- Boost-users - lists.boost.org

[Boost.Atomic] Wait-free multi-producer queue example bug?
by Ioannis Papadopoulos 16 May '13

16 May '13

Hi, In http://www.boost.org/doc/libs/1_53_0/doc/html/atomic/usage_examples.html#bo… I see that pop_all_reverse() is implemented as: node * pop_all_reverse(void) { return head_.exchange(0, boost::memory_order_consume); } Shouldn't the memory order be memory_order_acquire in this example? If I'm not mistaken, at least an acquire is required to see all the writes from the thread that did the release. GCC 4.8 also complains that memory_order_consume is not a correct memory ordering for exchange. Thanks

1 0

[Graph] Problem with large vertex ids
by Stephen Woodbridge 16 May '13

16 May '13

Hi all, I'm trying to understand/fix some bugs in pgRouting that use the boost libraries. I know C not not C++, so I can understand most of the more C-ish parts. We have an old bug that I believe is caused when we add an edge and it has large vertex ids or when there is a large number between the min and max vertex numbers. We currently identify the min id number and down shift the ids by that, but that does not help if I add an edge (1)->(100000000). I could renumber all the nodes and then un-renumber them when done if I have to. Questions: 1. is this a known limitiation in Boost graph? 2. is there a different type of adjacency list mechanism that does not have this issue 3. currently the code crashes the database backend, ideally I would like to catch any issues in the C++ wrapper, free memory and return an error code to the C caller so we don't kill the server or leak lots of memory. Here is a link to our boost wrapper code: https://github.com/pgRouting/pgrouting/blob/sew-devel-2_0/src/dijkstra/src/… We use that pattern for a lot of functions so if there is a better pattern, I might be able to update most(all?) of the functions that use this. I could really use some help resolving this issue. Meanwhile, I'm off to write a C main() to run this outside of the database and in gdb to see whats happening. Thanks, -Steve, trying to release pgRouting 2.0

3 4

Interprocess : vectors of vectors
by Oodini 16 May '13

16 May '13

Hello, I'd like to share a vector of vectors (jagged 2D array). The data come from a deserialization. I know the size of the "external" vector. The sizes of the "internal" vectors are discovered during deserialization. I read : http://www.boost.org/doc/libs/1_52_0/doc/html/interprocess/quick_guide.html… and regarding this line : managed_shared_memory segment(create_only, "MySharedMemory", //segment name 65536); how to set the size (3rd argument), when considering my use case ? Even with a simple vector, and if the number and the types of the elements are know? i woldn't know what size to set for truncation. On VS 2010, a sizeof(vector<int>) is 40; I suppose I have to add it sizeof(int)*nbElements to get the full size. 1. I don't know if this assumption is valid 2. If so, I don't know if it's still valid when using Boost.Interprocess (I suppose it isn't) 3. I don't know how to generalize that to a vector of vectors in order what should be the truncation size. Thanks for help. PS : I also read http://www.boost.org/doc/libs/1_52_0/doc/html/interprocess/managed_memory_s…, but I havn't been enlightened about these points.

4 14

Interprocess : offset_ptr used as array
by Oodini 16 May '13

16 May '13

Hello, I have an offset_ptr named tmpData defined like this : offset_ptr<const byte>* someVariable; indextype index; ... offset_ptr<const byte> tmpData = ((offset_ptr<const byte>*)someVariable)[index]; data points actually to the beginning of an array. The original code, without shared memory was : byte* data= ((byte**)m_someVariable)[index]; The problem is that I have to translate that : indextype* data = (indextype*)&(tmpData[1]); The adress of the second element is taken though the & adress. But when tmpData is an offset_ptr, the [] operator returns a reference, and I can apply the & operator. Is there any solution that copying the pointer to temporary variable, and increment this one ? Thanks.

