- Boost-users - lists.boost.org

Re: [Boost-users] boost vs std atomic sequential consistency semantics
by Isaac To 16 Apr '15

16 Apr '15

On Wed, Apr 15, 2015 at 4:34 PM, Olivier Tristan <o.tristan(a)uvi.net> wrote: Have you tried Boost LockFree ? > http://www.boost.org/doc/libs/1_55_0/doc/html/lockfree.html > which handle multiple producer/Multiple-consumer queue > I know about the lockfree library, I've read the book underlying it (the Art of Multiprocessor Programming), and think I've understood at least 80% of it. My problem is not multiple producer multiple consumer queue, so it doesn't work for me. In particular, although there are multiple log producers, there is no consumer. They log into a ring buffer, and normally nobody would care and the log entries get overwritten by new entries (I want to avoid having to wait when logging). When the reader wants to read, it simply dumps everything in the ring buffer, and won't change the status of the buffer (i.e., it doesn't "consume"). And I don't care if the reader found a few entries are busy being overwritten, it just has to know about that so as to avoid using the data (which leads to wrong output and could lead to crash). The problem is in fact a lot easier than any of the queues / stacks that is implemented by lockfree. I've also inspected the assembly generated by g++ over my code in x86-64, and I think it is okay (i.e., no reordering with non-atomic statement is done by the compiler -- together with the strong consistency guarantee under x86-64 the code should be okay I think). Of course, "okay for one version of a brand of compiler and library under one platform" does not mean my program is correct: a new version of the compiler / library could break it, which is my main worry. But I've got something more fundamental to ask: what the memory model really means, and what to do once we (inevitably) move away from Boost and join the C++1x/y/z family. It seems to me that the C++ standard provides something quite a bit weaker than what Boost promises, and I don't know how to get the Boost promise from the C++ standard. That's the other reason why I ask. > Le 15/04/2015 08:42, Isaac To a ?crit : > > > I'd like to write a C++ object where there are many logger threads > > logging to a large global (non-atomic) ring buffer, with an occasional > > reader thread which wants to read as much data in the buffer as > > possible. I ended up having a global atomic counter where loggers get > > locations to write to, and each logger increments the counter > > atomically before writing. The reader tries to read the buffer and > > per-logger local (atomic) variable to know whether particular buffer > > entries are busy being written by some logger, so as to avoid using them. > Regards, Isaac

1 0

Re: [Boost-users] Possible to use with JNI?
by Lorenzo Trivelli 16 Apr '15

16 Apr '15

At 06:30 13/04/2015, you wrote: >Hi, I really need to use Java with this library. Is it possible to use JNI on this library or do I need to convert this library to Java? I find the latter task is going to be really difficult so I hope that I can use JNI. > >Thank you for your help. Hi, AFAIK it is not easy to do that due to C++ name mangling and the absence of any ABI in C++ itself like C A possible but limited workaround should be to create a DLL with a C interface which uses the boost objects you need. Next, you can call it through JNI, but obviously you can pass only basic data types regardds Lorenzo

7 9

[asio][bind] Access violation with placeholders used by static objects
by Olivier Sauvageau 16 Apr '15

16 Apr '15

Hello, I had an access violation issue using boost::asio::placeholders::error as parameter to boost::bind in a class that was used as type for a static class member object. I realized that the placeholders, both those in Asio and those in Bind (e.g., _1) are themselves static objects, and thus found out that my problem was due to the famous undefined order of initialization of static objects. I was able to fix my problem simply by encapsulating my static object in a function and making sure it's called only after main() has started. However, I'm still wondering, has this issue ever been discussed and any correction or alternative proposed? I couldn't find anything on the topic. On the other hand, I noticed that Asio's placeholders are initialized from functions, e.g. from <boost/asio/placeholders.hpp>: boost::arg<1>& error = boost::asio::placeholders::detail::placeholder<1>::get(); Using that function instead could be an alternative, but since it is not in the "public" namespace and is not in the library documentation, using it would not be a good idea for maintainability. Finally, is there any documentation stating the fact that placeholders such as boost::asio::placeholders::error and _1 are static? I can't find any in the Asio or Bind documentation, and I feel this is a limitation that should at least be documented. Thank you, -- Olivier Sauvageau

