April 2008 - Boost-users - lists.preview.boost.org

Boost.Threads, Boost.Interprocess, mutex
by Axel 30 Apr '08

30 Apr '08

Hello, In a single process but multithreaded software, should I use Boost.Threads synchronization mechanisms or can I use Boost.Interprocess synchronization mechanisms without counterparts and without seeing any difference ? ( in particular boost::interprocess::interprocess_semaphore class ) When using interprocess_mutex, my programm threw a lock_exception. I changed to boost::mutex and didn't managed to reproduce it. So, could this be related or the lock_exception has same chances to happen with boost::mutex ? I'm using semaphore to synchronize threads, so is it reliable in a single process multithreaded software ? By the way, I'm currently using boost::mutex instances which I lock() and unlock() directly. What benefits could I have by using boost::unique_lock (with a deferred lock) rather than access directly to the mutex instance ? Thanks in advance. Axel

2 1

Building boost 1.35.0 using mingw-gcc-4.2.1 with shared runtime
by Lothar May 30 Apr '08

30 Apr '08

Hi, by default, the shared boost libraries 1.35.0 compiled with mingw gcc 4.2.1 link with the static C/C++ runtime, which means that exceptions cannot be thrown across dll boundaries. Is there a simple way to make boost link using the "-shared-libgcc -lstdc++_s -D_DLL" options? Best regards, Lothar

1 0

[filesystem] bug???
by Ovanes Markarian 30 Apr '08

30 Apr '08

Hello *, iterating through a directory (let's say in Windows c:\) with system files, to which only OS has access (like hiberfil.sys) causes the directory iterator to throw an exception. This is all fine, but how can I know what was the error? status function of the particular directory enty returns regular_file as file_status and system_error_type is 0. The exception thrown is as always (in the filesystem lib) very much saying: boost::directory_iterator. Is anybody aware of other ways to determine what happened? Just to ignore the exception while iterating through the directory is not always the best choice. Many Thanks, Ovanes

1 0

[thread][1.35.0] Updated win32 condition variable implementation
by Anthony Williams 30 Apr '08

30 Apr '08

Hi all, Apologies to those who have been experiencing problems using boost.thread condition variables on Windows. I have checked in a new tentative fix to SVN trunk, and would be grateful if people could put it through its paces. Thank you for your patience, Anthony -- Anthony Williams | Just Software Solutions Ltd Custom Software Development | http://www.justsoftwaresolutions.co.uk Registered in England, Company Number 5478976. Registered Office: 15 Carrallack Mews, St Just, Cornwall, TR19 7UL

5 9

[iostreams] zlib name clashes
by Robert Bielik 30 Apr '08

30 Apr '08

I have an application where I need the zlib iostreams filter, however, the zlib part of the application is compiled with Z_PREFIX (which prepends all zlib functions with z_ via defines). Since Z_PREFIX prepends via defines, the zlib_base functions that are named the same as their zlib counterpart are "renamed" so the compiler cannot find the functions z_* in zlib_base. My suggestion is to rename the zlib_base functions so that they won't be redefined if Z_PREFIX is defines. /R

2 1

FW: Using Boost.Build with Boost.Python: Three Questions; "Copy" Rule
by Bilokon, Paul 30 Apr '08

