On 29/07/2014 09:40 μμ, Marshall Clow wrote:

On Jul 29, 2014, at 10:06 AM, Marshall Clow wrote:

...

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/ Ok, ok - fine. 1.56.0

...

As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy.

This helps ensure the candidates build OK before we push them out to SourceForge.

The files (and associated md5s) are: MD5 (boost_1_56_0_rc1.7z) = 0ac22b92b24209286d48e4c41925cd9f MD5 (boost_1_56_0_rc1.tar.bz2) = 5b5bfa99244b1ac5737323decceb19c2 MD5 (boost_1_56_0_rc1.tar.gz) = 2047cdd4e09cda34e7f10fec0f821175 MD5 (boost_1_56_0_rc1.zip) = 0c544f6acc85550b15cdc9221f2ac0c9 I have successfully built using gcc 4.9.0 and clang 3.4 on Mac OS X 9.4.

Same here on ubuntu 13.10 with gcc 4.8 and clang 3.2. A lot of unused local typedef warnings on gcc 4.8 and A LOT of warnings with clang (almost 46 MB build log file). - m.

— Marshall

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

On Jul 29, 2014, at 11:40 AM, Marshall Clow wrote:

On Jul 29, 2014, at 10:06 AM, Marshall Clow wrote:

...

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/

Ok, ok - fine. 1.56.0

...

As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy.

This helps ensure the candidates build OK before we push them out to SourceForge.

The files (and associated md5s) are: MD5 (boost_1_56_0_rc1.7z) = 0ac22b92b24209286d48e4c41925cd9f MD5 (boost_1_56_0_rc1.tar.bz2) = 5b5bfa99244b1ac5737323decceb19c2 MD5 (boost_1_56_0_rc1.tar.gz) = 2047cdd4e09cda34e7f10fec0f821175 MD5 (boost_1_56_0_rc1.zip) = 0c544f6acc85550b15cdc9221f2ac0c9

I have successfully built using gcc 4.9.0 and clang 3.4 on Mac OS X 9.4.

Also successfully built with clang in C++11 mode and tip-of-tree clang in c++14 mode. — Marshall

Stephan T. Lavavej writes:

[Agustín K-ballo Bergé]

...

Builds successfully with msvc-12.0. However, the following trivial snippet results in an ICE:

Repros with both VC12 and VC14. Filed as DevDiv#996370 "ICE with Boost 1.56.0 RC1 - Assertion failed: is<T>(), trees.h 626".

Could that be the same bug as https://connect.microsoft.com/VisualStudio/feedback/details/915061/vc12- update-3-rc-boost-1-56-compiler-crash? That ICE is also caused by the Boost.Range free-standing operators. The mentioned workaround also applies to Boost.Lambda: replacing the anonymous enum in lambda/detail/operator_return_type_traits.hpp with static const ints makes the ICE go away.

[Agustín K-ballo Bergé]

Builds successfully with msvc-12.0. However, the following trivial snippet results in an ICE:

[STL]

Repros with both VC12 and VC14. Filed as DevDiv#996370 "ICE with Boost 1.56.0 RC1 - Assertion failed: is<T>(), trees.h 626".

The compiler team determined that this was a duplicate of DevDiv#980577/Connect#915061 "VC12 Update 3 RC: Boost 1.56 compiler crash": http://connect.microsoft.com/VisualStudio/feedbackdetail/view/915061/vc12-up... This bug is still active. STL

On 7/30/2014 1:04 AM, Andrey Semashev wrote:

On Wed, Jul 30, 2014 at 10:49 AM, Andrey Semashev wrote:

...

On Tue, Jul 29, 2014 at 9:06 PM, Marshall Clow wrote:

...

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/

I'm seeing compilation failures with gcc 4.4.7 in C++11 mode:

<snip>

...

"g++-4.4" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -pthread -fPIC -std=c++0x -DBOOST_ALL_NO_LIB=1 -DBOOST_GRAPH_DYN_LINK=1 -DBOOST_HAS_ICU=1 -DNDEBUG -I"." -I"libs/graph/src" -c -o "bin.v2/libs/graph/build/gcc-4.4/release/threading-multi/graphml.o" "libs/graph/src/graphml.cpp"

...failed gcc.compile.c++ bin.v2/libs/graph/build/gcc-4.4/release/threading-multi/graphml.o...

I'll see if I can come up with a patch.

https://github.com/boostorg/graph/pull/14

Thanks for the patch. My opinion is that we should hold on to this fix and consider rolling it into a point release shortly after 1.56.0. My reasoning is that the but doesn't have very broad scope for causing problems (it only affects GCC 4.4, which is over 4 years old). \e

On 7/30/2014 10:15 AM, Jason Roehm wrote:

On 07/30/2014 01:12 PM, Eric Niebler wrote:

...

