June 2012 - Boost-users - lists.preview.boost.org

mapped_region locks on multithread
by Mikhail Eremin 09 Jun '12

09 Jun '12

Hello, SETTING: - There is an application, written using Boost Template library, meant for QUICK processing of bulk text files (cca 50-100Gb each). - There is a huge, quick and expensive piece of hardware with HUGE amount of RAM and multiple CPU. - There is [theoretically] any possible UNIX-like OS, even Microsoft Windows (R) is considered. - Boost Thread Pool extension is used; previously memory mapped files through memory_segment have been used, now got rid of the entire Boost::interprocess. - There are NO explicit data items in the application's algorithm to be shared by threads, each has its own piece of input file, thus - there is NO explicit concurrency. PROBLEM: - Ensure fast processing without locks and threads sleeping. Currently the threads sleep on some internal mutex. We thought it's been boost::interprocess (specifically - mmap, wrapped by a mutex), but it apparently isn't so. SPECIFIC QUESTION: - How could we get rid of Boost locks? Mike

2 1

[Intrusive][Interprocess] Using Intrusive Containers in Managed Shared Memory
by Hering Cheng 09 Jun '12

09 Jun '12

Hi, Is it possible to use Boost.Intrusive containers in managed shared memory? If so, are there examples that I can learn from? Thanks.

2 1

define multi dimension array at run time
by Moataz Elmasry 08 Jun '12

08 Jun '12

Hello list I'm writing an n-dimensional histogram class and would like to use the boost multidimension array to store the elements of the histogram So the array is defined as a class member but is first initialized through the constructor, but according to boost example page its something like: typedef boost::multi_array<double, 3> array_type; but I'd like: typedef boost::multi_array<std::vector<MyObject>, n> array_type; where n is given through the constructor Is this somehow possible? Regards Moataz

3 2

C++Now! keynote videos are available
by Marshall Clow 08 Jun '12

08 Jun '12

Howard's was not recorded Daveed: http://www.youtube.com/watch?v=8SOCYQ033K8 Sean: http://www.youtube.com/watch?v=iGenpw2NeKQ The full list of videos is at: http://www.youtube.com/user/BoostCon (more will be added soon) -- Marshall Marshall Clow Idio Software <mailto:mclow.lists@gmail.com> A.D. 1517: Martin Luther nails his 95 Theses to the church door and is promptly moderated down to (-1, Flamebait). -- Yu Suzuki

1 0

Building PDF with Open Source Solution
by Christian Henning 08 Jun '12

08 Jun '12

Hi there, This is slightly off topic. I have a question how pdf's are build within the boost build system. As far as I could figure it out the build system uses a proprietary solution called RenderX XEP or so. I'm wondering if there are open source solutions out there that could be used to generate boost's documentation as pdf files? Thanks, Christian

8 13

[serialization] custom allocation/deletion
by Brian Budge 07 Jun '12

07 Jun '12

Hi all - Is there a way to use custom allocation while loading from an archive? I'm thinking like passing an allocator along with the archive during deserialization? Adding to this, is it possible to add a custom deleter to a shared_ptr during deserialization? I am thinking of using this with boost mpi, and if I send a shared_ptr over the wire with some kind of deleter, the same deleter object can't be used on the other side since it's in a different address space. Thanks, Brian

2 5

signal_set_service.ipp(247) error SIG_ERR undeclared
by Christopher Pisz 07 Jun '12

07 Jun '12

signal_set_service.ipp(247) error SIG_ERR undeclared I am getting this error when trying to compile a precompiled header that includes boost\asio.hpp I am using boost version 1.49.0 This did not occur with 1.40.0 Any ideas?

