December 2015 - Boost-users - lists.preview.boost.org

how to help an object to delete itself the boost way
by Ireneusz Szcześniak 18 Dec '15

18 Dec '15

Hi, In my code there are objects created, which delete themselves. They can do it in the constructor, or later in some function. This is ugly, and I'm getting some wierd behaviour, probably because of this bad design. Is there a good Boost way of resolving my problem? Thanks, Irek

6 10

Negative effect of expected-to-fail tests in compiler development projects.
by Sergey Sprogis 18 Dec '15

18 Dec '15

I’m wondering how difficult it would be to add into bjam one option, something like: -negative_tests=off which will not launch any of negative (expected to fail) tests. For people who use Boost to test newly developed compilers negative tests are quite a nuisance. I mean those tests which got “compile_fail” and “run_fail” status inside final *.xml file generated at the end of Boost regression testing. [ they are also marked similarly inside Jamfiles ] That’s because newly developed compilers in their early stages of implementation normally have bugs which produce a lot o false error messages for correct Boost code. Important task here is to extract those messages, to evaluate them, and to prove that they are indeed correct, but compiler is wrong. And when hundreds of such false error messages are mixing together with thousands of legitimate error messages produced by negative tests (there are > 700 of them in boost.1.59.0, for example) it becomes a non-trivial task to filter them out. So the natural desire is not to launch such tests at all. Another, slightly unpleasant effect of negative tests is that they appear during so called “test pass rate” calculations during compiler testing process. Typically managers responsible for compiler development want to know the progress in terms of what “test pass rate” produces new compiler for the whole Boost testing or for the testing of some specific libraries. Normally such pass rate is calculated as the ratio between number of passed tests and the total number of tests. But ideally, tests in those calculations should be correct, so if they fail, it 100% means compiler bugs. And here again, negative tests are not useful, and should be avoided to make calculations more accurate. On a side note, I think it could be also useful to add such total testing pass rates into the Boost Regression Dashboard, so the quality of every tested compiler could be easy to seen. Much more people could be interested in looking at such dashboard, I guess.

5 6

boost::mutex::scoped_lock crashes on OSX
by David Medine 17 Dec '15

17 Dec '15

I am using boost in a python extension and it is having some problems on OSX. Specifically, I use a mutex of type boost::mutex in both the constructor and destructor methods. The constructor goes fine and the destructor crashes. The code works perfectly well on both Windows and Linux. When my class is building itself, it calls functions like this one a bunch of times: void enter_headers_phase(bool phase_locked) { if (phase_locked) { boost::mutex::scoped_lock lock(phase_mut_); headers_to_finish_++; } } where phase_mut_ is of type boost::mutex. When debugging in XCode, I see nice looking members for phase_mut_ at this point in the program: phase_mut_ (boost::mutex) m (pthread_mutex_t) __sig (long) 1297437784 1297437784 __opaque char [56] "" Later, when I am destroying, I make a nearly identical call as before: ... { boost::mutex::scoped_lock lock(phase_mut_); shutdown_ = true; } ... but now, my phase_mut_ looks like it has an apparently empty _sig member: phase_mut_ (boost::mutex) m (pthread_mutex_t) __sig (long) __opaque (char [56]) "" When we go down the line of the mutex code in boost and finally arrive at line 62 in boost/thread/pthread/mutex.hpp: ret = ::pthread_mutex_lock(m); I get a segmentation fault: EXC_BAD_ACCESS It seems like this is becausem->_sig is now empty for some reason, but I can not determine why (XCode's debugger won't let me 'watch' this variable) but it should certainly be the case that it doesn't get touched between the apparently successful constructor routines and now. This all happens on OSX Yosemite. I am compiling my extension with LLVM6.1 and I have gnu++0x as the C++ language dialect and libc++ as the stdlib. I am linking statically to boost 1.57 which I built using this command: sudo ./b2 -a cxxflags="-stdlib=libc++" linkflags="-stdlib=libc++" toolset=clang link=static link=shared threading=multi install --prefix=/opt/local --layout-tagged THX! -David

