[clang] Revise the modules document for clarity (PR #90237)
Aaron Ballman via cfe-commits
cfe-commits at lists.llvm.org
Fri Apr 26 10:16:50 PDT 2024
https://github.com/AaronBallman created https://github.com/llvm/llvm-project/pull/90237
The intention isn't to add or change the information provided, but to improve clarity through some grammar fixes, improvements to the markdown, and so forth.
>From a75aa14fcad6f346a3073ae88e91fa890e803422 Mon Sep 17 00:00:00 2001
From: Aaron Ballman <aaron at aaronballman.com>
Date: Fri, 26 Apr 2024 13:12:51 -0400
Subject: [PATCH] Revise the modules document for clarity
The intention isn't to add or change the information provided, but to
improve clarity through some grammar fixes, improvements to the
markdown, and so forth.
---
clang/docs/StandardCPlusPlusModules.rst | 1133 +++++++++++------------
1 file changed, 561 insertions(+), 572 deletions(-)
diff --git a/clang/docs/StandardCPlusPlusModules.rst b/clang/docs/StandardCPlusPlusModules.rst
index ee57fb5da64857..e0ceb2582725c2 100644
--- a/clang/docs/StandardCPlusPlusModules.rst
+++ b/clang/docs/StandardCPlusPlusModules.rst
@@ -8,79 +8,60 @@ Standard C++ Modules
Introduction
============
-The term ``modules`` has a lot of meanings. For the users of Clang, modules may
-refer to ``Objective-C Modules``, ``Clang C++ Modules`` (or ``Clang Header Modules``,
-etc.) or ``Standard C++ Modules``. The implementation of all these kinds of modules in Clang
-has a lot of shared code, but from the perspective of users, their semantics and
-command line interfaces are very different. This document focuses on
-an introduction of how to use standard C++ modules in Clang.
-
-There is already a detailed document about `Clang modules <Modules.html>`_, it
-should be helpful to read `Clang modules <Modules.html>`_ if you want to know
-more about the general idea of modules. Since standard C++ modules have different semantics
-(and work flows) from `Clang modules`, this page describes the background and use of
-Clang with standard C++ modules.
-
-Modules exist in two forms in the C++ Language Specification. They can refer to
-either "Named Modules" or to "Header Units". This document covers both forms.
+The term ``modules`` has a lot of meanings. For Clang users, modules may refer
+to ``Objective-C Modules``, `Clang Modules <Modules.html>`_ (also called
+``Clang Header Modules``, etc.) or ``C++20 Modules`` (or
+``Standard C++ Modules``). The implementation of all these kinds of modules in
+Clang shares a lot of code, but from the perspective of users, their semantics
+and command line interfaces are very different. This document focuses on an
+introduction of focusing on the use of C++20 modules in Clang. In the remainder
+of this document, the term ``modules`` will refer to Standard C++20 modules and
+the term ``Clang modules`` will refer to the Clang modules extension.
+
+Modules exist in two forms in the C++ Standard. They can refer to either
+"Named Modules" or "Header Units". This document covers both forms.
Standard C++ Named modules
==========================
-This document was intended to be a manual first and foremost, however, we consider it helpful to
-introduce some language background here for readers who are not familiar with
-the new language feature. This document is not intended to be a language
-tutorial; it will only introduce necessary concepts about the
-structure and building of the project.
+In order to understand compiler behavior, it is helpful to introduce some
+language background here for readers who are not familiar with the C++ feature.
+This document is not a tutorial on C++; it only introduces necessary concepts
+to better understand use of modules for a project.
Background and terminology
--------------------------
-Modules
-~~~~~~~
-
-In this document, the term ``Modules``/``modules`` refers to standard C++ modules
-feature if it is not decorated by ``Clang``.
-
-Clang Modules
-~~~~~~~~~~~~~
-
-In this document, the term ``Clang Modules``/``Clang modules`` refer to Clang
-c++ modules extension. These are also known as ``Clang header modules``,
-``Clang module map modules`` or ``Clang c++ modules``.
-
Module and module unit
~~~~~~~~~~~~~~~~~~~~~~
-A module consists of one or more module units. A module unit is a special
-translation unit. Every module unit must have a module declaration. The syntax
-of the module declaration is:
+A module consists of one or more module units. A module unit is a special kind
+of translation unit. Every module unit must have a module declaration. The
+syntax of the module declaration is:
.. code-block:: c++
[export] module module_name[:partition_name];
-Terms enclosed in ``[]`` are optional. The syntax of ``module_name`` and ``partition_name``
-in regex form corresponds to ``[a-zA-Z_][a-zA-Z_0-9\.]*``. In particular, a literal dot ``.``
-in the name has no semantic meaning (e.g. implying a hierarchy).
-
-In this document, module units are classified into:
-
-* Primary module interface unit.
-
-* Module implementation unit.
+Terms enclosed in ``[]`` are optional. ``module_name`` and ``partition_name``
+are typical C++ identifiers, except that they may contain a period (``.``).
+Note that a ``.`` in the name has no semantic meaning (e.g. implying a
+hierarchy).
-* Module interface partition unit.
+In this document, module units are classified as:
-* Internal module partition unit.
+* Primary module interface unit
+* Module implementation unit
+* Module interface partition unit
+* Internal module partition unit
A primary module interface unit is a module unit whose module declaration is
-``export module module_name;``. The ``module_name`` here denotes the name of the
+``export module module_name;`` where ``module_name`` denotes the name of the
module. A module should have one and only one primary module interface unit.
A module implementation unit is a module unit whose module declaration is
-``module module_name;``. A module could have multiple module implementation
-units with the same declaration.
+``module module_name;``. Multiple module implementation units can be declared
+in the same translation unit.
A module interface partition unit is a module unit whose module declaration is
``export module module_name:partition_name;``. The ``partition_name`` should be
@@ -95,22 +76,23 @@ In this document, we use the following umbrella terms:
* A ``module interface unit`` refers to either a ``primary module interface unit``
or a ``module interface partition unit``.
-* An ``importable module unit`` refers to either a ``module interface unit``
- or a ``internal module partition unit``.
+* An ``importable module unit`` refers to either a ``module interface unit`` or
+ an ``internal module partition unit``.
* A ``module partition unit`` refers to either a ``module interface partition unit``
- or a ``internal module partition unit``.
+ or an ``internal module partition unit``.
-Built Module Interface file
-~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Binary Module Interface
+~~~~~~~~~~~~~~~~~~~~~~~
-A ``Built Module Interface file`` stands for the precompiled result of an importable module unit.
-It is also called the acronym ``BMI`` generally.
+A ``Binary Module Interface`` (or ``BMI``) is the precompiled result of an
+importable module unit.
Global module fragment
~~~~~~~~~~~~~~~~~~~~~~
-In a module unit, the section from ``module;`` to the module declaration is called the global module fragment.
+The ``global module fragment`` (or ``GMF``) is the code between the ``module;``
+and the module declaration within a module unit.
How to build projects using modules
@@ -138,7 +120,7 @@ Let's see a "hello world" example that uses modules.
return 0;
}
-Then we type:
+From the command line, enter:
.. code-block:: console
@@ -148,9 +130,9 @@ Then we type:
Hello World!
In this example, we make and use a simple module ``Hello`` which contains only a
-primary module interface unit ``Hello.cppm``.
+primary module interface unit named ``Hello.cppm``.
-Then let's see a little bit more complex "hello world" example which uses the 4 kinds of module units.
+A more complex "hello world" example which uses the 4 kinds of module units is:
.. code-block:: c++
@@ -192,7 +174,7 @@ Then let's see a little bit more complex "hello world" example which uses the 4
return 0;
}
-Then we are able to compile the example by the following command:
+Back on the command line, compile the example with:
.. code-block:: console
@@ -216,51 +198,56 @@ We explain the options in the following sections.
How to enable standard C++ modules
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Currently, standard C++ modules are enabled automatically
-if the language standard is ``-std=c++20`` or newer.
+Standard C++ modules are enabled automatically if the language standard is
+``-std=c++20`` or newer.
How to produce a BMI
~~~~~~~~~~~~~~~~~~~~
-We can generate a BMI for an importable module unit by either ``--precompile``
-or ``-fmodule-output`` flags.
+To generate a BMI for an importable module unit, use either the ``--precompile``
+or ``-fmodule-output`` command line option.
-The ``--precompile`` option generates the BMI as the output of the compilation and the output path
-can be specified using the ``-o`` option.
+The ``--precompile`` option generates the BMI as the output of the compilation
+and the output path can be specified using the ``-o`` option.
-The ``-fmodule-output`` option generates the BMI as a by-product of the compilation.
-If ``-fmodule-output=`` is specified, the BMI will be emitted the specified location. Then if
-``-fmodule-output`` and ``-c`` are specified, the BMI will be emitted in the directory of the
-output file with the name of the input file with the new extension ``.pcm``. Otherwise, the BMI
-will be emitted in the working directory with the name of the input file with the new extension
+The ``-fmodule-output`` option generates the BMI as a by-product of the
+compilation. If ``-fmodule-output=`` is specified, the BMI will be emitted to
+the specified location. If ``-fmodule-output`` and ``-c`` are specified, the
+BMI will be emitted in the directory of the output file with the name of the
+input file with the extension ``.pcm``. Otherwise, the BMI will be emitted in
+the working directory with the name of the input file with the extension
``.pcm``.
