Virtualization
/
meson
mirror of https://gitlab.com/qemu-project/meson.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
meson/docs/markdown/Reference-tables.md

331 lines
14 KiB
Markdown
Raw Normal View History

Added a page for reference tables.
2017-09-17 00:16:00 +08:00
# Reference tables
## Compiler ids
Add new compiler.get_argument_syntax method Some compilers try very had to pretend they're another compiler (ICC pretends to be GCC and Linux and MacOS, and MSVC on windows), Clang behaves much like GCC, but now also has clang-cl, which behaves like MSVC. This method provides an easy way to determine whether testing for MSVC like arguments `/w1234` or gcc like arguments `-Wfoo` are likely to succeed, without having to check for dozens of compilers and the host operating system, (as you would otherwise have to do with ICC).
2018-10-17 01:03:13 +08:00
These are return values of the `get_id` (Compiler family) and
`get_argument_syntax` (Argument syntax) method in a compiler object.
docs/reference-table: Add intel-cl Because the Intel compiler behaves significantly differently on windows than it does on Linux and MacOS I've decided it would be better to follow the clang/clang-cl split and make id "intel-cl" on windows (leaving "intel" alone on Linux and Mac). Since we've never supported ICL and it hasn't worked in the past I think this is an okay change to make.
2019-05-11 04:56:04 +08:00
| Value | Compiler family | Argument syntax |
| ----- | --------------- | --------------- |
| arm | ARM compiler | |
| armclang | ARMCLANG compiler | |
-Add xc16 and c2000 C,Cpp toolchain support
2020-03-21 04:13:42 +08:00
| c2000 | Texas Instruments C2000 compiler | |
Add support for the CompCert C Compiler * Add preliminary support for the CompCert C Compiler The intention is to use this with the picolibc, so some GCC flags are automatically filtered. Since CompCert uses GCC is for linking, those GCC-linker flags which are used by picolibc, are automatically prefixed with '-WUl', so that they're passed to GCC. Squashed commit of the following: commit 4e0ad66dca9de301d2e41e74aea4142afbd1da7d Author: Sebastian Meyer <meyer@absint.com> Date: Mon Aug 31 14:20:39 2020 +0200 remove '-fall' from default arguments, also filter -ftls-model=.* commit 41afa3ccc62ae72824eb319cb8b34b7e6693cb67 Author: Sebastian Meyer <meyer@absint.com> Date: Mon Aug 31 14:13:55 2020 +0200 use regex for filtering ccomp args commit d68d242d0ad22f8bf53923ce849da9b86b696a75 Author: Sebastian Meyer <meyer@absint.com> Date: Mon Aug 31 13:54:36 2020 +0200 filter some gcc arguments commit 982a01756266bddbbd211c54e8dbfa2f43dec38f Author: Sebastian Meyer <meyer@absint.com> Date: Fri Aug 28 15:03:14 2020 +0200 fix ccomp meson configuration commit dce0bea00b1caa094b1ed0c6c77cf6c12f0f58d9 Author: Sebastian Meyer <meyer@absint.com> Date: Thu Aug 27 13:02:19 2020 +0200 add CompCert to meson (does not fully work, yet) * remove unused import and s/cls/self/ fixes the two obvious LGTM warnings * CompCert: Do not ignore unsupported GCC flags Some are safe to ignore, however, as per https://github.com/mesonbuild/meson/pull/7674, they should not be ignored by meson itself. Instead the meson.build should take care to select only those which are actually supported by the compiler. * remove unused variable * Only add arguments once. * Apply suggestions from code review Co-authored-by: Dylan Baker <dylan@pnwbakers.com> * Remove erroneous ' ' from '-o {}'.format() As noticed by @dcbaker * added release note snippet for compcert * properly split parameters As suggested by @dcbaker, these parameters should be properly split into multiple strings. Co-authored-by: Dylan Baker <dylan@pnwbakers.com> * Update add_compcert_compiler.md Added a sentence about the state of the implementation (experimental); use proper markdown * properly separate arguments Co-authored-by: Dylan Baker <dylan@pnwbakers.com>
2020-09-16 02:51:21 +08:00
| ccomp | The CompCert formally-verified C compiler | |
docs/reference-table: Add intel-cl Because the Intel compiler behaves significantly differently on windows than it does on Linux and MacOS I've decided it would be better to follow the clang/clang-cl split and make id "intel-cl" on windows (leaving "intel" alone on Linux and Mac). Since we've never supported ICL and it hasn't worked in the past I think this is an okay change to make.
