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

pthread, futex_wait hang
by Matt Paindavoine. 30 Aug '04

30 Aug '04

Hello, I have been using the spirit parser library, and other boost pieces (date, test, variant, ...), along with Mozilla. I recently merged in some parts of the application that were running separately, and the program started hanging. I did a bit of searching, but couldn't find anything significantly wrong, especially since the individual parts are working correctly. The problem seems to be related to threads, as the stack indicates (cf stack at the end), and as the strace output shows here: parsing:*****************************: ) = 240 time([1093872322]) = 1093872322 futex(0x4045b860, FUTEX_WAIT, 2, NULL) = -1 EAGAIN (Resource temporarily unavailable) futex(0x4045b860, FUTEX_WAIT, 2, NULL) = -1 EAGAIN (Resource temporarily unavailable) [... ad vitam eternam] I compiled this program using gcc (GCC) 3.4.1 (Debian 3.4.1-7) using -DBOOST_MPL_NO_PREPROCESSED_HEADERS -DBOOST_SPIRIT_DEBUG -fPIC -I/usr/X11R6/include -frtti -fexceptions -Wall -Wconversion -Wpointer-arith -Wcast-align -Woverloaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-non-virtual-dtor -Wno-long-long -pedantic -fshort-wchar -pthread -pipe -DDEBUG -D_DEBUG -DDEBUG_root -DTRACING -g -fno-inline -I/usr/X11R6/include -DMOZILLA_CLIENT -include I don't know exactly what those threads are doing behind my back, but I'd appreciate if someone could provide some suggestion for fixing this problem. Thanks. Matt stack: #0 0x4040c791 in pthread_setcanceltype () from /lib/tls/libc.so.6 #1 0x4045b860 in __after_morecore_hook () from /lib/tls/libc.so.6 #2 0x4045aedc in ?? () from /lib/tls/libc.so.6 #3 0x00000031 in ?? () #4 0x4039ad63 in posix_memalign () from /lib/tls/libc.so.6 #5 0x00000012 in ?? () #6 0x409c91fc in ~lightweight_mutex (this=0x819eb40) at lwm_pthreads.hpp:53 #7 0x409c9d24 in ~sp_counted_base_impl (this=0x819eb40) at nsCXExpRecord.cpp:219 #8 0x409c96e2 in boost::detail::sp_counted_base::destruct (this=0x819eb40) at shared_count.hpp:115 #9 0x409b2a7e in boost::detail::sp_counted_base::weak_release (this=0x819eb40) at shared_count.hpp:173 #10 0x409bd6de in boost::detail::weak_count::operator= (this=0x40b3bb3c, r=@0x41749188) at shared_count.hpp:492 #11 0x409bd68f in boost::weak_ptr<boost::spirit::impl::grammar_helper<boost::spirit::grammar<cx::FooFooRecordGrammar<cx::FooConfigFooRecordGrammar>, boost::spirit::closure_context<cx::FooFooRecordClosure<cx::FooConfigFooRecordGrammar> > >, cx::FooFooRecordGrammar<cx::FooConfigFooRecordGrammar>, boost::spirit::scanner<__gnu_cxx::__normal_iterator<char const*, std::string>, boost::spirit::scanner_policies<boost::spirit::iteration_policy, boost::spirit::match_policy, boost::spirit::action_policy> > > >::operator=<boost::spirit::impl::grammar_helper<boost::spirit::grammar<cx::FooFooRecordGrammar<cx::FooConfigFooRecordGrammar>, boost::spirit::closure_context<cx::FooFooRecordClosure<cx::FooConfigFooRecordGrammar> > >, cx::FooFooRecordGrammar<cx::FooConfigFooRecordGrammar>, boost::spirit::scanner<__gnu_cxx::__normal_iterator<char const*, std::string>, boost::spirit::scanner_policies<boost::spirit::iteration_policy, boost::spirit::match_policy, boost::spirit::action_policy> > > > (this=0x40b3bb38, r=@0x41749184) at weak_ptr.hpp:87 #12 0x409bc67a in grammar_helper (this=0x41749170, p=@0x40b3bb38) at grammar.ipp:156 [ goes further up, I don't know if it will help... ]

