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

Graph lib: read_dimacs_max_flow segment fault
by Xian Chen 18 Feb '09

18 Feb '09

Hi All, I compile read_dimacs_max_flow example ( http://www.boost.org/doc/libs/1_38_0/libs/graph/example/read_write_dimacs-e…) with these warnnings. g++ -I/home/xianchen/opt/include/boost-1_38 -Wno-deprecated -c read.cpp -o objects/simu.o /home/xianchen/opt/include/boost-1_38/boost/graph/read_dimacs.hpp: In function 'int boost::read_dimacs_max_flow(Graph&, CapacityMap, ReverseEdgeMap, typename boost::graph_traits<G>::vertex_descriptor&, typename boost::graph_traits<G>::vertex_descriptor&, std::istream&) [with Graph = main()::Graph, CapacityMap = boost::adj_list_edge_property_map<boost::directed_tag, long int, long int&, unsigned int, boost::property<boost::edge_capacity_t, long int, boost::property<boost::edge_reverse_t, boost::detail::edge_desc_impl<boost::directed_tag, unsigned int>, boost::no_property> >, boost::edge_capacity_t>, ReverseEdgeMap = boost::adj_list_edge_property_map<boost::directed_tag, boost::detail::edge_desc_impl<boost::directed_tag, unsigned int>, boost::detail::edge_desc_impl<boost::directed_tag, unsigned int>&, unsigned int, boost::property<boost::edge_capacity_t, long int, boost::property<boost::edge_reverse_t, boost::detail::edge_desc_impl<boost::directed_tag, unsigned int>, boost::no_property> >, boost::edge_reverse_t>]': read.cpp:86: instantiated from here /home/xianchen/opt/include/boost-1_38/boost/graph/read_dimacs.hpp:108: warning: deprecated conversion from string constant to 'char*' /home/xianchen/opt/include/boost-1_38/boost/graph/read_dimacs.hpp:108: warning: deprecated conversion from string constant to 'char*' ...... My GCC is: Using built-in specs. Target: i686-pc-linux-gnu Configured with: ../gcc-4.2.4/configure --prefix=/home/xianchen/opt --enable-shared --enable-threads=posix --enable-languages=c,c++ --enable-checking=release --enable-mpfr=/home/xianchen/opt/ --with-system-zlib Thread model: posix gcc version 4.2.4 I run this program using data "boost/libs/graph/example/max_flow.dat" But get segment fault: *** glibc detected *** free(): invalid pointer: 0x08053ff4 *** Aborted Can any one tell me the why this happen? The program aborts under GCC.4.3.3 either. Which version of GCC is the most boost friendly? Thanks in advance, Xian

1 0

[serialization] export.hpp changes from 1.37 to 1.38
by Chard 18 Feb '09

18 Feb '09

The BOOST_CLASS_EXPORT.. macros have been moved into a detail namespace (1.38) from a (public) anonymous namespace (1.37). I'm guessing that the reasoning for this was in order to avoid many definitions of the class singletons. However, this appears to have enforced that all of the guids (for all archive types) now have to reside in a single compilation unit. The attached code illustrates the limitation; this compiles under 1.37 but results in duplicate symbol definitions in 1.38 (with MSVC7.1). It worked with 1.37 because the anonymous namespace ensured no symbol clashes, but now the symbols are all in the same namespace. The example shows a case where two archive types (xml and binary) might be in use. The obvious solution to this (simple) case would be to just use a single compilation unit, which brings in the guids for both binary and xml. However, turning a blind eye to the 'untidiness', this fix does not help with the project we're working on. In our project, there is not just a simple type being serialized; there are a lot more classes and a bit of mpl thrown in. This actually resulted in breaching the COFF symbol limit (approx. 65k) with MSVC7.1, so we were forced to split the serialization. In fact, we had to split both the save and load parts of the binary serialization. Our current workaround is to use the archive.register_type<...> method, and ignore the export macros, but this does mean ensuring all types are covered (as described in the documentation).