2019-05-11 04:56:04 +08:00
| ccrx | Renesas RX Family C/C++ compiler | |
| clang | The Clang compiler | gcc |
| clang-cl | The Clang compiler (MSVC compatible driver) | msvc |
| dmd | D lang reference compiler | |
Add basic Webassembly support via Emscripten.
2019-06-20 02:56:00 +08:00
| emscripten| Emscripten WASM compiler | |
docs/reference-table: Add intel-cl Because the Intel compiler behaves significantly differently on windows than it does on Linux and MacOS I've decided it would be better to follow the clang/clang-cl split and make id "intel-cl" on windows (leaving "intel" alone on Linux and Mac). Since we've never supported ICL and it hasn't worked in the past I think this is an okay change to make.
2019-05-11 04:56:04 +08:00
| flang | Flang Fortran compiler | |
| g95 | The G95 Fortran compiler | |
| gcc | The GNU Compiler Collection | gcc |
| intel | Intel compiler (Linux and Mac) | gcc |
| intel-cl | Intel compiler (Windows) | msvc |
| lcc | Elbrus C/C++/Fortran Compiler | |
| llvm | LLVM-based compiler (Swift, D) | |
| mono | Xamarin C# compiler | |
| msvc | Microsoft Visual Studio | msvc |
| nagfor | The NAG Fortran compiler | |
doc
2020-07-13 12:20:33 +08:00
| nvidia_hpc| NVidia HPC SDK compilers | |
docs/reference-table: Add intel-cl Because the Intel compiler behaves significantly differently on windows than it does on Linux and MacOS I've decided it would be better to follow the clang/clang-cl split and make id "intel-cl" on windows (leaving "intel" alone on Linux and Mac). Since we've never supported ICL and it hasn't worked in the past I think this is an okay change to make.
2019-05-11 04:56:04 +08:00
| open64 | The Open64 Fortran Compiler | |
| pathscale | The Pathscale Fortran compiler | |
-Add xc16 and c2000 C,Cpp toolchain support
2020-03-21 04:13:42 +08:00
| pgi | Portland PGI C/C++/Fortran compilers | |
docs/reference-table: Add intel-cl Because the Intel compiler behaves significantly differently on windows than it does on Linux and MacOS I've decided it would be better to follow the clang/clang-cl split and make id "intel-cl" on windows (leaving "intel" alone on Linux and Mac). Since we've never supported ICL and it hasn't worked in the past I think this is an okay change to make.
2019-05-11 04:56:04 +08:00
| rustc | Rust compiler | |
| sun | Sun Fortran compiler | |
| valac | Vala compiler | |
-Add xc16 and c2000 C,Cpp toolchain support
2020-03-21 04:13:42 +08:00
| xc16 | Microchip XC16 C compiler | |
Added a page for reference tables.
2017-09-17 00:16:00 +08:00
add compiler.get_linker_id() method this can be useful for if/elif where linker behaviors must be considered. For example, clang with "link" vs gcc with "ld.bfd" etc. ci for compiler.get_linker_id() method doc add @FeatureNew check Co-Authored-By: Daniel Mensinger <daniel@mensinger-ka.de>
2019-12-15 20:40:43 +08:00
## Linker ids
docs: Fix reference table for linker names These are pretty much all over the place because I never intended them to be exposed to the meson source language, they were meant just for documentation.
2020-02-19 06:22:13 +08:00
These are return values of the `get_linker_id` method in a compiler object.
linkers: Update the linker names to be more consistent This makes two basic changes, 1 it moves the name of the linker into the linker class, this should reduce the number of errors and typos, and ensure that a linker always has one name. This then renames the linkers to have more consistent names. Posix/gnu linkers are called ld.<name>: ld.gold, ld.lld, ld.solaris. Apple linkers are renamed ld64.
2020-02-20 03:02:12 +08:00
| Value | Linker family |
| ----- | --------------- |
compilers/linkers: Add a representation for wasm-ld Emscripten does have a stand alone linker, wasm-ld. This patch adds the linker, adds detection for the linker, and removes the IsLinkerMixin for emscripten. This is a little more correct, and makes the code a lot cleaner and more robust.
