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

thread safety of the mutex ctor
by James E Taylor 18 Aug '05

18 Aug '05

I hope this isn't a stupid question, but.... Is the boost mutex constructor thread-safe? I am writing a singleton template and trying to avoid the static initialisation/destruction order fiasco. I envisage two threads accessing a function like this concurrently: template<typename T> boost::shared_ptr<T> my_func() { static boost::mutex m; boost::mutex::scoped_lock l(m); static boost::shared_ptr<T> p(new T); return p; } and my worry is that both threads could attempt to do the static initialisation of m concurrently. Is there protection against this? Is it even possible to protect against this? I've tried to use boost::call_once, which would give sufficient guarantees but it's quite inflexible as nothing can be passed to or from the function being called. I know that integral types (like a pthread_t) of static storage are initialised before main() and before constructors of statically allocated objects are called, which means there would be no problem using a pthread_t above. However, constructors of objects of static storage are called at a time when there could be multiple threads running. Am I worrying about nothing? Or is there a problem here? Thanks, James -- James E Taylor ___________________________________ NOCC, http://nocc.sourceforge.net

3 2

Re: [Boost-users] Boost-users Digest, Vol 637, Issue 1
by James E Taylor 18 Aug '05

18 Aug '05

Thanks, Pablo, I'll have a look at it. James > FYI Matthew Wilson gave a technique (in "Imperfect C++&quot for handling > this, > by turning the mutex initialization into an integer initialization (that > doesn't have a problem with [multiple] simultaneous threads entering the > function) > > It uses STLSoft's spin_mutex, and looks pretty much like what you have > already: > > <code> > #include <platformstl/spin_mutex.hpp> > #include <stlsoft/lock_scope.hpp> > using platformstl::spin_mutex; > using stlsoft::lock_scope; > > template<typename T> > boost::shared_ptr<T> > my_func() > { > // static boost::mutex m; > // boost::mutex::scoped_lock l(m); > static int spin_count = 0; // No race condition initializing an > int > spin_mutex mutex(&spin_count); // Build the mutex on top of the > static > int > lock_scope lock(mutex); > > static boost::shared_ptr<T> p(new T); > return p; > } > </code> > > This assumes, of course, you can (and are willing to) use STLSoft... > > > HTH, > Pablo ___________________________________ NOCC, http://nocc.sourceforge.net

1 0

Random access to sequences
by Stephen Gross 18 Aug '05

18 Aug '05

I read up on boost::mpl::at<...>. It's pretty cool, but I wanted to make it a little easier to use. I found it a little annoying to have to type out 'int_<...>' for the index number, so I created a struct to automate that feature: ========================== template<typename SEQ, int IDX> struct atIdx { typedef typename boost::mpl::at<SEQ, boost::mpl::int_<IDX> >::type type; }; ========================== With the struct above, you can do the following: ============================= typedef boost::mpl::vector<int, char, int> t; atIdx<t, 0>::type i; // This line is shorter than the one below: /* atIdx<t, boost::mpl::int_<0> >::type i; */ ============================= Is this feature already available in boost somewhere? If I've reinvented the wheel, I'll be happy to use an existing implementation... Thanks, --Steve Stephen Gross Case Western School of Medicine Cleveland, OH "By Grabthar's hammer, by the sons of Worvan, you shall be avenged." - Dr. Lazarus

2 1

Re: [Boost-users] thread safety of the mutex ctor
by Christopher Hunt 17 Aug '05

17 Aug '05

On 18/08/2005, at 1:19 AM, boost-users-request(a)lists.boost.org wrote: > I envisage two threads accessing a function like this concurrently: > > template<typename T> > boost::shared_ptr<T> > my_func() > { > static boost::mutex m; > boost::mutex::scoped_lock l(m); > > static boost::shared_ptr<T> p(new T); > return p; > } > > and my worry is that both threads could attempt to do the static > initialisation of m concurrently. Is there protection against this? > Is it even possible to protect against this? The scope of m is actually outside of your function given that it is static. Thus it will be initialised outside of your function generally before any other threads get to start up. Cheers, -C

3 2

Wave capabilities
by Foster Gareth 17 Aug '05

17 Aug '05

Hello list, I've been reading through the wave docs, it looks very impressive. I was wondering, is it possible to use Wave to extract all the macro/class/function definitions with their respective line numbers from a source file? I noticed the mention of include paths etc however, does this mean that just throwing random source files from a large source tree at Wave, with no regard for whether or not all the types in it will be defined, is going to result in it throwing exceptions? Thanks for any help. Cheers, Gaz ********************************************** * http://www.mozilla.org/products/firefox/ * **********************************************