4 9

Boost and Real Time
by Brian Wrenn 18 Feb '09

18 Feb '09

Hi, I'm an experienced C++ programmer, but very new to Boost. I'm looking into making use of it for my next project, which will run on LynxOS, a real time system. Does anyone have any experience, knowledge of pitfalls, or words of wisdom to share about building and running with Boost, or more likely a subset of Boost's libraries, for a real time system? Thanks for any help. -Brian

2 1

[asio] io_service use in dll leads to service_registry crash
by joshr 18 Feb '09

18 Feb '09

Configuration: 1) Boost 1.37 2) Visual Studio 2008 3) Windows executable using boost's multi-threaded static library, using a static linked multi-threaded run time. 4) A dll I have as part of the project is compiled with the same settings. 5) MFC is not used. The executable creates an io_service (io) and pool of worker threads, each calling io.run(). Afterwards, the executable begins loading DLLs I've written to add specific functionality (i.e plugins). Each of these plugins are passed the executable's io_service so that they can perform asynchronous work against the thread pool. One of the DLLs creates a boost::asio::ip::tcp::acceptor object, passing the io_service to its constructor. This, in turn, eventually calls use_service, which creates a new service and associates it with the service_registry. This is all happening in the DLL. At shut down, I begin unloading the DLLs, giving each a chance to shut down everything (which they do). I then call FreeLibrary() on the DLL's HMODULE handle. After my application completes shutdown, all the destructors hit, as expected. When the destructor of the service_registry hits, I believe its linked list of services is still pointing to a service located in the memory space of the DLL. This results in a crash. I believe this is the case as I can track the pointer to the acceptor_service in a Watch Window. The pointers and the object remain valid until I call "FreeLibrary", at which point they become worthless. I think this is because the DLL and the executable have their own CRT memory space, due to the way I'm linking in the CRT. The service for the acceptor is being created in the DLL's space. When FreeLibrary is called, this space is released to the OS, making the pointer to the object worthless. If that's the case, I believe I'll need to supply my own service creation routine for the DLLs so that the services are created in my executable's memory space. Is this possible? Is there a way to get the service_registry to release a service when it's no longer in use anywhere, or to specify my own new/delete routines the use_service calls should use? Or, is using ASIO in a DLL like this a really really bad idea? If that's the case, how can I tell what other objects in boost could cause these sorts of problems? I'll try to get a project together that reproduces the issue as soon as I can. I can't use the code I have at the moment as it is proprietary. Thanks for any input anyone can provide. -- View this message in context: http://www.nabble.com/-asio--io_service-use-in-dll-leads-to-service_registr… Sent from the Boost - Users mailing list archive at Nabble.com.

2 3

[ASIO] Newbie; How can I access remote_endpoint data in an SSL server
by May James 18 Feb '09

18 Feb '09

Can anybody help a boost::newbie I have generated an ASIO SSL server based on the sample code "server.cpp" How can I determine the Client Address from the ssl_socket which is defined as typedef boost::asio::ssl::stream<boost::asio::ip::tcp::socket> ssl_socket; relevant code is: acceptor_.async_accept(new_session->socket(), boost::bind(&server::handle_accept, this, new_session, boost::asio::placeholders::error)); Thank You

2 1

[assign] Nested initialization of std containers
by Joseph Lisee 18 Feb '09

18 Feb '09

I am trying to use list_of and map_list_of to initialize a std::vector containing std::maps. I am using Boost 1.37 and Mac OS X 10.4 with Apple's G++ 4.0.1. Here is a complete example: #include <vector> #include <map> #include <string> #include <boost/assign/list_of.hpp> using namespace boost::assign; typedef std::map<int, std::string> dict; typedef std::vector<dict> list; int main() { // Works dict d = map_list_of(1,"B")(2,"A"); list l1 = list_of(d); // Does not compile list l2 = list_of(map_list_of(1,"B")(2,"A")); return 0; } Is there some way to get this to work?

