On Thu, 16 Dec 2021 at 00:28, Khem Raj <raj.k...@gmail.com> wrote:

> > Drop 0001-fiber-libs-Define-SYS_futex-if-it-does-not-exist.patch as
> > it is difficult to rebase and needs to land upstream first.
>
> It's not that rebasing is too hard for this patch but this patch is an
> incorrect way to fix what it's trying to solve and that's why I am
> fine with dropping it.
> landing upstream is not a precondition for submitting patches to
> OpenEmbedded but good to have. All distributions carry patches and as
> long as patches are working
> their way upstream asynchronously its ok.  Let's not make it too hard
> for contributors and chase them away.
>

Yes, rather than 'land upstream' I should've said 'submitted upstream' -
that patch is dated 16 October 2020, and it never happened. While myself
and everyone else enormously appreciate your toolchain and target work,
there's a problem with amassing invasive pending patches related to that:
no one except you truly understands them. So what I'm asking is just a bit
more rigor going forward: submit patches upstream at the same time you
submit them to core, and steadily work your way towards reducing the pile
that's already there.

Alex



>
> >
> > Signed-off-by: Alexander Kanavin <a...@linutronix.de>
> > ---
> >  .../{boost-1.77.0.inc => boost-1.78.0.inc}    |   2 +-
> >  ....cmake-allow-searching-for-python310.patch |  50 ------
> >  ...h-instruction-set-flags-we-do-that-o.patch |  15 +-
> >  ...efine-SYS_futex-if-it-does-not-exist.patch |  54 -------
> >  ...th_no_atomic_int-on-the-command-line.patch |  53 ------
> >  ...oft-failure-in-bernoulli_details_hpp.patch | 151 ------------------
> >  ...7e01635306085488290ea83de541ec393f8b.patch |  30 ++++
> >  meta/recipes-support/boost/boost_1.77.0.bb    |  12 --
> >  meta/recipes-support/boost/boost_1.78.0.bb    |   9 ++
> >  9 files changed, 50 insertions(+), 326 deletions(-)
> >  rename meta/recipes-support/boost/{boost-1.77.0.inc =>
> boost-1.78.0.inc} (90%)
> >  delete mode 100644
> meta/recipes-support/boost/boost/0001-BoostConfig.cmake-allow-searching-for-python310.patch
> >  delete mode 100644
> meta/recipes-support/boost/boost/0001-fiber-libs-Define-SYS_futex-if-it-does-not-exist.patch
> >  delete mode 100644
> meta/recipes-support/boost/boost/0002-math-allow-definition-of-boost_math_no_atomic_int-on-the-command-line.patch
> >  delete mode 100644
> meta/recipes-support/boost/boost/0003-math-make-no-atomics-a-soft-failure-in-bernoulli_details_hpp.patch
> >  create mode 100644
> meta/recipes-support/boost/boost/de657e01635306085488290ea83de541ec393f8b.patch
> >  delete mode 100644 meta/recipes-support/boost/boost_1.77.0.bb
> >  create mode 100644 meta/recipes-support/boost/boost_1.78.0.bb
> >
> > diff --git a/meta/recipes-support/boost/boost-1.77.0.inc
> b/meta/recipes-support/boost/boost-1.78.0.inc
> > similarity index 90%
> > rename from meta/recipes-support/boost/boost-1.77.0.inc
> > rename to meta/recipes-support/boost/boost-1.78.0.inc
> > index 6df06e76c7..729a47b54f 100644
> > --- a/meta/recipes-support/boost/boost-1.77.0.inc