-The style to generate BMIs by ``--precompile`` is called two-phase compilation since it takes
-2 steps to compile a source file to an object file. The style to generate BMIs by ``-fmodule-output``
-is called one-phase compilation respectively. The one-phase compilation model is simpler
-for build systems to implement and the two-phase compilation has the potential to compile faster due
-to higher parallelism. As an example, if there are two module units A and B, and B depends on A, the
-one-phase compilation model would need to compile them serially, whereas the two-phase compilation
-model may be able to compile them simultaneously if the compilation from A.pcm to A.o takes a long
-time.
-
-File name requirement
-~~~~~~~~~~~~~~~~~~~~~
+Generating BMIs with ``--precompile`` is called two-phase compilation because
+it takes two steps to compile a source file to an object file. Generating BMIs
+with ``-fmodule-output`` is called one-phase compilation. The one-phase
+compilation model is simpler for build systems to implement and the two-phase
+compilation has the potential to compile faster due to higher parallelism. As
+an example, if there are two module units ``A`` and ``B``, and ``B`` depends on
+``A``, the one-phase compilation model needs to compile them serially, whereas
+the two-phase compilation model may be able to compile them simultaneously if
+the compilation from ``A.pcm`` to ``A.o`` takes a long time.
+
+File name requirements
+~~~~~~~~~~~~~~~~~~~~~~
-The file name of an ``importable module unit`` should end with ``.cppm``
-(or ``.ccm``, ``.cxxm``, ``.c++m``). The file name of a ``module implementation unit``
-should end with ``.cpp`` (or ``.cc``, ``.cxx``, ``.c++``).
+By convention, ``importable module unit`` files should use ``.cppm`` (or
+``.ccm``, ``.cxxm``, or ``.c++m``) as a file extension.
+``Module implementation unit`` files should use ``.cpp`` (or ``.cc``, ``.cxx``,
+or ``.c++``) as a file extension.
-The file name of BMIs should end with ``.pcm``.
-The file name of the BMI of a ``primary module interface unit`` should be ``module_name.pcm``.
-The file name of BMIs of ``module partition unit`` should be ``module_name-partition_name.pcm``.
+A BMI should use ``.pcm`` as a file extension. The file name of the BMI for a
+``primary module interface unit`` should be ``module_name.pcm``. The file name
+of a BMI for a ``module partition unit`` should be
+``module_name-partition_name.pcm``.
-If the file names use different extensions, Clang may fail to build the module.
-For example, if the filename of an ``importable module unit`` ends with ``.cpp`` instead of ``.cppm``,
-then we can't generate a BMI for the ``importable module unit`` by ``--precompile`` option
-since ``--precompile`` option now would only run preprocessor, which is equal to `-E` now.
-If we want the filename of an ``importable module unit`` ends with other suffixes instead of ``.cppm``,
-we could put ``-x c++-module`` in front of the file. For example,
+Clang may fail to build the module if different extensions are used. For
+example, if the filename of an ``importable module unit`` ends with ``.cpp``
+instead of ``.cppm``, then Clang cannot generate a BMI for the
+``importable module unit`` with the ``--precompile`` option because the
+``--precompile`` option would only run the preprocessor (``-E``). If using a
+different extension than the conventional one for an ``importable module unit``
+you can specify ``-x c++-module`` before the file. For example,
.. code-block:: c++
@@ -279,8 +266,9 @@ we could put ``-x c++-module`` in front of the file. For example,
return 0;
}
-Now the filename of the ``module interface`` ends with ``.cpp`` instead of ``.cppm``,
-we can't compile them by the original command lines. But we are still able to do it by:
+In this example, the extension used by the ``module interface`` is ``.cpp``
+instead of ``.cppm``, so is cannot be compiled with the command lines from the
+previous example, but it can be compiled with:
.. code-block:: console
@@ -289,12 +277,12 @@ we can't compile them by the original command lines. But we are still able to do
$ ./Hello.out
Hello World!
-Module name requirement
-~~~~~~~~~~~~~~~~~~~~~~~
+Module name requirements
+~~~~~~~~~~~~~~~~~~~~~~~~
-[module.unit]p1 says:
+..
-.. code-block:: text
+ [module.unit]p1:
All module-names either beginning with an identifier consisting of std followed by zero
or more digits or containing a reserved identifier ([lex.name]) are reserved and shall not
@@ -302,7 +290,7 @@ Module name requirement
module-name is a reserved identifier, the module name is reserved for use by C++ implementations;
otherwise it is reserved for future standardization.
-So all of the following name is not valid by default:
+So none of the following names are valid by default:
.. code-block:: text
@@ -312,75 +300,76 @@ So all of the following name is not valid by default:
__test
// and so on ...
-If you still want to use the reserved module names for any reason, use
-``-Wno-reserved-module-identifier`` to suppress the warning.
+Using a reserved module name is strongly discouraged, but
+``-Wno-reserved-module-identifier`` can be used to suppress the warning.
-How to specify the dependent BMIs
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Specifying dependent BMIs
+~~~~~~~~~~~~~~~~~~~~~~~~~
-There are 3 methods to specify the dependent BMIs:
+There are 3 ways to specify a dependent BMI:
-* (1) ``-fprebuilt-module-path=<path/to/directory>``.
-* (2) ``-fmodule-file=<path/to/BMI>`` (Deprecated).
-* (3) ``-fmodule-file=<module-name>=<path/to/BMI>``.
+1. ``-fprebuilt-module-path=<path/to/directory>``.
+2. ``-fmodule-file=<path/to/BMI>`` (Deprecated).
+3. ``-fmodule-file=<module-name>=<path/to/BMI>``.
-The option ``-fprebuilt-module-path`` tells the compiler the path where to search for dependent BMIs.
-It may be used multiple times just like ``-I`` for specifying paths for header files. The look up rule here is:
+The ``-fprebuilt-module-path`` option specifies the path to search for
+dependent BMIs. Multiple paths may be specified, similar to using ``-I`` to
+specify a search path for header files. When importing a module ``M``, the
+compiler looks for ``M.pcm`` in the directories specified by
+``-fprebuilt-module-path``. Similarly, When importing a partition module unit
+``M:P``, the compiler looks for ``M-P.pcm`` in the directories specified by
+``-fprebuilt-module-path``.
-* (1) When we import module M. The compiler would look up M.pcm in the directories specified
- by ``-fprebuilt-module-path``.
-* (2) When we import partition module unit M:P. The compiler would look up M-P.pcm in the
- directories specified by ``-fprebuilt-module-path``.
-
-The option ``-fmodule-file=<path/to/BMI>`` tells the compiler to load the specified BMI directly.
-The option ``-fmodule-file=<module-name>=<path/to/BMI>`` tells the compiler to load the specified BMI
-for the module specified by ``<module-name>`` when necessary. The main difference is that
+The ``-fmodule-file=<path/to/BMI>`` option causes the compiler to load the
+specified BMI directly. The ``-fmodule-file=<module-name>=<path/to/BMI>``
+option causes the compiler to load the specified BMI for the module specified
+by ``<module-name>`` when necessary. The main difference is that
``-fmodule-file=<path/to/BMI>`` will load the BMI eagerly, whereas
-``-fmodule-file=<module-name>=<path/to/BMI>`` will only load the BMI lazily, which is similar
-with ``-fprebuilt-module-path``. The option ``-fmodule-file=<path/to/BMI>`` for named modules is deprecated
-and is planning to be removed in future versions.
+``-fmodule-file=<module-name>=<path/to/BMI>`` will only load the BMI lazily,
+which is similar to ``-fprebuilt-module-path``. The
+``-fmodule-file=<path/to/BMI>`` option for named modules is deprecated and will
+be removed in a future version of Clang.
-In case all ``-fprebuilt-module-path=<path/to/directory>``, ``-fmodule-file=<path/to/BMI>`` and
-``-fmodule-file=<module-name>=<path/to/BMI>`` exist, the ``-fmodule-file=<path/to/BMI>`` option
-takes highest precedence and ``-fmodule-file=<module-name>=<path/to/BMI>`` will take the second
-highest precedence.
+When these options are specified in the same invocation of the compiler, the
+``-fmodule-file=<path/to/BMI>`` option takes precedence over
+``-fmodule-file=<module-name>=<path/to/BMI>``, which takes precedence over
+``-fprebuilt-module-path=<path/to/directory>``.
-We need to specify all the dependent (directly and indirectly) BMIs.
-See https://github.com/llvm/llvm-project/issues/62707 for detail.
+Note: you must specify all the (directly or indirectly) dependent BMIs
+explicitly. See https://github.com/llvm/llvm-project/issues/62707 for details.
-When we compile a ``module implementation unit``, we must specify the BMI of the corresponding
-``primary module interface unit``.
-Since the language specification says a module implementation unit implicitly imports
-the primary module interface unit.
+When compiling a ``module implementation unit``, the BMI of the corresponding
+``primary module interface unit`` must be specified. This is because a module
+implementation unit implicitly imports the primary module interface unit.
[module.unit]p8
A module-declaration that contains neither an export-keyword nor a module-partition implicitly
imports the primary module interface unit of the module as if by a module-import-declaration.
-All of the 3 options ``-fprebuilt-module-path=<path/to/directory>``, ``-fmodule-file=<path/to/BMI>``
-and ``-fmodule-file=<module-name>=<path/to/BMI>`` may occur multiple times.
