November 2010 - Boost-users - lists.preview.boost.org

Understanding Threads: Slow on Win32 with DLL
by Simon Adler 01 Dec '10

01 Dec '10

Hello Boost-Users, I have a strange Problem and also a theorie to solve it. I do not realy understand the reason and hope for somebody how may explain the problem. I am using boost.threads to execute some algorithms in parallel. I will post source code in detail if this is required. I have some work to do which is similar for different Data. Concrete: I compute some Forces on a Deformationmodel of geometrical Edges. I create for n available Processors n-1 Threads so lets say we have 10.000 Edges 3 times 2500 are processed from boost threads and 2500 are processed by the main thread. Because this is a worker crew i join the threads and will be finished. All this is done in Win32 using Visual Studio Express and Boost. I create some wrapper Structure at the moment struct Wrappy { void operator()( EdgeProcessor * array, int count) { for (int i = 0; i < count; i++) array[i]->compute(); } } My main application is creating a EdgeProcessor Array of 10000 Elements. In boost::thread constructor I give a ptr to the elements the thread should compute and count is for every thread 2500. You'll see - everything is straight forward and worked fine in a lot of situations. Now EdgeProcessor is a class compiled in a seperate dll (Multithreaded DLL) doing a lot of calls - some recursive, Just basic C++ calls, no std contaiiners in use. If I process all Elements with the main application using no threads it takes 0,07 sec and i have 100% usage of one CPU (have an I5, so 25% in total) If I use threads, i got 0,19 sec - more than twice the time. If I use just one boost::thread - not processing in main app - again 0,07 sec. If I use all threads all four cpus are in 100% usage - they are all together working but require more than twice the time - Yes i am sure that every thread is just working on 2500 Elements. If more than one thread is calling the EdgeProcessor, the task is done very slow. The data for the EdgeProcessor is by the way parallel, so every EdgeProcessor has its own data and there are no intersections or synchronisations at all. I assume that there is a realy big overhead because of the dll. Maybe the access of the class can not be done in parallel or not as fast as in a static case. Maybe there is a hidden synchronisation. Do you have some explanation for this? I use the same approach on other parts of the application without problems. The only difference is a) using of a class Instance within a dll and b) the compute will result i a couple of recursive calls (traversing tree) I have experience in this, but this behavior is strange. I am sorry if this is a little win32 / dll / visual studio like, but i am not sure where to ask. Thanks for your ideas! Simon

2 1

Boost 1.45 compile error
by gast128 01 Dec '10

01 Dec '10

Hello all, just tried out Boost 1.45 on VStudio 2003 but I got a compile error by only including type_traits: #include <boost/type_traits.hpp> int main() { return 0; } C:\Work Sdk\boost_1_45_0\boost\typeof\msvc\typeof_impl.hpp(160) : error C2027: use of undefined type '<Unknown>' C:\Work Sdk\boost_1_45_0\boost\typeof\msvc\typeof_impl.hpp(143) : see declaration of '<Unknown>' C:\Work Sdk\boost_1_45_0\boost\typeof\msvc\typeof_impl.hpp(160) : see reference to class template instantiation '<Unknown>' being compiled C:\Work Sdk\boost_1_45_0\boost\type_traits\common_type.hpp(121) : see reference to class template instantiation '<Unknown>' being compiled C:\Work Sdk\boost_1_45_0\boost\type_traits\common_type.hpp(123) : see reference to class template instantiation 'boost::type_traits_detail::common_type_2<T,U>' being compiled C:\Work Sdk\boost_1_45_0\boost\typeof\msvc\typeof_impl.hpp(160) : error C2146: syntax error : missing ';' before identifier 'wrapped_type' C:\Work Sdk\boost_1_45_0\boost\typeof\msvc\typeof_impl.hpp(160) : error C2501: '<Unknown>' : missing storage-class or type specifiers C:\Work Sdk\boost_1_45_0\boost\typeof\msvc\typeof_impl.hpp(161) : error C2653: 'wrapped_type' : is not a class or namespace name afaik it seems located on 121 in 'boost/type_traits/coomon_type.hpp' Anyone can help?

5 4

Usign boost::bind with a call to member function
by przemyslaw.sliwa＠uk.bnpparibas.com 30 Nov '10

30 Nov '10

Hello, I would like to replace this trivial loop for(size_t i=0; i<intersection.size(); i++) { domCcyEventPtrs.push_back(events->event(domCcy, intersection[i])); } with a call to std::transform and boost::bind. The events object is actually a boost_shared_ptr<> of a certain type. domCcyEventPtrs and intersection are std::vector<> objects. Can this be done? Cheers ___________________________________________________________ This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is prohibited. Please refer to http://www.bnpparibas.co.uk/en/information/legal_information.asp?Code=ECAS-… for additional disclosures.