> > +++ b/meta/recipes-support/boost/boost-1.78.0.inc
> > @@ -12,7 +12,7 @@ BOOST_MAJ =
> "${@"_".join(d.getVar("PV").split(".")[0:2])}"
> >  BOOST_P = "boost_${BOOST_VER}"
> >
> >  SRC_URI = "
> https://boostorg.jfrog.io/artifactory/main/release/${PV}/source/${BOOST_P}.tar.bz2
> "
> > -SRC_URI[sha256sum] =
> "fc9f85fc030e233142908241af7a846e60630aa7388de9a5fafb1f3a26840854"
> > +SRC_URI[sha256sum] =
> "8681f175d4bdb26c52222665793eef08490d7758529330f98d3b29dd0735bccc"
> >
> >  UPSTREAM_CHECK_URI = "http://www.boost.org/users/download/";
> >  UPSTREAM_CHECK_REGEX = "release/(?P<pver>.*)/source/"
> > diff --git
> a/meta/recipes-support/boost/boost/0001-BoostConfig.cmake-allow-searching-for-python310.patch
> b/meta/recipes-support/boost/boost/0001-BoostConfig.cmake-allow-searching-for-python310.patch
> > deleted file mode 100644
> > index 0a9ee2cc95..0000000000
> > ---
> a/meta/recipes-support/boost/boost/0001-BoostConfig.cmake-allow-searching-for-python310.patch
> > +++ /dev/null
> > @@ -1,50 +0,0 @@
> > -From e193f080c7d209516ac9b712fa0c50bb08026fa2 Mon Sep 17 00:00:00 2001
> > -From: Martin Jansa <martin.ja...@gmail.com>
> > -Date: Tue, 19 Oct 2021 12:24:31 +0000
> > -Subject: [PATCH] BoostConfig.cmake: allow searching for python310
> > -
> > -* accept double digits in Python3_VERSION_MINOR
> > -
> > -* if someone is using e.g.:
> > -  find_package(Python3 REQUIRED)
> > -  find_package(Boost REQUIRED
> python${Python3_VERSION_MAJOR}${Python3_VERSION_MINOR})
> > -
> > -  with python-3.10 then it currently fails with:
> > -
> > -  -- Found PythonLibs: /usr/lib/libpython3.10.so (found version
> "3.10.0")
> > -  -- Found Python3: -native/usr/bin/python3-native/python3 (found
> version "3.10.0") found components: Interpreter
> > -  CMake Error at /usr/lib/cmake/Boost-1.77.0/BoostConfig.cmake:141
> (find_package):
> > -    Could not find a package configuration file provided by
> "boost_python310"
> > -    (requested version 1.77.0) with any of the following names:
> > -
> > -      boost_python310Config.cmake
> > -      boost_python310-config.cmake
> > -
> > -    Add the installation prefix of "boost_python310" to
> CMAKE_PREFIX_PATH or
> > -    set "boost_python310_DIR" to a directory containing one of the
> above files.
> > -    If "boost_python310" provides a separate development package or
> SDK, be
> > -    sure it has been installed.
> > -  Call Stack (most recent call first):
> > -    /usr/lib/cmake/Boost-1.77.0/BoostConfig.cmake:258
> (boost_find_component)
> > -    /usr/share/cmake-3.21/Modules/FindBoost.cmake:594 (find_package)
> > -    CMakeLists.txt:18 (find_package)
> > -
> > -Upstream-Status: Submitted [
> https://github.com/boostorg/boost_install/pull/53]
> > -Signed-off-by: Martin Jansa <martin.ja...@gmail.com>
> > ----
> > - tools/boost_install/BoostConfig.cmake | 2 +-
> > - 1 file changed, 1 insertion(+), 1 deletion(-)
> > -
> > -diff --git a/tools/boost_install/BoostConfig.cmake
> b/tools/boost_install/BoostConfig.cmake
> > -index fd17821..5dffa58 100644
