- Boost-users - lists.boost.org

RE: PP Library...
by Eric Robert 07 May '03

07 May '03

Ok, I see that there are 3 ways of doing this now. What is the recommendation for something small like 2 or 3 lines for each native types i.e. more or less 12 iterations? Eric "Paul Mensonides" <yg-boost-users(a)m.gmane.org> wrote > Tanton Gibbs wrote: > > You can use BOOST_PP_LIST_FOR_EACH > > > > #define ARG_LIST (char, (int, (unsigned int, (long, ...)))) > > #define DO_IT(r, data, elem) void foo( elem ); > > > > BOOST_PP_LIST_FOR_EACH( DO_IT, , ARG_LIST ) > ^ > This empty argument is undefined in C++. In C99, it is allowed and called a > "placemarker." Instead of passing nothing, just pass anything and ignore it. > > // ----- // > > #include <boost/preprocessor/list/for_each.hpp> > > #define ARG_LIST \ > (char, (int, (unsigned, (long, BOOST_PP_NIL)))) \ > /**/ > #define DO_IT(r, ignored, elem) \ > void foo( elem ); \ > /**/ > > BOOST_PP_LIST_FOR_EACH( DO_IT, ?, ARG_LIST ) > > #undef ARG_LIST > #undef DO_IT > > // ----- // > > Note that there are other ways to do this also. For instance, use of > "sequences" makes the definition of ARG_LIST easier: > > // ----- // > > #include <boost/preprocessor/seq/for_each.hpp> > > #define ARGS \ > (char)(int)(unsigned)(long) \ > /**/ > #define DO_IT(r, ignored, elem) \ > void foo( elem ); \ > /**/ > > BOOST_PP_SEQ_FOR_EACH( DO_IT, ?, ARGS ) > > #undef ARGS > #undef DO_IT > > // ----- // > > Of course, if what you want to generate is much larger, you might want to use > some form of vertical repetition: > > // ----- // > > #include <boost/preprocessor/iteration/local.hpp> > #include <boost/preprocessor/seq/elem.hpp> > > #define ARGS \ > (char)(int)(unsigned)(long) \ > /**/ > > #define BOOST_PP_LOCAL_MACRO(n) \ > void foo( BOOST_PP_SEQ_ELEM(n, ARGS) ); \ > /**/ > #define BOOST_PP_LOCAL_LIMITS \ > (0, BOOST_PP_SEQ_SIZE(ARGS) - 1) \ > /**/ > #include BOOST_PP_LOCAL_ITERATE() > > #undef ARGS > > // ----- // > > Regards, > Paul Mensonides [Non-text portions of this message have been removed]

3 8

RE: [Boost-Users] bug in pthreads implementation of recursive_mut ex
by Feldman, Yulik 06 May '03

06 May '03

Thanks Alexander for the pointer to this long and interesting discussion. I'll check my design whether I could get rid of the recursive mutexes in order to reduce the probability of not holding invariants when the wait() is called, especially since the code still doesn't behave correctly, even after the Boost bug is fixed :) William, unfortunately I don't have a small testcase for the bug and it seems that it will be too time-consuming for me to try to prepare a small "working" testcase. Thanks for confirming the bug report (it makes me feel better that the debugging time was not wasted). --Yulik. -----Original Message----- From: William E. Kempf [mailto:wekempf@cox.net] Sent: Monday, May 05, 2003 6:44 PM To: Boost-Users(a)yahoogroups.com Subject: Re: [Boost-Users] bug in pthreads implementation of recursive_mutex Feldman, Yulik said: > Hi, > > > > It looks that there is a bug in > .../boost_1_30_0/libs/thread/src/recursive_mutex.cpp. The bug shows in > the pthreads implementation of the recursive_mutex when the mutex is > used with the "condition" synchronization primitive. Here is the > description of the bug: You're evaluation of the bug is correct. I'll fix this in CVS shortly. Do you maybe have a test case for this that I could add to the regression tests? BTW, Mr. Terekhov does provide a good resource in his posting on this subject. I disagree with some points in the thread he posted, but it is quite true that you absolutely must gaurantee that invariants hold at the point cond.wait() is called, and that doing so is complicated by the use of recursive locks. Either avoid recursive locks entirely, avoid condition variables with recursive mutexes, or be very sure of the design and the state of your invariants. -- William E. Kempf Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