2 1

[BGL] Serializing a filtered_graph
by Cedric Laczny 30 Nov '10

30 Nov '10

Hi, I was wondering if there is a way to serialize filtered_graphs? My first attempts fail because keep_all seems not to be serializable: /usr/include/boost/serialization/access.hpp:109: error: ‘struct boost::keep_all’ has no member named ‘serialize’ I have no problems serializing even custom adjacency_lists using the same approach. Besides discussing if it actually makes sense to serialize a filtered_graph, I am interested in suggestions on how to achieve it (if it is indeed possible). Thank you. Best regards, Cedric

3 6

[Bind] Retrun-type of bind expressions
by Cedric Laczny 30 Nov '10

30 Nov '10

Hello, I am currently trying to find a way to conveniently combine different predicates for a boost::filtered_graph. boost::filtered_graph needs a specification of the template-arguments. Therefore it important to know the return type of a bound expression. I have read that actually the return type of bound expression is "undeterminated" and one should rather use boost::function. However, I could not come up with an idea to realize the following, either with bind or with function: using namespace boost; // Both predicates are function objects: Default Constructible, Copy Constructible and only having "template< class V > bool operator()( const V& v)" Pred1 pred1; Pred2 pred2; filtered_graph< MyGraph, keep_all, ?bind_return_type?> fg( g, keep_all(), bind<bool>(pred1, _1) && bind<bool>(pred2, _1)); The question is now how to specify the bind_return_type, or how to do this with boost::function? Besides that, I hope that the logical combination (both predicates should hold) and use of bind() in the example is correct?! Best, Cedric

1 0

Boost Graph Adjacency matrix to boost uBLAS integer matrix
by Masanao Yajima 30 Nov '10

30 Nov '10

Hi List I am wondering if there is any function that I can use to extract the actual adjacency matrix (square 0/1 matrix) from adjacency_matrix graph object and convert it into uBLAS integer matrix. I can create an zero matrix, and then get the edge list from the graph and go through it one by one to form such a matrix. But if anyone know any better way around, I would appreciate your input. Thank you. Masanao