2 2

BLL and member variables
by Mattias Brändström 30 Aug '04

30 Aug '04

Hello, I really like the BLL and would like to use it as much as possible. I have run into a small problem though. My code looks like this: #include <boost/lambda/bind.hpp> #include <boost/lambda/lambda.hpp> #include <map> #include <vector> class A { public: A(int a, int b) : a_(a), b_(b) { } int a_; int b_; }; int main() { using namespace boost::lambda; std::vector<A> v; v.push_back(A(1, 2)); typedef std::map<int, int> Map; Map m; for_each(v.begin(), v.end(), bind(&Map::insert, std::make_pair(bind(&A::a_, _1), bind(&A::b_, _1)))); return 0; } What I want to do is to populate a map with the information from the vetctor v. For each element in v I want to insert a value and a key into the map something like this: void operator()(const A& a) const { m_.insert(std::make_pair(a.a_, a.b_)); } I found some information on how to use member variables as targets in the BLL documentation and that is what I have tried to use in the code above. It won't compile however, using g++ I get the following error: g++ -c -o test.o test.cpp test.cpp: In function `int main()': test.cpp:27: error: no matching function for call to `bind(<unknown type>, std::pair<boost::lambda::lambda_functor<boost::lambda::lambda_functor_base<boost::lambda::action<2, boost::lambda::function_action<2, boost::lambda::detail::unspecified> >, boost::tuples::tuple<int A::*, const boost::lambda::lambda_functor<boost::lambda::placeholder<1> >, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type> > >, boost::lambda::lambda_functor<boost::lambda::lambda_functor_base<boost::lambda::action<2, boost::lambda::function_action<2, boost::lambda::detail::unspecified> >, boost::tuples::tuple<int A::*, const boost::lambda::lambda_functor<boost::lambda::placeholder<1> >, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type> > > >)' scons: *** [test.o] Error 1 scons: building terminated because of errors. If someone here has any ideas of how to accomplish this it would be really nice! Thanks in advance! Regards, Mattias

2 1

downloading problems
by Fred J. 30 Aug '04

30 Aug '04

Hello I am trying to download a boost library from http://sourceforge.net/project/showfiles.php?group_id=7586 where the library name suppost to be, there is [show only this release] instead, the wholl page, all the library names are not listed. any body is having the same problem or it is just me thanks _______________________________ Do you Yahoo!? Win 1 of 4,000 free domain names from Yahoo! Enter now. http://promotions.yahoo.com/goldrush

2 1

Re: [boost] Re: MPL vector size limit
by Aleksey Gurtovoy 30 Aug '04

30 Aug '04

Arkadiy Vertleyb writes: > Sorry to repeat this question, but I think I am missing somehing here: > > Do I understand correctly that it's not possible in the current > version of MPL to use mpl::vector of a size larger than 50? It is possible. As I tried to explain in my previous reply, you can pre-generate the coressponding headers up to whatever limit you choose. You can also generate the sequence "on the fly", although that's less efficient. > > If this is not correct, what exactly do I have to do to use, for > example, mpl::vector97<...>? #include "boost/mpl/vector/vector50.hpp" #include "boost/mpl/aux_/config/vector.hpp" #include "boost/mpl/aux_/config/ctps.hpp" #include "boost/preprocessor/iterate.hpp" #include "boost/config.hpp" namespace boost { namespace mpl { #define BOOST_PP_ITERATION_PARAMS_1 \ (3,(51, 100, "boost/mpl/vector/aux_/numbered.hpp")) #include BOOST_PP_ITERATE() }} > > If this _is_ correct, I would like to make a feature request for it > to be increased, since the limit of 50 doesnt seem to be high enough > (I am making this statement in conjunction with the need of my > typeof emulation lib, but I am kind of surprized this question > hasn't yet popped up in other contexts, or has it?). We are using sequences of ~70 elements here at work (mostly in context of FSMs), but then we simply pre-generate them locally. Increasing the default limit to, let's say, 100 elements means plus ~120 KB to the size of the distribution, and I'm not sure the need is widespread enough to make everybody pay for it. > In particular, isn't it possible to just get it fully controlled by > BOOST_MPL_LIMIT_VECTOR_SIZE without any limitations whatsoever? Once again, tweaking BOOST_MPL_LIMIT_VECTOR_SIZE is a dubious practice. You can get as long of a sequence as you want with the technique demostrated in the example above, with a caveat that it's going to be less efficient than using pre-generated headers. -- Aleksey Gurtovoy MetaCommunications Engineering

2 1

Using external property map with a mutable boost.graph
by Jeff Holle 30 Aug '04

30 Aug '04

I'm using a boost.graph via: typedef boost::adjacency_list<boost::vecS,boost::vecS,boost::bidirectionalS> GraphT; In my application, I've decided to use external maps. I have discovered a problem using vertex descriptors as a key of a map. Basically, in doing a remove_vertex sequence involving vertex descriptors 5, 3, 2, I discovered that the next vertex descriptor produced by add_vertex was 4. At this point the insert into my map failed because this element already existed. Given the mutable nature of vertex descriptors, I don't see how they can be used as a key in a map. I'm contemplating two options at this point. 1. Switch from map to vector in storing my vertex property externally. 2. Switch to internal properties. My questions are: 1. Is the first option viable? As I see things, vertex descriptors are always contiguous and in the range (0..num_vertices(graph)]. Is this true? 2. Do internal "maps" avoid this type of problem?