2 4

HP fortify Scan Issues
by Ganesh 17 Dec '15

17 Dec '15

I have run HP fortify scan on our code which uses Boos libraries for checking security vulnerabilities. During the scan I got issues reported in below two files. 1. compressed_pair.hpp2. next_prior.hpp The issue reported as below.1. compressed_pair.hpp (In line 154) LINE 153: compressed_pair_imp(first_param_type x, second_param_type y)LINE 154: : first_type(x), second_(y) {} Abstract: The program reads data from just outside the bounds of allocated memory. Explanation: Buffer overflow is probably the best known form of software security vulnerability. Most software developers know what a buffer overflow vulnerability is, but buffer overflow attacks against both legacy and newly-developed applications are still quite common. Part of the problem is due to the wide variety of ways buffer overflows can occur, and part is due to the error-prone techniques often used to prevent them. In a classic buffer overflow exploit, the attacker sends data to a program, which it stores in an undersized stack buffer. The result is that information on the call stack is overwritten, including the function's return pointer. The data sets the value of the return pointer so that when the function returns, it transfers control to malicious code contained in the attacker's data. 2. next_prior.hpp (In line 30) LINE 29: template <class T>LINE 30 inline T next(T x) { return ++x; } Abstract: The function next() in next_prior.hpp reads data from just outside the bounds of ~tp~1 on line 30. Explanation: Buffer overflow is probably the best known form of software security vulnerability. Most software developers know what a buffer overflow vulnerability is, but buffer overflow attacks against both legacy and newly-developed applications are still quite common. Part of the problem is due to the wide variety of ways buffer overflows can occur, and part is due to the error-prone techniques often used to prevent them. In a classic buffer overflow exploit, the attacker sends data to a program, which it stores in an undersized stack buffer. The result is that information on the call stack is overwritten, including the function's return pointer. The data sets the value of the return pointer so that when the function returns, it transfers control to malicious code contained in the attacker's data. Question is: Is this really an issue with the code? or boost has taken care of this out of bound in its code?

2 1

Size(Count of elements) in boost::multi_index_container?
by Amit Prakash Ambasta 17 Dec '15

17 Dec '15

Hi, How do I find out the size or count of elements added to a multi_index_container?

2 1

Re: [Boost-users] [Boost-announce] Boost 1.60.0 rc1 is now available
by Jan Mikkelsen 17 Dec '15