Thanks for the patch. My opinion is that we should hold on to this fix and consider rolling it into a point release shortly after 1.56.0. My reasoning is that the but doesn't have very broad scope for causing problems (it only affects GCC 4.4, which is over 4 years old).

Just making sure: is the problem in question only triggered in C++11 mode? Note that gcc 4.4 is the version provided by the RHEL6-based Linux distributions, which are still in pretty wide use. I doubt that many users are attempting to use "-std=c++0x" with that old of a compiler though.

The patch only changes code using rvalue references, so I would say yes, this only affects people using -std=c++0x. \e

On Jul 30, 2014, at 2:04 AM, Andrey Semashev wrote:

On Wed, Jul 30, 2014 at 10:49 AM, Andrey Semashev wrote:

...

On Tue, Jul 29, 2014 at 9:06 PM, Marshall Clow wrote:

...

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/

I'm seeing compilation failures with gcc 4.4.7 in C++11 mode:

I just added a tester for gcc-4.4.7 with c++0x and graph is now clean in develop. — Noel

On 7/30/2014 12:26 AM, Andrey Semashev wrote:

On Tue, Jul 29, 2014 at 9:06 PM, Marshall Clow wrote:

...

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/

As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy.

This helps ensure the candidates build OK before we push them out to SourceForge.

The files (and associated md5s) are: MD5 (boost_1_56_0_rc1.7z) = 0ac22b92b24209286d48e4c41925cd9f MD5 (boost_1_56_0_rc1.tar.bz2) = 5b5bfa99244b1ac5737323decceb19c2 MD5 (boost_1_56_0_rc1.tar.gz) = 2047cdd4e09cda34e7f10fec0f821175 MD5 (boost_1_56_0_rc1.zip) = 0c544f6acc85550b15cdc9221f2ac0c9

I also see compilation failures with gcc 4.6 in C++11 mode:

In file included from ./boost/thread/once.hpp:20:0, from ./boost/thread.hpp:17, from libs/coroutine/src/posix/stack_traits.cpp:22: ./boost/thread/pthread/once_atomic.hpp: In function ‘void boost::call_once(boost::once_flag&, Function&&, ArgTypes&& ...) [with Function = void (&)(long unsigned int*), ArgTypes = {long unsigned int*}]’: libs/coroutine/src/posix/stack_traits.cpp:57:46: instantiated from here ./boost/thread/pthread/once_atomic.hpp:127:9: error: call of overloaded ‘invoke(boost::decay::type, boost::decay::type)’ is ambiguous ./boost/thread/pthread/once_atomic.hpp:127:9: note: candidates are: ./boost/thread/detail/invoke.hpp:538:5: note: typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type boost::detail::invoke(Fp&&, Args&& ...) [with Ret = void, Fp = void (*)(long unsigned int*), Args = {long unsigned int*}, typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type = void] ./boost/thread/detail/invoke.hpp:544:5: note: Ret boost::detail::invoke(Ret (*)(Args ...), Args&& ...) [with Ret = void, Args = {long unsigned int*}] ./boost/thread/pthread/once_atomic.hpp: In function ‘void boost::call_once(boost::once_flag&, Function&&, ArgTypes&& ...) [with Function = void (&)(rlimit*), ArgTypes = {rlimit*}]’: libs/coroutine/src/posix/stack_traits.cpp:65:54: instantiated from here ./boost/thread/pthread/once_atomic.hpp:127:9: error: call of overloaded ‘invoke(boost::decay::type, boost::decay::type)’ is ambiguous ./boost/thread/pthread/once_atomic.hpp:127:9: note: candidates are: ./boost/thread/detail/invoke.hpp:538:5: note: typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type boost::detail::invoke(Fp&&, Args&& ...) [with Ret = void, Fp = void (*)(rlimit*), Args = {rlimit*}, typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type = void] ./boost/thread/detail/invoke.hpp:544:5: note: Ret boost::detail::invoke(Ret (*)(Args ...), Args&& ...) [with Ret = void, Args = {rlimit*}]

"g++-4.6" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -pthread -fPIC -std=c++0x -DBOOST_ALL_NO_LIB=1 -DBOOST_CHRONO_DYN_LINK=1 -DBOOST_CONTEXT_DYN_LINK=1 -DBOOST_COROUTINES_DYN_LINK=1 -DBOOST_COROUTINES_SOURCE -DBOOST_SYSTEM_DYN_LINK=1 -DBOOST_SYSTEM_NO_DEPRECATED -DBOOST_THREAD_BUILD_DLL=1 -DBOOST_THREAD_POSIX -DBOOST_THREAD_USE_DLL=1 -DNDEBUG -I"." -c -o "bin.v2/libs/coroutine/build/gcc-4.6/release/threading-multi/posix/stack_traits.o" "libs/coroutine/src/posix/stack_traits.cpp"

...failed gcc.compile.c++ bin.v2/libs/coroutine/build/gcc-4.6/release/threading-multi/posix/stack_traits.o...

