This is a second attempt at fixing this bug. The previous attempt was
on the good path, but a thinko got in the way, oops. We are not adding
the new binary of that bug report (which got re-opened) to the test
suite because it's too big and it takes too much time and memory to
abidw to analyze it.
* src/abg-ir.cc (type_base::get_canonical_type_for): Really
compare the size of the type to be canonicalized against the size
of the *current* potential canonical type of the same name.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This script parses the output of runtestreaddwarf (that output is
non-empty when the reference output *.abi files need updating, but
when there is no ABI change error). As a result, the script emits a
series of "cp <source-file> <destination-file>" command to issue, to
update the reference output of the runtestreaddwarf test program.
To use this script to update the reference output *.abi files for
runtestreaddwarf,
<build-dir>/tests/runtestreaddwarf > changed-output.txt
python update-test-read-dwarf-output.py changed-output.txt > shell-update-commands.sh
sh shell-update-commands.sh
* tests/update-test-read-dwarf-output.py: New helper python program.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* tests/test-read-dwarf.cc (for test test21-pr19092.so.abi): Fix
typo in the output path of that test.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is the last patch of a series of patches which aims at fixing bug
libabigail/19097. The short titles of the patches of the set are,
including this one:
Don't canonicalize types not added to their context in abixml reader
Support updating a class in the abixml reader
Fix emitting of referenced type in abixml writer
Use abidw --abidiff in test-read-dwarf.cc
Adjust regression tests reference output for the current patch set
Below is the cover letter of the last patch of the set.
The current patch set needs big reference output adjustments, that we
are doing at the end, here.
* tests/data/test-read-dwarf/test22-pr19097-libstdc++.so.6.0.17.so:
New test input binary.
* tests/data/test-read-dwarf/test22-pr19097-libstdc++.so.6.0.17.so.abi:
New test reference output.
* tests/data/Makefile.am: Add the new test files above to the
source distribution.
* tests/test-read-dwarf.cc (in_out_specs): Add the two new test
files above to the set of test input files.
* tests/data/test-read-dwarf/test10-pr18818-gcc.so.abi: Adjust.
* tests/data/test-read-dwarf/test12-pr18844.so.abi: Adjust.
* tests/data/test-read-dwarf/test13-pr18894.so.abi: Adjust.
* tests/data/test-read-dwarf/test14-pr18893.so.abi: Adjust.
* tests/data/test-read-dwarf/test15-pr18892.so.abi: Adjust.
* tests/data/test-read-dwarf/test16-pr18904.so.abi: Adjust.
* tests/data/test-read-dwarf/test17-pr19027.so.abi: Adjust.
* tests/data/test-read-dwarf/test18-pr19037-libvtkRenderingLIC-6.1.so.abi: Adjust.
* tests/data/test-read-dwarf/test19-pr19023-libtcmalloc_and_profiler.so.abi: Adjust.
* tests/data/test-read-dwarf/test20-pr19025-libvtkParallelCore-6.1.so.abi: Adjust.
* tests/data/test-read-dwarf/test21-pr19092.so.abi: Adjust.
* tests/data/test-read-dwarf/test9-pr18818-clang.so.abi: Adjust.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
That test was doing several sub-tests that amount to just calling
abidw --abidiff. So, let's use that, now that we have it.
* tests/test-read-dwarf.cc (handle_in_out_spec): Rather than
calling abilint on the abixml and abidiff-ing the .so file against
its .so.abi, call abidw --abidiff on the .so file and voila. Ok,
it does one extra save of abixml, but then that won't hurt. And
things are faster now than what they were anyway :-)
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
There are several issues with the abixml writer in how it handles the
process of emitting referenced types that are not directly reachable
but just walking the scopes (namespaces) of the translation units;
think about member types of a class A that are not necessarily present
in all the declarations of A, in all translation units, for instance.
This patch addresses them all because they are all intermingled.
* Use of canonical pointers in the hash map of referenced types
The abixml writer was using canonical types pointer values to hash
referenced types in a map. It was doing so "by hand"; and it was thus
messing things up for types without canonical types (like some class
declarations) etc.
This patch changes that by using the generic solution of
abigail::ir::hash_type_or_decl(), which also uses the same canonical
pointer type values. For types with no canonical types, that
functions knows has to gracefully fallback. At worst, it will just
make things slower, not wrong.
* Sorting of referenced types
The patch also changes the sorting function used for the hash map of
referenced types. The previous solution was sorting the pretty
representation of types; but then when two types have the same pretty
representation (think, typedefs, for instance) then their relative
position in the sorted result was random. This causes some stability
issues, in that emitting the abixml for the same binary several times
can lead to the some types being sorted differently -- they have the
same name, but not necessarily the same type *IDs*, as they are
different types.
The new sorting code handles this better; it also uses the pretty
representations of types, when they are equal, it uses the type IDs to
tell the types apart. At least this brings stability in the abixml
output, for a given binary.
* Avoiding duplicating declaration-only types when emitting the
context of referenced member types.
We don't keep track of declaration-only classes that are emitted.
This is because we allow a given class declaration (that carries no
definition) to appear several times in a given ABI corpus. So when a
referenced type is a class declaration, it always appears as if that
referenced type has not been emitted. So when we specifically emit
the not-emitted referenced types, it can happen that declaration-only
classes can appear a lot of times. This is unnecessary duplication,
aka bloat.
This patch thus introduces a new hash map that tracks emitted
declaration-only classes, so that we can allow duplication of class
declarations when they follow what's done in the IR read from DWARF,
and disallow that duplication when it's totally artificial and
useless.
* Better tracking of referenced types
We were blatantly forgetting to mark some referenced types as such.
So those were missing in some abixml output.
This patch fixes the spots where we were forgetting that important
information.
* Better representation of the scopes of the referenced types that
were specifically emitted.
The previous code was failing at properly representing the class scope
of some referenced types that were specifically emitted, or sometimes,
for member types, representing the scope would be so screwed that the
(referenced) member type itself wouldn't be emitted at all.