2 1

[BGL] Boost 1.45 implicit graph A* example crashes
by Szymon Gatner 30 Nov '10

30 Nov '10

Hi, I just tried running example included with BGL astar_maze.cpp and it crashes every time I do that. Call stack (VC++ 2008 SP1): > astar_maze.exe!boost::unordered_detail::hash_table<boost::unordered_detail::set<vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > >::find_iterator(boost::unordered_detail::hash_bucket<std::allocator<boost::array<unsigned int,2> > > * bucket=0xcfec9914, const boost::array<unsigned int,2> & k=0x002cf0d8) Line 55 + 0x3 bytes C++ astar_maze.exe!boost::unordered_detail::hash_table<boost::unordered_detail::set<vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > >::find(const boost::array<unsigned int,2> & k=0x002cf0d8) Line 583 + 0x10 bytes C++ astar_maze.exe!boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > >::find(const boost::array<unsigned int,2> & k=0x002cf0d8) Line 400 + 0x13 bytes C++ astar_maze.exe!boost::set_contains<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> >,boost::array<unsigned int,2> >(const boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > & s={...}, const boost::array<unsigned int,2> & x=0x002cf0d8) Line 21 + 0x23 bytes C++ astar_maze.exe!boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > >::operator()<boost::array<unsigned int,2> >(const boost::array<unsigned int,2> & x=0x002cf0d8) Line 60 + 0xf bytes C++ astar_maze.exe!boost::filter_iterator<boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > >,boost::transform_iterator<boost::detail::grid_graph_vertex_at<boost::grid_graph<2,unsigned int,unsigned int> >,boost::counting_iterator<unsigned int,boost::use_default,boost::use_default>,boost::use_default,boost::use_default> >::satisfy_predicate() Line 100 + 0x41 bytes C++ astar_maze.exe!boost::filter_iterator<boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > >,boost::transform_iterator<boost::detail::grid_graph_vertex_at<boost::grid_graph<2,unsigned int,unsigned int> >,boost::counting_iterator<unsigned int,boost::use_default,boost::use_default>,boost::use_default,boost::use_default> >::filter_iterator<boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > >,boost::transform_iterator<boost::detail::grid_graph_vertex_at<boost::grid_graph<2,unsigned int,unsigned int> >,boost::counting_iterator<unsigned int,boost::use_default,boost::use_default>,boost::use_default,boost::use_default> >(boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > f={...}, boost::transform_iterator<boost::detail::grid_graph_vertex_at<boost::grid_graph<2,unsigned int,unsigned int> >,boost::counting_iterator<unsigned int,boost::use_default,boost::use_default>,boost::use_default,boost::use_default> x={...}, boost::transform_iterator<boost::detail::grid_graph_vertex_at<boost::grid_graph<2,unsigned int,unsigned int> >,boost::counting_iterator<unsigned int,boost::use_default,boost::use_default>,boost::use_default,boost::use_default> end_={...}) Line 60 C++ astar_maze.exe!boost::vertices<boost::grid_graph<2,unsigned int,unsigned int>,boost::keep_all,boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > >(const boost::filtered_graph<boost::grid_graph<2,unsigned int,unsigned int>,boost::keep_all,boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > > & g={...}) Line 285 + 0x5e bytes C++ astar_maze.exe!boost::astar_search<boost::filtered_graph<boost::grid_graph<2,unsigned int,unsigned int>,boost::keep_all,boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > >,euclidean_heuristic,astar_goal_visitor,boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,boost::array<unsigned int,2> > > > >,boost::shared_array_property_map<double,boost::grid_graph_index_map<boost::grid_graph<2,unsigned int,unsigned int>,boost::array<unsigned int,2>,unsigned int> >,boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,double,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,double> > > >,boost::static_property_map<double>,boost::grid_graph_index_map<boost::grid_graph<2,unsigned int,unsigned int>,boost::array<unsigned int,2>,unsigned int>,boost::shared_array_property_map<enum boost::default_color_type,boost::grid_graph_index_map<boost::grid_graph<2,unsigned int,unsigned int>,boost::array<unsigned int,2>,unsigned int> >,std::less<double>,boost::closed_plus<double>,double,double>(const boost::filtered_graph<boost::grid_graph<2,unsigned int,unsigned int>,boost::keep_all,boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > > & g={...}, boost::array<unsigned int,2> s=0x002cf660, euclidean_heuristic h={...}, astar_goal_visitor vis={...}, boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,boost::array<unsigned int,2> > > > > predecessor={...}, boost::shared_array_property_map<double,boost::grid_graph_index_map<boost::grid_graph<2,unsigned int,unsigned int>,boost::array<unsigned int,2>,unsigned int> > cost={...}, boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,double,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,double> > > > distance={...}, boost::static_property_map<double> weight={...}, boost::grid_graph_index_map<boost::grid_graph<2,unsigned int,unsigned int>,boost::array<unsigned int,2>,unsigned int> index_map={...}, boost::shared_array_property_map<enum boost::default_color_type,boost::grid_graph_index_map<boost::grid_graph<2,unsigned int,unsigned int>,boost::array<unsigned int,2>,unsigned int> > color={...}, std::less<double> compare=less, boost::closed_plus<double> combine={...}, double inf=1.7976931348623157e+308, double zero=0.00000000000000000) Line 287 + 0x10 bytes C++ astar_maze.exe!boost::astar_search<boost::filtered_graph<boost::grid_graph<2,unsigned int,unsigned int>,boost::keep_all,boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > >,euclidean_heuristic,astar_goal_visitor,enum boost::graph_visitor_t,boost::bgl_named_params<boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,double,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,double> > > >,enum boost::vertex_distance_t,boost::bgl_named_params<boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,boost::array<unsigned int,2> > > > >,enum boost::vertex_predecessor_t,boost::bgl_named_params<boost::static_property_map<double>,enum boost::edge_weight_t,boost::no_property> > > >(const boost::filtered_graph<boost::grid_graph<2,unsigned int,unsigned int>,boost::keep_all,boost::is_not_in_subset<boost::unordered_set<boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<boost::array<unsigned int,2> > > > > & g={...}, boost::array<unsigned int,2> s=0x002cf8e8, euclidean_heuristic h={...}, const boost::bgl_named_params<astar_goal_visitor,enum boost::graph_visitor_t,boost::bgl_named_params<boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,double,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,double> > > >,enum boost::vertex_distance_t,boost::bgl_named_params<boost::associative_property_map<boost::unordered_map<boost::array<unsigned int,2>,boost::array<unsigned int,2>,vertex_hash,std::equal_to<boost::array<unsigned int,2> >,std::allocator<std::pair<boost::array<unsigned int,2> const ,boost::array<unsigned int,2> > > > >,enum boost::vertex_predecessor_t,boost::bgl_named_params<boost::static_property_map<double>,enum boost::edge_weight_t,boost::no_property> > > > & params={...}) Line 341 + 0x267 bytes C++ astar_maze.exe!maze::solve() Line 191 + 0xa2 bytes C++ astar_maze.exe!main(int argc=1, const char * * argv=0x00882610) Line 305 + 0xb bytes C++ astar_maze.exe!__tmainCRTStartup() Line 586 + 0x19 bytes C astar_maze.exe!mainCRTStartup() Line 403 C kernel32.dll!7593eccb() [Frames below may be incorrect and/or missing, no symbols loaded for kernel32.dll] ntdll.dll!7779d24d() ntdll.dll!7779d45f()

2 1

Parsing math expression
by Eloi Du Bois 30 Nov '10

30 Nov '10

Hi all, Is there a simple math expression parser in boost ? I only need to parse those kind of expression: "1+2/3". I have been thinking on boost::spirit, but it is quite hard for what I need. - Eloi.