Ouch. Can anybody else repro this? Oliver, thoughts? Eric

On Wednesday 30 July 2014 22:20:39 you wrote:

On Wednesday 30 July 2014 10:14:59 Eric Niebler wrote:

...

On 7/30/2014 12:26 AM, Andrey Semashev wrote:

...

On Tue, Jul 29, 2014 at 9:06 PM, Marshall Clow

wrote:

...

...

...

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/

As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy.

This helps ensure the candidates build OK before we push them out to SourceForge.>>

The files (and associated md5s) are: MD5 (boost_1_56_0_rc1.7z) = 0ac22b92b24209286d48e4c41925cd9f MD5 (boost_1_56_0_rc1.tar.bz2) = 5b5bfa99244b1ac5737323decceb19c2 MD5 (boost_1_56_0_rc1.tar.gz) = 2047cdd4e09cda34e7f10fec0f821175 MD5 (boost_1_56_0_rc1.zip) = 0c544f6acc85550b15cdc9221f2ac0c9

I also see compilation failures with gcc 4.6 in C++11 mode:

In file included from ./boost/thread/once.hpp:20:0,

from ./boost/thread.hpp:17,

from libs/coroutine/src/posix/stack_traits.cpp:22: ./boost/thread/pthread/once_atomic.hpp: In function ‘void boost::call_once(boost::once_flag&, Function&&, ArgTypes&& ...) [with Function = void (&)(long unsigned int*), ArgTypes = {long unsigned int*}]’: libs/coroutine/src/posix/stack_traits.cpp:57:46: instantiated from here ./boost/thread/pthread/once_atomic.hpp:127:9: error: call of overloaded ‘invoke(boost::decay::type, boost::decay::type)’ is ambiguous ./boost/thread/pthread/once_atomic.hpp:127:9: note: candidates are: ./boost/thread/detail/invoke.hpp:538:5: note: typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type boost::detail::invoke(Fp&&, Args&& ...) [with Ret = void, Fp = void (*)(long unsigned int*), Args = {long unsigned int*}, typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type = void] ./boost/thread/detail/invoke.hpp:544:5: note: Ret boost::detail::invoke(Ret (*)(Args ...), Args&& ...) [with Ret = void, Args = {long unsigned int*}] ./boost/thread/pthread/once_atomic.hpp: In function ‘void boost::call_once(boost::once_flag&, Function&&, ArgTypes&& ...) [with Function = void (&)(rlimit*), ArgTypes = {rlimit*}]’: libs/coroutine/src/posix/stack_traits.cpp:65:54: instantiated from here ./boost/thread/pthread/once_atomic.hpp:127:9: error: call of overloaded ‘invoke(boost::decay::type, boost::decay::type)’ is ambiguous ./boost/thread/pthread/once_atomic.hpp:127:9: note: candidates are: ./boost/thread/detail/invoke.hpp:538:5: note: typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type boost::detail::invoke(Fp&&, Args&& ...) [with Ret = void, Fp = void (*)(rlimit*), Args = {rlimit*}, typename boost::enable_if_c<(! boost::is_member_function_pointer<Fp>::value), Ret>::type = void] ./boost/thread/detail/invoke.hpp:544:5: note: Ret boost::detail::invoke(Ret (*)(Args ...), Args&& ...) [with Ret = void, Args = {rlimit*}]

"g++-4.6" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline

-Wall -pthread -fPIC -std=c++0x -DBOOST_ALL_NO_LIB=1 -DBOOST_CHRONO_DYN_LINK=1 -DBOOST_CONTEXT_DYN_LINK=1 -DBOOST_COROUTINES_DYN_LINK=1 -DBOOST_COROUTINES_SOURCE -DBOOST_SYSTEM_DYN_LINK=1 -DBOOST_SYSTEM_NO_DEPRECATED -DBOOST_THREAD_BUILD_DLL=1 -DBOOST_THREAD_POSIX -DBOOST_THREAD_USE_DLL=1 -DNDEBUG -I"." -c -o "bin.v2/libs/coroutine/build/gcc-4.6/release/threading-multi/posix/stack _t raits.o" "libs/coroutine/src/posix/stack_traits.cpp"

...failed gcc.compile.c++ bin.v2/libs/coroutine/build/gcc-4.6/release/threading-multi/posix/stack_ tr aits.o...

Ouch. Can anybody else repro this? Oliver, thoughts?

I think the problem is in Boost.Thread, boost/thread/detail/invoke.hpp. I emailed Vicente, no response so far.

I'll try to come up with a patch.

https://github.com/boostorg/thread/pull/21