This is because I thought that to emit a given member type, just
emitting its parent scope would be enough. I thought that would
automatically trigger emitting the member type itself. First, that
would emit too much information at times; the other members of the
scope are not necessarily needed. And second the "duplication
detection code" would sometime refuse to emit the scope class, because
it has already been emitted earlier! But the incarnation that got
emitted didn't have this member type as member, then. Yes, in DWARF,
the same class A can be declared several times with different member
types in it. The complete representation of A would be a union of all
those declarations of A that are seen.
This patch addresses this issue by carefully emitting just the
information that is needed from the scope of the referenced type.
Basically the scope is declared just to declare/define the type we are
interested in; period. The abixml reader is now properly geared to
re-construct the scope by merging its different parts that are now
scattered around, in the ABI corpus. That support is part of this
patch set.
instance, a member typedef would be emitted with the information of
its parent class badly formatted.
* src/abg-writer.cc (struct type_ptr_comp_functor): Remove this.
(sort_type_ptr_map): Likewise.
(write_context::record_type_as_referenced): Do not add the
canonical type of the type to record as referenced directly.
(write_context::type_is_referenced): Adjust accordingly.
(struct write_context::type_ptr_cmp): New comparison functor.
(write_context::sort_types): New sorting function.
(write_context::{record_decl_only_type_as_emitted,
decl_only_type_is_emitted}): New member functions.
(write_member_type_opening_tag): Factorize out of ...
(write_member_type): ... here.
(write_class_decl_opening_tag): Factorize out of ...
(write_class_decl): ... here. Now, keep track also of
declaration-only classes that are emitted.
(write_decl_in_scope): Use the new write_member_type_opening_tag
and write_class_decl_opening_tag. Now write class scopes
ourselves; they only contain the type declarations that we are
emitting.
(write_translation_unit): Use the new sorting code to sort the
referenced types to emit. Do not emit referenced types that are
declaration-only classes that have already been emitted. Handle
the fact that emitting the referenced types might make those
emitted type *reference* other types too! So handle those new
referenced types as such, and emit them too.
(write_qualified_type_def, write_typedef_decl, write_var_decl): Do
not forget to mark referenced types as such.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In DWARF, the same class declaration can be present several times but
with different "views", that is, it can be present in a first
translation unit, but without any member type; then in a subsequent
translation unit, its member types are defined. In another, it'll be
completely defined, with all its data members and base classes. The
DWARF reader knows how to amend the class to add new members to it, as
they show up in the debug information.
This patch adds the same functionality to the abixml reader. The
writer has already started to write class declarations with different
"views" too, since it's started to avoid duplicating full class
definitions in every translation unit that uses them.
Without this patch, abixml misses some class members, and that is a
bug.
* include/abg-ir.h (class_decl::{find_base_class,
find_member_type, find_data_member}): Declare new member functions ..
* src/abg-ir.cc (class_decl::{find_base_class,
find_member_type, find_data_member}): ... and define them.
* src/abg-reader.cc (build_class_decl): Add the ability to update
a class to add new data members, member types and base classes to
it, if necessary.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is the first patch of a series which aims at fixing bug
libabigail/19097.
The problem in that report is a result of several underlying issues.
This patch series address them in isolation. The patches do not
update the reference output of the regression test as they should,
because they are all needed to get to a sane working state. The test
reference output are thus adjusted in the last patch of the series.
Here are the short titles of the patches of the set, including this
one:
Don't canonicalize types not added to their context in abixml reader
Support updating a class in the abixml reader
Fix emitting of referenced type in abixml writer
Use abidw --abidiff in test-read-dwarf.cc
Adjust regression tests reference output for the current patch set
Below is the cover letter of the first patch of the set.
The abixml reader sometimes (wrongly) canonicalizes types that are not
(yet) added to their context. This can lead to comparison issues
because some information carried by some types are dependant on their
context (e.g, access specifiers) and can be important for type
comparison. Right now, access specifiers for member types are no more
taken into account when comparing member types because DWARF emitters
do not necessarily keep a correct track of those; but when they do, we
better be prepared. And in any case, it's wrong to have type
canonicalization happen on half backed types anyway.
So this patch fixes several spots where type canonicalization happens
on types that are not added to their scope.
* src/abg-reader.cc (read_context::maybe_canonicalize_type):
Assert that a class type that is scheduled for canonicalization
must be in a scope. We do this only for classes, for now. The
assert here helped to spot (and fix) a lot of places where we
were canonicalizing types without scope.
(read_context::build_or_get_type_decl): Canonicalize types here,
when they are built and (hopefully) added to their scope. There
might be cases here where we try to canonicalize types that are
not added to their scope. That should bomb in the assert above,
at least for class types, for now. We'll then fix the places where
the types are created, to make them properly scoped.
(build_type_decl, build_qualified_type_decl)
(build_pointer_type_def, build_reference_type_def)
(build_array_type_def, build_enum_type_decl, build_typedef_decl):
Do not try to canonicalize the types early, right when they are
created. Canonicalization should happen at the point where (or
after) they are added to their scope.
(build_class_decl): Likewise. Also, schedule member types for
canonicalization once they've been added to their scope.
(build_class_tdecl): Schedule the pattern of the class template
for canonicalization once it has been added to its scope. I am
not sure I should do this, as the pattern is not yet a real type,
but I am taking my bet.
(build_type_composition): Schedule the composed type for
canonicalization once it's been added to its scope.
(handle_type_decl, handle_qualified_type_decl)
(handle_pointer_type_def, handle_reference_type_def)
(handle_function_type, handle_array_type_def)
(handle_enum_type_decl, handle_typedef_decl, handle_class_decl):
At this point, we should know if the type is to be added to a
scope or not. If it's in a scope, then schedule for
canonicalization.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It appears that two different types from two different translation
units might have the same name in a DSO, like in the example of this
bug. This violates the One Definition Rule, which we rely on to go
fast, and more importantly, it introduces type canonicalization errors.
This patch recognizes more of these ODR violation cases by looking at
the size of the types. That is, if two types (from the same DSO) with
the same name have different sizes, then they are different.
* src/abg-ir.cc (type_base::get_canonical_type_for): Look at the
size of types with the same name which could be considered
ODR-equal, to spot possible violations that would induce a type
canonicalization error.
