June 2006 - Boost-users - lists.preview.boost.org

date_time microsec_clock::univeral_time() implementation
by Brian Neal 22 Jun '06

22 Jun '06

We have an application that was using our own custom functions for getting the OS time. We were basically representing time as the number of milliseconds since the UNIX epoch in an unsigned long, or as the number of nanoseconds since the epoch in a 64-bit type. We later discovered boost, and cut everything over to boost::date_time. This has worked out very, very well and has uncovered several time related bugs in our old code. However, we do have a small percentage of code that is time critical, and we have noticed that calling microsec_clock::univeral_time() is 2.6 times slower than our old function, and for this one bit of code that is significant. Looking at the implementation for UNIX like systems, I see that create_time() in microsec_time_clock.hpp is calling gettimeofday() followed by a call to gmtime() or localtime() as appropriate. I think it's this second call that is the difference between our old code and boost. In our time critical chunk of code we are just time tagging events, and we aren't concerned about timezone. I was thinking about adding a function to our code that called clock_gettime() (which is what our old code did), but then constructing a time_type out of that using the UNIX epoch for the date part and the results of clock_gettime() for the time_duration part, i.e, in pseudo code: time_type our_get_time() { clock_gettime(CLOCK_REALTIME, &timespec); time_duration td = massage(timespec); return time_type(unix_epoch, td); } This still won't be as fast as our old code, because we essentially ignored the date part of the times and simply worked with durations since the epoch. But I think the killer for us is the call to gmtime, which the above avoids. Comments? I'm curious why this approach wasn't taken in boost. Was it the timezone factor? I'm also curious why gettimeofday() was chosen over clock_gettime()? Is one more available than the other? Thanks.

2 2

Re: [Boost-users] warning C4561: '__fastcall' incompatible with the '/clr' option:, converting to '__stdcall'
by Sohail Somani 21 Jun '06

21 Jun '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Ovanes Markarian > Sent: Wednesday, June 21, 2006 6:48 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] warning C4561: '__fastcall' > incompatible with the '/clr' option:, converting to '__stdcall' > > I don not think you should mix up managed C++ with ISO C++. > Please read this critical reviews by > Danny Kalev about features of CLI and how they are compatible > with ISO C++. > > A Critical Overview of C++/CLI > > Part 1: > http://www.informit.com/guides/content.asp?g=cplusplus&seqNum=274 > > Part 2: > http://www.informit.com/guides/content.asp?g=cplusplus&seqNum=275 I had no idea... Thanks for the link

1 0

memory leak in VC6
by Elisabeth Benoit 21 Jun '06

21 Jun '06

Hello, I'm using boost::bind and boost::signal with VC6. VC6 is detecting some memory leak for a certain type of signal.connect(). I've done a little test program to check if the problem was due to some bug or special configuration in my main program, but the memory leaks are still detected in the test situation. Let me explain the structure of my test program. I have a main, and in the main, an object CCloneAppRecipe named clone. The class CCloneAppRecipe contains a member m_PtrTestMem (a shared pointer, but a raw point or an object does the same thing) of the class CTestMem. CTestMem contains 3 instances of the class CTestTab, and the class CTestTab contains a boost signal typedef boost::signal0<void> TypeSignalTest; TypeSignalTest m_SignalTest; In my main, I invoke clone.connectPtrTestMem(). clone.connectPtrTestMem() then invokes CTestMem::ConnectSignalTriplePtrTab(boost::bind(&CloneAppRecipe::test, this)); ConnectSignalTriplePtrTab() then invokes ConnectSignal(TypeSignalTest::slot_type slot) for each of the 3 instances of the class it contains. ConnectSignal then calls m_SignalTest.connect(slot). And I have a memory leak: 2 blocks of 16 bytes. If I connect only one instance of CTestTab, no leak. Two instances, one leak of 16 bytes. Three, two leaks of 16 bytes, and so on. I connect the m_SignalTest with a CCloneAppRecipe's slot. If I connect it to a CTestMem slot, I have no leak. Or if I place 3 instances of CTestTab directly in the CCloneAppRecipe class (so I don't use CTestMem anymore), I have no leak. But that's not what I need to do. I've tested plenty of variations (shared_pointer or object, etc.) Does anybody have a clue. Thanks elisabeth benoit

1 0

Re: [Boost-users] warning C4561: '__fastcall' incompatible with the '/clr' option:, converting to '__stdcall'
by Ovanes Markarian 21 Jun '06