17 Dec '15

Hi, (Resending to avoid bouncing on the lists.) I can build the libraries on FreeBSD 10.2 with clang 3.7 and gcc 5.2 in various modes. However, with both clang 3.7 and gcc 5.2 the following program fails to compile: #define BOOST_MOVE_USE_STANDARD_LIBRARY_MOVE #include <boost/thread.hpp> int main() { } This works fine with the same compilers and boost 1.59. The difference seems to be caused a change to boost/thread/detail/thread_group.hpp to use boost::csbl::unique_ptr<> instead of std::auto_ptr<>. Diff showing the change between 1.59 and 1.60 in thread_group.hpp and the compiler error messages below. Regards, Jan Mikkelsen $ diff -u thread/detail/thread_group.hpp ../../boost-1_60/boost/thread/detail/thread_group.hpp --- thread/detail/thread_group.hpp 2015-12-17 01:08:28.000000000 +1100 +++ ../../boost-1_60/boost/thread/detail/thread_group.hpp 2015-12-17 01:20:56.000000000 +1100 @@ -6,6 +6,7 @@ // (C) Copyright 2007-9 Anthony Williams #include <list> +#include <boost/thread/csbl/memory/unique_ptr.hpp> #include <boost/thread/shared_mutex.hpp> #include <boost/thread/mutex.hpp> #include <boost/thread/lock_guard.hpp> @@ -75,7 +76,7 @@ thread* create_thread(F threadfunc) { boost::lock_guard<shared_mutex> guard(m); - std::auto_ptr<thread> new_thread(new thread(threadfunc)); + boost::csbl::unique_ptr<thread> new_thread(new thread(threadfunc)); threads.push_back(new_thread.get()); return new_thread.release(); } $ g++5 -std=c++14 -isystem /usr/local/include/boost-1_60 twio/test_thread.cpp In file included from /usr/local/include/boost-1_60/boost/thread/csbl/memory/unique_ptr.hpp:16:0, from /usr/local/include/boost-1_60/boost/thread/detail/thread_group.hpp:9, from /usr/local/include/boost-1_60/boost/thread/thread.hpp:13, from /usr/local/include/boost-1_60/boost/thread.hpp:13, from twio/test_thread.cpp:2: /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp: In constructor 'boost::movelib::unique_ptr<T, D>::unique_ptr(boost::movelib::unique_ptr<T, D>&&)': /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:530:29: error: 'move_if_not_lvalue_reference' is not a member of 'boost' : m_data(u.release(), ::boost::move_if_not_lvalue_reference<D>(u.get_deleter())) ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:530:68: error: expected primary-expression before '>' token : m_data(u.release(), ::boost::move_if_not_lvalue_reference<D>(u.get_deleter())) ^ In file included from /usr/local/include/boost-1_60/boost/thread/csbl/memory/unique_ptr.hpp:16:0, from /usr/local/include/boost-1_60/boost/thread/detail/thread_group.hpp:9, from /usr/local/include/boost-1_60/boost/thread/thread.hpp:13, from /usr/local/include/boost-1_60/boost/thread.hpp:13, from twio/test_thread.cpp:2: /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp: In constructor 'boost::movelib::unique_ptr<T, D>::unique_ptr(boost::movelib::unique_ptr<Y, D>&&, typename boost::move_upd::enable_up_moveconv_constr<T, D, U, E>::type*)': /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:552:29: error: 'move_if_not_lvalue_reference' is not a member of 'boost' : m_data(u.release(), ::boost::move_if_not_lvalue_reference<E>(u.get_deleter())) ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:552:68: error: expected primary-expression before '>' token : m_data(u.release(), ::boost::move_if_not_lvalue_reference<E>(u.get_deleter())) ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp: In member function 'boost::movelib::unique_ptr<T, D>& boost::movelib::unique_ptr<T, D>::operator=(boost::movelib::unique_ptr<T, D>&&)': /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:581:26: error: 'move_if_not_lvalue_reference' is not a member of 'boost' m_data.deleter() = ::boost::move_if_not_lvalue_reference<D>(u.get_deleter()); ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:581:65: error: expected primary-expression before '>' token m_data.deleter() = ::boost::move_if_not_lvalue_reference<D>(u.get_deleter()); ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp: In member function 'typename boost::move_upd::enable_up_moveconv_assign<T, D, U, E, boost::movelib::unique_ptr<T, D>&>::type boost::movelib::unique_ptr<T, D>::operator=(boost::movelib::unique_ptr<Y, D>&&)': /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:603:26: error: 'move_if_not_lvalue_reference' is not a member of 'boost' m_data.deleter() = ::boost::move_if_not_lvalue_reference<E>(u.get_deleter()); ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:603:65: error: expected primary-expression before '>' token m_data.deleter() = ::boost::move_if_not_lvalue_reference<E>(u.get_deleter()); ^ $ clang++37 -std=c++14 -isystem /usr/local/include/boost-1_60 twio/test_thread.cpp In file included from twio/test_thread.cpp:2: In file included from /usr/local/include/boost-1_60/boost/thread.hpp:13: In file included from /usr/local/include/boost-1_60/boost/thread/thread.hpp:13: In file included from /usr/local/include/boost-1_60/boost/thread/detail/thread_group.hpp:9: In file included from /usr/local/include/boost-1_60/boost/thread/csbl/memory/unique_ptr.hpp:16: /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:530:67: error: 'D' does not refer to a value : m_data(u.release(), ::boost::move_if_not_lvalue_reference<D>(u.g... ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:353:26: note: declared here template <class T, class D = default_delete<T> > ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:530:38: error: no member named 'move_if_not_lvalue_reference' in namespace 'boost' : m_data(u.release(), ::boost::move_if_not_lvalue_reference<D>(u.g... ~~~~~~~~~^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:552:67: error: 'E' does not refer to a value : m_data(u.release(), ::boost::move_if_not_lvalue_reference<E>(u.g... ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:548:29: note: declared here template <class U, class E> ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:552:38: error: no member named 'move_if_not_lvalue_reference' in namespace 'boost' : m_data(u.release(), ::boost::move_if_not_lvalue_reference<E>(u.g... ~~~~~~~~~^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:581:64: error: 'D' does not refer to a value m_data.deleter() = ::boost::move_if_not_lvalue_reference<D>(u.get_... ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:353:26: note: declared here template <class T, class D = default_delete<T> > ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:581:35: error: no member named 'move_if_not_lvalue_reference' in namespace 'boost' m_data.deleter() = ::boost::move_if_not_lvalue_reference<D>(u.get_... ~~~~~~~~~^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:603:64: error: 'E' does not refer to a value m_data.deleter() = ::boost::move_if_not_lvalue_reference<E>(u.get_... ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:597:29: note: declared here template <class U, class E> ^ /usr/local/include/boost-1_60/boost/move/unique_ptr.hpp:603:35: error: no member named 'move_if_not_lvalue_reference' in namespace 'boost' m_data.deleter() = ::boost::move_if_not_lvalue_reference<E>(u.get_... ~~~~~~~~~^ 8 errors generated. > On 15 Dec 2015, at 02:54, Marshall Clow <mclow.lists(a)gmail.com <mailto:mclow.lists@gmail.com>> wrote: > > Release candidate files for 1.60.0 are available at: http://boost.cowic.de/rc/ <http://boost.cowic.de/rc/> > > As always, the release managers would appreciate it if you download > the candidate of your choice and give building it a try. Please report > both success and failure, and anything else that is noteworthy. > > This helps ensure the candidates build OK before we push them out to SourceForge. > > The files (and associated md5s) are: > MD5 (boost_1_60_0_rc1.7z) = 7ce7f5a4e396484da8da6b60d4ed7661 > MD5 (boost_1_60_0_rc1.tar.bz2) = 65a840e1a0b13a558ff19eeb2c4f0cbe > MD5 (boost_1_60_0_rc1.tar.gz) = 28f58b9a33469388302110562bdf6188 > MD5 (boost_1_60_0_rc1.zip) = 0cc5b9cf9ccdf26945b225c7338b4288 > > Thanks! > > -- The release managers > > P.S. Release notes are here: http://www.boost.org/users/history/version_1_60_0.html <http://www.boost.org/users/history/version_1_60_0.html> > > _______________________________________________ > Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost-announce <http://lists.boost.org/mailman/listinfo.cgi/boost-announce>