* tests/data/test-read-dwarf/test21-pr19092.so: New test input
binary.
* tests/data/test-read-dwarf/test21-pr19092.so.abi: New reference
abixml for the binary above.
* tests/data/Makefile.am: Add the new test input above to source
distribution.
* tests/data/test-read-dwarf/test9-pr18818-clang.so.abi: Adjust.
* tests/data/test-read-dwarf/test12-pr18844.so.abi: Likewise.
* tests/data/test-read-dwarf/test20-pr19025-libvtkParallelCore-6.1.so.abi:
Likewise.
* tests/test-read-dwarf.cc (int_out_specs): Add the two test input
above.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* include/abg-fwd.h: Remove unnecessary declaration of class
parameter.
* src/abg-ir.cc: Remove trailing space in a comment.
* src/abg-reader.cc: Fix a comment.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This patch makes abidw --stats emit statitics about class declarations
that got resolved, and those that we missed.
* src/abg-dwarf-reader.cc
(read_context::resolve_declaration_only_classes): Emit statistics
about resolved classes and the missed ones.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-reader.cc (read_corpus_from_input): Add the necessary
call to xmlTextReaderNext call after the xmlTextReaderExpand call.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Profiling as shown that we might gain some precious cycles by passing
some well chosen parameters by reference.
* include/abg-ir.h (operator==): For the type_base_sptr and
decl_base_sptr overloads, pass the parameters by reference.
({var,function}_decl::{set,get}_symbol): Pass the elf_symbol_ptr
by reference.
* src/abg-ir.cc (operator==): For the type_base_sptr and
decl_base_sptr overloads, pass the parameters by reference, now in
the definition.
({var,function}_decl::{set,get}_symbol): Pass the elf_symbol_ptr
by reference, now in the definition.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is the last patch of the series of patches whose titles are
(including this one):
Force late canonicalizing of function types read from abixml
Fix strip_typedef issues
Do not compare access specs for member types & functions
Fix "is-anonymous" abixml property impact on some tests
Fix const-ness of a function parameter
Handle aliased function decls when comparing decls in general
Make canonicalization non sensitive to struct-ness of subtypes
Set the corpus of all ABI artifact reads from abixml
Implement fast type lookup in a corpus
Accelerate a slow path in hash_type_or_decl()
A series of small speed optimizations here and there
Allow only one definition of a given type per corpus in abixml
Make abidw --abidiff not show definitely harmless changes
Adjust tests for the patchset
This patch carries the numerous adjustments necessary for the
regresion tests output after this patch set.
* tests/data/test-read-dwarf/test10-pr18818-gcc.so.abi: Adjust.
* tests/data/test-read-dwarf/test12-pr18844.so.abi: Likewise.
* tests/data/test-read-dwarf/test13-pr18894.so.abi: Likewise.
* tests/data/test-read-dwarf/test14-pr18893.so.abi: Likewise.
* tests/data/test-read-dwarf/test15-pr18892.so.abi: Likewise.
* tests/data/test-read-dwarf/test16-pr18904.so.abi: Likewise.
* tests/data/test-read-dwarf/test17-pr19027.so.abi: Likewise.
* tests/data/test-read-dwarf/test18-pr19037-libvtkRenderingLIC-6.1.so.abi: Likewise.
* tests/data/test-read-dwarf/test19-pr19023-libtcmalloc_and_profiler.so.abi: Likewise.
* tests/data/test-read-dwarf/test20-pr19025-libvtkParallelCore-6.1.so.abi: Likewise.
* tests/data/test-read-dwarf/test9-pr18818-clang.so.abi: Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
When comparing the ABI of the input ELF binary with that same ABI
saved to abixml and read back again, there can be some minor and
harmless changes that are seen, because libabigail makes some
approximations for performance reasons. For instance, if there are
two types that are equivalent, but have different names (because of
typedefs) then libabigail will consider that they are the same type,
and might save them (to abixml) and read them back (from abixml) in
different order.
That can lead to subtle changes that are reported (and filtered out)
by the command "abidw --abixml".
This patch arranges for abidw --abixml to avoid emitting a report
saying that a filtered out change was detected, as those cases are
considered OK.
The patch also fixes a little issue where abidw would abort because
the user forgot to provide the binary to analyze, on the command line.
* tools/abidw.cc (set_diff_context): New function.
(main): Use that new function. Do not show any output for
--abidiff if only compatible changes were detected. Also, do not
abort if no input binary was giving.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In abixml, the same type T can be defined in several translation
units. This leads to a lot of duplication and, for some big binaries,
to a *lot* of memory use. For instance, loading the abixml
representing the abi of the r300_dri.so library takes more that 10GB
or ram on a 64 bits system!
This patch addresses the issue by allowing declarations to be
duplicated, but by allowing only one definition per type, per corpus.
With it, loading the abixml of r300_dri.so now takes less than 2GB or
ram.
* src/abg-writer.cc (write_translation_unit): Do not clear some
important per-translation unit maps here. There are needed to
keep track of the emitted and referenced types through the entire
corpus. Avoid (wrongly) recording function types twice.
(write_array_type_def, write_function_decl, write_function_type):
Record referenced types.
(write_class_decl): Record referenced types, and, allow only
declarations to be duplicated in a corpus.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
These are small speed optimizations that are induced by some lesser
hot spot identified by profiling.
* src/abg-comparison.cc (var_diff::has_changes): Just compare the
two var_decl. It's (way) faster now than using recursive hashing
for that.
* src/abg-ir.cc (elf_symbol::does_alias): Get out early if the two
main symbols are equal.
(equals): In the overload for function_decl, start by comparing
types. This can be very fast for functions with different types,
as it amounts to a pointer comparison. In the overload for
class_decl, avoid a map lookup when it's not necessary.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Profiling shows that hash_type_or_decl() is very slow when hashing
function parameters and base class specifications. This is because in
those two cases we use the slow recursive hashing algorithm to hash
types, rather than using the faster one based on using the pointer
values of canonical types when possible.
This was making corpora comparison very slow, as it uses
hash_type_or_decl() to hash diffs of ABI artifacts.
This patch fixes that.
* include/abg-ir.h (is_function_parameter, is_class_base_spec):
Declare new functions.