1 0

Serialization of boost::shared_ptr - wrong ref. count?
by Josef Uher 16 Apr '15

16 Apr '15

Hi All, I am using serialisation of boost::shared_ptr objects. Saving/loading works fine, except that I found out that all shared_ptr objects have reference count (use_count) minimum of 2 even though they should have 1! Debugging into the boost’s code I found out that the ref. count is increased to value of 2 in “shared_ptr_helper.hpp”, line 181 where the pointer is stored inside map “m_o_sp”. This map however never gets deleted and hence the shared_ptr seems to be stored there for ever. Please, am I doing something wrong or is there a bug? How to fix this? The sample code I use is below. I would like to emphasise that I make sure the archives are deleted after being used. So it should release the resources. I am using Mac OS X 10.10.2, boost version 1.57.0 Thank you for help, Josef **************** #include <boost/iostreams/device/file_descriptor.hpp> #include <boost/iostreams/filtering_stream.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/serialization/shared_ptr.hpp> #include <boost/make_shared.hpp> class SOBJ { public: friend class boost::serialization::access; virtual void SerializationHelperFcn() {} SOBJ() { std::cout << "SOBJ Constructor" << std::endl; } virtual ~SOBJ() { std::cout << "SOBJ Destructor" << std::endl; } double value=0.; template<typename Archive> void serialize(Archive &ar, const unsigned int ver) { ar & boost::serialization::make_nvp("value", value); } }; void shared_ptr_serialization_test() { boost::shared_ptr<SOBJ> out=boost::make_shared<SOBJ>(); { boost::iostreams::filtering_ostream ofs; ofs.push(boost::iostreams::file_descriptor_sink("/Users/juhe/stest.xml")); boost::archive::xml_oarchive oa(ofs); oa << boost::serialization::make_nvp("tag", out); } boost::shared_ptr<SOBJ> in=nullptr; { boost::iostreams::filtering_istream ifs; ifs.push(boost::iostreams::file_descriptor_source("/Users/juhe/stest.xml")); boost::archive::xml_iarchive ia(ifs); ia >> boost::serialization::make_nvp("tag", in); } std::cout << "out.use_count()=" << out.use_count() << std::endl; std::cout << "in.use_count() =" << in.use_count() << std::endl; } ****************

2 3

Project linking problem with boost
by Max 16 Apr '15

16 Apr '15

Hello, The problem here relates to boost, but I'm aware it may of more general nature rather than boost specific. I had a weird problem with my project. It has a bunch of dependencies including many of the boost libraries. My platform is: Win7SP1 + VS2008SP1 + boost1.39.0. I have properly set the VC++ directories for both include and lib files pointing to the boost folder (1.39.0) and of course those for other dependencies. But I got the following message when I built the project: 1>Linking to GSL lib1>Linking to lib file: CGAL-vc90-mt.lib1>Linking to lib file: libboost_date_time-vc80-mt-1_34.lib1>Linking to lib file: libboost_regex-vc80-mt-1_34.lib I don't know why VS was attempting to link to libs of boost 1.34.0 built with VC80 (I'm using VC 90 - VS2008) Could anybody please give me some hints about this problem or where the message like "Linking to lib file: ..." came from? Thanks. Best regards, Max

3 2

Possible to use with JNI?
by Yolanda Septiana 16 Apr '15

16 Apr '15

Hi, I really need to use Java with this library. Is it possible to use JNI on this library or do I need to convert this library to Java? I find the latter task is going to be really difficult so I hope that I can use JNI. Thank you for your help.

2 2

boost vs std atomic sequential consistency semantics
by Isaac To 15 Apr '15

15 Apr '15

I've asked the question in stack overflow, but no luck there. I hope I will see some light here. I'd like to write a C++ object where there are many logger threads logging to a large global (non-atomic) ring buffer, with an occasional reader thread which wants to read as much data in the buffer as possible. I ended up having a global atomic counter where loggers get locations to write to, and each logger increments the counter atomically before writing. The reader tries to read the buffer and per-logger local (atomic) variable to know whether particular buffer entries are busy being written by some logger, so as to avoid using them. So I have to do synchronization between a pure reader thread and many writer threads. I sense that the problem can be solved without using locks, and I can rely on "happens after" relation to determine whether my program is correct. I've tried relaxed atomic operation, but it won't work: atomic variable stores are releases and loads are acquires, and the guarantee is that some acquire (and its subsequent work) always "happens after" some release (and its preceding work). That means there is no way for the reader thread (doing no store at all) to guarantee that something "happens after" the time it reads the buffer, which means I don't know whether some logger has overwritten part of the buffer when the thread is reading it. So I turned to sequential consistency. For me, "atomic" means Boost.Atomic, which notion of sequential consistency has a "pattern" documented <http://www.boost.org/doc/libs/1_57_0/doc/html/atomic/thread_coordination.ht…> : The third pattern for coordinating threads via Boost.Atomic uses seq_cst for coordination: If ... 1. thread1 performs an operation A, 2. thread1 subsequently performs any operation with seq_cst, 3. thread1 subsequently performs an operation B, 4. thread2 performs an operation C, 5. thread2 subsequently performs any operation with seq_cst, 6. thread2 subsequently performs an operation D, then either "A happens-before D" or "C happens-before B" holds. Note that the second and fifth lines say "any operation", without saying whether it modify anything, or what it operates on. This provides the guarantee that I wanted. All is happy until I watch the talk of Herb Sutter titled "atomic<> Weapons". What he implied is that seq_cst is just a acq_rel, with the additional guarantee of consistent atomic stores ordering. The suggestion that sequentially consistent atomic operations may be re-ordered with non-atomic operations only adds to my mystery. I turned to the cppreference.com <http://en.cppreference.com/w/cpp/atomic/memory_order>, which have similar descriptions. So my questions: 1. Does C++11 and Boost Atomic implement the same memory model? 2. If (1) is "yes", does it mean the "pattern" described by Boost is somehow implied by the C++11 memory model? How? Or does it mean the documentation of either Boost or C++11 in cppreference is wrong? 3. If (1) is "yes", but (2) is "Boost documentation is incorrect", is there any way to achieve the effect I want, namely to have guarantee that (the work subsequent to) some atomic store happens after (the work preceding) some atomic load? 4. If (1) is "no", is there any way to achieve the effect I want in C++11?