> > ---- a/tools/boost_install/BoostConfig.cmake
> > -+++ b/tools/boost_install/BoostConfig.cmake
> > -@@ -113,7 +113,7 @@ macro(boost_find_component comp required quiet)
> > -     set(_BOOST_REQUIRED REQUIRED)
> > -   endif()
> > -
> > --  if("${comp}" MATCHES "^(python|numpy|mpi_python)([1-9])([0-9])$")
> > -+  if("${comp}" MATCHES
> "^(python|numpy|mpi_python)([1-9])([0-9][0-9]?)$")
> > -
> > -     # handle pythonXY and numpyXY versioned components for
> compatibility
> > -
> > diff --git
> a/meta/recipes-support/boost/boost/0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch
> b/meta/recipes-support/boost/boost/0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch
> > index 91ab53efd9..67d5dff125 100644
> > ---
> a/meta/recipes-support/boost/boost/0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch
> > +++
> b/meta/recipes-support/boost/boost/0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch
> > @@ -1,4 +1,4 @@
> > -From 8e1d30454afde37eaa3c593ec19d108cd5ed10d0 Mon Sep 17 00:00:00 2001
> > +From 4d2a8fc8117e56bc283349e5f7f889ebbfc55c71 Mon Sep 17 00:00:00 2001
> >  From: Alexander Kanavin <alex.kana...@gmail.com>
> >  Date: Tue, 18 Dec 2018 15:42:57 +0100
> >  Subject: [PATCH] Don't set up arch/instruction-set flags, we do that
> > @@ -10,14 +10,14 @@ Signed-off-by: Christopher Larson <
> chris_lar...@mentor.com>
> >  Signed-off-by: Alexander Kanavin <alex.kana...@gmail.com>
> >
> >  ---
> > - tools/build/src/tools/gcc.jam | 139 ----------------------------------
> > - 1 file changed, 139 deletions(-)
> > + tools/build/src/tools/gcc.jam | 144 ----------------------------------
> > + 1 file changed, 144 deletions(-)
> >
> >  diff --git a/tools/build/src/tools/gcc.jam
> b/tools/build/src/tools/gcc.jam
> > -index ff3209f7b..70cbc39a5 100644
> > +index 47a113223..d77525724 100644
> >  --- a/tools/build/src/tools/gcc.jam
> >  +++ b/tools/build/src/tools/gcc.jam
> > -@@ -1217,142 +1217,3 @@ local rule cpu-flags ( toolset variable :
> architecture : instruction-set + :
> > +@@ -1122,147 +1122,3 @@ local rule cpu-flags ( toolset variable :
> architecture : instruction-set + :
> >
>  <architecture>$(architecture)/<instruction-set>$(instruction-set)
> >           : $(values) ;
> >   }
> > @@ -158,5 +158,10 @@ index ff3209f7b..70cbc39a5 100644
> >  -cpu-flags gcc OPTIONS : s390x : z13 : -march=z13 ;
> >  -cpu-flags gcc OPTIONS : s390x : z14 : -march=z14 ;
> >  -cpu-flags gcc OPTIONS : s390x : z15 : -march=z15 ;
> > +-# ARM
> > +-cpu-flags gcc OPTIONS : arm : cortex-a9+vfpv3 : -mcpu=cortex-a9
> -mfpu=vfpv3 -mfloat-abi=hard ;
> > +-cpu-flags gcc OPTIONS : arm : cortex-a53 : -mcpu=cortex-a53 ;
> > +-cpu-flags gcc OPTIONS : arm : cortex-r5 : -mcpu=cortex-r5 ;
> > +-cpu-flags gcc OPTIONS : arm : cortex-r5+vfpv3-d16 : -mcpu=cortex-r5
> -mfpu=vfpv3-d16 -mfloat-abi=hard ;
> >  -# AIX variant of RS/6000 & PowerPC
> >  -toolset.flags gcc AROPTIONS <address-model>64/<target-os>aix : "-X64" ;
> > diff --git
> a/meta/recipes-support/boost/boost/0001-fiber-libs-Define-SYS_futex-if-it-does-not-exist.patch