* src/abg-ir.cc (is_function_parameter, is_class_base_spec):
Define them.
(hash_type_or_decl): Handle hashing of function parameters are
class base specifications with the fast path of type hashing.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Profiling has shown that on libraries with a lot of class types
declarations (more than 10K types), the phase of resolving those
declarations to their definition was a hot spot. The lookup of the
type definition inside the entire corpus was the bottleneck.
This patch removes (or loosen) that bottleneck by doing away with the
graph-walking-based type lookup algorithm that was used. Rather, maps
of name -> types are maintained by each scope, in each translation
unit. Those maps are updated each time a type is added to a scope.
And looking up a type amounts to a lookup in a map. Way faster.
* include/abg-fwd.h (components_to_type_name): Declare new
function.
* include/abg-ir.h (string_type_base_wptr_map_type): New typedef.
(translation_unit::{get,set}_types): Declare new member functions.
* src/abg-ir.cc (translation_unit::priv::types_): New data member.
(translation_unit::{get,set}_types): Define these member
functions.
(maybe_update_types_lookup_map): Define new static function.
(components_to_type_name): Define new function.
(scope_decl::{add_member_decl, insert_member_decl}): Call the new
maybe_update_types_lookup_map.
(scope_decl::find_iterator_for_member): Fix logic.
(class_decl::set_is_declaration_only): When a class declaration
becomes a definition, update the name -> type map maintained in
the scope of the class.
(lookup_type_in_translation_unit): Use the hash map of qualified
name -> types that is now maintained in the translation unit.
This is way faster than the previous walking algorithm.
* src/abg-dwarf-reader.cc (build_translation_unit_and_add_to_ir):
When fixing up global variable declarations that need to be
re-added to the translation unit, use the new fast type lookup
function.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It turns out we were not setting the corpus for all ABI artifact read
from abixml. That was preventing the use of the ODR-based speed
optimization during type canonicalization, for corpora built from abixml.
Fixed thus.
* src/abg-reader.cc (read_translation_unit): Set the current
corpus to the current translation unit being built.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In a previous patch, we made canonicalization independant from
struct-ness of class types. This was in this commit:
0e3416e Bug 19023 - Type canonicalization is sensitive to struct-ness
But then, that didn't handle the case of composite types which have a
subtype of class type T, where the same T was declared as "struct" and
as "class" in the same binary.
This patch handles that case by passing a flag to the functions that
build the pretty representation of types. Note that the pretty
representation is used as a key in the hash map that contains
canonical types. That flag is passed all the way down to the function
that builds the pretty representation for class types, which decides
to use either "struct" or "class" as a previx for the representation.
The type canonicalization code then passes that flag (properly set) to
the pretty representation function.
* include/abg-fwd.h (get_type_name, get_function_type_name)
(get_method_type_name, get_pretty_representation): Add an
"internal" flag to all overoads.
* include/abg-ir.h
({type_or_decl_base, decl_base, type_decl, scope_type_decl,
qualified_type_def, array_type_def, enum_type_decl, typedef_decl,
var_decl, function_decl, function_decl::parameter, function_type,
method_type, class_decl}::get_pretty_representation): Add an
'internal' flag.
({decl_base, qualified_type_def, pointer_type_def,
reference_type_def, array_type_def, enum_type_decl::enumerator,
function_decl::parameter}::get_qualified_name): Likewise.
(qualified_type_def::build_name): Likewise.
* src/abg-ir.cc ({decl_base, qualified_type_def, pointer_type_def,
reference_type_def, array_type_def, enum_type_decl,
enum_type_decl::enumerator,
function_decl::parameter}::get_qualified_name): Take an "internal"
flag.
(qualified_type_def::build_name): Likewise.
({decl_base, type_decl, namespace_decl, array_type_def,
enum_type_decl, typedef_decl, var_decl, function_type,
method_type, function_decl,
class_decl}::get_pretty_representation): Likewise.
(get_type_name, get_function_type_name, get_method_type_name)
(get_pretty_representation): Likewise.
(type_base::get_canonical_type_for): Call
get_pretty_representation() with the "internal" flag set to
"true", to get a pretty representation that is independant from
the struct-ness of the subtypes of the type being canonicalized.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
When comparing two declarations, we look at their linkage name. When
the linkage names are different, then we infer that the two decls are
different.
But then, for *function* decls, it can happen that two different
linkage names are actually for different symbols that do alias; the
(ELF) symbols are different but they have the same address; so they
point to the same "thing". The two functions are not different, then.
And we were not supporting this last case of diffent linkage names
that are aliases of each other.
This patch adds support for that.
* include/abg-ir.h (is_function_decl): Add a const to the
reference parameter, making it comply with the definition.
* src/abg-ir.cc (equals): In the overload for decl_base, when the
two linkage names are different, consider the case of the decls
being aliased functions.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* include/abg-fwd.h (is_function_decl): Add a const to the
parameter to make it comply with the definition in abg-ir.cc.
Woops.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Since we started to rely on ODR for type canonicalization, we needed
to mark anonymous structures (and enums) as being anonymous, hence, a
new "is-anonymous" property was introduced in the abixml format.
While looking at something else, I noticed that some anonymous
structures in test files
tests/data/test-abidiff/test-corpus0-v{0,1}.so.abi were not marked as
anonymous, and that was causing some comparison issues. This patch
adjusts those abixml files. I forgot at the time to mention that
those files were coming from the libtirpc.so binary provided in bug
18166, so I am renaming the files now to reflect that. Also, I am
adding the binary here. I have thus re-generated a new abixml file
from that *.so file; it now has the proper "is-anonymous" properties.
* tests/data/test-abidiff/test-PR18166-libtirpc.so: New file.
* tests/data/test-abidiff/test-PR18166-libtirpc.so.abi: Likewise.
* tests/data/test-abidiff/test-corpus0-report0.txt: Renamed into
tests/data/test-abidiff/test-PR18166-libtirpc.so.report.txt.
* tests/data/test-abidiff/test-corpus0-v{0,1}.so.abi: Removed.
* tests/data/Makefile.am: Renamed test-corpus0-* files into
test-PR18166-libtirpc.so-* files.