21 Jun '06

Hello, > >When using the /clr option for generating managed code with Visual C++ >one gets the warning > >C4561: '__fastcall' incompatible with the '/clr' option:,converting to '__stdcall' > >Is there a way to compile files that include boost header files >such that __stdcall is the default, or is ignorng or switching >off the warning the "solution"? > >Thanks in advance, > >andreas I don not think you should mix up managed C++ with ISO C++. Please read this critical reviews by Danny Kalev about features of CLI and how they are compatible with ISO C++. A Critical Overview of C++/CLI Part 1: http://www.informit.com/guides/content.asp?g=cplusplus&seqNum=274 Part 2: http://www.informit.com/guides/content.asp?g=cplusplus&seqNum=275 Regards, Ovanes Markarian

1 0

strange compiler behavior...
by Ovanes Markarian 21 Jun '06

21 Jun '06

Dear Boosters, I have some strange problem and don't understand if this is really intended so by the C++ standard or is this a bug in the compiler of MS Visual C++ 2003 & 2005. I have a template class which receives an mpl::vector with possible storage types. This class contains an instance of boost::variant but is intended to wrap it with a little bit different functionality. //Solution A //A body of a _working__ solution: template<class Seq_> class my_union { public: template<class T> my_union(const T& rhs) {...} //... }; //Solution B //not working more general solution //this struct compiles fine template<T> struct const_ref : boost::add_const<typename boost::add_reference<T>::type> {}; template<class Seq_> class my_union { public: template<class T> my_union(typename const_ref<T>::type rhs) {...} //... }; now call of the class: //assume my_pair is defined earlier //assume classes A,B,C,D are defined upon typedef my_pair<0, A> t1; typedef my_pair<1, B> t2; typedef my_pair<2, C> t3; typedef my_pair<3, A> t4; typedef mpl::vector<t1,t2,t3,t4>::type types_vector; typedef my_union<types_vector> union_t; t1 t(A()); union_t u(t); //error! with Solution B, but fine with Solution A in the my_union I get an error stating: error C2664: 'my_union<Seq_>::my_union(const my_union<Seq_> &)' : cannot convert parameter 1 from 't1' to 'const my_union<Seq_> &' with [ Seq_=types_vector ] Reason: cannot convert from 't1' to 'const my_union<Seq_>' with [ Seq_=types_vector ] No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called Why does Solution A work fine and Solution B does not? Is this really intended by a standard? With Kind Regards, Ovanes Markarian

1 0

Templates in shared libraries
by Hughes, James 21 Jun '06

21 Jun '06

Hello all, I didn't get an answer to my question about a good C++/Templates mailing list (I am sure there is one, or maybe our c**p email system has culled any replies), so, since I am sure that the problem I am encountering must have been solved/avoided in the Boost libs I will ask here. I have a template, solely defined in a header file header.h. This template has a static factory function in case that's important. This is used by code in a shared library, shared.cpp->shared.so The main application (app) also uses the template. When I link my app without including the shared library, all is fine. When I start to use the shared library, I get a seg fault on app startup (doesn't even get to main()). I think this is a translation unit problem (Hope term is correct). When the shared.so is compiled, it must create instances/functions based on the template. When app is compiled, it also must be instantiating the same functions, since I don't think it can know about the ones in shared.so. So, when system attempts to start up there is some sort of conflict between the template instances in the shared lib and the ones in the app itself - there are effectively two sets of template code do the same thing with the same name. I have done some checking on the net, and have not yet seen how to avoid this problem, but of course it's quite possible I have got hold of completely the wrong end of the stick! I think I may be in the right track with -frepo and -fno-implicit-templates compiler flags, but I am not sure. Any help gratefully received. James This message (including any attachments) contains confidential and/or proprietary information intended only for the addressee. Any unauthorized disclosure, copying, distribution or reliance on the contents of this information is strictly prohibited and may constitute a violation of law. If you are not the intended recipient, please notify the sender immediately by responding to this e-mail, and delete the message from your system. If you have any questions about this e-mail please notify the sender immediately.