2 1

Problems with mpl::push_back
by joaquintides 06 May '03

06 May '03

#include <boost/mpl/list.hpp> #include <boost/mpl/push_back.hpp> int main() { typedef boost::mpl::list<> type0; typedef boost::mpl::push_back<type0,int>::type type1; typedef boost::mpl::push_back<type1,char>::type type2; return 0; } This fails in MSVC 6.5 and gcc 3.2. The error seems to indicate that push_back does not provide a type named 'type'. If I replace push_back with push_front, eveything works wonderfully. Am I doing any gross error, is the documentation wrong, ideas, clues? Thanx in advance. Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 1

boost::shared_ptr templated constructor?
by Lehrer, Joshua 06 May '03

06 May '03

WRT boost::shared_ptr<>, why not template the explicit constructor that takes a T*? This way, it could select the correct destructor at construction time: struct base { ~base(); }; struct sub : public base { ~sub() }; boost::shared_ptr<base>(new sub); The above code currently only calls ~base(); Templating the explicit single argument constructor would allow the above code to call ~sub(). The constructor now takes only a T*, and therefore populates the pointer to the destructor function with a call to ~T(), which is ~base() in the above example. If, however, the constructor took an X*, the above example would pass a sub*, and it could populate the default destructor function with a pointer to ~X(). Comments? Joshua Lehrer FactSet Research Systems NYSE:FDS

3 2

BGL leaky?
by Martin Okrslar 06 May '03

06 May '03

Dear Boost users, is the BGL leaky? For example the code added below produces memory leaks (as seen from also added valgrind output). When I comment out the line marked by /* THIS LINE DOES IT */, there are no leaks. I am running BGL 1.29 and gcc 3.2. Any ideas for a work-around? cheers Martin #include <iostream> // for std::cout #include <utility> // for std::pair #include <algorithm> // for std::for_each #include <boost/graph/graph_traits.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/dijkstra_shortest_paths.hpp> using namespace boost; int main(int,char*[]) { // create a typedef for the Graph type typedef adjacency_list<vecS, vecS, bidirectionalS> Graph; // Make convenient labels for the vertices enum { A, B, C, D, E, N }; const int num_vertices = N; // writing out the edges in the graph typedef std::pair<int, int> Edge; Edge edge_array[] = { Edge(A,B), Edge(A,D), Edge(C,A), Edge(D,C), Edge(C,E), Edge(B,D), Edge(D,E), }; int num_edges = sizeof(edge_array)/sizeof(edge_array[0]); // declare a graph object Graph* g = new Graph(num_vertices); // add the edges to the graph object for (int i = 0; i < num_edges; ++i) add_edge(edge_array[i].first, edge_array[i].second, *g); /* THIS LINE DOES IT */ delete g; return 0; } okrslar@liebrich:~/molgen/domclust/src/test> valgrind --leak-check=yes --show-reachable=yes ./domclust ==23473== valgrind-1.0.4, a memory error detector for x86 GNU/Linux. ==23473== Copyright (C) 2000-2002, and GNU GPL'd, by Julian Seward. ==23473== Estimated CPU clock rate is 2412 MHz ==23473== For more details, rerun with: -v ==23473== ==23473== ==23473== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0) ==23473== malloc/free: in use at exit: 384 bytes in 1 blocks. ==23473== malloc/free: 3 allocs, 2 frees, 544 bytes allocated. ==23473== For counts of detected errors, rerun with: -v ==23473== searching for pointers to 1 not-freed blocks. ==23473== checked 6470036 bytes. ==23473== ==23473== definitely lost: 0 bytes in 0 blocks. ==23473== possibly lost: 0 bytes in 0 blocks. ==23473== still reachable: 384 bytes in 1 blocks. ==23473== ==23473== 384 bytes in 1 blocks are still reachable in loss record 1 of 1 ==23473== at 0x4004A3A8: __builtin_new (vg_clientfuncs.c:125) ==23473== by 0x4004A3F6: operator new(unsigned) (vg_clientfuncs.c:138) ==23473== by 0x4035E4F1: ??? (bits/stl_alloc.h:482) ==23473== by 0x4035E40D: ??? (bits/stl_alloc.h:532) ==23473== ==23473== LEAK SUMMARY: ==23473== definitely lost: 0 bytes in 0 blocks. ==23473== possibly lost: 0 bytes in 0 blocks. ==23473== still reachable: 384 bytes in 1 blocks. ==23473== -- ----------------------------------------------------------------------------- Martin Okrslar MPI for Molecular Genetics phone: ++ 49 + 30 / 8413-1166 Computational Molecular Biology Fax: ++ 49 + 30 / 8413-1152 Ihnestrasse 73 email: okrslar(a)molgen.mpg.de D-14195 Berlin URL: http://cmb.molgen.mpg.de -----------------------------------------------------------------------------