> b/meta/recipes-support/boost/boost/0001-fiber-libs-Define-SYS_futex-if-it-does-not-exist.patch
> > deleted file mode 100644
> > index 523568e9bc..0000000000
> > ---
> a/meta/recipes-support/boost/boost/0001-fiber-libs-Define-SYS_futex-if-it-does-not-exist.patch
> > +++ /dev/null
> > @@ -1,54 +0,0 @@
> > -From d6f7b6064dc91d1d5fa18554b40b14822ab7a32b Mon Sep 17 00:00:00 2001
> > -From: Khem Raj <raj.k...@gmail.com>
> > -Date: Fri, 16 Oct 2020 11:13:22 -0700
> > -Subject: [PATCH] fiber,libs: Define SYS_futex if it does not exist
> > -
> > -__NR_futex is not defines by newer architectures e.g. arc, riscv32 as
> > -they only have 64bit variant of time_t. Glibc defines SYS_futex
> interface based on
> > -__NR_futex, since this is used in applications, such applications start
> > -to fail to build for these newer architectures. This patch defines a
> > -fallback to alias __NR_futex to __NR_futex_tim64 so SYS_futex keeps
> > -working
> > -
> > -Upstream-Status: Pending
> > -
> > -Signed-off-by: Khem Raj <raj.k...@gmail.com>
> > ----
> > - boost/fiber/detail/futex.hpp | 5 +++++
> > - libs/log/src/event.cpp       | 4 ++++
> > - 2 files changed, 9 insertions(+)
> > -
> > -diff --git a/boost/fiber/detail/futex.hpp b/boost/fiber/detail/futex.hpp
> > -index e64bd5990..16bee64f1 100644
> > ---- a/boost/fiber/detail/futex.hpp
> > -+++ b/boost/fiber/detail/futex.hpp
> > -@@ -17,6 +17,11 @@ extern "C" {
> > - #include <linux/futex.h>
> > - #include <sys/syscall.h>
> > - }
> > -+
> > -+#if !defined(SYS_futex) && defined(SYS_futex_time64)
> > -+#define SYS_futex SYS_futex_time64
> > -+#endif
> > -+
> > - #elif BOOST_OS_WINDOWS
> > - #include <windows.h>
> > - #endif
> > -diff --git a/libs/log/src/event.cpp b/libs/log/src/event.cpp
> > -index 5485154d7..2c7c0381f 100644
> > ---- a/libs/log/src/event.cpp
> > -+++ b/libs/log/src/event.cpp
> > -@@ -31,6 +31,10 @@
> > - #include <linux/futex.h>
> > - #include <boost/memory_order.hpp>
> > -
> > -+#if !defined(SYS_futex) && defined(SYS_futex_time64)
> > -+#define SYS_futex SYS_futex_time64
> > -+#endif
> > -+
> > - // Some Android NDKs (Google NDK and older Crystax.NET NDK versions)
> don't define SYS_futex
> > - #if defined(SYS_futex)
> > - #define BOOST_LOG_SYS_FUTEX SYS_futex
> > ---
> > -2.28.0
> > -
> > diff --git
> a/meta/recipes-support/boost/boost/0002-math-allow-definition-of-boost_math_no_atomic_int-on-the-command-line.patch
> b/meta/recipes-support/boost/boost/0002-math-allow-definition-of-boost_math_no_atomic_int-on-the-command-line.patch
> > deleted file mode 100644
> > index b05b795084..0000000000
> > ---
> a/meta/recipes-support/boost/boost/0002-math-allow-definition-of-boost_math_no_atomic_int-on-the-command-line.patch
> > +++ /dev/null
> > @@ -1,53 +0,0 @@
> > -From 32bd6197353f6ea8e5bef01f09e25c944141acfc Mon Sep 17 00:00:00 2001
> > -From: jzmaddock <j...@johnmaddock.co.uk>
> > -Date: Wed, 1 Sep 2021 18:54:54 +0100
> > -Subject: [PATCH] Allow definition of BOOST_MATH_NO_ATOMIC_INT on the
> command
> > - line. Allows us to test/emulate platforms with no atomic integers.
