Porting and Maintenance

Table of Contents

Configure and Build Hacking
Prerequisites
Overview
General Process
What Comes from Where
Configure
Storing Information in non-AC files (like configure.host)
Coding and Commenting Conventions
The acinclude.m4 layout
GLIBCXX_ENABLE, the --enable maker
Shared Library Versioning
Make
Writing and Generating Documentation
Introduction
Generating Documentation
Doxygen
Prerequisites
Generating the Doxygen Files
Debugging Generation
Markup
Docbook
Prerequisites
Generating the DocBook Files
Debugging Generation
Editing and Validation
File Organization and Basics
Markup By Example
Porting to New Hardware or Operating Systems
Operating System
CPU
Character Types
Thread Safety
Numeric Limits
Libtool
Testing
Test Organization
Directory Layout
Naming Conventions
Running the Testsuite
Basic
Variations
Permutations
Writing a new test case
Examples of Test Directives
Directives Specific to Libstdc++ Tests
Test Harness and Utilities
DejaGnu Harness Details
Utilities
Special Topics
Qualifying Exception Safety Guarantees
Overview
Existing tests
C++11 Requirements Test Sequence Descriptions
ABI Policy and Guidelines
The C++ Interface
Versioning
Goals
History
Prerequisites
Configuring
Checking Active
Allowed Changes
Prohibited Changes
Implementation
Testing
Single ABI Testing
Multiple ABI Testing
Outstanding Issues
API Evolution and Deprecation History
3.0
3.1
3.2
3.3
3.4
4.0
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5
Backwards Compatibility
First
No ios_base
No cout in <ostream.h>, no cin in <istream.h>
Second
Namespace std:: not supported
Illegal iterator usage
isspace from <cctype> is a macro
No vector::at, deque::at, string::at
No std::char_traits<char>::eof
No string::clear
Removal of ostream::form and istream::scan extensions
No basic_stringbuf, basic_stringstream
Little or no wide character support
No templatized iostreams
Thread safety issues
Third
Pre-ISO headers moved to backwards or removed
Extension headers hash_map, hash_set moved to ext or backwards
No ios::nocreate/ios::noreplace.
No stream::attach(int fd)
Support for C++98 dialect.
Support for C++TR1 dialect.
Support for C++11 dialect.
Container::iterator_type is not necessarily Container::value_type*

Configure and Build Hacking

Prerequisites

As noted previously, certain other tools are necessary for hacking on files that control configure (configure.ac, acinclude.m4) and make (Makefile.am). These additional tools (automake, and autoconf) are further described in detail in their respective manuals. All the libraries in GCC try to stay in sync with each other in terms of versions of the auto-tools used, so please try to play nicely with the neighbors.

Overview

General Process

The configure process begins the act of building libstdc++, and is started via:


configure

The configure file is a script generated (via autoconf) from the file configure.ac.

After the configure process is complete,


make all

in the build directory starts the build process. The all target comes from the Makefile file, which is generated via configure from the Makefile.in file, which is in turn generated (via automake) from the file Makefile.am.

What Comes from Where

Figure B.1. Configure and Build File Dependencies

Dependency Graph for Configure and Build Files

Regenerate all generated files by using the command autoreconf at the top level of the libstdc++ source directory.

Configure

Storing Information in non-AC files (like configure.host)

Until that glorious day when we can use AC_TRY_LINK with a cross-compiler, we have to hardcode the results of what the tests would have shown if they could be run. So we have an inflexible mess like crossconfig.m4.

Wouldn't it be nice if we could store that information in files like configure.host, which can be modified without needing to regenerate anything, and can even be tweaked without really knowing how the configury all works? Perhaps break the pieces of crossconfig.m4 out and place them in their appropriate config/{cpu,os} directory.

Alas, writing macros like "AC_DEFINE(HAVE_A_NICE_DAY)" can only be done inside files which are passed through autoconf. Files which are pure shell script can be source'd at configure time. Files which contain autoconf macros must be processed with autoconf. We could still try breaking the pieces out into "config/*/cross.m4" bits, for instance, but then we would need arguments to aclocal/autoconf to properly find them all when generating configure. I would discourage that.

Coding and Commenting Conventions

Most comments should use {octothorpes, shibboleths, hash marks, pound signs, whatever} rather than "dnl". Nearly all comments in configure.ac should. Comments inside macros written in ancillary .m4 files should. About the only comments which should not use #, but use dnl instead, are comments outside our own macros in the ancillary files. The difference is that # comments show up in configure (which is most helpful for debugging), while dnl'd lines just vanish. Since the macros in ancillary files generate code which appears in odd places, their "outside" comments tend to not be useful while reading configure.

Do not use any $target* variables, such as $target_alias. The single exception is in configure.ac, for automake+dejagnu's sake.

The acinclude.m4 layout

The nice thing about acinclude.m4/aclocal.m4 is that macros aren't actually performed/called/expanded/whatever here, just loaded. So we can arrange the contents however we like. As of this writing, acinclude.m4 is arranged as follows:

    GLIBCXX_CHECK_HOST
    GLIBCXX_TOPREL_CONFIGURE
    GLIBCXX_CONFIGURE
  

All the major variable "discovery" is done here. CXX, multilibs, etc.

    fragments included from elsewhere
  

Right now, "fragments" == "the math/linkage bits".