1 0

[pool] Questions about memory pool
by Junchao Zhang 07 Jun '12

07 Jun '12

Hello, I'm new to boost memory pool. I have two questions: 1) My program has multiple iterations. In each iteration, I need to allocate chunks(objects) from a pool. I don't need to allocate array of chunks. I don't free objects until the end of an iteration. Then what is the most efficient way to do that? I wrote the following code, but I'm not sure it is right. boost::pool<> p(sizeof(MyType)); for (...) { ... // many calls to p.malloc(); ... p.release_memory(); // Is it OK??? } Also, I was wondering, to improve locality, how to reuse the memory in the next iteration, instead of deallocating/allocating memory again and again. 2) How to inquiry a pool to know have many objects have been allocated from the pool? For example, if malloc() 2 times, free() 1 time, then the count is 1. Does boost pool have this interface? Thanks! -- Junchao Zhang

1 0

[MSM] cannot use state machine level internal_transition_table within the state machine which has Completion (anonymous) transitions in its transition_table
by James Zhang 06 Jun '12

06 Jun '12

Hi, I'm trying to setup a MSM state machine, according to : https://groups.google.com/group/boost-list/browse_thread/thread/bd9854b58a3… I can setup a state machine level internal_transition_table to handle the state-independent events. but the build failed if the transition_table has Completion (anonymous) transitions. I did following test: 1. if I remove the internal_transition_table, the build succeed. 2. if I give an event to the Rows, then the build succeed tool Row < Initialized , none , Synchronized >, Row < Synchronized , none , WaitResponse , sendPoll >, could anybody give me some clues about this? thanks. following is the code: #include <iostream> // back-end #include <boost/msm/back/state_machine.hpp> //front-end #include <boost/msm/front/state_machine_def.hpp> #include <boost/msm/front/functor_row.hpp> //#include <boost/msm/front/euml/common.hpp> // for And_ operator //#include <boost/msm/front/euml/operator.hpp> using namespace std; namespace msm = boost::msm; using namespace msm::front; namespace mpl = boost::mpl; // for And_ operator //using namespace msm::front::euml; namespace { // events struct evt_empty {}; struct evt_response{}; // front-end: define the FSM structure struct mystatemachine_ : public msm::front::state_machine_def<mystatemachine_> { // The list of FSM states struct Initialized : public msm::front::state<> { // every (optional) entry/exit methods get the event passed. template <class Event,class FSM> void on_entry(Event const&,FSM& ) { std::cout << "entering: Initialized" << std::endl; } template <class Event,class FSM> void on_exit(Event const&,FSM& ) {std::cout << "leaving: Initialized" << std::endl;} }; struct Synchronized : public msm::front::state<> { // every (optional) entry/exit methods get the event passed. template <class Event,class FSM> void on_entry(Event const&,FSM& ) { std::cout << "entering: Synchronized" << std::endl; } template <class Event,class FSM> void on_exit(Event const&,FSM& ) {std::cout << "leaving: Synchronized" << std::endl;} }; struct WaitResponse : public msm::front::state<> { // every (optional) entry/exit methods get the event passed. template <class Event,class FSM> void on_entry(Event const&,FSM& ) { std::cout << "entering: WaitBSPResponse" << std::endl; } template <class Event,class FSM> void on_exit(Event const&,FSM& ) {std::cout << "leaving: WaitBSPResponse" << std::endl;} }; // the initial state of the sas SM. Must be defined typedef Initialized initial_state; // actions struct sendPoll { template <class EVT,class FSM,class SourceState,class TargetState> void operator()(EVT const& ,FSM& ,SourceState& ,TargetState& ) { cout << "send sync poll" << endl; } }; // Transition table for sas struct transition_table : mpl::vector< // Start Event Next Action Guard // +----------------------+--------------------+----------------------+---------------+--------------------+ Row < Initialized , none , Synchronized >, Row < Synchronized , none , WaitResponse , sendPoll >, Row < WaitResponse , evt_response , WaitResponse , sendPoll , none > // +----------------------+--------------------+----------------------+---------------+--------------------+ > {}; struct internal_guard { template <class EVT,class FSM,class SourceState,class TargetState> bool operator()(EVT const& evt ,FSM&fsm,SourceState& ,TargetState& ) { std::cout << "Empty::internal_transition_table guard " <<std::endl; return true; } }; struct internal_action { template <class EVT,class FSM,class SourceState,class TargetState> void operator()(EVT const& ,FSM& fsm,SourceState& ,TargetState& ) { std::cout << "Empty::internal_transition_table action\n" <<std::endl; } }; struct internal_transition_table : mpl::vector< // Start Event Next Action Guard Internal < evt_empty , internal_action ,internal_guard > // +---------+-------------+---------+---------------------+----------------------+ > {}; // Replaces the default no-transition response. template <class FSM,class Event> void no_transition(Event const& e, FSM&,int state) { std::cout << "no transition from state " << state << " on event " << typeid(e).name() << std::endl; } }; // Pick a back-end typedef msm::back::state_machine<mystatemachine_> MyStateMachine; // // Testing utilities. // static char const* const state_names[] = { "initialized", "synchronized", "waitBSPResponse" }; void pstate(MyStateMachine const& p) { std::cout << " -> " << state_names[p.current_state()[0]] << std::endl; } void test() { MyStateMachine sm; sm.start(); sm.process_event(evt_empty()); pstate(sm); } }