-For example, the command line to compile ``M.cppm`` in
-the above example could be rewritten into:
+The ``-fprebuilt-module-path=<path/to/directory>``, ``-fmodule-file=<path/to/BMI>``,
+and ``-fmodule-file=<module-name>=<path/to/BMI>`` options may be specified
+multiple times. For example, the command line to compile ``M.cppm`` in
+the previous example could be rewritten as:
.. code-block:: console
$ clang++ -std=c++20 M.cppm --precompile -fmodule-file=M:interface_part=M-interface_part.pcm -fmodule-file=M:impl_part=M-impl_part.pcm -o M.pcm
When there are multiple ``-fmodule-file=<module-name>=`` options for the same
-``<module-name>``, the last ``-fmodule-file=<module-name>=`` will override the previous
-``-fmodule-file=<module-name>=`` options.
+``<module-name>``, the last ``-fmodule-file=<module-name>=`` overrides the
+previous ``-fmodule-file=<module-name>=`` option.
-``-fprebuilt-module-path`` is more convenient and ``-fmodule-file`` is faster since
-it saves time for file lookup.
+``-fprebuilt-module-path`` is more convenient while ``-fmodule-file`` is faster
+because it saves time for file lookup.
Remember that module units still have an object counterpart to the BMI
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-It is easy to forget to compile BMIs at first since we may envision module interfaces like headers.
-However, this is not true.
-Module units are translation units. We need to compile them to object files
-and link the object files like the example shows.
+It is easy to forget to compile BMIs at first because module interfaces seem
+like header files. However, this is not the case. Module units are translation
+units. They need to be compiled to object files and those object files need to
+be linked together, as the examples have shown.
For example, the traditional compilation processes for headers are like:
@@ -400,24 +389,27 @@ And the compilation process for module units are like:
mod1.cppm -> clang++ mod1.cppm ... -> mod1.pcm --,--> clang++ mod1.pcm ... -> mod1.o -+
src2.cpp ----------------------------------------+> clang++ src2.cpp -------> src2.o -'
-As the diagrams show, we need to compile the BMI from module units to object files and link the object files.
-(But we can't do this for the BMI from header units. See the later section for the definition of header units)
+As the diagrams show, we need to compile the BMI from module units to object
+files and then link the object files. (However, we can't do this for the BMI
+from header units. See the section on :ref:`header units <header-units>` for
+more details.
-If we want to create a module library, we can't just ship the BMIs in an archive.
-We must compile these BMIs(``*.pcm``) into object files(``*.o``) and add those object files to the archive instead.
+BMIs cannot be shipped in an archive to create a module library. Instead, the
+BMIs(``*.pcm``) are compiled into object files(``*.o``) and those object files
+are added to the archive instead.
-Consistency Requirement
-~~~~~~~~~~~~~~~~~~~~~~~
+Consistency Requirements
+~~~~~~~~~~~~~~~~~~~~~~~~
-If we envision modules as a cache to speed up compilation, then - as with other caching techniques -
-it is important to keep cache consistency.
-So **currently** Clang will do very strict check for consistency.
+If modules are thought of as a kind of cache to speed up compilation, then, as
+with other caching techniques, it is important to keep cache consistency.
+**Currently**, Clang does very strict checking for consistency.
Options consistency
^^^^^^^^^^^^^^^^^^^
-The language option of module units and their non-module-unit users should be consistent.
-The following example is not allowed:
+Language dialect compiler options for module units and their non-module-unit
+uses need to be consistent. Consider the following example:
.. code-block:: c++
@@ -432,9 +424,8 @@ The following example is not allowed:
$ clang++ -std=c++20 M.cppm --precompile -o M.pcm
$ clang++ -std=c++23 Use.cpp -fprebuilt-module-path=.
-The compiler would reject the example due to the inconsistent language options.
-Not all options are language options.
-For example, the following example is allowed:
+Clang rejects the example due to the inconsistent language standard modes. Not
+all compiler options are language dialect options. For example:
.. code-block:: console
@@ -444,9 +435,12 @@ For example, the following example is allowed:
# Inconsistent debugging level.
$ clang++ -std=c++20 -g Use.cpp -fprebuilt-module-path=.
-Although the two examples have inconsistent optimization and debugging level, both of them are accepted.
+Although the optimization and debugging levels are inconsistent, these
+compilations are accepted because the compiler options do not impact the
+language dialect.
-Note that **currently** the compiler doesn't consider inconsistent macro definition a problem. For example:
+Note that the compiler **currently** doesn't reject inconsistent macro
+definitions (this may change in the future). For example:
.. code-block:: console
@@ -454,43 +448,44 @@ Note that **currently** the compiler doesn't consider inconsistent macro definit
# Inconsistent optimization level.
$ clang++ -std=c++20 -O3 -DNDEBUG Use.cpp -fprebuilt-module-path=.
-Currently Clang would accept the above example. But it may produce surprising results if the
-debugging code depends on consistent use of ``NDEBUG`` also in other translation units.
-
-Definitions consistency
-^^^^^^^^^^^^^^^^^^^^^^^
-
-The C++ language defines that same declarations in different translation units should have
-the same definition, as known as ODR (One Definition Rule). Prior to modules, the translation
-units don't dependent on each other and the compiler itself can't perform a strong
-ODR violation check. With the introduction of modules, now the compiler have
-the chance to perform ODR violations with language semantics across translation units.
-
-However, in the practice, we found the existing ODR checking mechanism is not stable
-enough. Many people suffers from the false positive ODR violation diagnostics, AKA,
-the compiler are complaining two identical declarations have different definitions
-incorrectly. Also the true positive ODR violations are rarely reported.
-Also we learned that MSVC don't perform ODR check for declarations in the global module
-fragment.
-
-So in order to get better user experience, save the time checking ODR and keep consistent
-behavior with MSVC, we disabled the ODR check for the declarations in the global module
-fragment by default. Users who want more strict check can still use the
-``-Xclang -fno-skip-odr-check-in-gmf`` flag to get the ODR check enabled. It is also
-encouraged to report issues if users find false positive ODR violations or false negative ODR
-violations with the flag enabled.
+Currently, Clang accepts the above example though it may produce surprising
+results if the debugging code depends on consistent use of ``NDEBUG`` in other
+translation units.
+
+Object definition consistency
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The C++ language requires that the same declarations in different translation
+units have the same definition, which is known as the One Definition Rule (ODR).
+Prior to modules, the compiler cannot perform strong ODR violation checking
+because it only sees one translation unit at a time. With use of modules, the
+compiler can perform checks for ODR violations across translation units.
+
+However, the current ODR checking mechanisms are not perfect. There are a
+significant number of false positive ODR violation diagnostics, where the
+compiler incorrectly diagnoses two identical declarations as having different
+definitions. Further, true positive ODR violations are not always reported.
+Additionally, MSVC does not currently perform ODR checking for declarations in
+the global module fragment.
+
+To give a better user experience, improve compilation performance, and for
+consistentency with MSVC, ODR checking of declarations in the global module
+fragment is disabled by default. These checks can be enabled by specifying
+``-Xclang -fno-skip-odr-check-in-gmf`` when compiling. If the check is enabled
+and you encounter incorrect or missing diagnostics, please report them via the
+`community issue tracker <https://github.com/llvm/llvm-project/issues/>`_.
ABI Impacts
-----------
-This section describes the new ABI changes brought by modules.
-
-Only Itanium C++ ABI related change are mentioned
+This section describes the new ABI changes brought by modules. Only Itanium C++
+ABI related change are mentioned.
-Mangling Names
-~~~~~~~~~~~~~~
+Name Mangling
+~~~~~~~~~~~~~
-The declarations in a module unit which are not in the global module fragment have new linkage names.
+The declarations in a module unit which are not in the global module fragment
+have new linkage names.
For example,
@@ -501,22 +496,24 @@ For example,
export int foo();
}
-The linkage name of ``NS::foo()`` would be ``_ZN2NSW1M3fooEv``.
-This couldn't be demangled by previous versions of the debugger or demangler.
-As of LLVM 15.x, users can utilize ``llvm-cxxfilt`` to demangle this:
+The linkage name of ``NS::foo()`` is ``_ZN2NSW1M3fooEv``. This couldn't be
+demangled by previous versions of the debugger or demangler. As of LLVM 15.x,
+``llvm-cxxfilt`` can be used to demangle this:
.. code-block:: console
$ llvm-cxxfilt _ZN2NSW1M3fooEv
+ NS::foo at M()
+
+The result should be read as ``NS::foo()`` in module ``M``.
-The result would be ``NS::foo at M()``, which reads as ``NS::foo()`` in module ``M``.
-
-The ABI implies that we can't declare something in a module unit and define it in a non-module unit (or vice-versa),
-as this would result in linking errors.
+The ABI implies that something cannot be declared in a module unit and defined
+in a non-module unit (or vice-versa), as this would result in linking errors.
-If we still want to implement declarations within the compatible ABI in module unit,
-we can use the language-linkage specifier. Since the declarations in the language-linkage specifier
-is attached to the global module fragments. For example:
+Despite this, it is possbile to implement declarations with a compatible ABI in
+a module unit by using a language linkage specifier because the declarations in
+the language linkage specifier are attached to the global module fragment. For
+example:
.. code-block:: c++
@@ -530,43 +527,43 @@ Now the linkage name of ``NS::foo()`` will be ``_ZN2NS3fooEv``.
Module Initializers
~~~~~~~~~~~~~~~~~~~
-All the importable module units are required to emit an initializer function.
-The initializer function should contain calls to importing modules first and
-all the dynamic-initializers in the current module unit then.