2020-02-26 03:36:31 +08:00
| ld.bfd | The GNU linker |
linkers: Update the linker names to be more consistent This makes two basic changes, 1 it moves the name of the linker into the linker class, this should reduce the number of errors and typos, and ensure that a linker always has one name. This then renames the linkers to have more consistent names. Posix/gnu linkers are called ld.<name>: ld.gold, ld.lld, ld.solaris. Apple linkers are renamed ld64.
2020-02-20 03:02:12 +08:00
| ld.gold | The GNU gold linker |
| ld.lld | The LLVM linker, with the GNU interface |
| ld.solaris | Solaris and illumos |
compilers/linkers: Add a representation for wasm-ld Emscripten does have a stand alone linker, wasm-ld. This patch adds the linker, adds detection for the linker, and removes the IsLinkerMixin for emscripten. This is a little more correct, and makes the code a lot cleaner and more robust.
2020-02-26 03:36:31 +08:00
| ld.wasm | emscripten's wasm-ld linker |
linkers: Update the linker names to be more consistent This makes two basic changes, 1 it moves the name of the linker into the linker class, this should reduce the number of errors and typos, and ensure that a linker always has one name. This then renames the linkers to have more consistent names. Posix/gnu linkers are called ld.<name>: ld.gold, ld.lld, ld.solaris. Apple linkers are renamed ld64.
2020-02-20 03:02:12 +08:00
| ld64 | Apple ld64 |
| link | MSVC linker |
| lld-link | The LLVM linker, with the MSVC interface |
| xilink | Used with Intel-cl only, MSVC like |
| optlink | optlink (used with DMD) |
| rlink | The Renesas linker, used with CCrx only |
-Add xc16 and c2000 C,Cpp toolchain support
2020-03-21 04:13:42 +08:00
| xc16-ar | The Microchip linker, used with XC16 only |
| ar2000 | The Texas Instruments linker, used with C2000 only |
linkers: Update the linker names to be more consistent This makes two basic changes, 1 it moves the name of the linker into the linker class, this should reduce the number of errors and typos, and ensure that a linker always has one name. This then renames the linkers to have more consistent names. Posix/gnu linkers are called ld.<name>: ld.gold, ld.lld, ld.solaris. Apple linkers are renamed ld64.
2020-02-20 03:02:12 +08:00
| armlink | The ARM linker (arm and armclang compilers) |
| pgi | Portland/Nvidia PGI |
| nvlink | Nvidia Linker used with cuda |
Add support for the CompCert C Compiler * Add preliminary support for the CompCert C Compiler The intention is to use this with the picolibc, so some GCC flags are automatically filtered. Since CompCert uses GCC is for linking, those GCC-linker flags which are used by picolibc, are automatically prefixed with '-WUl', so that they're passed to GCC. Squashed commit of the following: commit 4e0ad66dca9de301d2e41e74aea4142afbd1da7d Author: Sebastian Meyer <meyer@absint.com> Date: Mon Aug 31 14:20:39 2020 +0200 remove '-fall' from default arguments, also filter -ftls-model=.* commit 41afa3ccc62ae72824eb319cb8b34b7e6693cb67 Author: Sebastian Meyer <meyer@absint.com> Date: Mon Aug 31 14:13:55 2020 +0200 use regex for filtering ccomp args commit d68d242d0ad22f8bf53923ce849da9b86b696a75 Author: Sebastian Meyer <meyer@absint.com> Date: Mon Aug 31 13:54:36 2020 +0200 filter some gcc arguments commit 982a01756266bddbbd211c54e8dbfa2f43dec38f Author: Sebastian Meyer <meyer@absint.com> Date: Fri Aug 28 15:03:14 2020 +0200 fix ccomp meson configuration commit dce0bea00b1caa094b1ed0c6c77cf6c12f0f58d9 Author: Sebastian Meyer <meyer@absint.com> Date: Thu Aug 27 13:02:19 2020 +0200 add CompCert to meson (does not fully work, yet) * remove unused import and s/cls/self/ fixes the two obvious LGTM warnings * CompCert: Do not ignore unsupported GCC flags Some are safe to ignore, however, as per https://github.com/mesonbuild/meson/pull/7674, they should not be ignored by meson itself. Instead the meson.build should take care to select only those which are actually supported by the compiler. * remove unused variable * Only add arguments once. * Apply suggestions from code review Co-authored-by: Dylan Baker <dylan@pnwbakers.com> * Remove erroneous ' ' from '-o {}'.format() As noticed by @dcbaker * added release note snippet for compcert * properly split parameters As suggested by @dcbaker, these parameters should be properly split into multiple strings. Co-authored-by: Dylan Baker <dylan@pnwbakers.com> * Update add_compcert_compiler.md Added a sentence about the state of the implementation (experimental); use proper markdown * properly separate arguments Co-authored-by: Dylan Baker <dylan@pnwbakers.com>
2020-09-16 02:51:21 +08:00
| ccomp | CompCert used as the linker driver |
docs: Fix reference table for linker names These are pretty much all over the place because I never intended them to be exposed to the meson source language, they were meant just for documentation.