> > -
> > -[buildr...@heine.tech:
> > -  - backport from boostorg/math 32bd6197353f6ea8e5bef01f09e25c944141acfc
> > -  - alter path to match boost release
> > -]
> > -Signed-off-by: Michael Nosthoff <buildr...@heine.tech>
> > ----
> > -Upstream-Status: Backport [
> https://github.com/boostorg/math/pull/684/commits/32bd6197353f6ea8e5bef01f09e25c944141acfc
> ]
> > - boost/math/tools/atomic.hpp | 10 +++++-----
> > - 1 file changed, 5 insertions(+), 5 deletions(-)
> > -
> > -diff --git a/boost/math/tools/atomic.hpp b/boost/math/tools/atomic.hpp
> > -index cc76ed269f..e3cbf5db89 100644
> > ---- a/boost/math/tools/atomic.hpp
> > -+++ b/boost/math/tools/atomic.hpp
> > -@@ -16,27 +16,27 @@
> > - namespace boost {
> > -    namespace math {
> > -       namespace detail {
> > --#if ATOMIC_INT_LOCK_FREE == 2
> > -+#if (ATOMIC_INT_LOCK_FREE == 2) && !defined(BOOST_MATH_NO_ATOMIC_INT)
> > -          typedef std::atomic<int> atomic_counter_type;
> > -          typedef std::atomic<unsigned> atomic_unsigned_type;
> > -          typedef int atomic_integer_type;
> > -          typedef unsigned atomic_unsigned_integer_type;
> > --#elif ATOMIC_SHORT_LOCK_FREE == 2
> > -+#elif (ATOMIC_SHORT_LOCK_FREE == 2) &&
> !defined(BOOST_MATH_NO_ATOMIC_INT)
> > -          typedef std::atomic<short> atomic_counter_type;
> > -          typedef std::atomic<unsigned short> atomic_unsigned_type;
> > -          typedef short atomic_integer_type;
> > -          typedef unsigned short atomic_unsigned_type;
> > --#elif ATOMIC_LONG_LOCK_FREE == 2
> > -+#elif (ATOMIC_LONG_LOCK_FREE == 2) &&
> !defined(BOOST_MATH_NO_ATOMIC_INT)
> > -          typedef std::atomic<long> atomic_unsigned_integer_type;
> > -          typedef std::atomic<unsigned long> atomic_unsigned_type;
> > -          typedef unsigned long atomic_unsigned_type;
> > -          typedef long atomic_integer_type;
> > --#elif ATOMIC_LLONG_LOCK_FREE == 2
> > -+#elif (ATOMIC_LLONG_LOCK_FREE == 2) &&
> !defined(BOOST_MATH_NO_ATOMIC_INT)
> > -          typedef std::atomic<long long> atomic_unsigned_integer_type;
> > -          typedef std::atomic<unsigned long long> atomic_unsigned_type;
> > -          typedef long long atomic_integer_type;
> > -          typedef unsigned long long atomic_unsigned_integer_type;
> > --#else
> > -+#elif !defined(BOOST_MATH_NO_ATOMIC_INT)
> > - #  define BOOST_MATH_NO_ATOMIC_INT
> > - #endif
> > -       } // Namespace detail
> > diff --git
> a/meta/recipes-support/boost/boost/0003-math-make-no-atomics-a-soft-failure-in-bernoulli_details_hpp.patch
> b/meta/recipes-support/boost/boost/0003-math-make-no-atomics-a-soft-failure-in-bernoulli_details_hpp.patch
> > deleted file mode 100644
> > index f69e4f21f3..0000000000
> > ---
> a/meta/recipes-support/boost/boost/0003-math-make-no-atomics-a-soft-failure-in-bernoulli_details_hpp.patch
> > +++ /dev/null
> > @@ -1,151 +0,0 @@
> > -From 7d482f6ebc356e6ec455ccb5f51a23971bf6ce5b Mon Sep 17 00:00:00 2001
> > -From: jzmaddock <j...@johnmaddock.co.uk>
> > -Date: Wed, 1 Sep 2021 20:31:53 +0100
> > -Subject: [PATCH] Make no atomics a soft failure in
> bernoulli_details.hpp.
> > - Include an "escape macro" so thread safety can be disabled if certain
> > - bernoulli features are to be used in a no-atomics environment. Fixes
> > - https://github.com/boostorg/math/issues/673.