-
-Translation units explicitly or implicitly importing named modules must call
-the initializer functions of the imported named modules within the sequence of
-the dynamic-initializers in the TU. Initializations of entities at namespace
-scope are appearance-ordered. This (recursively) extends into imported modules
-at the point of appearance of the import declaration.
+All importable module units are required to emit an initializer function. The
+initializer function emits calls to imported modules first followed by calls
+to all to dynamic initializers in the current module unit.
-It is allowed to omit calls to importing modules if it is known empty.
+Translation units that explicitly or implicitly import a named module must call
+the initializer functions of the imported named module within the sequence of
+the dynamic initializers in the translation unit. Initializations of entities
+at namespace scope are appearance-ordered. This (recursively) extends to
+imported modules at the point of appearance of the import declaration.
-It is allowed to omit calls to importing modules for which is known to be called.
+If the imported module is known to be empty, the call to its initializer may be
+omitted. Additionally, if the imported module is known to have already been
+imported, the call to its initializer may be omitted.
Reduced BMI
-----------
-To support the 2 phase compilation model, Clang chose to put everything needed to
-produce an object into the BMI. But every consumer of the BMI, except itself, doesn't
-need such informations. It makes the BMI to larger and so may introduce unnecessary
-dependencies into the BMI. To mitigate the problem, we decided to reduce the information
-contained in the BMI.
+To support the two-phase compilation model, Clang puts everything needed to
+produce an object into the BMI. However, other consumers of the BMI generally
+don't need that informations. This makes the BMI larger and may introduce
+unnecessary dependencies for the BMI. To mitigate the problem, Clang added a
+compiler option to reduce the information contained in the BMI. These two
+formats are known as Full BMI and Reduced BMI, respectively.
-To be clear, we call the default BMI as Full BMI and the new introduced BMI as Reduced
-BMI.
+Users can use the ``-fexperimental-modules-reduced-bmi`` option to produce a
+Reduced BMI.
-Users can use ``-fexperimental-modules-reduced-bmi`` flag to enable the Reduced BMI.
+For the one-phase compilation model (CMake implements this model), with
+``-fexperimental-modules-reduced-bmi``, the generated BMI will be a Reduced
+BMI automatically. (The output path of the BMI is specified by
+``-fmodule-output=`` as usual with the one-phase compilation model).
-For one phase compilation model (CMake implements this model), with
-``-fexperimental-modules-reduced-bmi``, the generated BMI will be Reduced BMI automatically.
-(The output path of the BMI is specified by ``-fmodule-output=`` as usual one phase
-compilation model).
-
-It is still possible to support Reduced BMI in two phase compilation model. With
-``-fexperimental-modules-reduced-bmi``, ``--precompile`` and ``-fmodule-output=`` specified,
-the generated BMI specified by ``-o`` will be full BMI and the BMI specified by
-``-fmodule-output=`` will be Reduced BMI. The dependency graph may be:
+It is also possible to produce a Reduced BMI with the two-phase compilation
+model. When ``-fexperimental-modules-reduced-bmi``, ``--precompile``, and
+``-fmodule-output=`` are specified, the generated BMI specified by ``-o`` will
+be a full BMI and the BMI specified by ``-fmodule-output=`` will be a Reduced
+BMI. The dependency graph in this case would look like:
.. code-block:: none
@@ -577,15 +574,16 @@ the generated BMI specified by ``-o`` will be full BMI and the BMI specified by
-> ...
-> consumer_n.cpp
-We don't emit diagnostics if ``-fexperimental-modules-reduced-bmi`` is used with a non-module
-unit. This design helps the end users of one phase compilation model to perform experiments
-early without asking for the help of build systems. The users of build systems which supports
-two phase compilation model still need helps from build systems.
+Clang does not emit diagnostics when ``-fexperimental-modules-reduced-bmi`` is
+used with a non-module unit. This design helps the end users of the one-phase
+compilation model to perform experiments without needing to modify the build
+system. The two-phase compilation module requires build system support.
-Within Reduced BMI, we won't write unreachable entities from GMF, definitions of non-inline
-functions and non-inline variables. This may not be a transparent change.
-`[module.global.frag]ex2 <https://eel.is/c++draft/module.global.frag#example-2>`_ may be a good
-example:
+In a Reduced BMI, Clang does not emit unreachable entities from the global
+module fragment, or definitions of non-inline functions and non-inline
+variables. This may not be a transparent change.
+
+[module.global.frag]ex2 provides a good example:
.. code-block:: c++
@@ -633,22 +631,25 @@ example:
// module M's interface, so is discarded
int c = use_h<int>(); // OK
-In the above example, the function definition of ``N::g`` is elided from the Reduced
-BMI of ``M.cppm``. Then the use of ``use_g<int>`` in ``M-impl.cpp`` fails
-to instantiate. For such issues, users can add references to ``N::g`` in the module purview
-of ``M.cppm`` to make sure it is reachable, e.g., ``using N::g;``.
-
-We think the Reduced BMI is the correct direction. But given it is a drastic change,
-we'd like to make it experimental first to avoid breaking existing users. The roadmap
-of Reduced BMI may be:
-
-1. ``-fexperimental-modules-reduced-bmi`` is opt in for 1~2 releases. The period depends
-on testing feedbacks.
-2. We would announce Reduced BMI is not experimental and introduce ``-fmodules-reduced-bmi``.
-and suggest users to enable this mode. This may takes 1~2 releases too.
-3. Finally we will enable this by default. When that time comes, the term BMI will refer to
-the reduced BMI today and the Full BMI will only be meaningful to build systems which
-loves to support two phase compilations.
+In the above example, the function definition of ``N::g`` is elided from the
+Reduced BMI of ``M.cppm``. Then the use of ``use_g<int>`` in ``M-impl.cpp``
+fails to instantiate. For such issues, users can add references to ``N::g`` in
+the module purview of ``M.cppm`` to ensure it is reachable, e.g.,
+``using N::g;``.
+
+Long-term, Clang is likely to make Reduced BMIs the default rather than Full
+BMIs. But because it is a drastic change, it is initially an experimental
+option to avoid breaking existing users. The expected roadmap for Reduced BMIs
+as of Clang 19.x is:
+
+1. ``-fexperimental-modules-reduced-bmi`` is opt-in for 1~2 releases. The period depends
+ on user feedback and may be extended.
+2. Announce that Reduced BMIs are no longer experimental and introduce
+ ``-fmodules-reduced-bmi`` as a new option, and recommend use of the new
+ option. This transition is expected to take 1~2 additional releases as well.
+3. Finally, ``-fmodules-reduced-bmi`` will be the default. When that time
+ comes, the term BMI will refer to the reduced BMI and the Full BMI will only
+ be meaningful to build systems which elect to support two-phase compilation.
Performance Tips
----------------
@@ -656,13 +657,10 @@ Performance Tips
Reduce duplications
~~~~~~~~~~~~~~~~~~~
-While it is legal to have duplicated declarations in the global module fragments
-of different module units, it is not free for clang to deal with the duplicated
-declarations. In other word, for a translation unit, it will compile slower if the
-translation unit itself and its importing module units contains a lot duplicated
-declarations.
-
-For example,
+While it is valid to have duplicated declarations in the global module fragments
+of different module units, it is not free for Clang to deal with the duplicated
+declarations. A translation unit will compile more slowly if it and its
+imported module units contain a lot of duplicated declarations. For example:
.. code-block:: c++
@@ -698,9 +696,9 @@ For example,
import M;
... // use declarations from module M.
-When ``big.header.h`` is big enough and there are a lot of partitions,
-the compilation of ``use.cpp`` may be slower than
-the following style significantly:
+When ``big.header.h`` is big enough and there are a lot of partitions, the
+compilation of ``use.cpp`` may be significantly slower than the following
+approach:
.. code-block:: c++
@@ -738,22 +736,21 @@ the following style significantly:
import M;
... // use declarations from module M.
-The key part of the tip is to reduce the duplications from the text includes.
+Reducing the duplication from textual includes is what improves compile-time
+performance.
-Ideas for converting to modules
--------------------------------
+Transitioning to modules
+------------------------
-For new libraries, we encourage them to use modules completely from day one if possible.
-This will be pretty helpful to make the whole ecosystems to get ready.
-
-For many existing libraries, it may be a breaking change to refactor themselves
-into modules completely. So that many existing libraries need to provide headers and module
+New code and libraries should use modules from day one if possible. However,
+it may be a breaking change for existing code or libraries to refactor to use
+modules. As a result, many existing libraries need to provide headers and module
interfaces for a while to not break existing users.
-Here we provide some ideas to ease the transition process for existing libraries.
-**Note that the this section is only about helping ideas instead of requirement from clang**.
-Let's start with the case that there is no dependency or no dependent libraries providing
-modules for your library.
+This section provides some ideas on how to ease the transition process for
+existing libraries. **Note that this information is only intended as helpful
+tips instead of requirements from Clang.** It presumes the project is starting
+with no module-based dependencies.
ABI non-breaking styles
~~~~~~~~~~~~~~~~~~~~~~~
@@ -776,9 +773,9 @@ export-using style
using decl_n;
}
-As the example shows, you need to include all the headers containing declarations needs
-to be exported and `using` such declarations in an `export` block. Then, basically,
-we're done.