1 0

multi_index::composite_key with char[]
by leyiliu 16 Dec '15

16 Dec '15

Hi,All: When I use the multi_index::composite_key with char[],I meet a error,when find a element。It seems to the make_tuple can't support the char[].Because the char[] more faster the std::string。So our project choose char[] as the column instead of std::string.So How should I do -- View this message in context: http://boost.2283326.n4.nabble.com/multi-index-composite-key-with-char-tp46… Sent from the Boost - Users mailing list archive at Nabble.com.

3 7

Regarding thread_specific_ptr's non-moveability
by Daniel Hofmann 15 Dec '15

15 Dec '15

thread_specific_ptr is currently (Boost 1.59) not moveable, since it defines a user-declared destructor and in addition (privately) user-declares the copy special member functions. Is this decision by design, or simply not yet updated to move semantics? >From what I can tell, a move should be possible along the lines of: auto* p = other.release(); // without invoking other's cleanup fn reset(p); // potentially differente cleanup fn associated The only problematic situation I can think of is a scenario with two semantically different custom cleanup function where a move eventually triggers only the second one. But this is behavior is consistent with move semantic. Cheers, Daniel J H References: > http://www.boost.org/doc/libs/1_59_0/doc/html/thread/thread_local_storage.h… > http://www.boost.org/doc/libs/1_59_0/boost/thread/tss.hpp

2 1

[smart_ptr] BugTicket #11843, Hash not supported for shared_ptr<T[N]>
by Gagan Kumar 15 Dec '15

15 Dec '15

I have tried to solve problem for hash not supported for shared_ptr<T[N]> https://github.com/boostorg/smart_ptr/pull/22 However, I wanted to know Is there a way to solve hash problem with shared_ptr<T[]>. Since hash here depends on all element and we can't get array size in this case. Still, does there exists way which I may be missing.

1 0

[PBGL] Very slow graph creation
by Quaglino Alessio 15 Dec '15

15 Dec '15

Hello, I am trying to parallelize my code and while the rest was easy, I have spent more than 1 month on the BGL part (despite it being only 2% of the runtime) without any success. I have written the following code: using namespace boost; using boost::graph::distributed::mpi_process_group; typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>, undirectedS> Graph; typedef iterator_property_map<std::vector<int>::iterator, property_map<Graph, vertex_index_t>::type> LocalMap; int getConnectedComponents(std::vector< std::pair<int,int> > &arcs, const std::vector<int> &isActive) { std::vector<int> localComponent(nV+1); std::vector<int> globalComponent(nV+1); Graph G(nV+1); // VERY SLOW!! // Add edges to the graph VERY SLOW!! for(std::vector<std::pair<int,int> >::iterator it = arcs.begin(); it != arcs.end(); ++it) if( isActive[arcCnt++]==0 ) add_edge(vertex(it->first,G),vertex(it->second,G),G); synchronize(G); // Compute number of connected components VERY SLOW with -np 2 LocalMap components(localComponent.begin(),get(vertex_index, G)); num = connected_components_ps(G, components); // Collect the global component vector components.set_max_ghost_cells(0); for( int i = 0; i < nV+1; i++ ) globalComponent[i] = get(components, vertex(i, G)); synchronize(components); } That parallelizes the following implementation: typedef adjacency_list <vecS, vecS, undirectedS> SerialGraph; int getConnectedComponents(std::vector< std::pair<int,int> > &arcs, const std::vector<int> &isActive) { int rank; std::vector<int> globalComponent(nV+1); SerialGraph G(nV+1); MPI_Comm_rank(PETSC_COMM_WORLD, &rank); // Only rank 0 has got the arc data structure if( rank == 0 ) { // Add edges to the graph for(std::vector<std::pair<int,int> >::iterator it = arcs.begin(); it != arcs.end(); ++it) if( isActive[arcCnt++]==0 ) boost::add_edge(it->first,it->second,G); num = connected_components(G, &globalComponent[0]); } // Propagate size for allocation MPI_Bcast( &num, 1, MPI_INT, 0, PETSC_COMM_WORLD ); // Propagate connected components MPI_Bcast( &globalComponent[0], nV+1, MPI_INT, 0, PETSC_COMM_WORLD ); However, the parallel version on a single core is about 5 times slower than the non-parallel one. Moreover, on two cores it becomes 15-20 times slower. What am I doing wrong? Regards, Alessio

1 0