* tests/test-abidiff.cc (specs): Adjust.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It turns that in some DWARF (e.g, from the r300_dri.so binary in bug
libabigail/19024) the same class Foo can be declared as a struct, and
later defined as a class. Or the other way around.
In some cases, Foo can be declared as a struct, have a member type
Foo::Type with no access specifier, and later that member type is
still present with no access specifier when Foo is defined as a class.
So when comparing Foo::Type (from struct Foo) against Foo::Type (from
class Foo) we must not consider the access specification of Type,
otherwise, as in the first case it's 'public' and in the second case
it's 'private', the two member types would be considered different.
And something similar happens for member function declarations too.
This patch thus avoids comparing access specifiers for member types
and functions. Though it can be considered as a regression compared
to what was being done before, access specifiers don't have an impact
on ABI per se. And they can cause noise in the result, as we are
seeing here.
* include/abg-fwd.h (is_function_decl): Declare a new overload.
* src/abg-ir.cc (is_function_decl): Define a new overload.
(equals): In the overload for decl_base, do not compare access
specifiers when comparing member functions and types.
* tests/data/test-diff-dwarf/test0-report.txt: Adjust.
* tests/data/test-diff-filter/test0-report.txt: Likewise.
* tests/data/test-diff-filter/test01-report.txt: Likewise.
* tests/data/test-diff-filter/test30-pr18904-rvalueref-report0.txt: Likewise.
* tests/data/test-diff-filter/test31-pr18535-libstdc++-report-0.txt: Likewise.
* tests/data/test-diff-filter/test4-report.txt: Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
strip_typedef currently has at least two issues. First, it was
triggering a potentially wrong early canonicalization. Second, it was
asserting too eagerly that a return type should not be nil; the truth
is that there can be a short period of time where a function has an
empty result type; that is usually during the building of said
function type, before the return type is fully built.
This patch addresses those two issues.
* src/abg-ir.cc (strip_typedef): Do not canonicalize
the return type of the method type to typedef-strip.
Acknowledge that the return type can be nil.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is the first patch of a series which aims at fixing:
Bug 19024 - abidw --abidiff fails and aborts when run against r300_dri.so
The issue reported in that bug is the manifestation of several
problems that different patches in the series address on a case by
case basis.
Suffice it to say abidw --abidiff on my X220 laptop was taking more
than 40 minutes, and north of 11GB or ram. An rather than yielding
the empty set, it was emitting lots of false postives!
The patchset thus applies a series of optimizations to reduce the time
and memory taken, so that I can at least debug the issues that prevent
abidw --abidiff from yielding the empty set, as it should. Then, with
those optimizations applied, I came up with a series of fixes.
With the series applies, abidw --abidiff now takes less than 8 minutes
and around of 4.8GB of ram.
The first seven patches are those fixes. The next five patches are
the time and size optimization that allowed me to work on the first
fixes. The thirteenth patch applies some needed modification (both
fixes and improvements) to abidw --abidiff itself. The last patch
carries the necessary adjustments to the regression tests output.
Here are the short titles of the patches of the set, including this one:
Force late canonicalizing of function types read from abixml
Fix strip_typedef issues
Do not compare access specs for member types & functions
Fix "is-anonymous" abixml property impact on some tests
Fix const-ness of a function parameter
Handle aliased function decls when comparing decls in general
Make canonicalization non sensitive to struct-ness of subtypes
Set the corpus of all ABI artifact reads from abixml
Implement fast type lookup in a corpus
Accelerate a slow path in hash_type_or_decl()
A series of small speed optimizations here and there
Allow only one definition of a given type per corpus in abixml
Make abidw --abidiff not show definitely harmless changes
Adjust tests for the patchset
We do not add the r300_dri.so library to the repository because of the
time it still takes to complete.
And now, here is the cover letter for this first patch.
When reading the abixml format, sometimes, function types can be
early-canonicalized. This can be wrong especially is the function
type has sub-types that are not canonicalized yet.
So this patch forces those to be late-canonicalized.
* src/abg-reader.cc (build_function_type): Late-canonicalize
function types.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
When abipkgdiff is invoked on a set of packages, the newer (second) one is also
inspected for files matching the pattern '*.abignore', whose contents are read
and interpreted as suppression specifications.
* tests/data/Makefile.am: Add new test material to the build system.
* tests/data/test-diff-pkg/dirpkg-{0-dir1,{1,2}-dir2}/dir.abignore:
A test suppression specification.
* tests/data/test-diff-pkg/dirpkg-{2,3}-dir2/.abignore: Likewise.
* tests/data/test-diff-pkg/dirpkg-3.suppr: Likewise.
* tests/data/test-diff-pkg/dirpkg-{1,2,3}-dir{1,2}/libobj-v0.so: New
binary test inputs.
* tests/data/test-diff-pkg/dirpkg-{1,2,3}-dir{1,2}/obj-v0.cc: New test
source files
* tests/data/test-diff-pkg/dirpkg-{1,2,3}-report-{0,1}.txt: New
reference outputs
* tests/test-diff-pkg.cc: Adjust to run the new tests.
* tools/abipkgdiff.cc (prog_options): New static pointer to struct
opts.
(file_tree_walker_callback_fn): Rename to
first_package_tree_walker_callback_fn.
(second_package_tree_walker_callback_fn): Check for ELF files just
like the previous function but additionally check for files
ending with ".abignore", unless disabled from the command line.
({create_maps_of_package,extract_package_and_map_its}_content):
Add a callback as a new argument.
(main) handle the new "--no-abignore" option, which turns off
the search for suppression files within the new package.
Signed-off-by: Ondrej Oprala <ooprala@redhat.com>
The for now, the underlying type of an enum type is always assumed to
be anonymous by libabigail. But then, the code of the DWARF reader
was failing to set the "is-anonymous" flag on it. So type
canonicalizing code was comparing the enum underlying types by looking
at their names; they all have the same name -- as we forget that they
are anonymous; so they (wrongly) all look the same, within the same
ABI corpus.
This patch sets properly sets the is-anonymous flag on enumerator
underlying types again.