1 0

Re: [Boost-users] boost::variant & c++11
by Antony Polukhin 16 May '13

16 May '13

2013/5/15 Bryan Catanzaro <bcatanzaro(a)acm.org>: > Hi Antony - > Perhaps it is compiler dependent. Which version of MSVC did you reproduce > the error with? Does that version have c++11 support enabled (in particular, > rvalue references) > As I said, with Boost 1.48.0, this code works with g++ 4.6 and 4.7. With > 1.53.0, it works with g++ 4.6 and 4.7 only if c++11 support is turned off. > If c++11 support is turned on, the code does not work. > I've attached a shell script that compiles the attached code with the cross > product of {(g++4.6, g++4.7), (boost1.48, boost1.53), (noc++11, yesc++11)}. > As you can see from test.txt, the error happens only with boost 1.53 and > c++11 support. > > > - bryan Exactly the same things happen on GCC-4.5 and Clang-3.0. But what is more interesting here: convertible c; const convertible& cc = c; // Always compiles, using any flags, compiler and Boost version my_variant y = cc; my_variant y1 = static_cast<const convertible&>(c); my_variant y2 = static_cast<const convertible>(c); // Always fail to compile, using any flags and Boost version // my_variant z = c; // my_variant y3 = static_cast<convertible&>(c); Tested on clang, gcc4.5, gcc 4.6, gcc4.7 on Boost 1.48 and Boost 1.53 with and without C++11 flags. If we modify `convertible`: struct convertible { operator my_variant() { return my_variant(); } operator my_variant() const { return my_variant(); } }; Then ALL the tests pass well, even the commented out ones. Even `my_variant y4 = static_cast<convertible&&>(c);` works well. I'll test it tomorrow on Visual Studio 11 once more and investigate issue futher. -- Best regards, Antony Polukhin

2 2

PROBLEM WITH BOOST MPI
by siva teja 15 May '13

15 May '13

Hi friends, I have to run distributed_delta_stepping with graphs in CSR fromat. So I tried with "distributed_shortest_path_test.cpp" located at "<top of boost source tree>/libs/graph_parallel/test/ ". It reads a random graph in adjacency list format. But I have to run DIMACS_USA graph in CSR format. So, I edited the code but it resulted in following errors. I guess I am facing problems while representing a graph in CSR format.. CODE : void test_distributed_shortest_paths(int n, double p, int c, int seed) { char *graf = "./dimacs_USA_sort_wdup" ; typedef compressed_sparse_row_graph<directedS, VertexProperties, WeightedEdge, no_property, distributedS<mpi_process_group> > Graph; typedef graph_traits<Graph>::vertex_descriptor vertex_descriptor; typedef graph_traits<Graph>::edge_descriptor edge_descriptor; typedef graph_traits<Graph>::vertex_iterator vertex_iterator; typedef graph_traits<Graph>::vertices_size_type vertices_size_type; typedef property_map<Graph, vertex_index_t>::type vertex_index_map; typedef std::pair<long int, long int> Edge; struct timeval starttime,endtime,timediff; gettimeofday(&starttime,0x0); char ch; long int num_nodes; long int num_arcs; Edge *edge_array; long int *weights; long int arc_index=0, wt_index=0; long int vert1, vert2, dist; std::fstream input; input.open(graf); if( !input.eof()) { input >> ch; if(ch == 'vertex') { input >> num_nodes; } } if(!input.eof()) { input >> ch; if(ch =='edge') { input >> num_arcs; } else std::cout << "Wrong graph format\n"; } edge_array=(Edge *)malloc(sizeof(Edge) * num_arcs); weights=(long int *)malloc(sizeof(long int) * num_arcs); while(!input.eof()) { input >> ch; if(ch == 'arc') { input >> vert1 >> vert2 >> dist; Edge temp; temp.first=vert1; temp.second=vert2; edge_array[arc_index++]=temp; weights[wt_index++]=dist; } } gettimeofday(&endtime, 0x0); Graph g(edge_array, edge_array + num_arcs, weights, num_nodes); vertex_descriptor s = vertex(1, g); graph::distributed::delta_stepping_shortest_paths(g, s, dummy_property_map(), get(&VertexProperties::distance, g), get(&WeightedEdge::weight, g)); } But when I run the command " mpirun -np 1 ./a.out" I face the following errors. **** Testing aborted. **** 1 error detected ------------------------------ -------------------------------------------- mpirun has exited due to process rank 0 with PID 19096 on node rise-02 exiting without calling "finalize". This may have caused other processes in the application to be terminated by signals sent by mpirun (as reported here). -------------------------------------------------------------------------- Thank you in advance.

