class_decl&)
const): During hashing, a base class should be complete.
* src/abg-reader.cc
(read_context::clear_per_translation_unit_data): Do not clear
id/xml node, and type maps here. Rather, clear it ...
(read_context::clear_per_corpus_data): ... here instead.
(read_translation_unit_from_input): Do not perform late
canonicalizing here. Rather, do it ...
(read_corpus_from_input): ... here. Also, call the new
read_context::clear_per_corpus_data() here.
(build_function_decl): Do not canonicalize function types here so
early. Rather, bind the life time of the function type to the
life time of the translation unit.
* src/abg-writer.cc (write_translation_unit): Do not clear the
type/ID map here.
* tests/data/test-read-dwarf/test2.so.abi: Adjust test input.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
For portability of the regression tests of libabigail, we need abidw
to avoid emitting information about the architecture of the ELF file
it's looking at. This patch thus adds a new --no-architecture option
to abidw for that purpose.
* tools/abidw.cc (options::write_architecture): New flag.
(options::options): Initialize it.
(display_usage): Add usage string for the new --no-architecture
option.
(parse_command): Parse the new --no-architecture command line
option.
(main): Ignore the architecture name if required by the user.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is useful to call is_type() under GDB.
* include/abg-fwd.h (is_type): Declare new overload that takes a
naked pointer.
* src/abg-ir.cc (is_type): Define new overload that takes a naked
pointer.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
I am seeing issues related to the fact that a declaration-only class A
would compare different to the full version of class A. This is due
to the fact that that the declaration-only A and the full A have
different hashes, even though they structurally compare equal. So
they have different canonical types, with the current code. This
patch arranges for declaration-only classes to have no canonical type,
forcing it to compare structurally to other types. Then the patch
adjusts strip_typedef() that used to expect that all types it sees
have canonical types. Then the patch changes the type hashing code to
avoid making it cache their hash, because otherwise, in some cases
when we hash a type (too) early, a temporary hash of it gets stored ad
infinitum, even after the type has been later updated. Last but not
least, the patch returns a zero hash for declaration-only classes.
* include/abg-fwd.h (keep_type_alive): Declare new function.
* src/abg-ir.cc (strip_typedef): Simplify logic. Support types
that are not canonicalized.
(type_base::get_canonical_type_for): For declaration-only classes,
return an empty canonical class, forcing the class to be compared
structurally.
(keep_type_alive): Define new function.
* src/abg-hash.cc ({decl_base, type_decl, scope_type_decl,
qualified_type_def, pointer_type_def, reference_type_def,
array_type_def, enum_type_decl, typedef_decl,
class_decl::member_class_template, class_decl, type_tparameter,
template_tparameter, }:#️⃣:operator()): Do not cache the
computed hash.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Propagating redundancy categorization is broken for cases where there
is a diff node that has children nodes carrying changes that are all
filtered out. In that case, if among those children changes there is
a redundant change, normally the parent diff node we are look at
should be marked redundant too. The bug is that it's not marked as
redundant at the moment. This patch fixes that.
* src/abg-comparison.cc (redundancy_marking_visitor::visit_end):
Consider the cases of changes that are a filtered out.
* tests/data/test-diff-filter/libtest27-redundant-and-filtered-children-nodes-v{0,1}.so:
New test binaries to use as test input.
* tests/data/test-diff-filter/test27-redundant-and-filtered-children-nodes-report-{0,1,2}.txt:
New test result baselines.
* tests/data/test-diff-filter/test27-redundant-and-filtered-children-nodes-v{0,1}.cc:
Source code for the test input binaries above.
* tests/test-diff-filter.cc (in_out_spec): Add the binaries to the
test inputs used for this test harness.
* tests/data/Makefile.am: Add the new test material above to the
distribution.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
From inside the comparison engine, I noticed that there were some
discrepancies between some comparison performed there and the
comparison performed from inside the internal representation of
abigail::ir. This can lead to some change reports in which the
reporter thinks there are changes in the IR where there actually are
not. This patch re-uses comparison operators from the generic IR, rather
than re-implementing them in the comparison engine.
* include/abg-ir.h (operator==(scope_decl_sptr, scope_decl_sptr)):
Declare.
(operator==(type_decl_sptr, type_decl_sptr)): Likewise.
(operator==(enum_type_decl_sptr, enum_type_decl_sptr)): Likewise.
* src/abg-comparison.cc (diff_length_of_decl_bases)
(diff_length_of_type_bases): Remove these static functions.
(class_diff::has_changes): Re-use the comparison operator for
class_decl_sptr.
(type_decl_diff::has_changes): Re-use the comparison operator for
type_decl_sptr.
* src/abg-ir.cc (operator==(scope_decl_sptr, scope_decl_sptr)):
Define.
(operator==(type_decl_sptr, type_decl_sptr)): Likewise.
(operator==(enum_type_decl_sptr, enum_type_decl_sptr)): Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
While diff-ing libstdc++ against EL 6.5 and EL 7, in the report for
virtual member functions of class types, I noticed that there were
some un-necessary white spaces emitted there. This patch fixes that.
* src/abg-comparison.cc (class_diff::report): When reporting
virtual member functions make sure to emit the newline only if one
report for member function has already been emitted.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This speeds up comparison of function types.
* src/abg-dwarf-reader.cc (build_function_decl): Call
maybe_canonicalize_type to canonicalize the function type.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