+This example shows how to include all the headers containing declarations which
+need to be exported and uses `using` declarations in an `export` block to
+produce the module interface. Then, basically, we're done.
export extern-C++ style
^^^^^^^^^^^^^^^^^^^^^^^
@@ -799,7 +796,7 @@ export extern-C++ style
#include "header_n.h"
}
-Then in your headers (from ``header_1.h`` to ``header_n.h``), you need to define the macro:
+Headers (from ``header_1.h`` to ``header_n.h``) need to define the macro:
.. code-block:: c++
@@ -809,9 +806,10 @@ Then in your headers (from ``header_1.h`` to ``header_n.h``), you need to define
#define EXPORT
#endif
-And you should put ``EXPORT`` to the beginning of the declarations you want to export.
+and put ``EXPORT`` on the declarations you want to export.
-Also it is suggested to refactor your headers to include thirdparty headers conditionally:
+Also, it is recommended to refactor headers to include third-party headers
+conditionally:
.. code-block:: c++
@@ -823,26 +821,25 @@ Also it is suggested to refactor your headers to include thirdparty headers cond
...
-This may be helpful to get better diagnostic messages if you forgot to update your module
-interface unit file during maintaining.
+This can be helpful because it gives better diagnostic messages if the module
+interface unit is not properly updated when modifying code.
-The reasoning for the practice is that the declarations in the language linkage are considered
-to be attached to the global module. So the ABI of your library in the modular version
-wouldn't change.
+This approach works because the declarations with language linkage are attached
+to the global module. Thus, the ABI of the modular form of the library does not
+change.
-While this style looks not as convenient as the export-using style, it is easier to convert
-to other styles.
+While this style is more involved than the export-using style, it makes it
+easier to further refactor the library to other styles.
ABI breaking style
~~~~~~~~~~~~~~~~~~
-The term ``ABI breaking`` sounds terrifying generally. But you may want it here if you want
-to force your users to introduce your library in a consistent way. E.g., they either include
-your headers all the way or import your modules all the way.
-The style prevents the users to include your headers and import your modules at the same time
-in the same repo.
+The term ``ABI breaking`` may sound like a bad approach. However, this style
+forces consumers of the library use it in a consistent way. e.g., either always
+include headers for the library or always import modules. The style prevents
+the ability to mix includes and imports for the library.
-The pattern for ABI breaking style is similar with export extern-C++ style.
+The pattern for ABI breaking style is similar to the export extern-C++ style.
.. code-block:: c++
@@ -875,11 +872,11 @@ The pattern for ABI breaking style is similar with export extern-C++ style.
}
#endif
-(And add `EXPORT` and conditional include to the headers as suggested in the export
-extern-C++ style section)
+(And add `EXPORT` and conditional include to the headers as suggested in the
+export extern-C++ style section.)
-Remember that the ABI get changed and we need to compile our source files into the
-new ABI format. This is the job of the additional part of the interface unit:
+The ABI with modules is different and so we need to compile the source files
+into the new ABI. This is done by an additional part of the interface unit:
.. code-block:: c++
@@ -900,16 +897,17 @@ new ABI format. This is the job of the additional part of the interface unit:
}
#endif
-In case the number of your source files are small, we may put everything in the private
-module fragment directly. (it is suggested to add conditional include to the source
-files too). But it will make the compilation of the module interface unit to be slow
-when the number of the source files are not small enough.
+If the number of source files is small, everything can be put in the private
+module fragment directly. (It is recommended to add conditional includes to the
+source files too). However, compile time performance will be slow if there are
+a lot of source files to compile.
-**Note that the private module fragment can only be in the primary module interface unit
-and the primary module interface unit containing private module fragment should be the only
-module unit of the corresponding module.**
+**Note that the private module fragment can only be in the primary module
+interface unit and the primary module interface unit containing the private
+module fragment should be the only module unit of the corresponding module.**
-In that case, you need to convert your source files (.cpp files) to module implementation units:
+In this case, source files (.cpp files) must be converted to module
+implementation units:
.. code-block:: c++
@@ -925,45 +923,41 @@ In that case, you need to convert your source files (.cpp files) to module imple
// Following off should be unchanged.
...
-The module implementation unit will import the primary module implicitly.
-We don't include any headers in the module implementation units
-here since we want to avoid duplicated declarations between translation units.
-This is the reason why we add non-exported using declarations from the third
-party libraries in the primary module interface unit.
+The module implementation unit will import the primary module implicitly. Do
+not include any headers in the module implementation units because that avoids
+duplicated declarations between translation units. This is why non-exported
+using declarations are added from third-party libraries in the primary module
+interface unit.
-And if you provide your library as ``libyour_library.so``, you probably need to
-provide a modular one ``libyour_library_modules.so`` since you changed the ABI.
+If the library is provided as ``libyour_library.so``, a modular library (e.g.,
+``libyour_library_modules.so``) may also need to be provided for ABI
+compatibility.
What if there are headers only inclued by the source files
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-The above practice may be problematic if there are headers only included by the source
-files. If you're using private module fragment, you may solve the issue by including them
-in the private module fragment. While it is OK to solve it by including the implementation
-headers in the module purview if you're using implementation module units, it may be
-suboptimal since the primary module interface units now containing entities not belongs
-to the interface.
-
-If you're a perfectionist, maybe you can improve it by introducing internal module partition unit.
+The above practice may be problematic if there are headers only included by the
+source files. If using a private module fragment, this issue may be solved by
+including those headers in the private module fragment. While it is OK to solve
+it by including the implementation headers in the module purview when using
+implementation module units, it may be suboptimal because the primary module
+interface units now contain entities that do not belong to the interface.
-The internal module partition unit is an importable module unit which is internal
-to the module itself. The concept just meets the headers only included by the source files.
-
-We don't show code snippet since it may be too verbose or not good or not general.
-But it may not be too hard if you can understand the points of the section.
+This can potentially be improved by introducing internal module partition unit.
+The internal module partition unit is an importable module unit which is
+internal to the module itself. However, this approach may not always be the
+best way forward.
Providing a header to skip parsing redundant headers
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-It is a problem for clang to handle redeclarations between translation units.
-Also there is a long standing issue in clang (`problematic include after import <https://github.com/llvm/llvm-project/issues/61465>`_).
-But even if the issue get fixed in clang someday, the users may still get slower compilation speed
-and larger BMI size. So it is suggested to not include headers after importing the corresponding
-library.
-
-However, it is not easy for users if your library are included by other dependencies.
-
-So the users may have to write codes like:
+Many redeclarations shared between translation units causes Clang to have
+slower compile-time performance. Further, there are known issues with
+`include after import <https://github.com/llvm/llvm-project/issues/61465>`_.
+Even when that issue is resolved, users may still get slower compilation speed
+and larger BMIs. For these reasons, it is recommended to not include headers
+after importing the corresponding module. However, it is not always easy if the
+library is included by other dependencies, as in:
.. code-block:: c++
@@ -977,9 +971,9 @@ or
import your_library;
#include "third_party/A.h" // #include "your_library/a_header.h"
-For such cases, we suggest the libraries providing modules and the headers at the same time
-to provide a header to skip parsing all the headers in your libraries. So the users can
-import your library as the following style to skip redundant handling:
+For such cases, it is best if the library providing both module and header
+interfaces also provides a header which skips parsing so that the library can
+be imported with the following approach that skips redundant redeclarations:
.. code-block:: c++
@@ -987,9 +981,9 @@ import your library as the following style to skip redundant handling:
#include "your_library_imported.h"
#include "third_party/A.h" // #include "your_library/a_header.h" but got skipped
-The implementation of ``your_library_imported.h`` can be a set of controlling macros or
-an overall controlling macro if you're using `#pragma once`. So you can convert your
-headers to:
+The implementation of ``your_library_imported.h`` can be a set of controlling
+macros or an overall controlling macro if using `#pragma once`. Then headers
+can be refactored to:
.. code-block:: c++
@@ -998,25 +992,23 @@ headers to:
...
#endif
-If the modules imported by your library provides such headers too, remember to add them to
-your ``your_library_imported.h`` too.
+If the modules imported by the library provide such headers, remember to add
+them to ``your_library_imported.h`` too.
Importing modules
~~~~~~~~~~~~~~~~~
-When there are dependent libraries providing modules, we suggest you to import that in
-your module.
-
-Most of the existing libraries would fall into this catagory once the std module gets available.
+When there are dependent libraries providing modules, they should be imported
+in your module as well. Many existing libraries will fall into this catagory
+once the ``std`` module is more widely available.
All dependent libraries providing modules
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-Life gets easier if all the dependent libraries providing modules.
+Life gets easier if all the dependent libraries provide modules.
-You need to convert your headers to include thirdparty headers conditionally.
-
-Then for export-using style:
+Headers need to be converted to include third-party headers conditionally. Then,
+for the export-using style:
.. code-block:: c++
@@ -1035,7 +1027,7 @@ Then for export-using style:
using decl_n;
}
-For export extern-C++ style:
+or, for the export extern-C++ style:
.. code-block:: c++
@@ -1049,7 +1041,7 @@ For export extern-C++ style:
#include "header_n.h"
}
-For ABI breaking style,
+or, for the ABI-breaking style,
.. code-block:: c++
@@ -1069,35 +1061,39 @@ For ABI breaking style,
#include "source_n.cpp"
#endif
-We don't need the non-exported using declarations if we're using implementation module
-units now. We can import thirdparty modules directly in the implementation module
-units.
+Non-exported using declarations are not needed if using implementation module
+units. Instead, third-party modules can be imported directly in implementation
+module units.
Partial dependent libraries providing modules
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-In this case, we have to mix the use of ``include`` and ``import`` in the module of our
-library. The key point here is still to remove duplicated declarations in translation
-units as much as possible. If the imported modules provide headers to skip parsing their
-headers, we should include that after the including. If the imported modules don't provide
-the headers, we can make it ourselves if we still want to optimize it.