> > -
> > -[buildr...@heine.tech:
> > -  - backport from boostorg/math 7d482f6ebc356e6ec455ccb5f51a23971bf6ce5b
> > -  - alter path to match boost release
> > -]
> > -Signed-off-by: Michael Nosthoff <buildr...@heine.tech>
> > ----
> > -Upstream-Status: Backport [
> https://github.com/boostorg/math/pull/684/commits/7d482f6ebc356e6ec455ccb5f51a23971bf6ce5b
> ]
> > - .../detail/bernoulli_details.hpp                 | 10 +++++++---
> > - libs/math/test/Jamfile.v2                        |  3 +++
> > - test/compile_test/bernoulli_no_atomic_d.cpp      | 14 ++++++++++++++
> > - test/compile_test/bernoulli_no_atomic_fail.cpp   | 15 +++++++++++++++
> > - test/compile_test/bernoulli_no_atomic_mp.cpp     | 16 ++++++++++++++++
> > - 5 files changed, 55 insertions(+), 3 deletions(-)
> > - create mode 100644 test/compile_test/bernoulli_no_atomic_d.cpp
> > - create mode 100644 test/compile_test/bernoulli_no_atomic_fail.cpp
> > - create mode 100644 test/compile_test/bernoulli_no_atomic_mp.cpp
> > -
> > -diff --git a/boost/math/special_functions/detail/bernoulli_details.hpp
> b/boost/math/special_functions/detail/bernoulli_details.hpp
> > -index cf35545264..8519b7c89c 100644
> > ---- a/boost/math/special_functions/detail/bernoulli_details.hpp
> > -+++ b/boost/math/special_functions/detail/bernoulli_details.hpp
> > -@@ -360,7 +360,7 @@ class bernoulli_numbers_cache
> > -          return out;
> > -       }
> > -
> > --      #ifndef BOOST_HAS_THREADS
> > -+      #if !defined(BOOST_HAS_THREADS) ||
> defined(BOOST_MATH_BERNOULLI_UNTHREADED)
> > -       //
> > -       // Single threaded code, very simple:
> > -       //
> > -@@ -382,6 +382,8 @@ class bernoulli_numbers_cache
> > -          *out = (i >= m_overflow_limit) ?
> policies::raise_overflow_error<T>("boost::math::bernoulli_b2n<%1%>(std::size_t)",
> 0, T(i), pol) : bn[i];
> > -          ++out;
> > -       }
> > -+      #elif defined(BOOST_MATH_NO_ATOMIC_INT)
> > -+      static_assert(sizeof(T) == 1, "Unsupported configuration: your
> platform appears to have no atomic integers.  If you are happy with
> thread-unsafe code, then you may define BOOST_MATH_BERNOULLI_UNTHREADED to
> suppress this error.");
> > -       #else
> > -       //
> > -       // Double-checked locking pattern, lets us access cached already
> cached values
> > -@@ -464,7 +466,7 @@ class bernoulli_numbers_cache
> > -          return out;
> > -       }
> > -
> > --      #ifndef BOOST_HAS_THREADS
> > -+      #if !defined(BOOST_HAS_THREADS) ||
> defined(BOOST_MATH_BERNOULLI_UNTHREADED)
> > -       //
> > -       // Single threaded code, very simple:
> > -       //
> > -@@ -494,6 +496,8 @@ class bernoulli_numbers_cache
> > -          }
> > -          ++out;
> > -       }
> > -+      #elif defined(BOOST_MATH_NO_ATOMIC_INT)
> > -+      static_assert(sizeof(T) == 1, "Unsupported configuration: your
> platform appears to have no atomic integers.  If you are happy with
> thread-unsafe code, then you may define BOOST_MATH_BERNOULLI_UNTHREADED to
> suppress this error.");
> > -       #else
> > -       //
> > -       // Double-checked locking pattern, lets us access cached already
> cached values
> > -@@ -555,7 +559,7 @@ class bernoulli_numbers_cache
> > -    // The value at which we know overflow has already occurred for the
> Bn:
> > -    std::size_t m_overflow_limit;
> > -
> > --   #ifdef BOOST_HAS_THREADS
> > -+   #if defined(BOOST_HAS_THREADS) && !defined(BOOST_MATH_NO_ATOMIC_INT)
> > -    std::mutex m_mutex;
> > -    atomic_counter_type m_counter, m_current_precision;
> > -    #else
> > -diff --git a/libs/math/test/Jamfile.v2 b/libs/math/test/Jamfile.v2
> > -index 52fb87f5e5..3ac63f9279 100644
> > ---- a/libs/math/test/Jamfile.v2
> > -+++ b/libs/math/test/Jamfile.v2
> > -@@ -1137,6 +1137,9 @@ test-suite misc :
> > -
> > - #   [ run __temporary_test.cpp test_instances//test_instances : : :
> <test-info>always_show_run_output <pch>off ]
> > -    [ compile test_no_long_double_policy.cpp ]
> > -+   [ compile compile_test/bernoulli_no_atomic_d.cpp ]
> > -+   [ compile compile_test/bernoulli_no_atomic_mp.cpp ]
> > -+   [ compile-fail compile_test/bernoulli_no_atomic_fail.cpp ]
> > - ;
> > -
> > - test-suite interpolators :
> > -diff --git a/test/compile_test/bernoulli_no_atomic_d.cpp
> b/test/compile_test/bernoulli_no_atomic_d.cpp
> > -new file mode 100644
> > -index 0000000000..61926f7e1f
> > ---- /dev/null
> > -+++ b/test/compile_test/bernoulli_no_atomic_d.cpp
> > -@@ -0,0 +1,14 @@
> > -+//  (C) Copyright John Maddock 2021.