2 1

[Signals] Boost.signals dll version crashes on OS X
by Chiedo Acholonu 17 Aug '05

17 Aug '05

I built the boost.signals library using bjam and am trying to use it in an xcode project named TestSignals. The project only consists of the signals library and the maximum.cpp example file (boost-root/libs/signals/example/maximum.cpp). The project builds but it crashes with a SIGBUS signal 10 when run. This only happens with the dynamic version of signals. I have defined BOOST_SIGNALS_DYN_LINK and built the library multithreaded, so I'm not sure what I'm doing wrong. I'm running OS X 10.4.2 with Xcode 2.1. Here is a snippet of the crash log (I can post the rest on request): Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000000 Thread 0 Crashed: 0 maximum.ob 0x0040c824 std::less<int>::operator()(int const&, int const&) const + 48 (stl_function.h:227) 1 maximum.ob 0x0040e0d8 boost::signals::detail::any_bridge_compare<std::less<int>, int>::operator()(boost::any const&, boost::any const&) const + 364 (named_slot_map.hpp:62) 2 maximum.ob 0x0040e148 boost::detail::function::function_obj_invoker2<boost::signals::detail::any_bridge_compare<std::less<int>, int>, bool, boost::any, boost::any>::invoke(boost::detail::function::any_pointer, boost::any, boost::any) + 76 (function_template.hpp:119) Any help would be much appreciated. Chiedo

1 0

boost:signals SIGSEV
by Massimiliano Liccardo 17 Aug '05

17 Aug '05

I opened the bug https://sourceforge.net/tracker/index.php?func=detail&aid=1262038&group_id=… concerning a segmentation fault using the boost:signals library. I attached the source i used. bye -- Massimiliano Liccardo (maX) <liccardo(a)acmesolutions.it> jid:coccolesto@jabber.org GnuPG public key available on wwwkeys.eu.pgp.net Key ID: D01F1CAD Key fingerprint: 992D 91B7 9682 9735 12C9 402D AD3F E4BB D01F 1CAD "la velocità induce all'oblio, la lentezza al ricordo"

1 0

[Regex] regex_match changed to match_any in 1.33?
by John Sichi 17 Aug '05

17 Aug '05

In boost/regex/v4/regex_match.hpp, a number of the overloads (the ones that don't take a result parameter) are now bitwise-oring in regex_constants::match_any. As a result, a sequence like boost::regexp exp("a(bc)?") char *value = "abcbc"; bool result = boost::regex_match(value, value + 5, exp); will now set result to true where it was previously false in 1.32. I fixed our code by passing in a dummy result object as an extra parameter. Was this change intentional? I couldn't find any documentation that mentions it. JVS

5 6

Re: [Boost-users] std::vector<std::string> pb
by Lana Abadie 17 Aug '05

17 Aug '05

Hi Joel, Thanks! It works now. Lana -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Joel de Guzman Sent: Tuesday, August 16, 2005 7:50 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] std::vector<std::string> pb Lana Abadie wrote: > > Hello, > Sorry to reiterate but if it's not the right forum to send this kind > of message, I would really appreciate that sb tells me where to post > it, because I'm stuck. > Thanks for helping > Lana > /*****************************/ Hi Lana, I'm sorry I wasn't able to reply as soon as possible. Yes, the proper forum is: C++-sig mailing list C++-sig(a)python.org http://mail.python.org/mailman/listinfo/c++-sig Anyway, here are some links that you might find useful: http://mail.python.org/pipermail/c++-sig/2004-March/007025.html http://mail.python.org/pipermail/c++-sig/2004-March/007028.html http://mail.python.org/pipermail/c++-sig/2004-March/007029.html Cheers, -- Joel de Guzman http://www.boost-consulting.com http://spirit.sf.net _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

[boost::program_options] bad_function_call and -g
by alexandre.ignjatovic＠insia.org 17 Aug '05

17 Aug '05

Hi, Does anyone know why i get a bad_function_call at the execution of this code (compiled without the '-g' option): namespace bpo = boost::program_options; _optDesc.add_options() ("version,v", "print version") ("help,h", "print this help message") ("config-file,f", bpo::value<std::string>(&_cfg)->default_value(_cfg), "specify configuration file") ("id,i", bpo::value<std::string>(&_id)->default_value(a_id), "specify server identifier"); bpo::variables_map vm; bpo::store(bpo::parse_command_line(argc, argv, _optDesc), vm); bpo::notify(vm); while it runs correctly with "-g" ? Compiler : gcc 343 (linux) or cc 57 (solaris) Boost version: 1.32.0 The exception seems to be trown by the notify function. By the way, it is quit hard to trace undebuggable compiled code (no -g, no way :)) Thanks in advance. -- Alexandre Ignjatovic

2 4