2 1

Re: [Boost-users] Unable to build old project using boost
by Max 14 Apr '15

14 Apr '15

----- Original Message ----- From: John Maddock <jz.maddock(a)googlemail.com> To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Unable to build old project using boost Date: 2015-04-13 00:08 On 12/04/2015 15:24, Juan Ramírez wrote: It seems that boost wasn't compiled with VC++ 9. The macro BOOST_NO_RVALUE_REFERENCES is not defined, which means boost thinks the compiler supports rvalue references. Since the code is header only, how Boost was built is irrelevant - but there must be some mismatch between what Boost.Config is doing and what type_traits expects. The only thing I can think of that would cause that are either: * A really really old version of Boost.Config with a newer version of type_traits - not sure how you would manage that though, or... * You have a custom config (in boost/config/user.hpp) which is setup for a newer MSVC version. If all else fails, try with a fresh boost source tree, HTH, John. ------------------------------------------------------- Thank you very much John. I didn't changed anything of the boost installation whatsoever, except for the build itself, before that problem.I was using exactly the same thing as I did 2 yearly ago, when it worked. Now it works, for boost 1.39.0, 1.49.0 and 1.57.0. But I cannot think of anything that I did made the difference. Best regards,Max On Sat, Apr 11, 2015 at 10:33 AM, Edward Diener <eldiener(a)tropicsoft.com> wrote: On 4/11/2015 6:33 AM, Max wrote: Hi guys, This must be a trivial question, but I just couldn't sort it out. I'm compiling a old project which used to work fine. But now I received a bunch of errors (this is not a surprise because the project have a lot of other dependencies and I'm in the way of setting them up) with the first one as: c:\lib\boost_1_49_0\boost\type_traits\add_lvalue_reference.hpp(19) : error C2143: syntax error : missing ',' before '&&' I'm using Win 7 SP1 + VS 2008 with SP + boost 1.49.0 (I'm using this because I previous used this. If this works fine, I will consider migrating to the latest version) Thanks for any hints. If you are using an older version of Boost and it was working before and now it is not working, something must have changed. Try to determine what might have changed. You also need to show source code you are compiling when the error occurs. _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users -- Juan :wq _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

Re: [Boost-users] Unable to build old project using boost
by Max 14 Apr '15

14 Apr '15

----- Original Message ----- From: Juan_Ramírez <juanantonio.ram(a)gmail.com> To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Unable to build old project using boost Date: 2015-04-12 22:24 It seems that boost wasn't compiled with VC++ 9. The macro BOOST_NO_RVALUE_REFERENCES is not defined, which means boost thinks the compiler supports rvalue references. On Sat, Apr 11, 2015 at 10:33 AM, Edward Diener <eldiener(a)tropicsoft.com> wrote: On 4/11/2015 6:33 AM, Max wrote: Hi guys, This must be a trivial question, but I just couldn't sort it out. I'm compiling a old project which used to work fine. But now I received a bunch of errors (this is not a surprise because the project have a lot of other dependencies and I'm in the way of setting them up) with the first one as: c:\lib\boost_1_49_0\boost\type_traits\add_lvalue_reference.hpp(19) : error C2143: syntax error : missing ',' before '&&' I'm using Win 7 SP1 + VS 2008 with SP + boost 1.49.0 (I'm using this because I previous used this. If this works fine, I will consider migrating to the latest version) Thanks for any hints. If you are using an older version of Boost and it was working before and now it is not working, something must have changed. Try to determine what might have changed. You also need to show source code you are compiling when the error occurs. _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users -- Juan:wq _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users Thank you. My boost was built with VC++ 9. And when add #define BOOST_NO_RVALUE_REFERENCES, other errors arose - actually, VS2008 supports rvalue references. Best regards,Max

1 0

building 64bit boost on aix 6.1
by 임영운 13 Apr '15

13 Apr '15

hi, I'm newbie. the only one thing that I wanted to ask is building boost 64bit on aix6.1. I try it a number of times. and I failed building 64bit boost objects. It built only 32bit objects. I used this option, address-model=64. but it didn't work. would you tell me how to build boost 64bit on aix6.1 thanks read this msg. 주식회사 테미 임영운(Yeong-woon Im) 305-764 대전광역시 유성구 대학로99 충남대학교 산학연교육연구관 807호 #807, TBI Center, Chungnam University 99 Daehak-ro, Yuseong-gu, Deajeon, 305-764 Korea Home URL : http://www.temee.co.kr Tel : 042-824-7014 Fax : 042-824-7015 Mobile : 010-6266-9960

1 0