2 1

date_time bug cvs latest vc7.1
by Gordon Smith 29 Aug '04

29 Aug '04

The following sample will not compile with the following settings (I know it has been discussed with regards to the regex library previously) - this code compiles fine in 1_31: "Character set": "Use Multi-Byte Character Set" (I am targetting win 95/98) "Treat wchar_t as Built-in Type (/Zc:wchar_t)": "No" The contrived example demonstrates why I can't turn "/Zc:wchar_t" on: #include "stdafx.h" #include <comutil.h> #include <boost/date_time/gregorian/gregorian.hpp> int _tmain(int argc, _TCHAR* argv[]) { _variant_t var("Narrow String"); return 0; } begin 666 boost_bug_001.zip M4$L#!!0````(`&A9$C&80G7MR@````\!```1````8F]O<W1?8G5G7S P,2YC M<'!5CD%+Q# 0A<\&\A^&>MD52>K5E071@R</ZDTD9-,T'6@G83I=%?&_V^Q% MO7WO8^;QK(5#SK.XPY)<VUZ94 I<PWWLD>(,,D2()/P))2,)])E/+F2:\QC! MES)B\(*9C%;6:J75.5(8ERY",TOG^P\S-/_T3<C3(CB:8?_7GF;8SDMT@E.T MB6/*C)Y^R0RE[&M7G>)D\DB;BIY3N 3W<O=P^W11T_'U;:O5EU9G[NCKISB! ME3;-HV?.[_ LC)2:[6X]X2@+$[0K?_\`4$L#!!0````(`%!6$C$O3O0P5P$` M`),#```1````8F]O<W1?8G5G7S P,2YS;&Z=D,UKPC 8QL\6^C^$GAR8DFCM MZF&'IA]CA\%0V([2CR(a)962)-*AO._WVI5ITZF'H(29XW[^]]\CRSHI)*SC1X M9:K..)CHNF023"2O-9,"I(S3'DAE]9&9-[12C1BX"-G62R7?::&[SBH@T2A* M2 `#XH40XQC!(,88(A2B:(1-U>NOG3OP`)Q<2J6G>3V?(H2=WHG@+HN%H3;Z M:NCA*"+W'D1ID$+/1Q@&89I GY!1WP_]88(':\>V.JV/B5G&7+>]QG1!14E% MP:AJ)B_,F+9D>A)1'K?9UD&RK4<N\XR;=]O##KU+)9)BQN9UE367#;RBNYII MZL34_,?(F[T1QI333%$CM:>MA2/ZV;36S?DP\Y/?TRY+RMV8<4/#7M)HMK?W M_<;$H'\UA]2,ERZZF=+F<.2GU6YF'3R=DO[-.OG45"B6,\[TU[:D_LCZ.DY8 MED_B,LQ>L:T?4$L#!!0````(`%!6$C&SGO.4R0```"P!```*````<W1D869X M+F-P<&60O6[#, R$=P-^AT,ZUT[7; 6"HEN&9.@6T!)=*U!(03]H_/:17:-+ MUR._XQW['BE;&A^="0$')"W1,$;G&7FB#"?&%\L)MY)RE;CNDUB*]F_4-GV/ M037EZU"^K_O]6Q?,A!_G/09>F1#YU>@]5%N+B<ER7*GMM@ZWWW6CDLG)?R;/ M@>O!4>.=LE-IF[9YV0)@M]E,NT6NMI?3\71 Y)$C2ZU#,H.L=0M)?@N0,&N! M,"]-<+X<WS^^NL^5KP4ANI2O25Q:W]$V3U!+`P04````" !05A(QJ,6F9;L` M```E`0``" ```'-T9&%F>"YH58_!BL,P#$3O@?S#P%Y+<R]+3WOO97] :\NQ M2V)G91F:OZ]<*&TO`FF>1J-I0E5/X7:,."%EMS3/"&FQ4L0TRI[$H^Y5>?T` MZF$<I@E&;5*N[!1U8Y="<I\8-)*"A-$J>P3A_\99E_V OZ8/CRZZ2'DV/>47 MT<5Q&(>O36A>"24[[KU-GB>^4ZDJ3.OY?:CF)L=X[JCY_UY^+B<(!Q8V"Y#W M25/)M" R>9:*O;3'([/0BAX@B46/MC .=U!+`0(4"Q0````(`&A9$C&80G7M MR@````\!```1``````````$`( ````````!B;V]S=%]B=6=?,# Q+F-P<%!+ M`0(4"Q0````(`%!6$C$O3O0P5P$``),#```1``````````$`( ```/D```!B M;V]S=%]B=6=?,# Q+G-L;E!+`0(4"Q0````(`%!6$C&SGO.4R0```"P!```* M``````````$`( ```'\"``!S=&1A9G@N8W!P4$L!`A0+% ````@`4%82,:C% MIF6[````)0$```@``````````0`@````< ,``'-T9&%F>"YH4$L%!@`````$ -``0`[ ```%$$```````` ` end