2020-02-19 06:22:13 +08:00
For languages that don't have separate dynamic linkers such as C# and Java, the
`get_linker_id` will return the compiler name.
Added a page for reference tables.
2017-09-17 00:16:00 +08:00
## Script environment variables
| Value | Comment |
| ----- | ------- |
| MESONINTROSPECT | Command to run to run the introspection command, may be of the form `python /path/to/meson introspect`, user is responsible for splitting the path if necessary. |
docs: Alphabetically sort reference table rows [skip ci]
2018-11-22 19:49:45 +08:00
| MESON_BUILD_ROOT | Absolute path to the build dir |
Add support for custom dist scripts.
2018-07-18 22:11:57 +08:00
| MESON_DIST_ROOT | Points to the root of the staging directory, only set when running `dist` scripts |
docs: Alphabetically sort reference table rows [skip ci]
2018-11-22 19:49:45 +08:00
| MESON_SOURCE_ROOT | Absolute path to the source dir |
| MESON_SUBDIR | Current subdirectory, only set for `run_command` |
Add support for custom dist scripts.
2018-07-18 22:11:57 +08:00
Added a page for reference tables.
2017-09-17 00:16:00 +08:00
## CPU families
These are returned by the `cpu_family` method of `build_machine`,
`host_machine` and `target_machine`. For cross compilation they are
set in the cross file.
Add m68k to known cpus for Motorola 68000 series processors Fixes #6492
2020-01-22 03:01:12 +08:00
| Value | Comment |
| ----- | ------- |
| aarch64 | 64 bit ARM processor |
| alpha | DEC Alpha processor |
| arc | 32 bit ARC processor |
| arm | 32 bit ARM processor |
Add AVR to cpu families. Closes #7085.
2020-05-10 23:16:23 +08:00
| avr | Atmel AVR processor |
-Add xc16 and c2000 C,Cpp toolchain support
2020-03-21 04:13:42 +08:00
| c2000 | 32 bit C2000 processor |
docs: Alphabetically sort the cpu_family reference table It's close, but not quite. This should make it easier to read
2020-06-20 01:54:18 +08:00
| dspic | 16 bit Microchip dsPIC |
| e2k | MCST Elbrus processor |
Add m68k to known cpus for Motorola 68000 series processors Fixes #6492
2020-01-22 03:01:12 +08:00
| ia64 | Itanium processor |
add loongarch support
2020-10-06 22:42:43 +08:00
| loongarch64 | 64 bit Loongson processor|
Add m68k to known cpus for Motorola 68000 series processors Fixes #6492
2020-01-22 03:01:12 +08:00
| m68k | Motorola 68000 processor |
| microblaze | MicroBlaze processor |
| mips | 32 bit MIPS processor |
| mips64 | 64 bit MIPS processor |
| parisc | HP PA-RISC processor |
docs: Alphabetically sort the cpu_family reference table It's close, but not quite. This should make it easier to read
2020-06-20 01:54:18 +08:00
| pic24 | 16 bit Microchip PIC24 |
Add m68k to known cpus for Motorola 68000 series processors Fixes #6492
2020-01-22 03:01:12 +08:00
| ppc | 32 bit PPC processors |
| ppc64 | 64 bit PPC processors |
| riscv32 | 32 bit RISC-V Open ISA |
| riscv64 | 64 bit RISC-V Open ISA |
| rl78 | Renesas RL78 |
| rx | Renesas RX 32 bit MCU |
| s390 | IBM zSystem s390 |
| s390x | IBM zSystem s390x |
envconfig: Add support SuperH SH-4 Fixes: #7358
2020-06-20 01:56:09 +08:00
| sh4 | SuperH SH-4 |
Add m68k to known cpus for Motorola 68000 series processors Fixes #6492
2020-01-22 03:01:12 +08:00
| sparc | 32 bit SPARC |
| sparc64 | SPARC v9 processor |
| wasm32 | 32 bit Webassembly |
| wasm64 | 64 bit Webassembly |
| x86 | 32 bit x86 processor |
| x86_64 | 64 bit x86 processor |
Added a page for reference tables.
