August 2013 - Boost - lists.preview.boost.org

Removing old config macro and increasing compiler requirements.
by Stephen Kelly 12 Sep '13

12 Sep '13

Hi there, I've been working on the modularization effort from a CMake point of view, and from an 'actual modularization' point of view, looking at interdependencies: http://thread.gmane.org/gmane.comp.programming.tools.ryppl.devel/9/focus=26 I'd like to start on reducing interdependencies, and I'd like to start reasonably small, as no-one in the boost community knows me and see how it goes from there. I'm very active in CMake, Qt and KDE. Currently, in the modularized boost repos, boost::config and boost::core depend on each other. I listed 3 ways of fixing this on the ryppl list: http://thread.gmane.org/gmane.comp.programming.tools.ryppl.devel/201 My preference is the removal of BOOST_NO_EXPLICIT_FUNCTION_TEMPLATE_ARGUMENTS, and requiring compilers to have the feature. That way boost/config/suffix.hpp will no longer need to include boost/{non_,}type.hpp. My preference is increasing the compiler requirement, because that may open up more similar opportunities for reducing dependencies and interdependencies throughout boost. Grepping indicates that that means increasing the compiler requirements to something like __DMC__ > 0x840, GCC > 3.2, BOOST_INTEL_CXX_VERSION > 500, VC++ > 7.0. Comments? Thanks, Steve.

15 82

[utility] Proposal to extract some components from Boost.Log
by Andrey Semashev 09 Sep '13

09 Sep '13

Hi, I'd like to extract some general-purpose components from Boost.Log to Boost.Utility so that they could be used in other libraries and user's code that is not related to logging. The components are mostly standalone already and are not tied to Boost.Log specifics. 1. BOOST_EXPLICIT_OPERATOR_BOOL() macro for defining explicit operator bool() for a class. For C++03 compilers the macro implements safe bool idiom or just a regular operator bool(), if the compiler doesn't handle safe bool well. The generated operator is implemented in terms of operator!(), which should be defined for the class: template< typename T > class my_ptr { T* m_p; public: BOOST_EXPLICIT_OPERATOR_BOOL() bool operator! () const { return !m_p; } }; Various different implementations of explicit/safe bool operator are present in different libraries. It would be good to unify them. Implementation: boost/log/utility/explicit_operator_bool.hpp 2. basic_string_literal< CharT, CharTraitsT > class template. String literal is similar to string_ref, but it only allows to be constructed from a string literal. Obviously, it doesn't allow to modify the literal and it doesn't allocate memory for the literal. It supports clear() though, which makes the literal object to refer to an empty literal. Otherwise it provides interface similar to std::string. Docs: < http://www.boost.org/doc/libs/release/libs/log/doc/html/log/detailed/utilit… > Implementation: boost/log/utility/string_literal.hpp 3. type_info_wrapper class. The class wraps a reference to std::type_info and provides equality and ordering operators. The wrapper implements value semantics (i.e. it can be constructed and copied), which allows it to be stored in containers, including as keys. Docs: < http://www.boost.org/doc/libs/release/libs/log/doc/html/log/detailed/utilit… > Implementation: boost/log/utility/type_info_wrapper.hpp 4. empty_deleter class. The class can be used as a deleter (e.g. for shared_ptr) that does nothing. It can be useful in some cases, when you need a shared_ptr but you know you will delete the referred object by other means. For example, with Boost.Log this deleter can be used to create shared_ptr< std::ostream > to std::cout. Another use case is to create a weak_ptr to an object to be able to test whether the object has already been deleted. Implementation: boost/log/utility/empty_deleter.hpp 5. BOOST_LOG_UNIQUE_IDENTIFIER_NAME(prefix) macro. This macro expands in a unique (within the current TU) token that starts with prefix. This can be useful for declaring unique variable and type names within other macros. Not that this macro has much value but it could be useful for other libraries that define macros. Implementation: boost/log/utility/unique_identifier_name.hpp

4 10

Re: [boost] [Boost.Interprocess] conditions variables get 10 times faster when opening a multiprocess browser
by Marcello Pietrobon 08 Sep '13

08 Sep '13

I've noticed the same 10 times acceleration even while I make a skyPE call... How can this have anything to do with interprocess programming? -- View this message in context: http://boost.2283326.n4.nabble.com/Boost-Interprocess-conditions-variables-… Sent from the Boost - Dev mailing list archive at Nabble.com.