    GLIBCXX_CHECK_COMPILER_FEATURES
    GLIBCXX_CHECK_LINKER_FEATURES
    GLIBCXX_CHECK_WCHAR_T_SUPPORT

Next come extra compiler/linker feature tests. Wide character support was placed here because I couldn't think of another place for it. It will probably get broken apart like the math tests, because we're still disabling wchars on systems which could actually support them.

    GLIBCXX_CHECK_SETRLIMIT_ancilliary
    GLIBCXX_CHECK_SETRLIMIT
    GLIBCXX_CHECK_S_ISREG_OR_S_IFREG
    GLIBCXX_CHECK_POLL
    GLIBCXX_CHECK_WRITEV

    GLIBCXX_CONFIGURE_TESTSUITE

Feature tests which only get used in one place. Here, things used only in the testsuite, plus a couple bits used in the guts of I/O.

    GLIBCXX_EXPORT_INCLUDES
    GLIBCXX_EXPORT_FLAGS
    GLIBCXX_EXPORT_INSTALL_INFO

Installation variables, multilibs, working with the rest of the compiler. Many of the critical variables used in the makefiles are set here.

    GLIBGCC_ENABLE
    GLIBCXX_ENABLE_C99
    GLIBCXX_ENABLE_CHEADERS
    GLIBCXX_ENABLE_CLOCALE
    GLIBCXX_ENABLE_CONCEPT_CHECKS
    GLIBCXX_ENABLE_CSTDIO
    GLIBCXX_ENABLE_CXX_FLAGS
    GLIBCXX_ENABLE_C_MBCHAR
    GLIBCXX_ENABLE_DEBUG
    GLIBCXX_ENABLE_DEBUG_FLAGS
    GLIBCXX_ENABLE_LONG_LONG
    GLIBCXX_ENABLE_PCH
    GLIBCXX_ENABLE_SYMVERS
    GLIBCXX_ENABLE_THREADS

All the features which can be controlled with enable/disable configure options. Note how they're alphabetized now? Keep them like that. :-)

    AC_LC_MESSAGES
    libtool bits

Things which we don't seem to use directly, but just has to be present otherwise stuff magically goes wonky.

GLIBCXX_ENABLE, the --enable maker

All the GLIBCXX_ENABLE_FOO macros use a common helper, GLIBCXX_ENABLE. (You don't have to use it, but it's easy.) The helper does two things for us:

  1. Builds the call to the AC_ARG_ENABLE macro, with --help text properly quoted and aligned. (Death to changequote!)

  2. Checks the result against a list of allowed possibilities, and signals a fatal error if there's no match. This means that the rest of the GLIBCXX_ENABLE_FOO macro doesn't need to test for strange arguments, nor do we need to protect against empty/whitespace strings with the "x$foo" = "xbar" idiom.

Doing these things correctly takes some extra autoconf/autom4te code, which made our macros nearly illegible. So all the ugliness is factored out into this one helper macro.

Many of the macros take an argument, passed from when they are expanded in configure.ac. The argument controls the default value of the enable/disable switch. Previously, the arguments themselves had defaults. Now they don't, because that's extra complexity with zero gain for us.

There are three "overloaded signatures". When reading the descriptions below, keep in mind that the brackets are autoconf's quotation characters, and that they will be stripped. Examples of just about everything occur in acinclude.m4, if you want to look.

    GLIBCXX_ENABLE (FEATURE, DEFAULT, HELP-ARG, HELP-STRING)
    GLIBCXX_ENABLE (FEATURE, DEFAULT, HELP-ARG, HELP-STRING, permit a|b|c)
    GLIBCXX_ENABLE (FEATURE, DEFAULT, HELP-ARG, HELP-STRING, SHELL-CODE-HANDLER)
  • FEATURE is the string that follows --enable. The results of the test (such as it is) will be in the variable $enable_FEATURE, where FEATURE has been squashed. Example: [extra-foo], controlled by the --enable-extra-foo option and stored in $enable_extra_foo.

  • DEFAULT is the value to store in $enable_FEATURE if the user does not pass --enable/--disable. It should be one of the permitted values passed later. Examples: [yes], or [bar], or [$1] (which passes the argument given to the GLIBCXX_ENABLE_FOO macro as the default).

    For cases where we need to probe for particular models of things, it is useful to have an undocumented "auto" value here (see GLIBCXX_ENABLE_CLOCALE for an example).

  • HELP-ARG is any text to append to the option string itself in the --help output. Examples: [] (i.e., an empty string, which appends nothing), [=BAR], which produces --enable-extra-foo=BAR, and [@<:@=BAR@:>@], which produces --enable-extra-foo[=BAR]. See the difference? See what it implies to the user?

    If you're wondering what that line noise in the last example was, that's how you embed autoconf special characters in output text. They're called quadrigraphs and you should use them whenever necessary.