30 Apr '08

Hi, Forwarding this question to boost-users as it applies to Boost.Python and Boost.Build in equal measure. Regards, Paul > ______________________________________________ > From: Bilokon, Paul > Sent: Wednesday, April 30, 2008 3:07 PM > To: 'boost-build(a)lists.boost.org' > Subject: Using Boost.Build with Boost.Python: Three Questions; > "Copy" Rule > > Hi, > > I would be very much obliged if you could share your insights on the > following three questions: > > (1) Is it at all possible to change the location of the "bin" folder > for my project or is it hardcoded? (bjam generates a number of files > under (project root)/bin/msvc-8.0/debug and (project > root)/bin/msvc-8.0/release. If so, what is the "correct" (good > practice) way of achieving this? > > (2) I am using Boost.Build with Boost.Python. The generated files are > linked to boost_python-vc80-mt-gd-1_34_1.dll (debug) and > boost_python-vc80-mt-gd-1_34_1.dll (release). But then I'm using > > install bin : hello : <location>$(dist-dir)/debug <variant>debug ; > install bin : hello : <location>$(dist-dir)/release <variant>release ; > > in Jamroot to dist the files. The DLLs do not get copied to the dist > location (nor do they appear under "bin"). How can I instruct bjam to > copy the DLLs? > > (3) On a related note, is there a "copy" rule in Boost.Build (very > much along the lines of the similar Ant rule), which would copy my > supporting Python files to $(dist-dir)? > > Sorry if my questions have obvious answers; I did read through the > documentation and check the related resources on the Web. > > Many thanks for your help. > > Regards, > Paul > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - This message is intended only for the personal and confidential use of the designated recipient(s) named above. If you are not the intended recipient of this message you are hereby notified that any review, dissemination, distribution or copying of this message is strictly prohibited. This communication is for information purposes only and should not be regarded as an offer to sell or as a solicitation of an offer to buy any financial product, an official confirmation of any transaction, or as an official statement of Lehman Brothers. Email transmission cannot be guaranteed to be secure or error-free. Therefore, we do not represent that this information is complete or accurate and it should not be relied upon as such. All information is subject to change without notice.

1 0

Re: [Boost-users] [variant, mpl] passing variable list of types
by Neal Meyer 30 Apr '08

30 Apr '08

This is what I use for the same thing typedef typename boost::make_variant_over< Types >::type variant_t; Where Types is a mpl::vector<> of the types I want to put in the vector. -Neal Date: Wed, 30 Apr 2008 17:28:30 +0300 From: Igor R. <igor_rubinov(a)hotmail.com> Subject: [Boost-users] [variant, mpl] passing variable list of types as a template parameter To: <boost-users(a)lists.boost.org> Message-ID: <BAY109-W253127985653CE5AF34F9AF8D80(a)phx.gbl> Content-Type: text/plain; charset="windows-1255" Hello, I've got a class template that one of its members is boost::variant: template<?> class MyClass { boost::variant</*bounded types*/> var_; }; At the stage of MyClass instantiation, I'd like to specify variant' bounded types list - which is variable. What's the most simple way to do this? Can I use mpl::list somehow - both as MyClass template parameter and in the variant specialization? Thank you.

2 1

[statechart] using the same event handler in multiple states.
by Paul Heil 30 Apr '08

30 Apr '08

I'm trying to implement a state machine using the boost statechart library. It has 3 main states (call them "State 1", "State 2", and "State 3") and 3 classes of events ("Class 1", "Class 2", and "Class 3"). State 1 should have event handlers for all Class 1 events. State 2 should have handlers for Class 1 and 2 events. State 3 should have handlers for Class 1, 2, and 3 events. The handling of each class of events is essentially the same regardless of what state the machine is in. Is there a way I can derive State 3 from States 1 and 2 so that I'm not repeating the Class 1 and 2 event handling code in State 3? What is the best way to architect something like this? Thanks, PaulH