> > -+//  Use, modification and distribution are subject to the
> > -+//  Boost Software License, Version 1.0. (See accompanying file
> > -+//  LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
> > -+
> > -+#define BOOST_MATH_NO_ATOMIC_INT
> > -+
> > -+#include <boost/math/special_functions/bernoulli.hpp>
> > -+#include "test_compile_result.hpp"
> > -+
> > -+void compile_and_link_test()
> > -+{
> > -+   check_result<double>(boost::math::bernoulli_b2n<double>(4));
> > -+}
> > -diff --git a/test/compile_test/bernoulli_no_atomic_fail.cpp
> b/test/compile_test/bernoulli_no_atomic_fail.cpp
> > -new file mode 100644
> > -index 0000000000..bbd7152412
> > ---- /dev/null
> > -+++ b/test/compile_test/bernoulli_no_atomic_fail.cpp
> > -@@ -0,0 +1,15 @@
> > -+//  (C) Copyright John Maddock 2021.
> > -+//  Use, modification and distribution are subject to the
> > -+//  Boost Software License, Version 1.0. (See accompanying file
> > -+//  LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
> > -+
> > -+#define BOOST_MATH_NO_ATOMIC_INT
> > -+
> > -+#include <boost/math/special_functions/bernoulli.hpp>
> > -+#include <boost/multiprecision/cpp_bin_float.hpp>
> > -+#include "test_compile_result.hpp"
> > -+
> > -+void compile_and_link_test()
> > -+{
> > -+
>  
> check_result<boost::multiprecision::cpp_bin_float_50>(boost::math::bernoulli_b2n<boost::multiprecision::cpp_bin_float_50>(4));
> > -+}
> > -diff --git a/test/compile_test/bernoulli_no_atomic_mp.cpp
> b/test/compile_test/bernoulli_no_atomic_mp.cpp
> > -new file mode 100644
> > -index 0000000000..8d5a6e78e6
> > ---- /dev/null
> > -+++ b/test/compile_test/bernoulli_no_atomic_mp.cpp
> > -@@ -0,0 +1,16 @@
> > -+//  (C) Copyright John Maddock 2021.
> > -+//  Use, modification and distribution are subject to the
> > -+//  Boost Software License, Version 1.0. (See accompanying file
> > -+//  LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
> > -+
> > -+#define BOOST_MATH_NO_ATOMIC_INT
> > -+#define BOOST_MATH_BERNOULLI_UNTHREADED
> > -+
> > -+#include <boost/math/special_functions/bernoulli.hpp>
> > -+#include <boost/multiprecision/cpp_bin_float.hpp>
> > -+#include "test_compile_result.hpp"
> > -+
> > -+void compile_and_link_test()
> > -+{
> > -+
>  
> check_result<boost::multiprecision::cpp_bin_float_50>(boost::math::bernoulli_b2n<boost::multiprecision::cpp_bin_float_50>(4));
> > -+}
> > diff --git
> a/meta/recipes-support/boost/boost/de657e01635306085488290ea83de541ec393f8b.patch
> b/meta/recipes-support/boost/boost/de657e01635306085488290ea83de541ec393f8b.patch
> > new file mode 100644
> > index 0000000000..867ec03ba6
> > --- /dev/null
> > +++
> b/meta/recipes-support/boost/boost/de657e01635306085488290ea83de541ec393f8b.patch
> > @@ -0,0 +1,30 @@
> > +From de657e01635306085488290ea83de541ec393f8b Mon Sep 17 00:00:00 2001
> > +From: Leonardo Neumann <leona...@neumann.dev.br>
> > +Date: Mon, 13 Dec 2021 01:07:20 -0300
> > +Subject: [PATCH] Fix missing sys/stat.h include on musl-based systems
> > +
> > +Boost 1.78.0 fails to build on musl-based systems because musl does
> > +not include sys/stat.h by default.