* src/abg-dwarf-raeder.cc (build_enum_type): Set the is-anonymous
flag on the underlying type of the enum.
* tests/data/test-read-dwarf/test0.abi: Adjust.
* tests/data/test-read-dwarf/test9-pr18818-clang.so.abi: Likewise.
* tests/data/test-read-dwarf/test10-pr18818-gcc.so.abi: Likewise.
* tests/data/test-read-dwarf/test11-pr18828.so.abi: Likewise.
* tests/data/test-read-dwarf/test12-pr18844.so.abi: Likewise.
* tests/data/test-read-dwarf/test13-pr18894.so.abi: Likewise.
* tests/data/test-read-dwarf/test14-pr18893.so.abi: Likewise.
* tests/data/test-read-dwarf/test15-pr18892.so.abi: Likewise.
* tests/data/test-read-dwarf/test16-pr18904.so.abi: Likewise.
* tests/data/test-read-dwarf/test17-pr19027.so.abi: Likewise.
* tests/data/test-read-dwarf/test18-pr19037-libvtkRenderingLIC-6.1.so.abi: Likewise.
* tests/data/test-read-dwarf/test19-pr19023-libtcmalloc_and_profiler.so.abi: Likewise.
* tests/data/test-read-dwarf/test20-pr19025-libvtkParallelCore-6.1.so.abi: Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
When a member type (a type that is a member of a class) M is
referenced by some types emitted by abixml, but the context of M (the
class type which M is a member of) is not itself referenced by any ABI
artifact, then abixml forgets to emit the context of M and thus M
itself.
With this patch, when the abixml writer has emitted almost all ABI
artifacts for the current translation unit, it looks for types that
have been referenced by the emitted ABI artifacts, but that haven't
been emitted themselves.
It then emits those referenced-but-not-emitted types, and makes sure
their contexts are emitted as well.
* include/abg-fwd.h (is_namespace): Fix prototype.
* src/abg-writer.cc (struct type_ptr_comp_functor): New internal
type.
(sort_type_ptr_map): New static function.
(write_context::m_referenced_types_map): Renamed
m_referenced_fntypes_map data member into this.
(write_context::get_referenced_types): New member function.
(write_context::record_type_as_referenced): Renamed
record_fntype_as_referenced member function into this. Adjust.
(write_context::type_is_referenced): Renamed fntype_is_referenced
into this.
(write_context::clear_referenced_types_map): Renamed
clear_referenced_fntypes_map member function into this. Adjust.
(write_decl_in_scope): New static function.
(write_translation_unit): Use it here to emit types that are
referenced by other types in the TU, but that are not emitted.
Adjust.
(write_pointer_type_def, write_reference_type_def)
(write_typedef_decl): Record the underlying types referenced by
the emitted types as being, well, referenced.
* tests/data/test-read-dwarf/test20-pr19025-libvtkParallelCore-6.1.so:
New test binary input.
* tests/data/test-read-dwarf/test20-pr19025-libvtkParallelCore-6.1.so.abi:
New reference output of the binary input above.
* tests/data/Makefile.am: Add the new test material above to the
source distribution.
* tests/test-read-dwarf.cc (in_out_spec): Add the new test inputs.
* tests/data/test-read-dwarf/test9-pr18818-clang.so.abi: Adjust.
* tests/data/test-read-dwarf/test10-pr18818-gcc.so.abi: Likewise.
* tests/data/test-read-dwarf/test12-pr18844.so.abi: Likewise.
* tests/data/test-read-dwarf/test13-pr18894.so.abi: Likewise.
* tests/data/test-read-dwarf/test14-pr18893.so.abi: Likewise.
* tests/data/test-read-dwarf/test15-pr18892.so.abi: Likewise.
* tests/data/test-read-dwarf/test16-pr18904.so.abi: Likewise.
* tests/data/test-read-dwarf/test17-pr19027.so.abi: Likewise.
* tests/data/test-read-dwarf/test18-pr19037-libvtkRenderingLIC-6.1.so.abi:
Likewise.
* tests/data/test-read-dwarf/test19-pr19023-libtcmalloc_and_profiler.so.abi:
Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* tests/Makefile.am: Link runtestreaddwarf with libpthread.
* tests/test-read-dwarf.cc (main) Create worker threads corresponding
to the number of CPUs online, add a "--no-parallel" option and move
the main loop...
(handleInOutSpec) ...here.
Signed-off-by: Ondrej Oprala <ooprala@redhat.com>
In some debug info of some shared library, the same type can be
present as a struct in some translation units, and as a class in
others. As we are using the "pretty representation" of types to hash
types during type canonicalization, a "class foo" and "struct foo"
are (wrongly) considered different, because those pretty
representations are different.
This patch changes the canonicalization code to make it independent
from the struct-ness of the class being canonicalized.
* include/abg-ir.h (class_decl::is_struct): Declare a setter for the
"is-struct" property.
* src/abg-ir.cc (class_decl::is_struct): And define that setter
here.
(type_base::get_canonical_type_for): Temporarily set the
'is-struct' flag of the class type to 'false' before building its
pretty representation.
* tests/data/test-read-dwarf/test19-pr19023-libtcmalloc_and_profiler.so:
New test input binary.
* tests/data/test-read-dwarf/test19-pr19023-libtcmalloc_and_profiler.so.abi:
New test reference output.
* tests/data/Makefile.am: Add the new test material above to the
source distribution.
* tests/test-read-dwarf.cc (in_out_specs): Add the two new test
inputs to the list of test inputs to consider.
* tests/data/test-read-dwarf/test14-pr18893.so.abi: Adjust.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
It turns out that, in DWARF, there can be function template
instantiations foo<int>(int) and foo<TypedefOfInt>(TypedefOfInt) which
have the same symbol name, if TypedefOfInt is a typedef of int.
An ABI corpus retains only one function declaration per symbol
name. So in the example of the bug the input DWARF has the two
instantiations, but libabigail is just keeping one of the two; so the
abixml only has one of the two template instantiations.
This patch changes the ABI corpus model so that it represents the fact
that there can be several function declarations for a given symbol.
The patch then adjust the comparison engine to make it know about this
new model.