2017-09-17 00:16:00 +08:00
Any cpu family not listed in the above list is not guaranteed to
remain stable in future releases.
Capitalize "Meson" consistently as it is a proper name. [skip ci]
2021-02-01 00:00:24 +08:00
Those porting from autotools should note that Meson does not add
docs: Add note to cpu_family table about endianess [skip ci] Autotools will add endianess to the name of the architecture in some cases (such as ppc64le vs ppc64) meson doesn't do this. It's worth noting this in the documentation.
2018-10-03 01:14:21 +08:00
endianness to the name of the cpu_family. For example, autotools
Capitalize "Meson" consistently as it is a proper name. [skip ci]
2021-02-01 00:00:24 +08:00
will call little endian PPC64 "ppc64le", Meson will not, you must
docs: Add note to cpu_family table about endianess [skip ci] Autotools will add endianess to the name of the architecture in some cases (such as ppc64le vs ppc64) meson doesn't do this. It's worth noting this in the documentation.
2018-10-03 01:14:21 +08:00
also check the `.endian()` value of the machine for this information.
Added a page for reference tables.
2017-09-17 00:16:00 +08:00
## Operating system names
These are provided by the `.system()` method call.
| Value | Comment |
| ----- | ------- |
Add android to os list informally. Closes #6233. [skip ci]
2020-06-03 03:08:00 +08:00
| android | By convention only, subject to change |
docs: Add 'cygwin' to reference table for .system() This aligns this table with the set of values tested for by test case common/140
2017-11-17 20:58:05 +08:00
| cygwin | The Cygwin environment for Windows |
docs: Alphabetically sort reference table rows [skip ci]
2018-11-22 19:49:45 +08:00
| darwin | Either OSX or iOS |
docs: Add dragonfly BSD to OS reference table
2017-12-19 08:25:40 +08:00
| dragonfly | DragonFly BSD |
Add basic Webassembly support via Emscripten.
2019-06-20 02:56:00 +08:00
| emscripten | Emscripten's Javascript environment |
docs: Alphabetically sort reference table rows [skip ci]
2018-11-22 19:49:45 +08:00
| freebsd | FreeBSD and its derivatives |
Add gnu hurd to *_machine information
2018-10-09 05:24:59 +08:00
| gnu | GNU Hurd |
docs: Alphabetically sort reference table rows [skip ci]
2018-11-22 19:49:45 +08:00
| haiku | |
| linux | |
| netbsd | |
Add missing openbsd to the operating system names table [skip ci]
2019-11-12 04:41:09 +08:00
| openbsd | |
docs: Alphabetically sort reference table rows [skip ci]
2018-11-22 19:49:45 +08:00
| windows | Any version of Windows |
docs: Add sunos to the reference table I originally liked "solaris", but I've changed my mind. Both illumos (the open-source fork of OpenSolaris) and Oracle's closed-source Solaris are identified by the same token, and there are differences between them; so using "sunos" as a sort "supertype" for both makes sense to me.
2019-08-21 04:42:41 +08:00
| sunos | illumos and Solaris |
Added a page for reference tables.
2017-09-17 00:16:00 +08:00
Any string not listed above is not guaranteed to remain stable in
docs: Add haiku to reference table
2017-11-21 08:59:21 +08:00
future releases.