2 3

boost::thread callback copy problem
by Howard Cole 27 Aug '04

27 Aug '04

Hi, I am new to boost::thread but have come across the following problem and wondered if anyone could help... The problem is the copy overhead used by the boost::thread / boost::function0 code. The following code actually creates multiple copies of myclass rather than using a reference to myclass and due to the size of the class, this leads to a considerable overhead. I only noticed this problem because the thread kept failing due to the fact that my constructor held some socket initialisation code, which was being called multiple times. Is there a way that I can force the thread class to use references rather than copying? Looking at the code for the function library, I see that you can use boost::ref for function classes, any ideas as to how it can be used in the thread class? Thanks in advance Howard Cole www.selestial.com Example code. class myclass { public: // Lots of local data here void operator() () { do_thread_stuff(); }; // CTOR myclass(){ do_some_initialisation_stuff(); }; // DTOR ~myclass(){ do_some_cleanup_stuff(); }; // Lots of member funcs ... }; void main (void){ myclass test_instance; boost::thread test_thread(test_instance); test_thread.join(); }

3 3

RE: [Boost-users] Re: Guidelines for effective exception usage
by Mark Storer 26 Aug '04

26 Aug '04

> No it is not. Even if you aren't throwing due to memory starvation, > you could run out of memory during unwinding, when the exception is > copied. That leads you directly to terminate(). Do not pass Go; do > not collect $200. Venturing a little far afield, but one technique I've heard of is to grab a block, so you can release it in low-memory situations (or just every time you throw). This ensures (in theory) that you'll have enough memory to copy your exception, create your strings, and so on. If you manage to recover, you just grab another block ("prime the pump") and soldier bravely forward. Given STL's habit of allowing people to specify an allocator, you could even go so far as to build a custom allocator that worked on some pre-allocated block of memory for your exceptions (strings, vectors, whatever you deem necessary). Either way, you'll run out of memory a little earlier (not a big issue in many environments), but you _will_ get to pocket the $200. It's not hard to imagine an environment where grabbing a couple extra Kb would be a Bad Thing, but I'd think this trick would be fine on PCs and the like. --Mark Storer Senior Software Engineer Verity, Inc.

4 3

Generative Programming book
by Samuel 26 Aug '04

26 Aug '04

The ISBN for the "Generative Programming" book is wrong. 0-210-30977-7 is incorrect 0-201-30977-7 is correect That book is used and described in the Boost "Feature Model Diagrams in text and HTML" page. Also, the Generative Programming web site isn't working for me, but perhaps the problem is only temporary. Note that I am not asking for help with any of that. I obviously already know what the correct ISBN is. I am posting this in case it helps others; the web page needs to be updated for the ISBN at least. The Boost "What to do about Boost bugs" page says to submit "a bug report to either boost-users list, boost mailing list, or our bug tracking facility at SourceForge" and that "either of the mailing lists is a preferred way". However it does not describe what a bug report is. If there is not a formal format, then I hope the above is adequate.

2 1

/usr/include/boost/mpl/if.hpp:75: internal compiler error: Segmentation fault
by Nils Wagner 26 Aug '04

26 Aug '04

Dear experts, I am going to install VPython http://www.vpython.org/linux_download.html on SuSE 9.1 which makes use of the boost library. However I got a segmentation fault. /usr/include/boost/mpl/if.hpp:75: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <URL:http://www.suse.de/feedback> for instructions. Preprocessed source stored into /tmp/ccrXaXyw.out file, please attach this to your bugreport. Any pointer would be appreciated. Nils

2 2