  • HELP-STRING is what you think it is. Do not include the "default" text like we used to do; it will be done for you by GLIBCXX_ENABLE. By convention, these are not full English sentences. Example: [turn on extra foo]

With no other arguments, only the standard autoconf patterns are allowed: "--{enable,disable}-foo[={yes,no}]" The $enable_FEATURE variable is guaranteed to equal either "yes" or "no" after the macro. If the user tries to pass something else, an explanatory error message will be given, and configure will halt.

The second signature takes a fifth argument, "[permit a | b | c | ...]" This allows a or b or ... after the equals sign in the option, and $enable_FEATURE is guaranteed to equal one of them after the macro. Note that if you want to allow plain --enable/--disable with no "=whatever", you must include "yes" and "no" in the list of permitted values. Also note that whatever you passed as DEFAULT must be in the list. If the user tries to pass something not on the list, a semi-explanatory error message will be given, and configure will halt. Example: [permit generic|gnu|ieee_1003.1-2001|yes|no|auto]

The third signature takes a fifth argument. It is arbitrary shell code to execute if the user actually passes the enable/disable option. (If the user does not, the default is used. Duh.) No argument checking at all is done in this signature. See GLIBCXX_ENABLE_CXX_FLAGS for an example of handling, and an error message.

Shared Library Versioning

The libstdc++.so shared library must be carefully managed to maintain binary compatible with older versions of the library. This ensures a new version of the library is still usable by programs that were linked against an older version.

Dependent on the target supporting it, the library uses ELF symbol versioning for all exported symbols. The symbol versions are defined by a linker script that assigns a version to every symbol. The set of symbols in each version is fixed when a GCC release is made, and must not change after that.

When new symbols are added to the library they must be added to a new symbol version, which must be created the first time new symbols are added after a release. Adding a new symbol version involves the following steps:

  • Edit acinclude.m4 to update the "revision" value of libtool_VERSION, e.g. from 6:22:0 to 6:23:0, which will cause the shared library to be built as libstdc++.so.6.0.23.

  • Regenerate the configure script by running the autoreconf tool from the correct version of the Autoconf package (as dictated by the GCC prerequisites).

  • Edit the file config/abi/pre/gnu.ver to add a new version node after the last new node. The node name should be GLIBCXX_3.4.X where X is the new revision set in acinclude.m4, and the node should depend on the previous version e.g.

        GLIBCXX_3.4.23 {
    
        } GLIBCXX_3.4.22;
    

    For symbols in the ABI runtime, libsupc++, the symbol version naming uses CXXABI_1.3.Y where Y increases monotonically with each new version. Again, the new node must depend on the previous version node e.g.

        CXXABI_1.3.11 {
    
        } CXXABI_1.3.10;
    

  • In order for the check-abi test target to pass the testsuite must be updated to know about the new symbol version(s). Edit the file testsuite/util/testsuite_abi.cc file to add the new versions to the known_versions list, and update the checks for the latest versions that set the latestp variable).

Once the new symbol version has been added you can add the names of your new symbols in the new version node:

    GLIBCXX_3.4.23 {

      # basic_string<C, T, A>::_Alloc_hider::_Alloc_hider(C*, A&&)
      _ZNSt7__cxx1112basic_stringI[cw]St11char_traitsI[cw]ESaI[cw]EE12_Alloc_hiderC[12]EP[cw]OS3_;

    } GLIBCXX_3.4.22;

You can either use mangled names, or demangled names inside an extern "C++" block. You might find that the new symbol matches an existing pattern in an old symbol version (causing the check-abi test target to fail). If that happens then the existing pattern must be adjusted to be more specific so that it doesn't match the new symbol.

For an example of these steps, including adjusting old patterns to be less greedy, see https://gcc.gnu.org/ml/gcc-patches/2016-07/msg01926.html and the attached patch.

If it wasn't done for the last release, you might also need to regenerate the baseline_symbols.txt file that defines the set of expected symbols for old symbol versions. A new baseline file can be generated by running make new-abi-baseline in the libbuildir/testsuite directory. Be sure to generate the baseline from a clean build using unmodified sources, or you will incorporate your local changes into the baseline file.

Make

The build process has to make all of object files needed for static or shared libraries, but first it has to generate some include files. The general order is as follows:

  1. make include files, make pre-compiled headers

  2. make libsupc++

    Generates a libtool convenience library, libsupc++convenience with language-support routines. Also generates a freestanding static library, libsupc++.a.

  3. make src

    Generates two convenience libraries, one for C++98 and one for C++11, various compatibility files for shared and static libraries, and then collects all the generated bits and creates the final libstdc++ libraries.

    1. make src/c++98

      Generates a libtool convenience library, libc++98convenience with language-support routines. Uses the -std=gnu++98 dialect.

    2. make src/c++11

      Generates a libtool convenience library, libc++11convenience with language-support routines. Uses the -std=gnu++11 dialect.

    3. make src

      Generates needed compatibility objects for shared and static libraries. Shared-only code is seggregated at compile-time via the macro _GLIBCXX_SHARED.

      Then, collects all the generated convenience libraries, adds in any required compatibility objects, and creates the final shared and static libraries: libstdc++.so and libstdc++.a.