[skip ci] docs: Add reference table for language args parameter names (#3126)
2018-02-25 20:02:34 +08:00
## Language arguments parameter names
These are the parameter names for passing language specific arguments to your build target.
docs: Add link args to the table of language args [skip ci]
2019-01-09 03:35:34 +08:00
| Language | compiler name | linker name |
| ------------- | ------------- | ----------------- |
| C | c_args | c_link_args |
| C++ | cpp_args | cpp_link_args |
| C# | cs_args | cs_link_args |
| D | d_args | d_link_args |
| Fortran | fortran_args | fortran_link_args |
| Java | java_args | java_link_args |
| Objective C | objc_args | objc_link_args |
| Objective C++ | objcpp_args | objcpp_link_args |
| Rust | rust_args | rust_link_args |
| Vala | vala_args | vala_link_args |
Add method to check for C/C++ function attributes It's fairly common on Linux and *BSD platforms to check for these attributes existence, so it makes sense to me to have this checking build into meson itself. Autotools also has a builtin for handling these, and by building them in we can short circuit cases that we know that these don't exist (MSVC). Additionally this adds support for two common MSVC __declspec attributes, dllimport and dllexport. This implements the declspec version (even though GCC has an __attribute__ version that both it and clang support), since GCC and Clang support the MSVC version as well. Thus it seems reasonable to assume that most projects will use the __declspec version over teh __attribute__ version.
2017-09-30 02:52:06 +08:00
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
All these `<lang>_*` options are specified per machine. See in
[specifying options per
machine](Builtin-options.md#Specifying-options-per-machine) for on how
to do this in cross builds.
[skip ci] docs: Bring up to date on options per machine I tried to fix the docs when the stuff was added, but it turns out I missed things, as noted in https://github.com/mesonbuild/meson/issues/7284#issuecomment-641641177
2020-06-10 08:38:46 +08:00
[skip ci] Document CFLAGS and friends in main docs Currently this is only partially documented in the quick-start guide, include this in the main docs and document all the environment argumens used.
2018-11-18 22:35:25 +08:00
## Compiler and linker flag environment variables
These environment variables will be used to modify the compiler and
linker flags.
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
It is recommended that you **do not use these**. They are provided
purely to for backwards compatibility with other build systems. There
are many caveats to their use, especially when rebuilding the project.
It is **highly** recommended that you use [the command line
arguments](#language-arguments-parameter-names) instead.
docs: document that using environment variables is discouraged [skip ci]
2019-01-09 03:31:58 +08:00
Format Compiler and linker flag table [skip ci]
2018-12-13 16:11:53 +08:00
| Name | Comment |
| ----- | ------- |
[skip ci] Document CFLAGS and friends in main docs Currently this is only partially documented in the quick-start guide, include this in the main docs and document all the environment argumens used.
2018-11-18 22:35:25 +08:00
| CFLAGS | Flags for the C compiler |
| CXXFLAGS | Flags for the C++ compiler |
| OBJCFLAGS | Flags for the Objective C compiler |
| FFLAGS | Flags for the Fortran compiler |
| DFLAGS | Flags for the D compiler |
| VALAFLAGS | Flags for the Vala compiler |
| RUSTFLAGS | Flags for the Rust compiler |
| LDFLAGS | The linker flags, used for all languages |
Add method to check for C/C++ function attributes It's fairly common on Linux and *BSD platforms to check for these attributes existence, so it makes sense to me to have this checking build into meson itself. Autotools also has a builtin for handling these, and by building them in we can short circuit cases that we know that these don't exist (MSVC). Additionally this adds support for two common MSVC __declspec attributes, dllimport and dllexport. This implements the declspec version (even though GCC has an __attribute__ version that both it and clang support), since GCC and Clang support the MSVC version as well. Thus it seems reasonable to assume that most projects will use the __declspec version over teh __attribute__ version.
2017-09-30 02:52:06 +08:00
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
N.B. these settings are specified per machine, and so the environment
varibles actually come in pairs. See the [environment variables per
[skip ci] docs: Bring up to date on options per machine I tried to fix the docs when the stuff was added, but it turns out I missed things, as noted in https://github.com/mesonbuild/meson/issues/7284#issuecomment-641641177
2020-06-10 08:38:46 +08:00
machine](#Environment-variables-per-machine) section for details.
Add method to check for C/C++ function attributes It's fairly common on Linux and *BSD platforms to check for these attributes existence, so it makes sense to me to have this checking build into meson itself. Autotools also has a builtin for handling these, and by building them in we can short circuit cases that we know that these don't exist (MSVC). Additionally this adds support for two common MSVC __declspec attributes, dllimport and dllexport. This implements the declspec version (even though GCC has an __attribute__ version that both it and clang support), since GCC and Clang support the MSVC version as well. Thus it seems reasonable to assume that most projects will use the __declspec version over teh __attribute__ version.
