[Interprocess Test]Boost Interprocess tests failed to compile on solaris 9
Hi,
The interprocess test compilation fails on solaris 9. Error is thrown for
solaris atomic operations.
Following the command used and log -
./bjam
./libs/interprocess/test/
...patience...
...patience...
...found 1953
targets...
...updating 473
targets...
common.mkdir
bin.v2/libs/interprocess
common.mkdir
bin.v2/libs/interprocess/test
common.mkdir
bin.v2/libs/interprocess/test/intersegment_ptr_test.test
common.mkdir
bin.v2/libs/interprocess/test/intersegment_ptr_test.test/gcc-3.4.2
common.mkdir
bin.v2/libs/interprocess/test/intersegment_ptr_test.test/gcc-3.4.2/debug
common.mkdir
bin.v2/libs/interprocess/test/intersegment_ptr_test.test/gcc-3.4.2/debug/threading-multi
gcc.compile.c++
bin.v2/libs/interprocess/test/intersegment_ptr_test.test/gcc-3.4.2/debug/threading-multi/intersegment_ptr_test.o
In file included from
./boost/interprocess/detail/managed_open_or_create_impl.hpp:20,
from
./boost/interprocess/detail/managed_multi_shared_memory.hpp:30,
from
libs/interprocess/test/intersegment_ptr_test.cpp:17:
./boost/interprocess/detail/atomic.hpp: In function `boost::uint32_t
boost::interprocess::detail::atomic_add32(volatile boost::uint32_t*,
boost::uint32_t)':
./boost/interprocess/detail/atomic.hpp:331: error: invalid conversion from
`volatile uint32_t*' to
`uint32_t*'
./boost/interprocess/detail/atomic.hpp:331: error: initializing argument 1
of `uint32_t atomic_add_32_nv(uint32_t*,
int32_t)'
./boost/interprocess/detail/atomic.hpp: In function `boost::uint32_t
boost::interprocess::detail::atomic_cas32(volatile boost::uint32_t*,
boost::uint32_t, boost::uint32_t)':
./boost/interprocess/detail/atomic.hpp:341: error: `atomic_cas_32'
undeclared (first use this
function)
./boost/interprocess/detail/atomic.hpp:341: error: (Each undeclared
identifier is reported only once for each function it appears
in.)
./boost/interprocess/detail/atomic.hpp: In function `boost::uint32_t
boost::interprocess::detail::atomic_inc32(volatile
boost::uint32_t*)':
./boost/interprocess/detail/atomic.hpp:347: error: invalid conversion from
`volatile uint32_t*' to
`uint32_t*'
./boost/interprocess/detail/atomic.hpp:347: error: initializing argument 1
of `uint32_t atomic_add_32_nv(uint32_t*,
int32_t)'
./boost/interprocess/detail/atomic.hpp: In function `boost::uint32_t
boost::interprocess::detail::atomic_dec32(volatile
boost::uint32_t*)':
./boost/interprocess/detail/atomic.hpp:353: error: invalid conversion from
`volatile uint32_t*' to
`uint32_t*'
./boost/interprocess/detail/atomic.hpp:353: error: initializing argument 1
of `uint32_t atomic_add_32_nv(uint32_t*,
int32_t)'
"g++" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -pthreads -fPIC
-DBOOST_ALL_NO_LIB=1 -DBOOST_THREAD_POSIX -DBOOST_THREAD_USE_DLL=1 -I"." -c
-o
"bin.v2/libs/interprocess/test/intersegment_ptr_test.test/gcc-3.4.2/debug/threading-multi/intersegment_ptr_test.o"
"libs/interprocess/test/intersegment_ptr_test.cpp"
...failed gcc.compile.c++
bin.v2/libs/interprocess/test/intersegment_ptr_test.test/gcc-3.4.2/debug/threading-multi/intersegment_ptr_test.o...
common.mkdir
bin.v2/libs/thread/build/gcc-3.4.2/debug
common.mkdir
bin.v2/libs/thread/build/gcc-3.4.2/debug/threading-multi
common.mkdir
bin.v2/libs/thread/build/gcc-3.4.2/debug/threading-multi/pthread
gcc.compile.c++
bin.v2/libs/thread/build/gcc-3.4.2/debug/threading-multi/pthread/thread.o
gcc.compile.c++
bin.v2/libs/thread/build/gcc-3.4.2/debug/threading-multi/pthread/exceptions.o
gcc.compile.c++
bin.v2/libs/thread/build/gcc-3.4.2/debug/threading-multi/pthread/once.o
gcc.link.dll
bin.v2/libs/thread/build/gcc-3.4.2/debug/threading-multi/libboost_thread-gcc34-mt-d-1_39.so.1.39.0
...skipped
d s escribió:
Hi,
The interprocess test compilation fails on solaris 9. Error is thrown for solaris atomic operations.
I have checked online, it suggests that solaris 9 atomic operations are not available for user code. It can only be used by kernel. Kindly suggest what may be wrong, or any workaround for the same.
Then Solaris 9 is not supported by Interprocess. Atomic operations are needed, so I'm afraid there is no easy fix. Sorry for the bad news, Ion
participants (2)
-
d s
-
Ion Gaztañaga