3 2

1.50.0 beta 1 Beta release available
by Beman Dawes 06 Jun '12

06 Jun '12

Boost release 1.50.0 beta 1 is now available from SourceForge See http://sourceforge.net/projects/boost/files/ This release contains 4 new libraries (Algorithm, Functional/OverloadedFunction, LocalFunction, Utility/IdentityType) and numerous enhancements and bug fixes for existing libraries. For details of what's in the release, see http://beta.boost.org/users/history/version_1_50_0.html. Note that the links to files on this web page are for the final release - use the SourceForge link above to get the beta files. Please download the beta, give it a try, and report any problems you encounter. Thanks, -- The Boost Release Team

2 1

(no subject)
by Avishay Ben Shabtai 05 Jun '12

05 Jun '12

> > > Hi > > > > I am having compilation error when calling copy_graph. > > The compilation error is regarding vertex_all for default copy of > vertices. > > Where can I find info regarding this property and how to resolve the > issue? > > What graph type are you using? If it is custom, you might have an easier > time passing in your own vertex and edge copying functions, which avoids > the need to have vertex_all or edge_all properties. See > <URL:http://www.boost.org/doc/libs/1_49_0/libs/graph/doc/copy_graph.html> > for information on those parameters. > > -- Jeremiah Willcock I don't understand what is custom graph but my graph definition is: typedef boost::adjacency_list<boost::listS //container class of the edges. ,boost::vecS ,boost::bidirectionalS ,MCVertexProp ,boost::property<boost::edge_weight_t,MCEdgeWeight>s ,MCGraphProp> MCBaseGraph; do i need to declare vertex_all as property and if so what is its type?

2 11

[Boost.Random] Discrete distribution behavior
by sguazt 05 Jun '12

05 Jun '12

Dear Booster, I have a problem in understanding how the boost::random::discrete_distribution works. As far as I know, given the following events: x1, x2 and x2 and weights: 1, 4, and 5, * the associated PDF of the discrete distribution would be: \Pr{X=x1}=1/10 \Pr{X=x2}=4/10 \Pr{X=x3}=5/10 * and the CDF would be: \Pr{X<=x1}=1/10 \Pr{X<=x2}=5/10 \Pr{X<=x3}=1 So, if my random generator generates the value 0.814724, the discrete random variate would return the event x3 (if I'm not wrong). Instead, it seems that boost::random::discrete_distribution does not work in this way. To test it I have implemented a small program (see attachment). Moreover, I have also implemented a naive version of discrete distribution (called my::discrete_distribution), which simply draws a uniform [0,1] random number and select the first event whose CDF is greater than or equal to the drawn value. I have also added some debug prints in other boost classes (just to get the value of the generated random numbers). I run it in Linux with Boost devel version + GCC 4.6.3 (with flags: -Wall -ansi -pedantic) Here's the output: Testing Boost... [boost::random::mersenne_twister] 3499211612 [boost::random::mersenne_twister] 581869302 [boost::random::discrete_distribution] test=0.814724 - result=0 Random Event: 1 [boost::random::mersenne_twister] 3890346734 [boost::random::mersenne_twister] 3586334585 [boost::random::discrete_distribution] test=0.905792 - result=2 Random Event: 2 [boost::random::mersenne_twister] 545404204 [boost::random::mersenne_twister] 4161255391 [boost::random::discrete_distribution] test=0.126987 - result=2 Random Event: 2 Testing My own version... [boost::random::mersenne_twister] 3499211612 [my::discrete_distribution::rand] RNG: 0.814724 Random Event: 2 [boost::random::mersenne_twister] 581869302 [my::discrete_distribution::rand] RNG: 0.135477 Random Event: 1 [boost::random::mersenne_twister] 3890346734 [my::discrete_distribution::rand] RNG: 0.905792 Random Event: 2 As you can see, the sequence of Random Event is different. Looking at the output of my::distribution, it seems it is working OK. Can you help me in understanding what is going on? Thank you very much for the help!! Best, -- Marco