3 3

boost MPI on FreeBSD 4.11
by Dheeraj Gautam 30 Apr '08

30 Apr '08

Hi, I am trying to compile Boost.MPI on FreeBSD 4.11 machine. But its failing with following error: Found MPICH wrapper compiler: mpic++ MPI compilation flags: -I/home/eqsupport/mpich2_install/include -L/home/eqsupport/mpich2_install/lib -lmpichcxx -lmpich MPI link flags: -I/home/eqsupport/mpich2_install/include -L/home/eqsupport/mpich2_install/lib -lmpichcxx -lmpich MPI build features: <include>/home/eqsupport/mpich2_install/include <library-path>/home/eqsupport/mpich2_install/lib <find-shared-library>mpichcxx <find-shared-library>mpich <include>/home/eqsupport/mpich2_install/include <library-path>/home/eqsupport/mpich2_install/lib < find-shared-library>mpichcxx <find-shared-library>mpich MPI launcher: mpirun -np ==================================================== warning: No toolsets are configured. warning: Configuring default toolset "gcc". warning: If the default is wrong, your build may not work correctly. warning: Use the "--toolset=xxxxx" option to override our guess. warning: For more configuration options, please consult warning: http://boost.org/boost-build2/doc/html/bbv2/advanced/configuration.html notice: will use 'g++' for gcc, condition <toolset>gcc-2.95.4 notice: using gcc libraries :: <toolset>gcc-2.95.4 :: /usr/bin /usr/lib /usr/lib32 /usr/lib64 notice: using gcc archiver :: <toolset>gcc-2.95.4 :: /usr/bin/ar warning: toolset gcc initialization: can't find tool windres warning: initialized from notice: using rc compiler :: <toolset>gcc-2.95.4 :: /usr/bin/as ...patience... ...found 1889 targets... ...updating 29 targets... gcc.compile.c++ bin.v2/libs/mpi/build/gcc-2.95.4/release/threading-multi/broadcast.o In file included from boost/serialization/nvp.hpp:34, from boost/serialization/serialization.hpp:24, from boost/archive/detail/oserializer.hpp:64, from boost/mpi/detail/mpi_datatype_oarchive.hpp:12, from boost/mpi/detail/mpi_datatype_cache.hpp:13, from boost/mpi/datatype.hpp:26, from boost/mpi/communicator.hpp:20, from boost/mpi/collectives.hpp:21, from boost/mpi/collectives_fwd.hpp:19, from boost/mpi/collectives/broadcast.hpp:11, from libs/mpi/src/broadcast.cpp:9: boost/serialization/base_object.hpp: In function `static const void * boost::serialization::detail::base_register<Base,Derived >::invoke()': boost/serialization/base_object.hpp:56: syntax error before `::' boost/serialization/base_object.hpp:59: use of class template `template <class T> boost::is_polymorphic<T>' as expression In file included from boost/mpi/detail/mpi_datatype_oarchive.hpp:12, from boost/mpi/detail/mpi_datatype_cache.hpp:13, from boost/mpi/datatype.hpp:26, from boost/mpi/communicator.hpp:20, from boost/mpi/collectives.hpp:21, from boost/mpi/collectives_fwd.hpp:19, from boost/mpi/collectives/broadcast.hpp:11, from libs/mpi/src/broadcast.cpp:9: boost/archive/detail/oserializer.hpp: In function `static const class boost::archive::detail::basic_pointer_oserializer * boos t::archive::detail::save_pointer_type<Archive,TPtr>::abstract<T>::register_type(Archive &)': boost/archive/detail/oserializer.hpp:329: syntax error before `::' boost/archive/detail/oserializer.hpp:330: template argument 1 is invalid boost/archive/detail/oserializer.hpp:330: confused by earlier errors, bailing out "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -fPIC -pthread -DBOOST_ALL_NO_LIB=1 -DBOOST_MPI_DYN_ LINK=1 -DBOOST_MPI_SOURCE=1 -DNDEBUG -I"." -I"/home/eqsupport/mpich2_install/include" -c -o "bin.v2/libs/mpi/build/gcc-2.95.4 /release/threading-multi/broadcast.o" "libs/mpi/src/broadcast.cpp" ...failed gcc.compile.c++ bin.v2/libs/mpi/build/gcc-2.95.4/release/threading-multi/broadcast.o... gcc.compile.c++ bin.v2/libs/mpi/build/gcc-2.95.4/release/threading-multi/communicator.o In file included from boost/serialization/nvp.hpp:34, from boost/serialization/serialization.hpp:24, from boost/archive/detail/oserializer.hpp:64, from boost/mpi/detail/mpi_datatype_oarchive.hpp:12, from boost/mpi/detail/mpi_datatype_cache.hpp:13, ========================================================================= Almost all compilations for MPI are failing with similar error. It seems some C++ support is not there. I am new to C++ and Boost. Any help will be of great help. Dheeraj Gautam

3 2

Serialization of boost::thread::detail::singleton
by Campbell Morrison 30 Apr '08

30 Apr '08

I need to serialize a singleton, but have no idea how to go about it. I have made several attempts to date with very little success. I am new to using Boost and would greatly appreciate any help. regards, Campbell.

1 0