1 0

Statechart timer extension and new version of visualizer
by Petr Silhavik 15 May '13

15 May '13

Hi, I have implemented timers for library Boost Statechart. It was tested only on Linux because it uses POSIX threads. It is extension that replaces fifo_scheduler and fifo_worker with only one class Scheduler. That means that it is suitable only for asynchronous state machines. The source codes with example and short description can be downloaded here: https://www.dropbox.com/s/ubtep8f7hs6g5xl/boostTimers.tar.gz I would like to also inform you about new version of Boost Statechart viewer. Now it is plugin for compiler Clang and it was tested vitw version 3.0 and higher - (mostly tested using version 3.1). You can find more information and news about it on the web page of this project where you can also download the source codes and see some examples of the output pictures. http://rtime.felk.cvut.cz/statechart-viewer/ Best regards Petr

2 2

boost::variant & c++11
by Bryan Catanzaro 15 May '13

15 May '13

Hi - I seem to have run into a regression with boost::variant from 1.48 to 1.53. With C++11 enabled, using g++, the following code seems to confuse variant's constructor selection machinery: if you look at the line after #ifndef WAR, the implicit conversion isn't happening, and so boost::variant attempts to convert construct from extract<my_variant> to A or B. It should instead copy construct from the output of extract<my_variant>'s implicit conversion to my_variant. If I force an explicit conversion, this problem doesn't occur. If I don't enable c++11, this problem doesn't occur. With Boost 1.48, this problem doesn't occur. Any advice from the list as to why this behavior changed? Thanks, bryan #include <boost/variant.hpp> //A simplified wrapper template<typename T> struct wrap { T t; wrap(T& _t) : t(_t) {} operator T() const { return t; } T force() const { return t; } }; struct A{}; struct B{}; typedef boost::variant<A, B> my_variant; int main() { A a; my_variant x(a); #ifndef WAR my_variant y = wrap<my_variant>(x); #else my_variant y = wrap<my_variant>(x).force(); #endif } And the error: /home/bcatanzaro/boost_1_53_0/boost/variant/variant.hpp:1574:9: error: no matching function for call to ‘boost::variant<A, B>::initializer::initialize(void*, std::remove_reference<wrap<boost::variant<A, B> >&>::type)’

3 3

Constraining unordered template constructors
by Michael Marcin 14 May '13

14 May '13

I have a class that I want to have accept 0-5 arguments in any order for it's constructor. Each argument is a const references to one of 5 possible class types. All arguments must be distinct in type, passing the same type twice in an error. Passing an argument that is not one of the 5 supported types is also an error. How do you go about implementing something like this? I figure first you have to create a constructor for each arity but then how do you constrain the constructors as described above? I imagine it would involve enable_if. class widget { widget(); template< class A > widget( const A& a ); template< class A, class B > widget( const A& a, const B& b ); template< class A, class B, class C > widget( const A& a, const B& b, const C& c ); template< class A, class B, class C, class D > widget( const A& a, const B& b, const C& c, const D& d ); template< class A, class B, class C, class D, class E > widget( const A& a, const B& b, const C& c, const D& d, const E& e ); };

4 3

directory monitoring with dynamically added dirs
by Richard Langly 14 May '13

14 May '13

I'm trying to dynamically add directories from a config file before I enter my programs main loop using the ASIO dir monitor example. I've not been able to understand the internals yet. Below is my current iteration of what I'm trying. I have been able to get the simple examples running just like in the async.cpp file. But I can't get adding dirs dynamically working and was hoping someone could offer suggestions. Any help much appreciated. void my_file_handler(const boost::system::error_code &ec, const boost::asio::dir_monitor_event &ev) { // event stuff ... return; } boost::asio::dir_monitor *dm; boost::asio::io_service io_service; vector<boost::asio::dir_monitor *> vecdm; for (vector<Directories>::iterator it = dirs->begin(); it != dirs->end(); ++it) { dm = new boost::asio::dir_monitor(io_service); dm->add_directory((*it).abs_dir_path); vecdm.push_back(dm); } boost::thread t = boost::thread(boost::bind(&boost::asio::io_service::run, boost::ref(io_service))); while (true) { // MAIN PROGRAM LOOP for (vector<boost::asio::dir_monitor *>::iterator it = vecdm.begin(); it != vecdm.end(); ++it) { (*it)->async_monitor(my_file_handler); } io_service.reset(); // DO PROGRAM STUFF ... } ~ ~

1 0