3 4

accumulator operations
by Olaf Peter 04 Jun '12

04 Jun '12

Hi, I like to collects some stats. There are several data slices where I can partially/separated collect them (each of them). Finaly I like to combine these stats to a single one to get summarized informations. How to? May this work: accumulator_set<..> acc1; accumulator_set<..> acc2; ... accumulator_set<..> acc_all; ... acc1(42); ... acc_all += acc1; ... acc_all += acc2; Because did not found informations about. The same problem woulde be probably for a threaded implementation too. Thanks, Olaf

4 4

[uBlas] Linkage problems in assignment.hpp
by Jérémy Coulon 04 Jun '12

04 Jun '12

Hi, I upgraded to Boost v1.49 recently. I am affected by the following bug : https://svn.boost.org/trac/boost/ticket/6010 According to trac, the bug should be fixed in 1.48, right ? Then why is it still there in 1.49 ? Jeremy

1 1

[BGL] copy_graph compilation issue
by Avishay Ben Shabtai 04 Jun '12

04 Jun '12

Hi I am having compilation error when calling copy_graph. The compilation error is regarding vertex_all for default copy of vertices. Where can I find info regarding this property and how to resolve the issue? Sent from my iPad

2 1

[multi_array] operator= requires resize - why?
by Markus Werle 04 Jun '12

04 Jun '12

Hi! IMHO the behaviour of operator= for multi_array is a little bit broken. A lot of preconditions required: template <typename ConstMultiArray> multi_array_ref& operator=(const ConstMultiArray& other) { [...] // make sure the dimensions agree assert(other.num_dimensions() == this->num_dimensions()); assert(std::equal(other.shape(),other.shape()+this->num_dimensions(), this->shape())); Besides the fact that I get mad about asserts instead of exceptions (please change this to a way users may choose) why do we have such an odd behaviour here? >From page 1, line 3 of the docs I cite: "The interface design is in line with the precedent set by the C++ Standard Library containers" - this is not true for operator=, since std::<container>::operator= works fine for me all the time _without_ any resize. Could we please change the behaviour of multi_array::operator= to take care of resize where in need to? IMHO operator= should result in a true copy ignoring previous state. Markus

6 18

Boost logger parameters.
by Mangesh Sawant 04 Jun '12

04 Jun '12

When I use boost for asynchrous logging, application crashes due to low memory. It seems internal queue maintained by boost during asynch logging is causing this problem. Is there any way to limit size of internal queue, during asynch logging. If I switch to synchronous logging there is no crash. Thanks, Mangesh Sawant.

1 0

Using boost for asynchronous logging.
by Mangesh Sawant 04 Jun '12

04 Jun '12

Hi , Using boost logger for asynchronous logging. Application crashes with bas_alloc exception. When I switch to synchronous logging mode there is no crash. It seems in asynchronous mode log queue size keeps growing leading to low memory. Is there any way(i.e a parameter in setting file) to control max log queue size in asynchronous logging. Thanks, Mangesh Sawant.

1 0

[boost] [review] Multiprecision review scheduled for June 8th - 17th, 2012
by Jeffrey Lee Hellrung, Jr. 04 Jun '12