3 8

Fwd: boost::interprocess_exception::library_error
by rohini chandra pallapothu 30 Nov '10

30 Nov '10

Hi, I forgot to mention that the exception was from the receive( ) method. Thanks & Regards, Rohini Chandra P ---------- Forwarded message ---------- From: rohini chandra pallapothu <rohinichandrap(a)gmail.com> Date: Mon, Nov 29, 2010 at 7:14 PM Subject: boost::interprocess_exception::library_error To: boost-users(a)lists.boost.org Hi, I am trying to use boost message queues for interprocess communincation between a command and a server. I had been receiving the below exception in the server. *boost::interprocess_exception::library_error* Below is the server code I had problems with. The exception was consistent when the maximum message size argument of the method was greater than the size of the data I was trying to send. I am trying to send an integer. When I assigned the max_msg_size variable a size greater than the size of int, it threw an exception. When I assigned the size of int, it is working fine. I got this working just before I was about to send this mail. But I want to understand the correct reason behind the exception. Is it not allowed to send a message which is less than the maximum message size because the name maximum message size itself says that it should only not exceed. I am working on a 64 bit SUSE 11. I am new to boost and Linux. I had even tried to change the /proc/sys/kernel/shmmax value after reading one of the discussion on the same exception in the BOOST archives which did not work. Could anyone please help. Please let me know if you need more information. *std::size_t max_msg_size=sizeof(int);* ============testsvc================= #include <boost/interprocess/sync/named_mutex.hpp> #include <boost/interprocess/ipc/message_queue.hpp> #include <iostream> using namespace std; using namespace boost::interprocess; int main(){ cout<<"in main"<<endl; cout<<"size of int is "<<sizeof(int)<<endl; //4 on my platform unsigned int priority; std::size_t recvd_size; std::size_t max_msg_size=sizeof(int); //not working when assigned a larger value than I was sending. std::size_t max_no_msgs=10; int i; message_queue::remove("TestQueue"); cout<<"message queue removed if any"<<endl; try{ //create a message queue. message_queue mq(create_only,"TestQueue",max_no_msgs,max_msg_size); cout<<"message queue created"<<endl; mq.receive(&i,sizeof(i),recvd_size,priority); } catch(interprocess_exception &ex){ std::cout <<ex.what() << std::endl; return 1; } cout<<"message received"<<endl; cout<<"message received is"<<i<<endl; return 0; } ==================== Thanks in advance. Best Regards, Rohini Chandra

1 0

[boost-users] [boost_thread] Condition
by Eloi Du Bois 30 Nov '10

30 Nov '10

Hi all, I would like to know how can I do the following: I have n working threads. Each one has a loop preceded by a wait condition (1). When the wait condition is unlocked, the thread takes a task on a queue, and process the task. I need to make something in the submit task function that wait until all processes are on the wait condition (1). How can I do that ? I tried a while(!_nWorking) {} but this is not clean. Here follows my threads function: void ProcessingThread::dispatcher( const int idx ) { using namespace boost; mutex::scoped_lock readyLocker( _dispatchMutex[idx] ); while( !_bDispStop ) { --_nWorking; // Wait for a task _condProcessIn[idx].wait( readyLocker ); ++_nWorking; while( !_bDispStop && !_pendingQueues[idx].empty() ) { _pendingQueuesMutex[idx].lock(); Task task = _pendingQueues[idx].front(); _pendingQueues[idx].pop_front(); _pendingQueuesMutex[idx].unlock(); // Process node assert( task._proc ); assert( task._procData ); task._proc->execute( task._range, *task._procData ); } } } Launched via: _nWorking = _nCpu; for( std::size_t i = 0; i < _nCpu; ++i ) { // Create dispatcher thread _dispThrdGroup.create_thread( boost::bind( &ProcessingThread::dispatcher, this, i ) ); } Submit task function: void ProcessingThread::submitProcess( ProcessLauncher * proc, const FrameRange & range, ProcessData & procData ) { // Get the index of the less used queue std::size_t idx = getLessUsedQueueIdx(); _pendingQueuesMutex[idx].lock(); // Append a task to the less used queue _pendingQueues[ idx ].push_back( Task( proc, range, procData ) ); // Signal that a task is ready _condProcessIn[idx].notify_one(); _pendingQueuesMutex[idx].unlock(); /// UGLY ! while( _nWorking ) { } } Thank you in advance, Eloi Du Bois.

2 3