2017-09-30 02:52:06 +08:00
## Function Attributes
These are the parameters names that are supported using
`compiler.has_function_attribute()` or
`compiler.get_supported_function_attributes()`
docs: Fix gcc attribute list link [skip ci]
2018-10-28 06:11:44 +08:00
### GCC `__attribute__`
Add method to check for C/C++ function attributes It's fairly common on Linux and *BSD platforms to check for these attributes existence, so it makes sense to me to have this checking build into meson itself. Autotools also has a builtin for handling these, and by building them in we can short circuit cases that we know that these don't exist (MSVC). Additionally this adds support for two common MSVC __declspec attributes, dllimport and dllexport. This implements the declspec version (even though GCC has an __attribute__ version that both it and clang support), since GCC and Clang support the MSVC version as well. Thus it seems reasonable to assume that most projects will use the __declspec version over teh __attribute__ version.
2017-09-30 02:52:06 +08:00
These values are supported using the GCC style `__attribute__` annotations,
which are supported by GCC, Clang, and other compilers.
compiler: add 'force_align_arg_pointer' function attribute
2020-06-22 00:31:59 +08:00
| Name |
|--------------------------|
| alias |
| aligned |
| alloc_size |
| always_inline |
| artificial |
| cold |
| const |
| constructor |
| constructor_priority |
| deprecated |
| destructor |
| error |
| externally_visible |
| fallthrough |
| flatten |
| format |
| format_arg |
| force_align_arg_pointer³ |
| gnu_inline |
| hot |
| ifunc |
| malloc |
| noclone |
| noinline |
| nonnull |
| noreturn |
| nothrow |
| optimize |
| packed |
| pure |
| returns_nonnull |
| unused |
| used |
| visibility* |
| visibility:default† |
| visibility:hidden† |
| visibility:internal† |
| visibility:protected† |
| warning |
| warn_unused_result |
| weak |
| weakreaf |
Add method to check for C/C++ function attributes It's fairly common on Linux and *BSD platforms to check for these attributes existence, so it makes sense to me to have this checking build into meson itself. Autotools also has a builtin for handling these, and by building them in we can short circuit cases that we know that these don't exist (MSVC). Additionally this adds support for two common MSVC __declspec attributes, dllimport and dllexport. This implements the declspec version (even though GCC has an __attribute__ version that both it and clang support), since GCC and Clang support the MSVC version as well. Thus it seems reasonable to assume that most projects will use the __declspec version over teh __attribute__ version.
2017-09-30 02:52:06 +08:00
Split attribute visibility * c_function_attributes: remove 'protected' from 'visibility' This doesn't exist on macos with the apple compiler, which always causes failures. Fixes #5530 * compilers: Add split visibility checks to has_function_attribute These check for a single visibility at a time, rather than all four at once. This allows for finer grained searches, and should make using these checks safer across operating systems.
2019-06-25 05:02:26 +08:00
\* *Changed in 0.52.0* the "visibility" target no longer includes
"protected", which is not present in Apple's clang.
† *New in 0.52.0* These split visibility attributes are preferred to the plain
"visibility" as they provide narrower checks.
compiler: add 'force_align_arg_pointer' function attribute
2020-06-22 00:31:59 +08:00
³ *New in 0.55.0*
Add method to check for C/C++ function attributes It's fairly common on Linux and *BSD platforms to check for these attributes existence, so it makes sense to me to have this checking build into meson itself. Autotools also has a builtin for handling these, and by building them in we can short circuit cases that we know that these don't exist (MSVC). Additionally this adds support for two common MSVC __declspec attributes, dllimport and dllexport. This implements the declspec version (even though GCC has an __attribute__ version that both it and clang support), since GCC and Clang support the MSVC version as well. Thus it seems reasonable to assume that most projects will use the __declspec version over teh __attribute__ version.
2017-09-30 02:52:06 +08:00
### MSVC __declspec
These values are supported using the MSVC style `__declspec` annotation,
which are supported by MSVC, GCC, Clang, and other compilers.
| Name |
|----------------------|
| dllexport |
| dllimport |
Add table for dependency method types. [skip ci]
2019-03-03 00:17:16 +08:00
## Dependency lookup methods
These are the values that can be passed to `dependency` function's
`method` keyword argument.
| Name | Comment |
| ----- | ------- |
| auto | Automatic method selection |
| pkg-config | Use Pkg-Config |
| cmake | Look up as a CMake module |
| config-tool | Use a custom dep tool such as `cups-config` |
| system | System provided (e.g. OpenGL) |
| extraframework | A macOS/iOS framework |
docs: Add Environment variables for compiler/linker selection We really should be documenting these in an easy to find and reference place.