+If the library has to mix the use of ``include`` and ``import`` in its module,
+the key point is still to remove duplicated declarations in translation units
+as much as possible. If the imported modules provide headers to skip parsing
+their headers, those should be included after the import. If the imported
+modules don't provide such a header, one can be made manually for improved
+compile time performance.
Known Problems
--------------
-The following describes issues in the current implementation of modules.
-Please see https://github.com/llvm/llvm-project/labels/clang%3Amodules for more issues
-or file a new issue if you don't find an existing one.
-If you're going to create a new issue for standard C++ modules,
-please start the title with ``[C++20] [Modules]`` (or ``[C++23] [Modules]``, etc)
-and add the label ``clang:modules`` (if you have permissions for that).
+The following describes issues in the current implementation of modules. Please
+see
+`the issues list for moduled <https://github.com/llvm/llvm-project/labels/clang%3Amodules>`_
+for a list of issues or to file a new issue if you don't find an existing one.
+When creating a new issue for standard C++ modules, please start the title with
+``[C++20] [Modules]`` (or ``[C++23] [Modules]``, etc) and add the label
+``clang:modules`` if possible.
-For higher level support for proposals, you could visit https://clang.llvm.org/cxx_status.html.
+A high-level overview of support for standards features, including modules, can
+be found on the `C++ Feature Status <https://clang.llvm.org/cxx_status.html>`_
+page.
Including headers after import is problematic
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-For example, the following example can be accept:
+The following example is accepted:
.. code-block:: c++
@@ -1110,8 +1106,8 @@ For example, the following example can be accept:
return 0;
}
-but it will get rejected if we reverse the order of ``#include <iostream>`` and
-``import foo;``:
+but if the order of ``#include <iostream>`` and ``import foo;`` is reversed,
+then the code is currently rejected:
.. code-block:: c++
@@ -1126,33 +1122,31 @@ but it will get rejected if we reverse the order of ``#include <iostream>`` and
Both of the above examples should be accepted.
-This is a limitation in the implementation. In the first example,
-the compiler will see and parse <iostream> first then the compiler will see the import.
-So the ODR Checking and declarations merging will happen in the deserializer.
-In the second example, the compiler will see the import first and the include second.
-As a result, the ODR Checking and declarations merging will happen in the semantic analyzer.
-
-So there is divergence in the implementation path. It might be understandable that why
-the orders matter here in the case.
-(Note that "understandable" is different from "makes sense").
-
-This is tracked in: https://github.com/llvm/llvm-project/issues/61465
-
-Ignored PreferredName Attribute
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+This is a limitation in the implementation. In the first example, the compiler
+will see and parse ``<iostream>`` first then it will see the ``import``. In
+this case, ODR checking and declaration merging will happen in the
+deserializer. In the second example, the compiler will see the ``import`` first
+and the ``#include`` second which results in ODR checking and declarations
+merging happening in the semantic analyzer. This is due to a divergence in the
+implementation path. This is tracked by
+`#61465 <https://github.com/llvm/llvm-project/issues/61465>`_.
-Due to a tricky problem, when Clang writes BMIs, Clang will ignore the ``preferred_name`` attribute, if any.
-This implies that the ``preferred_name`` wouldn't show in debugger or dumping.
+Ignored ``prefered_name`` Attribute
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-This is tracked in: https://github.com/llvm/llvm-project/issues/56490
+When Clang writes BMIs, it will ignore the ``preferred_name`` attribute on
+declarations which use it. Thus, the preferred name will not be displayed in
+the debugger as expected. This is tracked by
+`#56490 <https://github.com/llvm/llvm-project/issues/56490>`_.
Don't emit macros about module declaration
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-This is covered by P1857R3. We mention it again here since users may abuse it before we implement it.
+This is covered by `P1857R3 <https://wg21.link/P1857R3>`_. It is mentioned here
+because we want users to be aware that we don't yet implement it.
-Someone may want to write code which could be compiled both by modules or non-modules.
-A direct idea would be use macros like:
+A direct approach to write code that can be compiled by both modules and
+non-module builds may look like:
.. code-block:: c++
@@ -1162,39 +1156,37 @@ A direct idea would be use macros like:
IMPORT header_name
EXPORT ...
-So this file could be triggered like a module unit or a non-module unit depending on the definition
-of some macros.
-However, this kind of usage is forbidden by P1857R3 but we haven't implemented P1857R3 yet.
-This means that is possible to write illegal modules code now, and obviously this will stop working
-once P1857R3 is implemented.
-A simple suggestion would be "Don't play macro tricks with module declarations".
+With the idea being that this file can be compiled like a module unit or a
+non-module unit depending on the definition of some macros. However, this kind
+of usage is forbidden by P1857R3 which is not yet implemented in Clang. This
+means that is possible to write invalid modules which no longer be accepted
+once P1857R3 is implemented. This is tracked by
+`#56917 <https://github.com/llvm/llvm-project/issues/56917>`_.
+
+Until then, it is recommended not to mix macros with module declarations.
-This is tracked in: https://github.com/llvm/llvm-project/issues/56917
In consistent filename suffix requirement for importable module units
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Currently, clang requires the file name of an ``importable module unit`` should end with ``.cppm``
-(or ``.ccm``, ``.cxxm``, ``.c++m``). However, the behavior is inconsistent with other compilers.
-
-This is tracked in: https://github.com/llvm/llvm-project/issues/57416
+Currently, Clang requires the file name of an ``importable module unit`` to
+have ``.cppm`` (or ``.ccm``, ``.cxxm``, ``.c++m``) as the file extension.
+However, the behavior is inconsistent with other compilers. This is tracked by
+`#57416 <https://github.com/llvm/llvm-project/issues/57416>`_.
clang-cl is not compatible with the standard C++ modules
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Now we can't use the `/clang:-fmodule-file` or `/clang:-fprebuilt-module-path` to specify
-the BMI within ``clang-cl.exe``.
-
-This is tracked in: https://github.com/llvm/llvm-project/issues/64118
+``/clang:-fmodule-file`` and ``/clang:-fprebuilt-module-path`` cannot be used
+to specify the BMI with ``clang-cl.exe``. This is tracked by
+`#64118 <https://github.com/llvm/llvm-project/issues/64118>`_.
false positive ODR violation diagnostic due to using inconsistent qualified but the same type
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-ODR violation is a pretty common issue when using modules.
-Sometimes the program violated the One Definition Rule actually.
-But sometimes it shows the compiler gives false positive diagnostics.
-
-One often reported example is:
+ODR violations are a common issue when using modules. Sometimes it's a true
+positive result where the program violates the One Definition Rule, other times
+it is a false positive result. One often reported example is:
.. code-block:: c++
@@ -1222,33 +1214,31 @@ One often reported example is:
export module repro;
export import :part;
-Currently the compiler complains about the inconsistent definition of `fun()` in
-2 module units. This is incorrect. Since both definitions of `fun()` has the same
-spelling and `T` refers to the same type entity finally. So the program should be
-fine.
-
-This is tracked in https://github.com/llvm/llvm-project/issues/78850.
+Currently the compiler incorrectly diagnoses the inconsistent definition of
+``fun()`` in two module units. Because both definitions of ``fun()`` have the
+same spelling and ``T`` refers to the same type entity, so there is no ODR
+violation. This is tracked by
+`#78850 <https://github.com/llvm/llvm-project/issues/78850>`_.
Using TU-local entity in other units
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-Module units are translation units. So the entities which should only be local to the
-module unit itself shouldn't be used by other units in any means.
+Module units are translation units, so the entities which should be local to
+the module unit itself should never be used by other units.
-In the language side, to address the idea formally, the language specification defines
-the concept of ``TU-local`` and ``exposure`` in
+The C++ standard defines the concept of ``TU-local`` and ``exposure`` in
`basic.link/p14 <https://eel.is/c++draft/basic.link#14>`_,
`basic.link/p15 <https://eel.is/c++draft/basic.link#15>`_,
`basic.link/p16 <https://eel.is/c++draft/basic.link#16>`_,
-`basic.link/p17 <https://eel.is/c++draft/basic.link#17>`_ and
+`basic.link/p17 <https://eel.is/c++draft/basic.link#17>`_, and
`basic.link/p18 <https://eel.is/c++draft/basic.link#18>`_.
-However, the compiler doesn't support these 2 ideas formally.
-This results in unclear and confusing diagnostic messages.
-And it is worse that the compiler may import TU-local entities to other units without any
-diagnostics.
+However, Clang doesn't formally support these two ideas. This results in
+unclear or confusing diagnostic messages. Further, Clang may import TU-local
+entities to other units without any diagnostics. This is tracked by
+`#78173 <https://github.com/llvm/llvm-project/issues/78173>`_.
-This is tracked in https://github.com/llvm/llvm-project/issues/78173.
+.. _header-units:
Header Units
============
@@ -1258,15 +1248,14 @@ How to build projects using header unit
.. warning::
- The user interfaces of header units is highly experimental. There are still
- many unanswered question about how tools should interact with header units.
- The user interfaces described here may change after we have progress on how
- tools should support for header units.
+ The user interfaces of header units are experimental. There are still many
+ unanswered question about how tools should interact with header units. The
+ user interfaces described here may change in the future.