2 1

More Mac Compiler Woes - Boost 1.30.0 - Thread and type_traits
by Bobby Thomale 05 May '03

05 May '03

Hey guys, I am having some more problems getting Boost 1.30 to compile on the Mac. I am using CW Pro 8.3, compiling a TARGET_CARBON project. I am having trouble getting the MP implementation of threads to compile. It seems to be having trouble with function pointers where the syntax function<type> is used. I remember at one point it was mentioned that: function<type, type...> was deprecated and we were to use either the more universally supported: functionN<type,type...> or the cleaner: function<void myfuncSig()> instead if our compiler supported it. (Mine doesn't support that last one yet I don't think.) I am guessing that the oldest syntax was removed and no one has bothered to update the Mac carbon thread implementaton to compile again since then? That one seems pretty easy to fix - just update to a supported syntax? (I could do this, update to the functionN<type,type...> syntax since that is what my compiler supports. If someone wants me to make and submit the changes to them for inclusion I'd be happy to, if I knew who to send them to...) Or is there a better/easier workaround? Secondly, the Signals library is not compiling. It's actually not a Signals problem, I think it is a type_traits problem. I am getting a bunch of strange compile errors, in a boost::type_traits file called: is_base_and_derived.hpp It looks like there is a template near the top that is the first thing that fails. There is a define that goes into an alternate implementation for certain broken compilers. Perhaps CodeWarrior needs to be added to the list of compilers that compiles the "broken compiler" version of the stuff in this header? Thanks in advance for your help! -- Bobby --------------------------------------------------------------------- Bobby Thomale Senior Software Developer Inoveon Corporation http://www.inoveon.com/ ---------------------------------------------------------------------