04 Jun '12

Hi all, The review of the proposed Boost.Multiprecision library authored by John Maddock and Christopher Kormanyos has been scheduled for June 8th - June 17th, 2012 and will be managed by myself. >From the Introduction: -------- "The Multiprecision Library provides *User-defined* integer, rational and floating-point C++ types which try to emulate as closely as practicable the C++ built-in types, but provide for more range and precision. Depending upon the number type, precision may be arbitrarily large (limited only by available memory), fixed at compile time values, for example 50 decimal digits, or a variable controlled at run-time by member functions. The types are expression-template-enabled for better performance than naive user-defined types." -------- And from the original formal review request from John: -------- Features: * Expression template enabled front end. * Support for Integer, Rational and Floating Point types. Supported Integer backends: * GMP. * Libtommath. * cpp_int. cpp_int is an all C++ Boost licensed backend, supports both arbitrary precision types (with Allocator support), and signed and unsigned fixed precision types (with no memory allocation). There are also some integer specific functions - for Miller Rabin testing, bit fiddling, random numbers. Plus interoperability with Boost.Rational (though that loses the expression template frontend). Supported Rational Backends: * GMP * libtommath * cpp_int (as above) Supported Floating point backends: * GMP * MPFR * cpp_dec_float cpp_dec_float is an all C++ Boost licensed type, adapted from Christopher Kormanyos' e_float code (published in TOMS last year). All the floating point types, have full std lib support (cos sin exp, pow etc), as well as full interoperability with Boost.Math. There's nothing in principal to prevent extension to complex numbers and interval arithmetic types (plus any other number types I've forgotten!), but I've run out of energy for now ;-) Code is in the sandbox under /big_number/. Docs can be viewed online here: http://svn.boost.org/svn/boost/sandbox/big_number/libs/multiprecision/doc/h… -------- I hope everyone interested can reserve some time to read through the documentation, try the code out, and post a formal review, either during the formal review window or before. I expect to conduct the review in the "traditional" manner, i.e., entirely within the regular boost developers' mailing list (boost(a)lists.boost.org) I will send a reminder of review process details closer to the review window. Thanks! - Jeff

3 4

Boost threads with arguments
by Jamie Lahowetz 04 Jun '12

04 Jun '12

I am trying to pass two arguments to a boost thread using this call: boost::thread workerthread(&SshClient::SSHconnect,&sshclt,1,provec); the method for SSHconnect is: int SSHconnect(int,vector<string>); When I try to make my program I get the errors: /usr/include/boost/bind/bind.hpp: In member function ‘void boost::_bi::list2<A1, A2>::operator()(boost::_bi::type<void>, F&, A&, int) [with F = int (SshClient::*)(int, std::vector<std::basic_string<char> >), A = boost::_bi::list0, A1 = boost::_bi::value<SshClient*>, A2 = boost::_bi::value<int>]’: /usr/include/boost/bind/bind_template.hpp:20:59: instantiated from ‘boost::_bi::bind_t<R, F, L>::result_type boost::_bi::bind_t<R, F, L>::operator()() [with R = void, F = int (SshClient::*)(int, std::vector<std::basic_string<char> >), L = boost::_bi::list2<boost::_bi::value<SshClient*>, boost::_bi::value<int> >, boost::_bi::bind_t<R, F, L>::result_type = void]’ /usr/include/boost/thread/detail/thread.hpp:61:17: instantiated from ‘void boost::detail::thread_data<F>::run() [with F = boost::_bi::bind_t<void, int (SshClient::*)(int, std::vector<std::basic_string<char> >), boost::_bi::list2<boost::_bi::value<SshClient*>, boost::_bi::value<int> > >]’ terminal.cpp:271:1: instantiated from here /usr/include/boost/bind/bind.hpp:313:9: error: no match for call to ‘(boost::_mfi::dm<int(int, std::vector<std::basic_string<char> >), SshClient>) (SshClient*&, int&)’ /usr/include/boost/bind/mem_fn.hpp:314:34: note: candidates are: /usr/include/boost/bind/mem_fn.hpp:340:9: note: R& boost::_mfi::dm<R, T>::operator()(T*) const [with R = int(int, std::vector<std::basic_string<char> >), T = SshClient] /usr/include/boost/bind/mem_fn.hpp:340:9: note: candidate expects 1 argument, 2 provided /usr/include/boost/bind/mem_fn.hpp:345:15: note: const R& boost::_mfi::dm<R, T>::operator()(const T*) const [with R = int(int, std::vector<std::basic_string<char> >), T = SshClient] /usr/include/boost/bind/mem_fn.hpp:345:15: note: candidate expects 1 argument, 2 provided /usr/include/boost/bind/mem_fn.hpp:350:33: note: template<class U> const R& boost::_mfi::dm::operator()(const U&) const [with U = U, R = int(int, std::vector<std::basic_string<char> >), T = SshClient] /usr/include/boost/bind/mem_fn.hpp:357:9: note: R& boost::_mfi::dm<R, T>::operator()(T&) const [with R = int(int, std::vector<std::basic_string<char> >), T = SshClient] /usr/include/boost/bind/mem_fn.hpp:357:9: note: candidate expects 1 argument, 2 provided /usr/include/boost/bind/mem_fn.hpp:362:15: note: const R& boost::_mfi::dm<R, T>::operator()(const T&) const [with R = int(int, std::vector<std::basic_string<char> >), T = SshClient] /usr/include/boost/bind/mem_fn.hpp:362:15: note: candidate expects 1 argument, 2 provided make[2]: *** [terminal.o] Error 1 make[2]: Leaving directory `/home/deadpickle/hprcc_terminal/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/deadpickle/hprcc_terminal' make: *** [all] Error 2 Quite a long error. Not sure what to make of it and could use some help. thank you. -- Jamie Ryan Lahowetz University of Nebraska - Lincoln Graduate Student - Earth and Atmospheric Sciences 402.304.0766 jlahowetz(a)gmail.com