1 0

warning C4561: '__fastcall' incompatible with the '/clr' option:, converting to '__stdcall'
by Andreas Fabri 20 Jun '06

20 Jun '06

Hello, When using the /clr option for generating managed code with Visual C++ one gets the warning C4561: '__fastcall' incompatible with the '/clr' option:,converting to '__stdcall' Is there a way to compile files that include boost header files such that __stdcall is the default, or is ignorng or switching off the warning the "solution"? Thanks in advance, andreas

2 1

[serialization] link problem
by picca＠synchrotron-soleil.fr 20 Jun '06

20 Jun '06

Hello My post is closely related to this one. http://lists.boost.org/boost-users/2004/12/8768.php I am trying to build a library. so I put a serialize template in most of my classes. I have no problem to create the static library. Now I have a test suit for my library. if I build it with the the serialization test. diffractometer::twoC::Vertical d_ref, d; stringstream flux; diffractometer::twoC::Vertical const & work_around = d_ref; boost::archive::text_oarchive oa(flux); oa << work_around; boost::archive::text_iarchive ia(flux); ia >> d; Everything is ok but when I comment the oa << work_around and the the ia >> d The compilation fail with tone of error like this: debug/linux2/src/libhkl.a(affinement_simplex.o): In function `boost::archive::detail::load_non_pointer_type<boost::archive::text_iarchive, hkl::affinement::Simplex>::load_standard::invoke(boost::archive::text_iarchive&, hkl::affinement::Simplex&)': /usr/include/boost/archive/detail/iserializer.hpp:375: undefined reference to `boost::archive::detail::basic_iarchive::load_object(void*, boost::archive::detail::basic_iserializer const&)' debug/linux2/src/libhkl.a(affinement_simplex.o): In function `boost::archive::detail::pointer_iserializer<hkl::affinement::Simplex, boost::archive::text_iarchive>::load_object_ptr(boost::archive::detail::basic_iarchive&, void*&, unsigned int) const': /usr/include/boost/archive/detail/iserializer.hpp:309: undefined reference to `boost::archive::detail::basic_iarchive::next_object_pointer(void*)' debug/linux2/src/libhkl.a(affinement_simplex.o): In function `~void_caster_primitive': /usr/include/boost/serialization/void_cast.hpp:132: undefined reference to `boost::serialization::void_cast_detail::void_caster::~void_caster()' /usr/include/boost/serialization/void_cast.hpp:132: undefined reference to `boost::serialization::void_cast_detail::void_caster::~void_caster()' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `boost::serialization::detail::extended_type_info_typeid_1<hkl::geometry::Kappa const>::export_register(char const*)': /usr/include/boost/serialization/extended_type_info_typeid.hpp:101: undefined reference to `boost::serialization::extended_type_info::key_register(char const*)' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `pointer_oserializer': /usr/include/boost/archive/detail/oserializer.hpp:226: undefined reference to `boost::archive::detail::archive_pointer_oserializer<boost::archive::text_oarchive>::archive_pointer_oserializer(boost::serialization::extended_type_info const&)' /usr/include/boost/archive/detail/oserializer.hpp:236: undefined reference to `boost::archive::detail::archive_pointer_oserializer<boost::archive::text_oarchive>::~archive_pointer_oserializer()' /usr/include/boost/archive/detail/oserializer.hpp:226: undefined reference to `boost::archive::detail::archive_pointer_oserializer<boost::archive::text_oarchive>::archive_pointer_oserializer(boost::serialization::extended_type_info const&)' /usr/include/boost/archive/detail/oserializer.hpp:236: undefined reference to `boost::archive::detail::archive_pointer_oserializer<boost::archive::text_oarchive>::~archive_pointer_oserializer()' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `pointer_iserializer': /usr/include/boost/archive/detail/iserializer.hpp:332: undefined reference to `boost::archive::detail::archive_pointer_iserializer<boost::archive::text_iarchive>::archive_pointer_iserializer(boost::serialization::extended_type_info const&)' /usr/include/boost/archive/detail/iserializer.hpp:341: undefined reference to `boost::archive::detail::archive_pointer_iserializer<boost::archive::text_iarchive>::~archive_pointer_iserializer()' /usr/include/boost/archive/detail/iserializer.hpp:332: undefined reference to `boost::archive::detail::archive_pointer_iserializer<boost::archive::text_iarchive>::archive_pointer_iserializer(boost::serialization::extended_type_info const&)' /usr/include/boost/archive/detail/iserializer.hpp:341: undefined reference to `boost::archive::detail::archive_pointer_iserializer<boost::archive::text_iarchive>::~archive_pointer_iserializer()' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `~pointer_oserializer': /usr/include/boost/archive/detail/oserializer.hpp:184: undefined reference to `boost::archive::detail::archive_pointer_oserializer<boost::archive::text_oarchive>::~archive_pointer_oserializer()' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `~pointer_iserializer': /usr/include/boost/archive/detail/iserializer.hpp:197: undefined reference to `boost::archive::detail::archive_pointer_iserializer<boost::archive::text_iarchive>::~archive_pointer_iserializer()' /usr/include/boost/archive/detail/iserializer.hpp:197: undefined reference to `boost::archive::detail::archive_pointer_iserializer<boost::archive::text_iarchive>::~archive_pointer_iserializer()' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `~pointer_oserializer': /usr/include/boost/archive/detail/oserializer.hpp:184: undefined reference to `boost::archive::detail::archive_pointer_oserializer<boost::archive::text_oarchive>::~archive_pointer_oserializer()' debug/linux2/src/libhkl.a(geometry_kappa.o): In function `boost::archive::detail::pointer_iserializer<hkl::geometry::Kappa, boost::archive::text_iarchive>::load_object_ptr(boost::archive::detail::basic_iarchive&, void*&, unsigned int) const': /usr/include/boost/archive/detail/iserializer.hpp:309: undefined reference to `boost::archive::detail::basic_iarchive::next_object_pointer(void*)' collect2: ld returned 1 exit status So the question is how can I avoid all this ? Have a nice day.