5 14

Problems with yield_k workaround. Still too slow.
by Marcello Pietrobon 07 Sep '13

07 Sep '13

Hello Boost team, As exposed by Gav Wood in http://boost.2283326.n4.nabble.com/Boost-Interprocess-Chronic-performance-o… Windows doesn't handle well the contest switching between threads. Basically if a thread is given a time to sleep for 1ms then the OS puts that thread on sleep for its entire timeslice (~20ms), no matter if the concurrent threads need a much shorter time to complete their job. The effect of this choice is that if you need to have two threads talking with each other through an emulated synchronization object, they take well 2 seconds to do just 100 back and forth. At least it seems to me that this problem arises only when *emulating * a synchronization object, which I understand is necessary, for example, in the interprocess library. It's not a problem of easy solution. There is an ongoing discussion in the interprocess library on how to fix this (see for example http://boost.2283326.n4.nabble.com/Boost-Interprocess-conditions-variables-… ) but clearly this is not a problem limited to it. The workaround http://www.boost.org/doc/libs/1_54_0/boost/smart_ptr/detail/yield_k.hpp improves the situation, yet only a bit. In order to isolate the problem I've run some tests using the interprocess library, with the example comp_doc_anonymous_conditionA.cpp (and B). This test is slowed down by this time of a factor ranging from 2 to 10, 100 or even 1000 or more times depending on how soon that infamous Sleep(1) is triggered. *The bottom line is that in yield_k.cpp we need to change the constant in the code * from 32 to *1000 * at least. I call this constant kMax: Here some test results, running XPsp3 on Xeon E5450 - dual socket 32bit CPU quad cores. 2GHz core speed. For test I changed comp_doc_anonymous_conditionA.cpp : that is 100000 iterations, while in comp_doc_anonymous_conditionB.cpp I've added the line just to emulate a job done by the thread while owning the mutex. That is: The results are (these numbers are representative of many tests I've run) : kMax = 32, jMax = 0 : time = 00:00:01.125000 kMax = 50, jMax = 0 : time = 00:00:00.403750 kMax = 100, jMax = 0 : time = 00:00:00.2803750 kMax = 65535, jMax = 0 : time = 00:00:00.203125 kMax ~= 64 is good kMax = 32, jMax = 10 : time = 00:00:32.843750 kMax = 100, jMax = 10 : time = 00:00:02.750000 kMax = 1000, jMax = 10 : time = 00:00:00.859375 kMax = 65535, jMax = 10 : time = 00:00:00.328125 kMax ~= 1000 is good kMax = 32, jMax = 100 : time = 00:04:01.953125 kMax = 50, jMax = 100 : time = 00:00:04.296875 kMax = 1000, jMax = 100 : time = 00:00:02.062500 kMax = 65535, jMax = 100 : time = 00:00:01.593750 kMax ~= 1000 is good If the job takes longer we need to increase kMax even more, but it seems that a value of 65535 cuts the problem and makes sure that a locked thread doesn't take most of the cpu time for more than one timeslice. -- View this message in context: http://boost.2283326.n4.nabble.com/Problems-with-yield-k-workaround-Still-t… Sent from the Boost - Dev mailing list archive at Nabble.com.

4 6

[boost.chrono] Slow down process (if specifing c++11)
by Gaetano Mendola 06 Sep '13

06 Sep '13

Hi have tried the following code compiling it with gcc 4.6.3, with four combinations: 1) -O3 2) -O3 (commeting out the chrono call) 3) -O3 -std=c++0x 4) -O3 -std=c++0x (commenting out the chrono call) and I'm obtaining the following runtime values: 1) 8921 ms 2) 8915 ms 3) 9400 ms 4) 8933 ms as you can see the combination: "c++11 and chrono call" slows down the entire process. I have tried the same with gcc 4.8.1 and all times are more or less the same. At this point I'm not sure if it's a problem of chrono with gcc 4.6.3 or simply an issue of 4.6.3 disappeared in 4.8.1 series. Regards Gaetano Mendola Here the code I have used: #include <iostream> #include <boost/chrono.hpp> #include <sys/time.h> #include <math.h> int main() { cpu_set_t myAffinityMask; CPU_ZERO( &myAffinityMask ); CPU_SET(0, &myAffinityMask ); sched_setaffinity(0, sizeof(myAffinityMask), &myAffinityMask); volatile float* myMemoryA = new float[(1<<24)]; volatile float* myMemoryB = new float[(1<<24)]; struct timeval myStart; struct timeval myStop; struct timeval myResult; boost::chrono::time_point<boost::chrono::steady_clock> t1 = boost::chrono::high_resolution_clock::now(); gettimeofday(&myStart, 0); for (size_t i = 0; i < (1<<24); ++i) { myMemoryA[i] = i; myMemoryB[i] = i+1; } delete []myMemoryA; delete []myMemoryB; for (size_t j = 0; j < 100; ++j) { volatile float* myMemoryA = new float[(1<<24)]; volatile float* myMemoryB = new float[(1<<24)]; for (size_t i = 0; i < (1<<24); ++i) { myMemoryA[i] *= sqrtf(myMemoryB[i]); } delete []myMemoryA; delete []myMemoryB; } gettimeofday(&myStop, 0); timersub(&myStop,&myStart,&myResult); std::cout << "Time: " << myResult.tv_sec*1000 + myResult.tv_usec/1000.0 << std::endl; std::cout << "t1: " << t1 << std::endl; }