5 6

bug in pthreads implementation of recursive_mutex
by Feldman, Yulik 05 May '03

05 May '03

Hi, It looks that there is a bug in .../boost_1_30_0/libs/thread/src/recursive_mutex.cpp. The bug shows in the pthreads implementation of the recursive_mutex when the mutex is used with the "condition" synchronization primitive. Here is the description of the bug: When the client code calls wait() function on a condition sending as a parameter a lock object associated with a recursive_mutex ("m_Condition.wait(Lock)"), the following functions get called: boost::condition::wait(L& Lock) at condition.hpp:91 boost::condition::do_wait(M& mutex) at condition.hpp:137 boost::detail::thread::lock_ops<boost::recursive_mutex>::unlock(Mutex& m, lock_state& state) at lock.hpp:54 boost::recursive_mutex::do_unlock(cv_state& state) The implementation of the last function on "winthreads" and "mptasks" resets the 'm_count' data member to zero. The "pthreads" implementation at recursive_mutex.cpp:399 doesn't reset the 'm_count' to zero. As a result, when the calling thread enters pthread_cond_wait() and the mutex is unlocked by the pthreads the 'm_count' still remains non-zero. Starting from this point, the things may go bad. Consider, for example, that another thread tries to acquire the mutex. The thread calls: boost::recursive_mutex::scoped_lock Lock(m_Mutex) boost::detail::thread::scoped_lock<boost::recursive_mutex>::scoped_lock<boos t::recursive_mutex>(Mutex& mx, initially_locked = true) at lock.hpp:66 boost::detail::thread::scoped_lock<boost::recursive_mutex>::lock() at lock.hpp:76 boost::detail::thread::lock_ops<boost::recursive_mutex>::lock(Mutex& m) at lock.hpp:34 boost::recursive_mutex::do_lock() at recursive_mutex.cpp:307 The last function locks the mutex again, and, if BOOST_HAS_PTHREAD_MUTEXATTR_SETTYPE is defined, increments the counter to 2. Now, since the count is more than 1, the function unlocks the mutex. As a result, the mutex remains unlocked, after the lock function is called. The same problem exists in recursive_try_mutex and recursive_timed_mutex. I've tried to fix the code as follows, which seemed to help in my particular case: ---------------------------------------------------------------------------- ------------------------------------ *** recursive_mutex.cpp.orig Sun Feb 23 19:38:11 2003 --- recursive_mutex.cpp Sat May 3 17:32:44 2003 *************** *** 412,417 **** --- 412,418 ---- state.pmutex = &m_mutex; state.count = m_count; + m_count = 0; } recursive_try_mutex::recursive_try_mutex() *************** *** 608,613 **** --- 609,615 ---- state.pmutex = &m_mutex; state.count = m_count; + m_count = 0; } recursive_timed_mutex::recursive_timed_mutex() *************** *** 787,792 **** --- 789,795 ---- state.pmutex = &m_mutex; state.count = m_count; + m_count = 0; } #elif defined(BOOST_HAS_MPTASKS) ---------------------------------------------------------------------------- ------------------------------------ Any comments? Is this indeed a bug? Is the fix correct? Thanks, Yulik. [Non-text portions of this message have been removed]

3 2

uBlas - matrix comparison A == B -> how-to guide ...
by geert_ceuppens 05 May '03

05 May '03

Sharing my effort in working with ublas ... Forgive me if this is described already somewhere - I could not find it immediately in the doc so I looked around in the code ; ) Matrix comparison is implemented using the equals() method defined in matrix_assign.hpp: bool equals (const matrix_expression<E1> &e1, const matrix_expression<E2> &e2) Code snippet: ublas::matrix<double> A(5,5), B(5,5); // Initialise A somehow init_matrix(A); B = A; // This does not compile ... bool equal = ( A == B ); // But this does (and it works ...) bool equal = ublas::equals ( A, B );

1 0

RE: [Boost-Users] How To Build Boost 1.30 With Dev-CPP?
by bjorn.karlsson＠readsoft.com 05 May '03

05 May '03

> From: firingme [mailto:yg-boost-users@m.gmane.org] > E:\LY\boost_1_30_0>bjam sTOOS=mingw This should read sTOOLS, not sTOOS. Bjorn

5 5

RE: [Boost-Users] Code Fail
by Jim.Hyslop 03 May '03

03 May '03

Hossein Haeri [mailto:powerprogman@yahoo.com] wrote: > There is a tiny cpp file attached to this message > that raises the following exception while running: No, there isn't ;-) The email list strips out attachments. If it truly is tiny, then paste it inline. If it's not-so-tiny, then you can upload it to the files area: http://groups.yahoo.com/group/Boost-Users/files/ -- Jim

1 0