Add --merge-translation-units option.
It is not always necessary to know in which translation unit a type,
variable or function was defined first. Provide an option to simply
merge all translation units for the same language (and address size).
This also reduces the size of the XML representation of the corpus by
~10% on a simple C only library.
Currently only implemented for the DWARF reader and abidw tool.
* doc/manuals/abidw.rst: Document --merge-translation-units.
* include/abg-dwarf-reader.h (set_merge_translation_units):
New function.
* src/abg-dwarf-reader.cc (read_context): Add
merge_translation_units_ bool, merge_translation_units and
merge_translation_units methods.
(set_merge_translation_units): New function.
(read_context::build_translation_unit_and_add_to_ir): Check
merge_translation_units to see how to find an existing
(compatible) translation_unit to use. Drop path and
compilation_dir when merging translation_units.
* src/abg-reader.cc (read_context): Add m_merge_translation_units
bool, merge_translation_units and merge_translation_units methods.
* tools/abidw.cc (options): Add merge_translation_units bool.
(display_usage): Describe --merge-translation-units.
(parse_command_line): Parse --merge-translation-units.
(main): Call set_merge_translation_units.
Signed-off-by: Mark Wielaard <mark@klomp.org>
Change-Id: Ic4553900277ba032a4b9481f1950ad497decfd28
Signed-off-by: Giuliano Procida <gprocida@google.com>
diff --git a/doc/manuals/abidw.rst b/doc/manuals/abidw.rst
index a67e5fa..4af497b 100644
--- a/doc/manuals/abidw.rst
+++ b/doc/manuals/abidw.rst
@@ -167,6 +167,13 @@
representation build by Libabigail to represent the ABI and will
not end up in the abi XML file.
+ * ``--merge-translation-units``
+
+ With this option translation units for the same language (and
+ address size) will be merged together as if the functions,
+ variables and types were all defined together. Note that this
+ also drops the compilation paths used.
+
* ``--no-linux-kernel-mode``
Without this option, if abipkgiff detects that the binaries it is
diff --git a/include/abg-dwarf-reader.h b/include/abg-dwarf-reader.h
index a328fb0..13a5d5e 100644
--- a/include/abg-dwarf-reader.h
+++ b/include/abg-dwarf-reader.h
@@ -178,6 +178,10 @@
bool f);
void
+set_merge_translation_units(read_context& ctxt,
+ bool f);
+
+void
set_do_log(read_context& ctxt, bool f);
void
diff --git a/src/abg-dwarf-reader.cc b/src/abg-dwarf-reader.cc
index 253abc5..f3534e1 100644
--- a/src/abg-dwarf-reader.cc
+++ b/src/abg-dwarf-reader.cc
@@ -2125,6 +2125,7 @@
corpus::exported_decls_builder* exported_decls_builder_;
options_type options_;
bool drop_undefined_syms_;
+ bool merge_translation_units_;
read_context();
private:
@@ -2268,6 +2269,7 @@
options_.load_in_linux_kernel_mode = linux_kernel_mode;
options_.load_all_types = load_all_types;
drop_undefined_syms_ = false;
+ merge_translation_units_ = false;
load_in_linux_kernel_mode(linux_kernel_mode);
}
@@ -2355,6 +2357,22 @@
drop_undefined_syms(bool f)
{drop_undefined_syms_ = f;}
+ /// Setter for the flag that tells us if we are merging translation
+ /// units.
+ ///
+ /// @param f the new value of the flag.
+ void
+ merge_translation_units(bool f)
+ {merge_translation_units_ = f;}
+
+ /// Getter for the flag that tells us if we are merging translation
+ /// units.
+ ///
+ /// @return true iff we are merging translation units.
+ bool
+ merge_translation_units() const
+ {return merge_translation_units_;}
+
/// Getter of the suppression specifications to be used during
/// ELF/DWARF parsing.
///
@@ -6148,6 +6166,17 @@
set_drop_undefined_syms(read_context& ctxt, bool f)
{ctxt.drop_undefined_syms(f);}
+/// Setter of the "merge_translation_units" flag.
+///
+/// This flag tells if we should merge translation units.
+///
+/// @param ctxt the read context to consider for this flag.
+///
+/// @param f the value of the flag.
+void
+set_merge_translation_units(read_context& ctxt, bool f)
+{ctxt.merge_translation_units(f);}
+
/// Setter of the "do_log" flag.
///
/// This flag tells if we should emit verbose logs for various
@@ -10899,29 +10928,50 @@
}
string compilation_dir = die_string_attribute(die, DW_AT_comp_dir);
- // See if the same translation unit exits already in the current
- // corpus. Sometimes, the same translation unit can be present
- // several times in the same debug info. The content of the
- // different instances of the translation unit are different. So to
- // represent that, we are going to re-use the same translation
- // unit. That is, it's going to be the union of all the translation
- // units of the same path.
