nuttx/libs/libxx
Gustavo Henrique Nihei 0bd436c8db libcxx: Fix libc++ for archs without conditional load/store support
This commit brings a patch already merged to upstream LLVM project:
https://reviews.llvm.org/D118391

Signed-off-by: Gustavo Henrique Nihei <gustavo.nihei@espressif.com>
2022-03-30 11:19:29 +08:00
..
libcxxmini libcxx-mini: operator new will assert when alloc failed. 2022-01-11 10:49:44 +01:00
.gitignore libc:libxx: add new low level c++ standard library support 2021-08-26 19:32:24 +08:00
0001-libc-abi-avoid-the-waring-__EXCEPTIONS-is-not-define.patch libcxxabi: Avoid the waring: "__EXCEPTIONS" is not defined, evaluates to 0 [-Werror=undef] 2021-08-28 17:13:43 -03:00
0001-libc-avoid-the-waring-__EXCEPTIONS-is-not-defined-ev.patch libcxx: Avoid the waring: "__EXCEPTIONS" is not defined, evaluates to 0 [-Werror=undef] 2021-08-28 17:13:43 -03:00
0001-libcxx-Rename-PS-macro-to-avoid-clashing-with-Xtensa.patch libcxx: Add patch for preventing redefinition of PS macro on Xtensa 2022-03-30 11:19:29 +08:00
0001-Remove-the-locale-fallback-for-NuttX.patch Remove the locale fallback from libcxx 2021-07-30 09:43:26 -03:00
0001-uClibcxx-basic_definitions-fix-GCC-specific-definiti.patch libxx:uClibcxx:fix-GCC-specific-definitiion error 2021-08-31 11:23:29 +08:00
0001-uclibxx-use-overload-constructor-of-filebuf-ostream.patch libxx:uclibxx: add patch use overload constructor of filebuf & ostream 2022-01-24 23:26:59 +08:00
0002-Omit-atomic_-un-signed_lock_free-if-unsupported.patch libcxx: Fix libc++ for archs without conditional load/store support 2022-03-30 11:19:29 +08:00
Kconfig libxx: Add CXX_RTTI for enabling RTTI support for C++ applications 2022-03-26 05:56:37 +02:00
libcxx.defs libcxx: Fix libc++ for archs without conditional load/store support 2022-03-30 11:19:29 +08:00
libcxxabi.defs libs/libxx: Enforce RTTI for building libcxxabi 2021-09-01 21:02:17 +08:00
libcxxmini.defs libxx: Move the partial c++ abi implementation to libcxxmini sub folder 2021-08-31 06:52:10 -03:00
libxx_cxa_atexit.cxx libxx: Move the partial c++ abi implementation to libcxxmini sub folder 2021-08-31 06:52:10 -03:00
libxx_eabi_atexit.cxx libcxx: : Author Gregory Nutt: update licenses to Apache 2021-02-25 01:44:30 -08:00
libxx_impure.cxx Fix libsupc++.a(vterminate.o): undefined reference to `_impure_ptr' 2020-08-17 13:45:56 -03:00
libxx.hxx include: fix double include pre-processor guards 2022-01-16 11:11:14 -03:00
Makefile libxx: Move the partial c++ abi implementation to libcxxmini sub folder 2021-08-31 06:52:10 -03:00
README.txt libxx: Stop mentioning libs not integrated with this version of libxx 2020-10-26 22:07:39 -07:00
system_configuration.h libxx: Integrate uClibc++ latest official release 2020-07-17 21:40:34 +01:00
uClibc++.defs libxx:uclibxx: add patch use overload constructor of filebuf & ostream 2022-01-24 23:26:59 +08:00

libs/libxx/README.txt
^^^^^^^^^^^^^^^^^^^^^

This directory contains three C++ library:

 - A fragmentary C++ library that will allow to build only the simplest of
   C++ applications. In the deeply embedded world, that is probably all
   that is necessary.

   At present, only the following are supported here:

     - void *operator new(std::size_t nbytes);
     - void operator delete(void* ptr);
     - void operator delete[](void *ptr);
     - void __cxa_pure_virtual(void);
     - int __aeabi_atexit(void* object, void (*destroyer)(void*),
                          void *dso_handle);
     - int __cxa_atexit(__cxa_exitfunc_t func, FAR void *arg,
                        FAR void *dso_handle);

   This implementation is selected when neither of the following
   two options are enabled.

 - LLVM "libc++" C++ library (http://libcxx.llvm.org/)
   This implementation is selected with CONFIG_LIBCXX=y.

 - uClibc++ C++ library (http://cxx.uclibc.org/)
   This implementation is selected with CONFIG_UCLIBCXX=y.

operator new
------------

  This operator should take a type of size_t.  But size_t has an unknown underlying
  type.  In the nuttx sys/types.h header file, size_t is typed as uint32_t
  (which is determined by architecture-specific logic).  But the C++
  compiler may believe that size_t is of a different type resulting in
  compilation errors in the operator.  Using the underlying integer type
  instead of size_t seems to resolve the compilation issues. Need to
  REVISIT this.

  Once some C++ compilers, this will cause an error:

    Problem:     "'operator new' takes size_t ('...') as first parameter"
    Workaround:  Add -fpermissive to the compilation flags