Quick Start
~~~~~~~~~~~
-For the following example,
+The following example:
.. code-block:: c++
@@ -1275,7 +1264,7 @@ For the following example,
std::cout << "Hello World.\n";
}
-we could compile it as
+could be compiled with:
.. code-block:: console
@@ -1285,14 +1274,14 @@ we could compile it as
How to produce BMIs
~~~~~~~~~~~~~~~~~~~
-Similar to named modules, we could use ``--precompile`` to produce the BMI.
-But we need to specify that the input file is a header by ``-xc++-system-header`` or ``-xc++-user-header``.
+Similar to named modules, ``--precompile`` can be used to produce a BMI.
+However, that requires specifying that the input file is a header by using
+``-xc++-system-header`` or ``-xc++-user-header``.
-Also we could use `-fmodule-header={user,system}` option to produce the BMI for header units
-which has suffix like `.h` or `.hh`.
-The value of `-fmodule-header` means the user search path or the system search path.
-The default value for `-fmodule-header` is `user`.
-For example,
+The ``-fmodule-header={user,system}`` option can also be used to produce the BMI
+for header units which have a file extension like `.h` or `.hh`. The argument to
+``-fmodule-header`` specifies either the user search path or the system search
+path. The default value for ``-fmodule-header`` is ``user``. For example:
.. code-block:: c++
@@ -1308,16 +1297,16 @@ For example,
Hello();
}
-We could compile it as:
+could be compiled with:
.. code-block:: console
$ clang++ -std=c++20 -fmodule-header foo.h -o foo.pcm
$ clang++ -std=c++20 -fmodule-file=foo.pcm use.cpp
-For headers which don't have a suffix, we need to pass ``-xc++-header``
-(or ``-xc++-system-header`` or ``-xc++-user-header``) to mark it as a header.
-For example,
+For headers which do not have a file extension, ``-xc++-header`` (or
+``-xc++-system-header``, ``-xc++-user-header``) must be used to specify the
+file as a header. For example:
.. code-block:: c++
@@ -1335,19 +1324,22 @@ For example,
How to specify the dependent BMIs
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-We could use ``-fmodule-file`` to specify the BMIs, and this option may occur multiple times as well.
+``-fmodule-file`` can be used to specify a dependent BMI (or multiple times for
+more than one dependent BMI).
-With the existing implementation ``-fprebuilt-module-path`` cannot be used for header units
-(since they are nominally anonymous).
-For header units, use ``-fmodule-file`` to include the relevant PCM file for each header unit.
+With the existing implementation, ``-fprebuilt-module-path`` cannot be used for
+header units (because they are nominally anonymous). For header units, use
+``-fmodule-file`` to include the relevant PCM file for each header unit.
-This is expect to be solved in future editions of the compiler either by the tooling finding and specifying
-the -fmodule-file or by the use of a module-mapper that understands how to map the header name to their PCMs.
+This is expect to be solved in a future version of Clang either by the compiler
+finding and specifying ``-fmodule-file`` automatically, or by the use of a
+module-mapper that understands how to map the header name to their PCMs.
Don't compile the BMI
~~~~~~~~~~~~~~~~~~~~~
-Another difference with modules is that we can't compile the BMI from a header unit.
+Because of the semantics of header units, which are just like headers, another
+difference with modules is that we can't compile the BMI from a header unit.
For example:
.. code-block:: console
@@ -1356,15 +1348,13 @@ For example:
# This is not allowed!
$ clang++ iostream.pcm -c -o iostream.o
-It makes sense due to the semantics of header units, which are just like headers.
-
Include translation
~~~~~~~~~~~~~~~~~~~
-The C++ spec allows the vendors to convert ``#include header-name`` to ``import header-name;`` when possible.
-Currently, Clang would do this translation for the ``#include`` in the global module fragment.
-
-For example, the following two examples are the same:
+The C++ standard allows vendors to convert ``#include header-name`` to
+``import header-name;`` when possible. Currently, Clang does this translation
+for the ``#include`` in the global module fragment. For example, the following
+example:
.. code-block:: c++
@@ -1375,7 +1365,7 @@ For example, the following two examples are the same:
std::cout << "Hello.\n";
}
-with the following one:
+is the same as this example:
.. code-block:: c++
@@ -1391,17 +1381,17 @@ with the following one:
$ clang++ -std=c++20 -xc++-system-header --precompile iostream -o iostream.pcm
$ clang++ -std=c++20 -fmodule-file=iostream.pcm --precompile M.cppm -o M.cpp
-In the latter example, the Clang could find the BMI for the ``<iostream>``
-so it would try to replace the ``#include <iostream>`` to ``import <iostream>;`` automatically.
+In the latter example, Clang can find the BMI for ``<iostream>`` and so it
+tries to replace the ``#include <iostream>`` with ``import <iostream>;``
+automatically.
Relationships between Clang modules
-----------------------------------
-Header units have pretty similar semantics with Clang modules.
-The semantics of both of them are like headers.
-
-In fact, we could even "mimic" the sytle of header units by Clang modules:
+Header units have similar semantics to Clang modules. The semantics of both are
+like headers. In fact, header units can be mimicked by Clang modules as in the
+following example:
.. code-block:: c++
@@ -1414,46 +1404,43 @@ In fact, we could even "mimic" the sytle of header units by Clang modules:
$ clang++ -std=c++20 -fimplicit-modules -fmodule-map-file=.modulemap main.cpp
-It would be simpler if we are using libcxx:
+It is simpler when using libc++:
.. code-block:: console
$ clang++ -std=c++20 main.cpp -fimplicit-modules -fimplicit-module-maps
-Since there is already one
-`module map <https://github.com/llvm/llvm-project/blob/main/libcxx/include/module.modulemap.in>`_
-in the source of libcxx.
-
-Then immediately leads to the question: why don't we implement header units through Clang header modules?
+because libc++ already supplies a
+`module map <https://github.com/llvm/llvm-project/blob/main/libcxx/include/module.modulemap.in>`_.
-The main reason for this is that Clang modules have more semantics like hierarchy or
-wrapping multiple headers together as a big module.
-However, these things are not part of Standard C++ Header units,
-and we want to avoid the impression that these additional semantics get interpreted as Standard C++ behavior.
+This raises the question: why are header units not implemented through Clang
+modules?
-Another reason is that there are proposals to introduce module mappers to the C++ standard
-(for example, https://wg21.link/p1184r2).
-If we decide to reuse Clang's modulemap, we may get in trouble once we need to introduce another module mapper.
+The main reason is because Clang modules have more hierarchical semantics when
+wrapping multiple headers together as one module, which is not supported by
+Standard C++ Header units. We want to avoid the impression that these
+additional semantics get interpreted as Standard C++ behavior.
-So the final answer for why we don't reuse the interface of Clang modules for header units is that
-there are some differences between header units and Clang modules and that ignoring those
-differences now would likely become a problem in the future.
+Another reason is that there are proposals to introduce module mappers to the
+C++ standard (for example, https://wg21.link/p1184r2). Reusing Clang's
+modulemap may be more difficult if we need to introduce another module mapper.
Discover Dependencies
=====================
-Prior to modules, all the translation units can be compiled parallelly.
-But it is not true for the module units. The presence of module units requires
-us to compile the translation units in a (topological) order.
+Prior to modules, all the translation units in a project can be compiled in
+parallel. The same is not true when using module units. The presence of module
+units requires compiling the translation units in a topological order.
-The clang-scan-deps scanner implemented
+The ``clang-scan-deps`` scanner implemented
`P1689 paper <https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2022/p1689r5.html>`_
-to describe the order. Only named modules are supported now.
+to describe the order. Only named modules are supported currently.
-We need a compilation database to use clang-scan-deps. See
+A compilation database is needed when using ``clang-scan-deps``. See
`JSON Compilation Database Format Specification <JSONCompilationDatabase.html>`_
-for example. Note that the ``output`` entry is necessary for clang-scan-deps
-to scan P1689 format. Here is an example:
+for more information about compilation databases. Note that the ``output``
+entry is necessary for ``clang-scan-deps`` to scan using the P1689 format. For
+example:
.. code-block:: c++
@@ -1533,13 +1520,13 @@ And here is the compilation database:
}
]
-And we can get the dependency information in P1689 format by:
+To get the dependency information in P1689 format, use:
.. code-block:: console
$ clang-scan-deps -format=p1689 -compilation-database P1689.json
-And we will get:
+to get:
.. code-block:: text
@@ -1619,14 +1606,14 @@ And we will get:
See the P1689 paper for the meaning of the fields.
-And if the user want a finer-grained control for any reason, e.g., to scan the generated source files,
-the user can choose to get the dependency information per file. For example:
+Getting dependency information per file with finer-grained control (such as
+scanning generated source files) is possible. For example:
.. code-block:: console
$ clang-scan-deps -format=p1689 -- <path-to-compiler-executable>/clang++ -std=c++20 impl_part.cppm -c -o impl_part.o
-And we'll get:
+will get:
.. code-block:: text
@@ -1652,22 +1639,23 @@ And we'll get:
"version": 1
}
-In this way, we can pass the single command line options after the ``--``.
-Then clang-scan-deps will extract the necessary information from the options.
-Note that we need to specify the path to the compiler executable instead of saying
-``clang++`` simply.
+Individual command line options can be specified after the ``--``. Then
+``clang-scan-deps`` will extract the necessary information from the specified
+options. Note that the path to the compiler executable needs to be specified
+explicitly instead of using ``clang++`` directly.
-The users may want the scanner to get the transitional dependency information for headers.
-Otherwise, the users have to scan twice for the project, once for headers and once for modules.