3 5

[thread] Can boost::this_thread::sleep_for() totally stop the thread?
by Klaim - Joël Lamotte 03 Sep '13

03 Sep '13

I'm having, sometime, trouble with a rare heisenbug (rare but appearing several times a week). I described it there: http://stackoverflow.com/questions/17918548/can-windows-totally-stop-a-thre… There I ask if it's possible for windows to totally stop a thread that is sleeping too often. As I'm using boost::thread for sleeping this specific thread that, for unknown reason, sometime totally stop working, I'm also wondering if the boost::this_thread() implementation could not be doing something that the OS might not like? ----- I'll report the content of the page here: Can Windows totally stop a thread if it's sleeping too often? I have a rare heisenbug in a multi-threaded application where the main thread, and only this thread, will just do nothing. As it's an heisenbug it's really hard to understand why this is happening. The main thread is basically just looping. In the loop, it check several concurrent priority queues which contain tasks ordered by time to be executed. It pop a task, see if it's time to execute it. If it's time, it will just schedule it into TBB's task scheduler (using a root task which is the parent of all other tasks). If it's not time, the task is pushed again in the priority queue. That's for one cycle. At the end of the cycle, the main thread is put to sleep for a very short time that I expect will be longer in practice but it's not really a problem, I just don't want it to use too much resources when not necessary. Litterally: static const auto TIME_SCHEDULED_TASKS_SPAWN_FREQUENCY = microseconds(250); while( !m_task_scheduler.is_exiting() ) // check if the application should exit { m_clock_scheduler.spawn_realtime_tasks(); // here we spawn tasks if it's time this_thread::sleep_for( TIME_SCHEDULED_TASKS_SPAWN_FREQUENCY ); } m_clock_scheduler.clear_tasks(); m_root_task.wait_for_all(); I have a special task that just log a "TICK" message each second. It is automatically rescheduling until the end of the program. However, when the heisenbug appear, I can see the "TICK" disappearing and the application not doing anything else than the work that occurs in non-main threads. So it appear that only the main thread is touched. The problem can come from different places, maybe in the scheduling logic, but then it's also the only thread that have a sleep call. That sleep is a boost::this_thread::sleep_for(). My question is: **Is it possible that Windows (7 64bit) consider the main thread to be sleeping often and decide that it should sleep for a longer period of time than asked or be definitely ended?** I expect that it is not possible but I would like to be sure. I didn't find any precision on this in online documentation so far. ----- I really suspect that the problem can't come from the sleeping function but I need to eliminate each potential source of problem to isolate it. Thanks for your time. Joel Lamotte

2 9

[git] b2 headers not producing headers
by Ahmed Charles 02 Sep '13

02 Sep '13

I've tried following the instructions here: https://svn.boost.org/trac/boost/wiki/TryModBoost The 'b2 headers' command doesn't see to actually produce a boost directory. Is this supposed to work or is it just too soon?