3 4

boost::serialization unregistered_class exceptions with DLL.
by Arjuna Balasingham 20 Jun '06

20 Jun '06

Re: unregistered_class exceptions are thrown when the derived class are in an explicitly loaded DLL. Hello, In a nutshell, the problem that I have is that an "unregistered_class" exception is thrown when the derived class is defined/declared in a DLL that is explicitly loaded. The problem as far as I can tell is that the derived class in dll and the derived class in the framework are being "registered" in two different instances (m_self->m_map.insert(eti); in "extended_type_info.cpp"). Exception is thrown when trying to serialize the foobar_derived_dll. It looks for "foobar_derived_dll" in the wrong instance of the m_map defined in the extened_type_info.cpp and then throws an exception. Can anybody please help The Application Structure is more or less as follows. Common Framework Library Contains. class foobar_base : public foobar_base { protected: bool m_attr0; public: foobar_base() {} template<class Archive> void serialize(Archive & ar, const unsigned int /* file_version */){ std::cout << "serialize foobar_base\n"; ar & BOOST_SERIALIZATION_NVP(m_attr0); } } class foobar_derived : public foobar_base { protected: bool m_attr1; public: foobar_derived() {} template<class Archive> void serialize(Archive & ar, const unsigned int /* file_version */){ std::cout << "serialize foobar_derived_dll\n"; ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(foobar_base); ar & BOOST_SERIALIZATION_NVP(m_attr1); } } BOOST_CLASS_EXPORT_GUID(foobar_derived, "foobar_derived") Dynamic Link Library Contains (Includes Common Framework Library) class foobar_derived_dll { protected: bool m_attr2; public: foobar_derived_dll() {} template<class Archive> void serialize(Archive & ar, const unsigned int /* file_version */){ std::cout << "serialize foobar_derived_dll\n"; ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(foobar_base); ar & BOOST_SERIALIZATION_NVP(m_attr2); } } BOOST_CLASS_EXPORT_GUID(foobar_derived_dll, "foobar_derived_dll") Main Application (includes Common framework library & explictly lods Dynamic Link Library) class foo { public: std::list<boost::shared_ptr<foobar_base> > m_foovec; public: foo ( ) { }; ~foo ( ) { }; template<class Archive> void serialize(Archive &ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP(m_foovec); } }; (..Include Framwork..) int main() { // load the DLL.. LoadLibrary(... DLL)" foo f; f.m_foovec.push_back( boost::shared_ptr<foobar_base>(new foobar_derived(.... f.m_foovec.push_back( boost::shared_ptr<foobar_base>(new foobar_derived_dll(.... std::string filename("c:\\tmp.txt"); try { // write to file std::ofstream ofs(filename.c_str()); assert(ofs.good()); boost::archive::text_oarchive oa(ofs); oa << BOOST_SERIALIZATION_NVP(f); ofs.close(); } catch (std::exception &e) { std::cout << e.what() << std::endl; } }

5 9

Boost.Thread locking across two functions
by Peteris Krumins [Newsgroups] 20 Jun '06

20 Jun '06

Hello! I am using a library which itself internally does not use any thread sync mechanisms but it provides callback functions lib_lock() and lib_unlock() which I can specify if the lib is used in multithread environment. lib_lock() would get called before critical segment and after it lib_unlock() would get called. I read Boost.Thread documentation and I understand that lock concept is not thread-safe itself and should be used only at block scope. This makes me think that I have no way to use locking provided by Boost to acquire a lock in lib_lock() function and release it in lib_unlock()? But I am not sure. Do you have any suggestions if this can be done using Boost? Thanks! P.Krumins

3 2