2 1

[mpi] multithreaded/funneled messaging
by Brian Budge 03 Jun '12

03 Jun '12

Hi all - Is boost.mpi able to work with MPI_THREAD_MULTIPLE? If so, how is it specified? Thanks, Brian

2 1

[GIL] color_converted_view usage problem
by Marc-Antoine ARNAUD 02 Jun '12

02 Jun '12

Hi guys, I currently working on TuttleOFX projet, and using Boost GIL to perform some image process. I wan't using color_converted_view to process clamping values on a view, witth a specific ColorConvert (CC) functor. But with the actual code [ https://github.com/tuttleofx/TuttleOFX/blob/master/libraries/terry/src/terr…] , the pixel_clamp contructor was corretly called, but our functor (void operator()( const SPixel& src, DPixel& dst ) const) wasn't. Does someone have the same bug ? Are they more documentation anywhere on this function ( I'm refering on this link: http://stlab.adobe.com/gil/html/group___image_view_transformations_color_co… ) Thanks, -- Marc-Antoine

2 3

Add a compile flag to bjam?
by Neal Becker 01 Jun '12

01 Jun '12

Using bjam, is there an easy way (preferably on the command line) to add a flag or 2 to the compile (using gcc)? I'd like to add an optimization or 2.

2 1

Re: [Boost-users] problem with boost compilation
by Martin Elzen 01 Jun '12

01 Jun '12

Ohad Barta <sohadb1357(a)gmail.com>wrote: > I installed and compiled boost for my project a long time ago, however now > I fail to repeat on this success on another computer.... > what is the corresponding cmd compilation command? I tried to specify all > the corresponding flags but for some reason it doesn't work,(for example- > despite using the flag --address-model=64, the folder "address-model-64" > wasn't constructed in my folders hierarchy tree, so I guess that this flag > was ignored) There may be other errors, but what jumps out at me right now is that the "address-model=" part does NOT require two minus-signs in front of it...

3 4

[Asio] Compiling DLL project with stdcall calling convention
by Da-Huntha 01 Jun '12

01 Jun '12

I'm using Boost.Asio in a windows DLL project and had to change the calling convention to stdcall, now when I try to build I get the following build errors: 1>C:\boost_1_49_0\boost/asio/ detail/impl/signal_set_service.ipp(74): error C2664: 'signal' : cannot convert parameter 2 from 'void (__stdcall *)(int)' to 'void (__cdecl *)(int)' 1> None of the functions with this name in scope match the target type 1>C:\boost_1_49_0\boost/asio/detail/impl/signal_set_service.ipp(247): error C2664: 'signal' : cannot convert parameter 2 from 'void (__stdcall *)(int)' to 'void (__cdecl *)(int)' 1> None of the functions with this name in scope match the target type I'm using MSVC10. Can anyone help me with these errors? I have no idea what to do with them.

2 2