3 2

[smart_ptr] Proposal to extract intrusive_ref_counter from Boost.Log
by Andrey Semashev 01 Sep '13

01 Sep '13

Hi, I'd like to move intrusive_ref_counter class from Boost.Log to Boost.SmartPtr. The class is intended to be used as a base class for a user's class that is intended to be used with intrusive_ptr. The class implements a reference counter and the related intrusive_ptr_add_ref/intrusive_ptr_release operations, so the user doesn't have to do that himself. Usage example: class my_class : public intrusive_ref_counter { }; intrusive_ptr< my_class > p = new my_class(); Besides just simplifying intrusive_ptr usage, the base class can be used as a sort of "void" as in shared_ptr< void >. I.e. intrusive_ptr< intrusive_ref_counter > can refer to object of any class that follows this protocol. Implementation: boost/log/utility/intrusive_ref_counter.hpp

5 10

[multimethods] Proposal: Open Multi-Methods
by Jean-Louis Leroy 01 Sep '13

01 Sep '13

Hi, I am working on library that pretty much implements what Pirkelbauer, Solodkyy and Stroustrup describe in their paper "Open Multi-Methods for C++" - although there are some divergences. I have posted it on GitHub: https://github.com/jll63/multimethods. Please examine the files in the "examples" directory to get an idea of how to use it. Salient features are: * syntax: is relatively uncluttered. There are no limitations on the number of virtual arguments. Virtual and non-virtual arguments can be arbitrarily mixed. Multiple inheritance is supported, with some limitations - see below. * speed: close to a virtual function call when the hierarchies involved in the virtual arguments collaborate with the library. Calling a method that does nothing, with a single virtual argument in a single inheritance hierarchy is 33% slower than the equivalent virtual function call. The difference becomes unnoticeable if the functions perform a few simple maths operations. See tests/benchmarks.cpp. * size: the dispatch table is constructed in terms of class groups. This results, most of the time, in a table devoid of redundancies. The size is bounded by the product of the number of specializers - not classes - in each dimension. * support for "foreign" class hierarchies: the library can be used without modifications to existing classes, at the cost of lower performance. Collaborating and foreign arguments can be freely mixed. Performance is still quite good, see the benchmarks. * next: a pointer to the next most specialized method is available inside method specializations - see examples/next.cpp. Alternatively, it is possible to call a specialization directly. Current limitations are: * restrictions on hierarchy: the classes involved in a method call must have a single root. Repeated inheritance is not supported. Multiple inheritance is supported, provided that the two conditions above are respected. * no overloading: multi-methods are implemented as constexpr function objects. This does not allow for overloading. This restriction can be easily side-stepped by wrapping multi-methods with different names in a set of overloaded functions. I know how to make multi-methods functions instead of objects, at the cost of a slightly less nice syntax for specializers. This is open to discussion. Besides, the library differs from Stroustrup et al's paper on the following points: * a specializer specializes only one multi-method: what would "next" mean otherwise? What would its static type be? * pure multi-methods are allowed: if called, an exception is thrown. This seems a better choice given that most of the time the declared virtual arguments will be abstract classes anyway. The library is essentially feature complete (if anything ever is). I am now going to work on the documentation, a build system, some clean up and compatibility (I tested it only with g++ 4.7.2 only so far). Is there interest in the Boost community for this stuff? If yes, I will adapt it to the guidelines and submit it for formal review. Of course, suggestions are welcome. Thank you for your time, Jean-Louis Leroy

5 10

[thread] MinGW/gcc compile error in thread.cpp
by Emil Dotchevski 01 Sep '13

01 Sep '13

Building the trunk version, I get this on MinGW: gcc.compile.c++ ..\..\..\bin.v2\libs\thread\build\gcc-mingw-4.7.2\debug\link-sta tic\threading-multi\win32\thread.o ..\..\..\libs\thread\src\win32\thread.cpp: In member function 'bool boost::threa d::start_thread_noexcept()': ..\..\..\libs\thread\src\win32\thread.cpp:238:9: error: reference to 'uintptr_t' is ambiguous In file included from c:\mingw\bin\../lib/gcc/mingw32/4.7.2/include/stdint.h:3:0 , from c:\mingw\bin\../lib/gcc/mingw32/4.7.2/../../../../include/ wchar.h:479, from c:\mingw\bin\../lib/gcc/mingw32/4.7.2/include/c++/cwchar:4 6, from c:\mingw\bin\../lib/gcc/mingw32/4.7.2/include/c++/bits/pos types.h:42, from c:\mingw\bin\../lib/gcc/mingw32/4.7.2/include/c++/bits/cha r_traits.h:42, from c:\mingw\bin\../lib/gcc/mingw32/4.7.2/include/c++/string:4 2, from ..\..\../boost/date_time/c_time.hpp:18, from ..\..\../boost/date_time/time_clock.hpp:16, from ..\..\../boost/thread/thread_time.hpp:9, from ..\..\../boost/thread/win32/thread_data.hpp:10, from ..\..\../boost/thread/thread_only.hpp:15, from ..\..\..\libs\thread\src\win32\thread.cpp:17: Emil

3 2