> > +
> > +Fixes #161 ("Boost compiler error")
> > +Upstream-Status: Submitted [
> https://github.com/boostorg/interprocess/pull/162]
> > +Signed-off-by: Alexander Kanavin <alex.kana...@gmail.com>
> > +---
> > + include/boost/interprocess/permissions.hpp | 4 ++++
> > + 1 file changed, 4 insertions(+)
> > +
> > +diff --git a/boost/interprocess/permissions.hpp
> b/boost/interprocess/permissions.hpp
> > +index ab55411e..0b21a685 100644
> > +--- a/boost/interprocess/permissions.hpp
> > ++++ b/boost/interprocess/permissions.hpp
> > +@@ -29,6 +29,10 @@
> > +
> > + #include <boost/interprocess/detail/win32_api.hpp>
> > +
> > ++#else
> > ++
> > ++#include <sys/stat.h>
> > ++
> > + #endif
> > +
> > + #endif   //#ifndef BOOST_INTERPROCESS_DOXYGEN_INVOKED
> > diff --git a/meta/recipes-support/boost/boost_1.77.0.bb
> b/meta/recipes-support/boost/boost_1.77.0.bb
> > deleted file mode 100644
> > index bde6b14a79..0000000000
> > --- a/meta/recipes-support/boost/boost_1.77.0.bb
> > +++ /dev/null
> > @@ -1,12 +0,0 @@
> > -require boost-${PV}.inc
> > -require boost.inc
> > -
> > -SRC_URI += "file://boost-CVE-2012-2677.patch \
> > -           file://boost-math-disable-pch-for-gcc.patch \
> > -
>  file://0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch \
> > -           file://0001-dont-setup-compiler-flags-m32-m64.patch \
> > -
>  file://0001-fiber-libs-Define-SYS_futex-if-it-does-not-exist.patch \
> > -
>  file://0001-BoostConfig.cmake-allow-searching-for-python310.patch \
> > -
>  
> file://0002-math-allow-definition-of-boost_math_no_atomic_int-on-the-command-line.patch
> \
> > -
>  
> file://0003-math-make-no-atomics-a-soft-failure-in-bernoulli_details_hpp.patch
> \
> > -           "
> > diff --git a/meta/recipes-support/boost/boost_1.78.0.bb
> b/meta/recipes-support/boost/boost_1.78.0.bb
> > new file mode 100644
> > index 0000000000..f4b6ed8576
> > --- /dev/null
> > +++ b/meta/recipes-support/boost/boost_1.78.0.bb
> > @@ -0,0 +1,9 @@
> > +require boost-${PV}.inc
> > +require boost.inc
> > +
> > +SRC_URI += "file://boost-CVE-2012-2677.patch \
> > +           file://boost-math-disable-pch-for-gcc.patch \
> > +
>  file://0001-Don-t-set-up-arch-instruction-set-flags-we-do-that-o.patch \
> > +           file://0001-dont-setup-compiler-flags-m32-m64.patch \
> > +           file://de657e01635306085488290ea83de541ec393f8b.patch \
> > +           "
> > --
> > 2.20.1
> >
>
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#159773): 
https://lists.openembedded.org/g/openembedded-core/message/159773
Mute This Topic: https://lists.openembedded.org/mt/87755576/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to