- {
- const string& abs_path =
- compilation_dir.empty() ? path : compilation_dir + "/" + path;
- result = ctxt.current_corpus()->find_translation_unit(abs_path);
- }
+ uint64_t lang = 0;
+ die_unsigned_constant_attribute(die, DW_AT_language, lang);
+ translation_unit::language language = dwarf_language_to_tu_language(lang);
+
+ corpus_sptr corp = ctxt.current_corpus();
+
+ if (ctxt.merge_translation_units())
+ {
+ // See if there is already a translation for the address_size
+ // and language. If so, just reuse that one.
+ for (const auto& tu : corp->get_translation_units())
+ {
+ if (tu->get_address_size() == address_size
+ && tu->get_language() == language)
+ {
+ result = tu;
+ break;
+ }
+ }
+ }
+ else
+ {
+ // See if the same translation unit exits already in the current
+ // corpus. Sometimes, the same translation unit can be present
+ // several times in the same debug info. The content of the
+ // different instances of the translation unit are different. So to
+ // represent that, we are going to re-use the same translation
+ // unit. That is, it's going to be the union of all the translation
+ // units of the same path.
+ const std::string& abs_path =
+ compilation_dir.empty() ? path : compilation_dir + "/" + path;
+ result = corp->find_translation_unit(abs_path);
+ }
if (!result)
{
result.reset(new translation_unit(ctxt.env(),
- path,
+ (ctxt.merge_translation_units()
+ ? "" : path),
address_size));
- result->set_compilation_dir_path(compilation_dir);
+ if (!ctxt.merge_translation_units())
+ result->set_compilation_dir_path(compilation_dir);
ctxt.current_corpus()->add(result);
- uint64_t l = 0;
- die_unsigned_constant_attribute(die, DW_AT_language, l);
- result->set_language(dwarf_language_to_tu_language(l));
+ result->set_language(language);
}
ctxt.cur_transl_unit(result);
diff --git a/src/abg-reader.cc b/src/abg-reader.cc
index 42985e5..cc19567 100644
--- a/src/abg-reader.cc
+++ b/src/abg-reader.cc
@@ -118,6 +118,7 @@
suppr::suppressions_type m_supprs;
bool m_tracking_non_reachable_types;
bool m_drop_undefined_syms;
+ bool m_merge_translation_units;
read_context();
@@ -129,7 +130,8 @@
m_corp_node(),
m_exported_decls_builder(),
m_tracking_non_reachable_types(),
- m_drop_undefined_syms()
+ m_drop_undefined_syms(),
+ m_merge_translation_units()
{}
/// Getter for the flag that tells us if we are tracking types that
@@ -167,6 +169,22 @@
drop_undefined_syms(bool f)
{m_drop_undefined_syms = f;}
+ /// Getter for the flag that tells us if we are merging translation
+ /// units.
+ ///
+ /// @return true iff we are merging translation units.
+ bool
+ merge_translation_units() const
+ {return m_merge_translation_units;}
+
+ /// Setter for the flag that tells us if we are merging translation
+ /// units.
+ ///
+ /// @param f the new value of the flag.
+ void
+ merge_translation_units(bool f)
+ {m_merge_translation_units = f;}
+
/// Getter of the path to the ABI file.
///
/// @return the path to the native xml abi file.
diff --git a/tools/abidw.cc b/tools/abidw.cc
index 22f640b..c20bb99 100644
--- a/tools/abidw.cc
+++ b/tools/abidw.cc
@@ -102,6 +102,7 @@
bool do_log;
bool drop_private_types;
bool drop_undefined_syms;
+ bool merge_translation_units;
type_id_style_kind type_id_style;
options()
@@ -126,6 +127,7 @@
do_log(),
drop_private_types(false),
drop_undefined_syms(false),
+ merge_translation_units(false),
type_id_style(SEQUENCE_TYPE_ID_STYLE)
{}
@@ -160,6 +162,7 @@
<< " --short-locs only print filenames rather than paths\n"
<< " --drop-private-types drop private types from representation\n"
<< " --drop-undefined-syms drop undefined symbols from representation\n"
+ << " --merge-translation-units merge translation units for same language\n"
<< " --no-comp-dir-path do not show compilation path information\n"
<< " --no-elf-needed do not show the DT_NEEDED information\n"
<< " --no-write-default-sizes do not emit pointer size when it equals"
@@ -324,6 +327,8 @@
opts.drop_private_types = true;
else if (!strcmp(argv[i], "--drop-undefined-syms"))
opts.drop_undefined_syms = true;
+ else if (!strcmp(argv[i], "--merge-translation-units"))
+ opts.merge_translation_units = true;
else if (!strcmp(argv[i], "--no-linux-kernel-mode"))
opts.linux_kernel_mode = false;
else if (!strcmp(argv[i], "--abidiff"))
@@ -811,6 +816,7 @@
opts.linux_kernel_mode);
read_context& ctxt = *c;
set_drop_undefined_syms(ctxt, opts.drop_undefined_syms);
+ set_merge_translation_units(ctxt, opts.merge_translation_units);
set_show_stats(ctxt, opts.show_stats);
set_suppressions(ctxt, opts);
abigail::dwarf_reader::set_do_log(ctxt, opts.do_log);
