mirror of
git://sourceware.org/git/libabigail.git
synced 2024-12-16 06:54:37 +00:00
The Git repository of the Libabigail Project
77bc4b7b15
This is from a problem report originating from Red Hat bugzilla at https://bugzilla.redhat.com/show_bug.cgi?id=1925876 I some C programs, the same type can be defined more than once in a binary, as there is no "One Definition Rule[1]" in C. [1]: https://en.wikipedia.org/wiki/One_Definition_Rule The definition of those two types can be slightly different and yet be equivalent. For instance, the data member of a struct S might be defined once as having a type which is a typedef Foo of, say, "long int" and that struct S might be defined again elsewhere with a data member of type typedef Bar of "long int" as well. With the current code, because Foo and Bar have different names, they are are going to compare different; so the two struct S are doing to compare different even though they are equivalent. Down the road, this is likely going to imply that the several 'struct S' that are declaration-only will not be resolved as being declarations of the definition of "struct S", because they is more than one such definition, and those definitions are different. This is going to lead to spurious (and hard to debug) type differences that are going to be detected and reported by libabigail later down the road. This patch addresses the problem by not taking typedef names into account when comparing typedefs before canonicalization. That allows the comparison of classes and structs that happens during the resolution of declaration-only classes to correctly deduce their equivalence even in cases like the one exposed above. It thus reduces the amount of declaration-only classes that are unnecessarily considered to be different from the definition they ought to equal. * include/abg-ir.h (maybe_compare_as_member_decls): Declare new function. Make it a friend of class decl_base. * src/abg-dwarf-reader.cc (maybe_canonicalize_type): Don't early canonicalize typedefs because they can be "part" of a type that is not yet completed, especially considering that class declaration resolution is part of type building, stricto sensu. * src/abg-ir.cc (maybe_compare_as_member_decls): Factorize this out of ... (equals): ... the overload for decl_base. Use it in the overload for typedef_decl. * tests/data/test-diff-pkg/nmap-7.70-5.el8_testjcc.x86_64-self-check-report-0.txt: New test reference output. * tests/data/test-diff-pkg/nmap-7.70-5.el8_testjcc.x86_64.rpm: New binary input. * tests/data/test-diff-pkg/nmap-debuginfo-7.70-5.el8_testjcc.x86_64.rpm: Likewise. * tests/data/Makefile.am: Add these new testing material to source distribution. * tests/test-diff-pkg.cc (in_out_specs): Add the new test input to the harness. * tests/data/test-diff-pkg/spice-server-0.12.4-19.el7.x86_64-0.12.8-1.el7.x86_64-report-0.txt: Adjust. * tests/data/test-diff-suppr/test39-opaque-type-report-0.txt: Adjust. Signed-off-by: Dodji Seketeli <dodji@redhat.com> |
||
|---|---|---|
| autoconf-archive | ||
| bash-completion | ||
| doc | ||
| include | ||
| m4 | ||
| relicensing-scripts | ||
| scripts | ||
| src | ||
| tests | ||
| tools | ||
| .clang-format | ||
| .gitignore | ||
| abigail.m4 | ||
| AUTHORS | ||
| ChangeLog | ||
| COMMIT-LOG-GUIDELINES | ||
| COMPILING | ||
| configure.ac | ||
| CONTRIBUTING | ||
| default.abignore | ||
| gen-changelog.py | ||
| install-sh | ||
| libabigail.pc.in | ||
| license-change-2020.txt | ||
| LICENSE.txt | ||
| ltmain.sh | ||
| Makefile.am | ||
| NEWS | ||
| README | ||
| release-text-template.txt | ||
| update-copyright.sh | ||
| VISIBILITY | ||
This is the Application Binary Interface Generic Analysis and Instrumentation Library. It aims at constructing, manipulating, serializing and de-serializing ABI-relevant artifacts. The set of artifacts that we are intersted is made of quantities like types, variable, fonctions and declarations of a given library or program. For a given library or program this set of quantities is called an ABI corpus. This library aims at (among other things) providing a way to compare two ABI Corpora (apparently the plural of corpus is copora, heh, that's cool), provide detailed information about their differences, and help build tools to infer interesting conclusions about these differences. You are welcome to contribute to this project after reading the files CONTRIBUTING and COMMIT-LOG-GUIDELINES files in the source tree. Communicating with the maintainers of this project -- including sending patches to be include to the source code -- happens via email at libabigail@sourceware.org.