* include/abg-corpus.h
(corpus::exported_decls_builder::str_{fn,var}_ptr_map_type):
Remove these typedefs from here as they only used internally in
abg-corpus.cc. So we move them there instead.
* src/abg-corpus.cc (str_fn_ptrs_map_type): New typedef.
(str_var_ptr_map_type): Moved the typedef that was in
corpus::exported_decls_builder here.
(corpus::exported_decls_builder::id_fns_map_): Rename the fns_
data member into this. Make it have a str_fn_ptrs_map_type as a
type.
(corpus::exported_decls_builder::id_fns_map): Renamed the
fns_map() accessor into this one.
(corpus::exported_decls_builder::{fn_id_is_in_id_fns_map,
fn_is_in_fns}): New member functions.
(corpus::exported_decls_builder::fn_is_in_id_fns_map): Rename
fn_is_in_map into this.
(corpus::exported_decls_builder::add_fn_to_id_fns_map): Rename
add_fn_to_map into this.
(corpus::exported_decls_builder::add_fn_to_exported): Adjust.
(corpus::exported_decls_builder::maybe_add_fn_to_exported_fns):
Adjust.
* src/abg-comparison.cc (function_decl_diff::report): Emit reports
about function name changes (for a given function ID) only if
there are sub-type changes to be reported for the function. In
that case, do not forget to emit the sub-type changes after the
name changes have been reported.
(corpus_diff::priv::ensure_lookup_tables_populated): Several
functions of the same ID can be removed or added from/to the
corpus.
* tests/data/test-read-dwarf/test18-pr19037-libvtkRenderingLIC-6.1.so:
New test input binary.
* tests/data/test-read-dwarf/test18-pr19037-libvtkRenderingLIC-6.1.so.abi:
New test output reference.
* tests/data/Makefile.am: Add the new test materials to the source
distribution.
* tests/test-read-dwarf.cc (in_out_specs): Adjust to add the new
test inputs above.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-reader.cc (read_context): Abort if we run into an
unsupported access specifier.
(build_class_decl) Default to public access for the children
of a struct.
Signed-off-by: Ondrej Oprala <ooprala@redhat.com>
* doc/Makefile.am: Prefix the path for DOXY_WEBSITE_SRC_CFG and
DOXY_WEBSITE_BLD_{CFG,DIR} with "/doc" to protect it against make
clean.
Signed-off-by: Ondrej Oprala <ooprala@redhat.com>
* src/abg-reader.cc (read_is_struct): Fix comment.
(build_type_decl): Use type_decl_sptr rather than
shared_ptr<type_decl>.
(build_type_decl): Use typedef_decl_sptr rather than
shared_ptr<typedef_decl>.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
This is the last patch of the series of 11 patches that started at the
patch with the subject:
constify is_class_type()
And below starts the cover letter of this patch.
While analyzing some libraries like libmozjs.so[1] it appeared that
type canonicalization takes a significant time to comparing composite
types that are re-defined in each translation units again and again.
The One Definition Rule[2] says that two types with the same name
shall designate the same thing; so when a type T being canonicalized
has the same name of a canonical type C in the same ABI corpus, then
this patch considers C as being the canonical type of T, without
comparing T and C structurally. This saves us from comparing T and C.
Before this patch, `abidw --noout libmozjs.so` was taking
approximatively 5 minutes; with the patch, it takes 1 minutes and 30
seconds.
To do this, the patch changes ABI artifacts to carry a pointer to the
corpus it belongs to. Whenever an ABI artifact is added to a given
context, the corpus of that context is propagated to the artifact;
that is now possible as the artifact now carries the property of the
corpus it belongs to.
During type canonicalization the ODR-based optimization outlined above
is performed as we can now compare the corpus of a given type again
the one of another type; it's now possible to know if two types come
from the same corpus.
There are a few cases though were the optimization is not performed:
- anonymous struct; when a struct is anonymous (it has no name, as
described in the DWARF), the DWARF reader gives it a name
nonetheless, so that diagnostics can refer to that anonymous type.
But then all anonymous types in the system have the same name. So
when faced with two anonymous types (with the same name) from the
same corpus, it's wrong to consider that they name the same thing.
The patch added an "is_anonymous" property to types created by the
DWARF reader so that such anonymous types can be detected by the
type canonicalizer; they are thus not involved in this
optimization. Note that the abixml writer and reader have been
updated to emit and read this property.
- typedefs. I have seen in some boost code two typedefs of the same
name refer to different underlying types. I believe this is a
violation of ODR. I'll need to investigate on this later. And I
think we really need to detect these ODR violations as part of
this enhancement request:
https://sourceware.org/bugzilla/show_bug.cgi?id=18941.
- pointers, references, arrays and function types, as they can refer
to the two exceptions above.
This is the last patch of the series which aimed at speeding up type
canonicalization in the context of types being re-defined a lot in
translation units.
[1]: Instruction to build libmozjs.so from the mongodb sources:
- git clone https://github.com/mongodb/mongo.git
- cd mongo
- scons --link-model=dynamic build/opt/third_party/mozjs-38/libmozjs.so
[2] One Definition Rule: https://en.wikipedia.org/wiki/One_Definition_Rule
* include/abg-fwd.h (class corpus): Forward-declare this.
(is_anonymous_type): Declare this new function.
* include/abg-ir.h (corpus_sptr, corpus_wptr): Declare these
typedefs here too.
(translation_unit::{g,s}et_corpus): Declare new member functions.
(type_or_decl_base::{g,s}et_corpus): Likewise.
* src/abg-ir.cc (translation_unit::priv::corpus): New data member.
(translation_unit::priv::priv): Initialize it.
(translation_unit::{g,s}et_corpus): Define new accessors.
(translation_unit::get_global_scope): Propagate the corpus of the
translation unit to its newly created global scope.
(translation_unit::bind_function_type_life_time): Propagate the
corpus of the translation_unit to the added function type.
(type_or_decl_base::priv::corpus_): Add new data member.
(type_or_decl_base::priv::priv): Initialize it.
(type_or_decl_base::{g,s}et_corpus): Define new accessors.
(scope_decl::{add,insert}_member_decl): Propagate the context's
corpus to the member added to the context.