2020-03-13 02:13:17 +08:00
## Compiler and Linker selection variables
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
N.B. these settings are specified per machine, and so the environment
varibles actually come in pairs. See the [environment variables per
[skip ci] docs: Bring up to date on options per machine I tried to fix the docs when the stuff was added, but it turns out I missed things, as noted in https://github.com/mesonbuild/meson/issues/7284#issuecomment-641641177
2020-06-10 08:38:46 +08:00
machine](#Environment-variables-per-machine) section for details.
docs: Add Environment variables for compiler/linker selection We really should be documenting these in an easy to find and reference place.
2020-03-13 02:13:17 +08:00
| Language | Compiler | Linker | Note |
References-tables.md: repair broken table At https://mesonbuild.com/Reference-tables.html#compiler-and-linker-selection-variables, the table is broken and it seems that the only difference with other well-working tables on this page is the semi-column ":" that are present under Language.
2020-04-12 00:41:06 +08:00
|---------------|----------|-----------|---------------------------------------------|
docs: Add Environment variables for compiler/linker selection We really should be documenting these in an easy to find and reference place.
2020-03-13 02:13:17 +08:00
| C | CC | CC_LD | |
| C++ | CXX | CXX_LD | |
| D | DC | DC_LD | Before 0.54 D_LD* |
| Fortran | FC | FC_LD | Before 0.54 F_LD* |
| Objective-C | OBJC | OBJC_LD | |
| Objective-C++ | OBJCXX | OBJCXX_LD | Before 0.54 OBJCPP_LD* |
| Rust | RUSTC | RUSTC_LD | Before 0.54 RUST_LD* |
| Vala | VALAC | | Use CC_LD. Vala transpiles to C |
| C# | CSC | CSC | The linker is the compiler |
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
*The old environment variales are still supported, but are deprecated
Capitalize "Meson" consistently as it is a proper name. [skip ci]
2021-02-01 00:00:24 +08:00
and will be removed in a future version of Meson.*
[skip ci] docs: Bring up to date on options per machine I tried to fix the docs when the stuff was added, but it turns out I missed things, as noted in https://github.com/mesonbuild/meson/issues/7284#issuecomment-641641177
2020-06-10 08:38:46 +08:00
## Environment variables per machine
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
Since *0.54.0*, Following Autotool and other legacy build systems,
environment variables that affect machine-specific settings come in
pairs: for every bare environment variable `FOO`, there is a suffixed
`FOO_FOR_BUILD`, where `FOO` just affects the host machine
configuration, while `FOO_FOR_BUILD` just affects the build machine
configuration. For example:
[skip ci] docs: Bring up to date on options per machine I tried to fix the docs when the stuff was added, but it turns out I missed things, as noted in https://github.com/mesonbuild/meson/issues/7284#issuecomment-641641177
2020-06-10 08:38:46 +08:00
- `PKG_CONFIG_PATH_FOR_BUILD` controls the paths pkg-config will search for
just `native: true` dependencies (build machine).
- `PKG_CONFIG_PATH` controls the paths pkg-config will search for just
`native: false` dependencies (host machine).
Capitalize "Meson" consistently as it is a proper name. [skip ci]
2021-02-01 00:00:24 +08:00
This mirrors the `build.` prefix used for (built-in) Meson options,
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
which has the same meaning.
[skip ci] docs: Bring up to date on options per machine I tried to fix the docs when the stuff was added, but it turns out I missed things, as noted in https://github.com/mesonbuild/meson/issues/7284#issuecomment-641641177
2020-06-10 08:38:46 +08:00
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
This is useful for cross builds. In the native builds, build = host,
and the unsuffixed environment variables alone will suffice.
docs: Add Environment variables for compiler/linker selection We really should be documenting these in an easy to find and reference place.
2020-03-13 02:13:17 +08:00
Rewrap long text lines in docs. [skip ci]
2021-01-30 19:20:59 +08:00
Prior to *0.54.0*, there was no `_FOR_BUILD`-suffixed variables, and
most environment variables only effected native machine
configurations, though this wasn't consistent (e.g. `PKG_CONFIG_PATH`
still affected cross builds).