-To address the requirement, clang-scan-deps will recognize the specified preprocessor options
-in the given command line and generate the corresponding dependency information. For example,
+Users may want the scanner to get the transitional dependency information for
+headers. Otherwise, the project has to be scanned twice, once for headers and
+once for modules. To address this, ``clang-scan-deps`` will recognize the
+specified preprocessor options in the given command line and generate the
+corresponding dependency information. For example:
.. code-block:: console
$ clang-scan-deps -format=p1689 -- ../bin/clang++ -std=c++20 impl_part.cppm -c -o impl_part.o -MD -MT impl_part.ddi -MF impl_part.dep
$ cat impl_part.dep
-We will get:
+will produce:
.. code-block:: text
@@ -1679,41 +1667,41 @@ We will get:
/usr/include/bits/types/__locale_t.h \
...
-When clang-scan-deps detects ``-MF`` option, clang-scan-deps will try to write the
+When ``clang-scan-deps`` detects the ``-MF`` option, it will try to write the
dependency information for headers to the file specified by ``-MF``.
Possible Issues: Failed to find system headers
----------------------------------------------
-In case the users encounter errors like ``fatal error: 'stddef.h' file not found``,
-probably the specified ``<path-to-compiler-executable>/clang++`` refers to a symlink
-instead a real binary. There are 4 potential solutions to the problem:
-
-* (1) End users can resolve the issue by pointing the specified compiler executable to
- the real binary instead of the symlink.
-* (2) End users can invoke ``<path-to-compiler-executable>/clang++ -print-resource-dir``
- to get the corresponding resource directory for your compiler and add that directory
- to the include search paths manually in the build scripts.
-* (3) Build systems that use a compilation database as the input for clang-scan-deps
- scanner, the build system can add the flag ``--resource-dir-recipe invoke-compiler`` to
- the clang-scan-deps scanner to calculate the resources directory dynamically.
- The calculation happens only once for a unique ``<path-to-compiler-executable>/clang++``.
-* (4) For build systems that invokes the clang-scan-deps scanner per file, repeatedly
- calculating the resource directory may be inefficient. In such cases, the build
- system can cache the resource directory by itself and pass ``-resource-dir <resource-dir>``
- explicitly in the command line options:
+If encountering an error like ``fatal error: 'stddef.h' file not found``,
+the specified ``<path-to-compiler-executable>/clang++`` probably refers to a
+symlink instead a real binary. There are four potential solutions to the
+problem:
-.. code-block:: console
+1. Point the specified compiler executable to the real binary instead of the
+ symlink.
+2. Invoke ``<path-to-compiler-executable>/clang++ -print-resource-dir`` to get
+ the corresponding resource directory for your compiler and add that
+ directory to the include search paths manually in the build scripts.
+3. For build systems that use a compilation database as the input for
+ ``clang-scan-deps``, the build system can add the
+ ``--resource-dir-recipe invoke-compiler`` option when executing
+ ``clang-scan-deps`` to calculate the resource directory dynamically.
+ The calculation happens only once for a unique ``<path-to-compiler-executable>/clang++``.
+4. For build systems that invoke ``clang-scan-deps`` per file, repeatedly
+ calculating the resource directory may be inefficient. In such cases, the
+ build system can cache the resource directory and specify
+ ``-resource-dir <resource-dir>`` explicitly, as in:
- $ clang-scan-deps -format=p1689 -- <path-to-compiler-executable>/clang++ -std=c++20 -resource-dir <resource-dir> mod.cppm -c -o mod.o
+ .. code-block:: console
+
+ $ clang-scan-deps -format=p1689 -- <path-to-compiler-executable>/clang++ -std=c++20 -resource-dir <resource-dir> mod.cppm -c -o mod.o
Import modules with clang-repl
==============================
-We're able to import C++20 named modules with clang-repl.
-
-Let's start with a simple example:
+``clang-repl`` supports importing C++20 named modules. For example:
.. code-block:: c++
@@ -1723,7 +1711,7 @@ Let's start with a simple example:
return "Hello Interpreter for Modules!";
}
-We still need to compile the named module in ahead.
+The named module still needs to be compiled ahead of time.
.. code-block:: console
@@ -1731,10 +1719,9 @@ We still need to compile the named module in ahead.
$ clang++ M.pcm -c -o M.o
$ clang++ -shared M.o -o libM.so
-Note that we need to compile the module unit into a dynamic library so that the clang-repl
-can load the object files of the module units.
-
-Then we are able to import module ``M`` in clang-repl.
+Note that the module unit needs to be compiled as a dynamic library so that
+``clang-repl`` can load the object files of the module units. Then it is
+possible to import module ``M`` in clang-repl.
.. code-block:: console
@@ -1753,17 +1740,18 @@ Possible Questions
How modules speed up compilation
--------------------------------
-A classic theory for the reason why modules speed up the compilation is:
-if there are ``n`` headers and ``m`` source files and each header is included by each source file,
-then the complexity of the compilation is ``O(n*m)``;
-But if there are ``n`` module interfaces and ``m`` source files, the complexity of the compilation is
-``O(n+m)``. So, using modules would be a big win when scaling.
-In a simpler word, we could get rid of many redundant compilations by using modules.
+A classic theory for the reason why modules speed up the compilation is: if
+there are ``n`` headers and ``m`` source files and each header is included by
+each source file, then the complexity of the compilation is ``O(n*m)``. But if
+there are ``n`` module interfaces and ``m`` source files, the complexity of the
+compilation is ``O(n+m)``. So, using modules would be a big win when scaling.
+In a simpler word, we could get rid of many redundant compilations by using
+modules.
-Roughly, this theory is correct. But the problem is that it is too rough.
-The behavior depends on the optimization level, as we will illustrate below.
+Roughly, this theory is correct. But the problem is that it is too rough. The
+behavior depends on the optimization level, as illustrated below.
-First is ``O0``. The compilation process is described in the following graph.
+First is ``-O0``. The compilation process is described in the following graph.
.. code-block:: none
@@ -1785,18 +1773,17 @@ First is ``O0``. The compilation process is described in the following graph.
│ │
└--------┘
-Here we can see that the source file (could be a non-module unit or a module unit) would get processed by the
-whole pipeline.
-But the imported code would only get involved in semantic analysis, which is mainly about name lookup,
-overload resolution and template instantiation.
-All of these processes are fast relative to the whole compilation process.
-More importantly, the imported code only needs to be processed once in frontend code generation,
-as well as the whole middle end and backend.
-So we could get a big win for the compilation time in O0.
-
-But with optimizations, things are different:
+In this case, the source file (which could be a non-module unit or a module
+unit) would get processed by the whole pipeline. But the imported code would
+only get involved in semantic analysis, which is mainly about name lookup,
+overload resolution, and template instantiation. All of these processes are
+fast relative to the whole compilation process. More importantly, the imported
+code only needs to be processed once during frontend code generation, as well
+as the whole middle end and backend. So we could get a big win for the
+compilation time in ``-O0``.
-(we omit ``code generation`` part for each end due to the limited space)
+But with optimizations, things are different (the ``code generation`` part for
+each end is omitted due to limited space):
.. code-block:: none
@@ -1817,27 +1804,29 @@ But with optimizations, things are different:
│ │
└---------------------------------------┘
-It would be very unfortunate if we end up with worse performance after using modules.
-The main concern is that when we compile a source file, the compiler needs to see the function body
-of imported module units so that it can perform IPO (InterProcedural Optimization, primarily inlining
-in practice) to optimize functions in current source file with the help of the information provided by
-the imported module units.
-In other words, the imported code would be processed again and again in importee units
-by optimizations (including IPO itself).
-The optimizations before IPO and the IPO itself are the most time-consuming part in whole compilation process.
-So from this perspective, we might not be able to get the improvements described in the theory.
-But we could still save the time for optimizations after IPO and the whole backend.
-
-Overall, at ``O0`` the implementations of functions defined in a module will not impact module users,
-but at higher optimization levels the definitions of such functions are provided to user compilations for the
-purposes of optimization (but definitions of these functions are still not included in the use's object file)-
-this means the build speedup at higher optimization levels may be lower than expected given ``O0`` experience,
-but does provide by more optimization opportunities.
+It would be very unfortunate if we end up with worse performance when using
+modules. The main concern is that when a source file is compiled, the compiler
+needs to see the body of imported module units so that it can perform IPO
+(InterProcedural Optimization, primarily inlining in practice) to optimize
+functions in the current source file with the help of the information provided
+by the imported module units. In other words, the imported code would be
+processed again and again in importee units by optimizations (including IPO
+itself). The optimizations before IPO and IPO itself are the most time-consuming
+part in whole compilation process. So from this perspective, it might not be
+possible to get the compile time improvements described in the theory, but
+there could be time savings for optimizations after IPO and the whole backend.
+
+Overall, at ``-O0`` the implementations of functions defined in a module will
+not impact module users, but at higher optimization levels the definitions of
+such functions are provided to user compilations for the purposes of
+optimization (but definitions of these functions are still not included in the
+use's object file) -- this means the build speedup at higher optimization
+levels may be lower than expected given ``-O0`` experience, but does provide
+more optimization opportunities.
Interoperability with Clang Modules
-----------------------------------
-We **wish** to support clang modules and standard c++ modules at the same time,
-but the mixed using form is not well used/tested yet.
-
-Please file new github issues as you find interoperability problems.
+We **wish** to support Clang modules and standard C++ modules at the same time,
+but the mixing them together is not well used/tested yet. Please file new
+github issues as you find interoperability problems.
More information about the cfe-commits
mailing list