2 1

No packets received with boost::asio::ip::udp::socket set to multicast group
by Lennyk 18 Feb '09

18 Feb '09

Hi, I'm using boost 1.35 on a Linux machine. I've written an application which creates a boost::asio::ip::udp::socket, then binds to a certain local address & port - and finally I've created an option to join a boost::asio::ip::multicast::join_group. All seems well, and I see the registration request through a network sniffer. BUT, when I try to receive packets after successfully joining the udp multicast group - I get NOTHING! Needless to say, I see tons of packets streaming through with the sniffer. What am I doing WRONG? Here's my code (just a test application): using namespace std; using namespace boost::asio; int main(int argc, char *argv[]) { if (argc != 4) { // Test for correct number of parameters cerr << "Usage: " << argv[0] << " <Local Address> <Multicast Address> <Port>" << endl; exit(1); } io_service m_io_service; ip::udp::socket m_socket(m_io_service); ip::address localAddress(ip::address_v4::from_string(argv[1])); ip::address multicastAddress(ip::address_v4::from_string(argv[2])); unsigned short port = atoi(argv[3]); // Second arg: port boost::system::error_code error; m_socket.open(ip::udp::v4()); m_socket.bind(ip::udp::endpoint(localAddress, port), error); if (error) { cout << "bind error!" << endl; exit(1); } ip::multicast::join_group option(multicastAddress.to_v4(), localAddress.to_v4()); m_socket.set_option(option); vector<unsigned char> m_receiveBuffer; mutable_buffers_1 buf = buffer((void*)&m_receiveBuffer[0], 1024); int size = (int)m_socket.receive(buf, 0, error); if (error) { cout << "recv error!" << endl; exit(1); } return 0; } Thanks! Lenny.

1 0

Setting thread affinity on Linux using boost::thread
by Lennyk 18 Feb '09

18 Feb '09

Hi, I'm running on Linux and I would like to set a specific thread affinity to each thread I create (i.e. setting each thread to run on a different core/CPU). How can this be done using the boost::thread library? Thanks! Lenny.

4 6

Do the Range Adaptors really depend upon the RegEx static lib?
by Nate Foreman 18 Feb '09

18 Feb '09

Hello, I'm using the range adaptors from the latest version of the range library in the vault. Namely: boost::adaptors::filtered boost::adaptors::transformed The Visual Studio linker was kind enough to inform me that these depend upon the regex static library (libboost_regex-vc80-mt-sgd-1_38.lib). Does anybody know why? I'd rather not build the boost regex lib. Is there any way to avoid this dependency?

2 2

[program options] access violation with _SECURE_SCL turned off
by klaus triendl 18 Feb '09

18 Feb '09

Hi, I'm using the boost 1.36 program options library on windows with msvc++2008, statically linked to my applications. In "debug mode" everything works fine, however, in "release mode" I encounter an access violation when parsing the c++ commandline passed to main() or when parsing the windows commandline with split_winmain(). Note that this happens only if and only if the preprocessor macro _SECURE_SCL gets turned off (_SECURE_SCL=0). scenario 1) Parsing the main() options, even the simple example from the tutorial, section "getting started", terminates with an access violation when calling po::store() without a usable stack trace; scenario 2) A simple call to po::split_winmain(L"\"--test\"") produces a crash. Note the use of a wide character string. The stack trace: boost::program_options::detail::utf8_codecvt_facet::do_in() boost::detail::convert<wchar_t,char,boost::_bi::bind_t<int,boost::_mfi::cmf7<int,std::codecvt<wchar_t,char,int>,int &,char const *,char const *,char const * &,wchar_t *,wchar_t *,wchar_t * &>,boost::_bi::list8<boost::_bi::value<std::codecvt<wchar_t,char,int> const *>,boost::arg<1>,boost::arg<2>,boost::arg<3>,boost::arg<4>,boost::arg<5>,boost::arg<6>,boost::arg<7> > > >() boost::from_8_bit() boost::from_utf8() boost::program_options::split_winmain() wmain(int argc=3407872, wchar_t * * argv=0x00000000) Did somebody encounter this issue? Any solutions? Thanks, Klaus Triendl

3 4