(decl_base::priv::is_anonymous_): Add new data member.
(decl_base::priv::priv): Initialize it.
(decl_base::{s,g}et_is_anonymous): Define accessors.
(is_anonymous_type): Define a new test function.
(decl_base::set_name): Update the "is_anonymous" property.
(type_base::get_canonical_type_for): Implement the ODR-based
optimization to type canonicalization.
* src/abg-corpus.cc (corpus::add): When a translation unit is
added to a corpus, set the corpus of the translation unit.
* src/abg-dwarf-reader.cc (build_enum_type)
(build_class_type_and_add_to_ir): Set the "is_anonymous" flag on
anonymous enums and classes.
* src/abg-reader.cc (read_is_anonymous): Define new static
function.
(build_type_decl, build_enum_type, build_class_decl): Call the new
read_is_anonymous function and set the "is_anonymous" property on
the built type declaration.
* src/abg-writer.cc (write_is_anonymous): Define new static
function.
(write_type_decl, write_enum_type_decl, write_class_decl): Write
the "is_anonymous" property.
* tests/data/test-diff-filter/test31-pr18535-libstdc++-report-0.txt:
Adjust.
* tests/data/test-read-dwarf/test9-pr18818-clang.so.abi: Likewise.
* tests/data/test-read-dwarf/test10-pr18818-gcc.so.abi: Likewise.
* tests/data/test-read-dwarf/test11-pr18828.so.abi: Likewise.
* tests/data/test-read-dwarf/test12-pr18844.so.abi: Likewise.
* tests/data/test-read-dwarf/test13-pr18894.so.abi: Likewise.
* tests/data/test-read-dwarf/test14-pr18893.so.abi: Likewise.
* tests/data/test-read-dwarf/test15-pr18892.so.abi: Likewise.
* tests/data/test-read-dwarf/test16-pr18904.so.abi: Likewise.
* tests/data/test-read-dwarf/test17-pr19027.so.abi: Likewise.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
Until now, the DWARF reader would late canonicalize typedefs to
classes, as well as classes. That is not enough. Let's also
late-canonicalize pointers, references and array of classes too. This
is because classes that might not be finished yet might be referenced
by those types, and so we want to wait until they are finished before
we canonicalize them.
* include/abg-fwd.h (peel_array_type): Declare new function.
* src/abg-ir.cc (peel_array_type): Define it.
(peel_typedef_pointer_or_reference_type): Peel arrays too, to get
the type of its element.
* src/abg-dwarf-reader.cc (maybe_canonicalize_type): If a pointer,
reference, array or typedef references a class, then do
late-canonicalize this type.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-ir.cc (peel_typedef_pointer_or_reference_type): Make
sure the variable tested in the condition is the one updated by
the loop.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
In hash_type_or_decl, when we encounter a declaration-only class
(those have no canonical type), we not trying to get the canonical
type of the definition, when the class had a definition. We were
instead going straight to the slow path of computing the recursive
hash of the type.
This patch tries to get the canonical type of the class definition,
when it exists.
* src/abg-ir.cc (hash_type_or_decl): When a declaration-only
class has a definition, then use the canonical type of that
definition as a hash value. If the class no definition, only
then, use the slow patfh of computing the recursive progressive
hash value of the type.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
When the abixml writer emits function types, it puts function
types that are referenced by pointers or references into a map on the
side. Unfortunately, that map hashes types by recursively calculating
a progressive hash value. That is dog slow and we avoid that
throughout the code base.
This patch changes that to use the numerical values of the canonical
type pointer of the function type as a hash, making abixml fast again,
again on big library as libmozjs.so.
* src/abg-writer.cc (typedef fn_shared_ptr_map): Remove.
(write_context::m_referenced_fntypes_map): Change the type of this
into type_ptr_map.
(write_context::{record_fntype_as_referenced,
fntype_is_referenced}): Use the pointer value of the canonical
type of the referenced type as key for the map.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
I noticed that in some cases in build_function_type, when building the
sub-types of the function type, the construction of a function type
for the same DIE could be triggered. This happens frequently for
aggregate types that happen to be recursive. In those cases, we must
arrange for the construction of the function type for the same DIE to
return the same type that is being currently built by
build_function_type; otherwise, several types are going to be built
for the same DIE, and only one of them is going to be canonicalized.
build_function_type was just not prepared for this.
This patch fixes that.
Please note that the patch changes the test output
/home/dodji/git/libabigail.git/merge/build/tests/output/test-read-dwarf/test12-pr18844.so.abi
but it's a later patch that adjust that file because several patches
are going to require an update to that file. We are going to update
that patch in one go at the end of the patch series.
* src/abg-dwarf-reader.cc (build_function_type): Associate the
type being built with its DIE, before starting to build the
sub-types. The current type is then amended with the sub-types
that are built later.
(build_ir_node_from_die): In the case for DW_TAG_subroutine_type,
do not associate the type to the DIE here, as it's been done in
build_function_type.
* src/abg-ir.cc (function_type::set_parameters): Adjust the index
of the parameters being set to the function: they start at 1,
unless the first parameter is artificial, in which case its index
starts at zero. This is just like what is done when the function
type is constructed directly with the parameters passed as an
argument to the constructor.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* src/abg-comparison.cc (pointer_diff::has_changes): Just
comparing the underlying type might not be enough. Let's just
compare the pointer itself. Now that we have canonical types,
comparing the pointer itself is not slower.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
strip_typedef() canonicalizes the stripped typed, even if the input
type was not canonicalized. This can lead to early canonicalization
that is not warranted. For instance, is_compatible_with_class_type()
calls strip_typedef() and can be called during DWARF reading on types
that haven't been canonicalized yet; this was triggering a
canonicalization what was happening too early.
With this patch, strip_typedef() does not canonicalize a stripped type
if the input type wasn't itself canonicalized.
* src/abg-ir.cc (strip_typedef): Do not canonicalize the stripped
type if the input one is not canonicalized.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
* include/abg-ir.h (operator==): In the overloads for type_decl,
enum and class_decl, turn the shared_ptr parameter into a const
reference to the shared_ptr.
* src/abg-ir.cc (operator==): Do the same in the definitions.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
